1
00:00:00,810 --> 00:00:06,770
DR. NILESH MEHTA: Hello. Welcome to day five of
the NIH Malnutrition in Clinical Settings,

2
00:00:06,770 --> 00:00:12,210
Research Gaps and Opportunities Workshop.
My name is Dr. Nilesh Mehta, I'm a professor

3
00:00:12,210 --> 00:00:19,920
of Anesthesia Critical Care at Harvard Medical
School and Faculty in the Critical Care Division

4
00:00:19,920 --> 00:00:25,529
in the Department of Anesthesia at Boston
Children's Hospital. On behalf of my co-chairs,

5
00:00:25,529 --> 00:00:32,630
Dr. Gail Cresci, Dr. Todd Rice, Dr. Alison
Steiber, and Dr. David Seres, and on behalf

6
00:00:32,630 --> 00:00:38,170
of the NIH scientific leadership, Dr. Christopher
Lynch, Dr. Charlotte Pratt, and Dr. Ashley

7
00:00:38,170 --> 00:00:44,129
Vargas, It is my pleasure to welcome you to
the second day of the early life malnutrition

8
00:00:44,129 --> 00:00:49,750
sessions. We are grateful for this opportunity
to contribute to this unique workshop, and

9
00:00:49,750 --> 00:00:54,719
we are truly impressed by the foresight, time,
investment, and commitment to malnutrition

10
00:00:54,719 --> 00:00:59,949
research by the NIH scientific leadership.
We are also grateful to Kimberly Barch for

11
00:00:59,949 --> 00:01:04,400
keeping us organized over the past month,
for Mark Dennis, for all the logistics

12
00:01:04,400 --> 00:01:09,720
support, and special thanks to Brett Long,
along with his colleagues, my colleague Christi

13
00:01:09,720 --> 00:01:15,689
Jules, and Israel Gonzalez for their expert
help with this digital platform Labroots during

14
00:01:15,689 --> 00:01:19,660
all our sessions.
Thank you to our expert speakers who have

15
00:01:19,660 --> 00:01:25,370
generously given their time to present on
a variety of topics over these two days. We

16
00:01:25,370 --> 00:01:31,090
started day four yesterday with an introduction
to early-life malnutrition. We discussed its

17
00:01:31,090 --> 00:01:36,570
burden, the epidemiology, we talked about
tools of screening, and concepts of diagnosing

18
00:01:36,570 --> 00:01:43,510
malnutrition. We alluded to research considerations
along each of these topics. We reviewed malnutrition

19
00:01:43,510 --> 00:01:49,860
in preterm infants, neonates, infants undergoing
cardiac surgery, children dependent on food

20
00:01:49,860 --> 00:01:55,719
systems. We did some common themes that were
alluded to in the first three days of this

21
00:01:55,719 --> 00:02:01,960
workshop in the adult malnutrition topics.
We summarize the evidence, current guidelines

22
00:02:01,960 --> 00:02:07,710
pertaining to malnutrition. And in our final
session, we highlighted disease-specific malnutrition

23
00:02:07,710 --> 00:02:12,810
in select conditions. Not exhaustive, but
very comprehensive. And in these conditions,

24
00:02:12,810 --> 00:02:20,840
we identified gaps in research and look for
potential advances for the future. After a

25
00:02:20,840 --> 00:02:26,110
very full and exciting day one, we are once
again fortunate to be joined by a group of

26
00:02:26,110 --> 00:02:31,100
clinical experts, thought leaders, and
research leaders in our field.

27
00:02:31,100 --> 00:02:38,260
Once again, thank you to all of you who have
joined us over the past five days of this

28
00:02:38,260 --> 00:02:43,750
workshop and the last two days of the early
Life malnutrition workshop sessions. We look

29
00:02:43,750 --> 00:02:50,580
forward to our discussions during the day.
I hope that you will consider during this

30
00:02:50,580 --> 00:02:57,409
session to share your questions for the presenters.
You can use the chat box on your screen and

31
00:02:57,409 --> 00:03:04,620
if you send these in, I will collate and try
to get to as many as we can during the Q&A

32
00:03:04,620 --> 00:03:12,920
session. We have an exciting day ahead. So,
let's get started once again with a warm welcome

33
00:03:12,920 --> 00:03:21,350
to you all with Session nine. Session nine
today is Early Life Malnutrition Research

34
00:03:21,350 --> 00:03:28,100
Considerations Part two. In this session,
we will first describe some updates in malnutrition-related

35
00:03:28,100 --> 00:03:35,720
measures in pediatrics. To that end, Dr. Leanne
Redman from Pennington Biomedical Research

36
00:03:35,720 --> 00:03:42,730
Center in Baton Rouge, Louisiana, will speak
on the assessment of energy needs, advances,

37
00:03:42,730 --> 00:03:48,280
and future directions. This will be followed
by an update on body composition assessment

38
00:03:48,280 --> 00:03:55,060
in pediatrics by Ms. Stephanie Merlino Barr,
who's in the neonatal intensive care unit

39
00:03:55,060 --> 00:03:58,650
at MetroHealth Medical Center, Cleveland,
Ohio.

40
00:03:58,650 --> 00:04:05,980
This will be followed by a discussion, a presentation
on pediatric malnutrition, specifically addressing

41
00:04:05,980 --> 00:04:13,210
outcomes that we would want to measure, that
we measure currently. This presentation will

42
00:04:13,210 --> 00:04:18,079
be by Dr. Vijay Srinivasan, who's a faculty
in critical care medicine at the Children's

43
00:04:18,079 --> 00:04:24,410
Hospital of Philadelphia and University of
Pennsylvania. Following this group of presentations,

44
00:04:24,410 --> 00:04:30,440
we will then move to discussing disparities,
family and system level, determinants of early

45
00:04:30,440 --> 00:04:36,940
life, malnutrition, risk access, and treatment.
In that session, in that group, we will first

46
00:04:36,940 --> 00:04:43,220
be started on a discussion and presentation
on impact of social determinants of health

47
00:04:43,220 --> 00:04:47,690
other than food insecurity on low birth weight,
growth faltering, and anemia among children.

48
00:04:47,690 --> 00:04:52,560
This presentation will be given by Dr. Deborah
Frank from Boston University. This will be

49
00:04:52,560 --> 00:04:58,100
followed by Dr. Laurie Nommsen-Rivers from
the University of Cincinnati, who will discuss

50
00:04:58,100 --> 00:05:04,979
excess weight loss and failure to thrive in
the context of exclusively breastfed newborns.

51
00:05:04,979 --> 00:05:08,919
Once again, highlighting research gaps in
prevention, identification and treatment,

52
00:05:08,919 --> 00:05:16,480
and inequities. After all these talks, which
will follow in tandem right after this introduction,

53
00:05:16,480 --> 00:05:23,919
we will come back at 1pm for the Q&A session.
I will be the moderator for the Q&A session

54
00:05:23,919 --> 00:05:30,590
and I will be assisted by my colleague, Dr.
Todd Rice, and our NIH program lead, Dr. Ashley

55
00:05:30,590 --> 00:05:37,720
Vargas. After the Q&A session, we will take
a short break and then move on to session

56
00:05:37,720 --> 00:05:44,539
ten. So, once again, I hope that you will
enjoy the presentations. I thank once again

57
00:05:44,539 --> 00:05:49,830
all the expert speakers and I look forward
to receiving your questions during the session

58
00:05:49,830 --> 00:05:55,080
and see you on the other side for the Q&A
session. So, let's kick this off now with

59
00:05:55,080 --> 00:06:00,759
the first presentation, which will be on assessment
of energy needs by Dr. Leanne Redman. Thank

60
00:06:00,759 --> 00:06:01,759
you.

61
00:06:01,759 --> 00:06:12,010
MS. STEPHANIE MERLINO BARR: Good afternoon. My
name is Stephanie Merlino Barr. I am a registered

62
00:06:12,010 --> 00:06:15,940
dietitian in the neonatal intensive care unit
at MetroHealth Medical Center in Cleveland,

63
00:06:15,940 --> 00:06:20,580
Ohio, and a PhD candidate in the Department
of Population and Quantitative Health Sciences

64
00:06:20,580 --> 00:06:25,220
at Case Western Reserve University's School
of Medicine. Today, I will be speaking about

65
00:06:25,220 --> 00:06:31,319
body composition assessment in pediatrics.
My objectives today will be to identify methods

66
00:06:31,319 --> 00:06:35,700
of body composition assessment and explore
their validation in the literature focusing

67
00:06:35,700 --> 00:06:40,099
on the infant population. I will describe
the challenges of these measurements and their

68
00:06:40,099 --> 00:06:47,360
implementation into the clinical environment.
Body composition assessment in infants and

69
00:06:47,360 --> 00:06:51,560
young children is challenging to perform due
to rapid growth and substantial changes of

70
00:06:51,560 --> 00:06:56,509
body composition that occur. Preterm infants
have unique body composition development compared

71
00:06:56,509 --> 00:07:01,319
to their term infant counterparts with lower
lean body mass secretion, resulting in relatively

72
00:07:01,319 --> 00:07:04,379
higher percent body fat at term corrected
age.

73
00:07:04,379 --> 00:07:09,190
Peak percent fat mass occurs at approximately
six months of age before independent locomotion

74
00:07:09,190 --> 00:07:15,330
begins and then slowly declines throughout
early childhood. These distinct periods of

75
00:07:15,330 --> 00:07:20,819
rapid growth are influenced by important variables,
including sex, ethnicity and race, and disease

76
00:07:20,819 --> 00:07:26,419
states. Understanding, body composition, development
and utilizing its assessment as a part of

77
00:07:26,419 --> 00:07:30,330
standard clinical care may allow for a better
understanding of the relationship between

78
00:07:30,330 --> 00:07:35,009
disease growth and nutrition, allowing for
the promotion of optimal growth and health

79
00:07:35,009 --> 00:07:41,319
outcomes through nutrition and non-nutritional
strategies. The most clinically feasible tools

80
00:07:41,319 --> 00:07:45,530
for body composition assessment are anthropometric
measurements. There is not one measurement

81
00:07:45,530 --> 00:07:50,980
that can fully divulge all there is to know
about an individual's growth and body composition.

82
00:07:50,980 --> 00:07:54,909
Accurate measurements obtained using the appropriate
tools and assessed on standardized growth

83
00:07:54,909 --> 00:07:59,940
charts are necessary for these data to be
useful in a clinical setting. Weight, length

84
00:07:59,940 --> 00:08:05,449
and head circumference are already used clinically.
In premature infants, birth weight is predictive

85
00:08:05,449 --> 00:08:11,139
of lean body mass at birth. However, weight
on its own does not allow clinicians to assess

86
00:08:11,139 --> 00:08:16,789
the quality of growth over time. Head circumference
for age Z scores can help identify infants

87
00:08:16,789 --> 00:08:21,700
at risk for long-term neurodevelopmental impairment
as head circumference is indicative of brain

88
00:08:21,700 --> 00:08:26,949
size. Smaller head circumferences have been
observed in pediatric patients classified

89
00:08:26,949 --> 00:08:31,960
as malnourished using this subjective global
nutritional assessment and deviation from

90
00:08:31,960 --> 00:08:37,039
the WHO growth standards. Head circumference
is not as sensitive an indicator of short-term

91
00:08:37,039 --> 00:08:43,880
nutritional status, as brain growth is protected
in acute nutritional crises. The most important,

92
00:08:43,880 --> 00:08:48,130
cost-effective, and clinically feasible measurement
that I will discuss today is the crown-heel

93
00:08:48,130 --> 00:08:54,029
length measurement on a recumbent length board.
Unlike weight, linear growth is not influenced

94
00:08:54,029 --> 00:08:58,750
by fluid status. In the infant population,
linear growth has been associated with lean

95
00:08:58,750 --> 00:09:04,750
body mass, protein accretion, and organ growth.
Linear growth has proven to be a useful and

96
00:09:04,750 --> 00:09:10,329
important marker of nutritional status in
the management of infants with bronchopulmonary

97
00:09:10,329 --> 00:09:14,860
dysplasia. For all my pediatric clinicians
looking to improve your overall approach to

98
00:09:14,860 --> 00:09:21,080
growth assessment, ensuring appropriate length
using a length board is the place to start.

99
00:09:21,080 --> 00:09:24,980
Other anthropometric measurements like mid-upper
arm circumferences, and flank skin folds

100
00:09:24,980 --> 00:09:29,210
are more frequently seen in research settings
but have begun to be utilized in clinical

101
00:09:29,210 --> 00:09:35,230
care. Mid-upper arm circumferences have been
reliable predictors of fat mass in preterm

102
00:09:35,230 --> 00:09:39,760
infants in pediatric populations. MUAC is
not as influenced by fluid status as other

103
00:09:39,760 --> 00:09:44,741
areas of the body and therefore can be well
implemented in an ICU environment. Mid-upper

104
00:09:44,741 --> 00:09:48,760
arm circumference growth trends for preterm
infants in other pediatric populations have

105
00:09:48,760 --> 00:09:55,760
been described over the past several decades
by researchers, the CDC, and the WHO. However,

106
00:09:55,760 --> 00:10:01,460
consensus on ideal mid-upper arm circumference
growth in modern pediatric populations, standardization

107
00:10:01,460 --> 00:10:06,140
of measurement tools, and the incorporation
of MUAC growth charts into electronic medical

108
00:10:06,140 --> 00:10:11,720
records are needed for clinical implementation.
Skinfold measurements like (INAUDIBLE) skinfold

109
00:10:11,720 --> 00:10:15,839
are performed with calipers and are most often
incorporated into predictive equations of

110
00:10:15,839 --> 00:10:21,010
infant body composition, which will be discussed
shortly. Skinfold measurements are correlated

111
00:10:21,010 --> 00:10:27,850
with fat and fat-free mass but have relatively
low explanatory value compared to weight and

112
00:10:27,850 --> 00:10:32,510
length. Proportionality indices like weight
for length can be useful tools in a clinical

113
00:10:32,510 --> 00:10:37,400
setting to illustrate and discuss proportional
growth. However, in the pre-term infant and

114
00:10:37,400 --> 00:10:42,519
pediatric populations, the utility of weight
and length indices is inconsistent in predicting

115
00:10:42,519 --> 00:10:48,889
fat and fat-free mass. Other indices not commonly
used in a clinical setting include our muscle

116
00:10:48,889 --> 00:10:53,709
area, which uses mid-upper arm circumference and
triceps skinfold has been studied in older

117
00:10:53,709 --> 00:10:58,420
children and has a fair correlation with fat
mass. However, this calculation has not been

118
00:10:58,420 --> 00:11:04,130
found to be a reliable predictor of fat or
fat-free mass in neonates. The mid-upper arm

119
00:11:04,130 --> 00:11:07,810
circumference to head circumference ratio
has been suggested to be a useful tool for

120
00:11:07,810 --> 00:11:11,980
body composition prediction but has not been
validated.

121
00:11:11,980 --> 00:11:16,180
Some of skinfolds was not shown to be a good
estimate of fat mass in preterm infants as

122
00:11:16,180 --> 00:11:22,029
compared to DEXA and ADP measurements. Finally,
predictive equations utilizing anthropometric

123
00:11:22,029 --> 00:11:27,480
measurements have been validated against body
composition analyses. While utilization of

124
00:11:27,480 --> 00:11:31,640
predictive equations may be a more feasible
method of implementing measures of fat in

125
00:11:31,640 --> 00:11:38,950
fat-free mass into clinical practice, these
equations may have limited external validity.

126
00:11:38,950 --> 00:11:43,269
Secondary body composition assessment using
techniques that estimate four, three and two-compartment

127
00:11:43,269 --> 00:11:48,520
models have primarily existed in the research
realm. For techniques like MRI and isotope

128
00:11:48,520 --> 00:11:53,520
dilution, this may be where they remain due
to cost, scale, and general feasibility that

129
00:11:53,520 --> 00:11:58,770
make them likely incompatible with the clinical
setting, particularly for infants. Dual x-ray

130
00:11:58,770 --> 00:12:03,220
absorptiometry or DEXA is the only assessment
method that measures bone mineral content

131
00:12:03,220 --> 00:12:08,079
and regional body composition estimates. However,
a lack of standardization of this technique

132
00:12:08,079 --> 00:12:14,220
and a functional limit of serial scanning
are important shortcomings to recognize.

133
00:12:14,220 --> 00:12:17,760
Quantitative nuclear magnetic resonance is
a relatively new technology and human body

134
00:12:17,760 --> 00:12:23,070
composition analysis that can be used from
infancy to adulthood. qNMR has shown to give

135
00:12:23,070 --> 00:12:28,260
a precise measurement of fat and fat-free
mass, but it is largely cost-prohibitive and

136
00:12:28,260 --> 00:12:33,880
requires further validation. Methods that
seem most feasible to translate to clinical

137
00:12:33,880 --> 00:12:39,790
practice include bioelectric impedance analysis,
which can be used in predictive equations

138
00:12:39,790 --> 00:12:44,880
and typically performed best in the populations
they were validated in. BIA is a useful tool

139
00:12:44,880 --> 00:12:49,910
for epidemiological settings but may have
limited usefulness in clinical practice. Ultrasound

140
00:12:49,910 --> 00:12:56,290
is a non-invasive method of body composition
analysis that utilizes technology that hospitals

141
00:12:56,290 --> 00:13:01,921
typically have, making it a strong candidate
for use in clinical assessment. This technique

142
00:13:01,921 --> 00:13:06,310
has been used in pre-term infant populations
and was able to observe accretion rates in

143
00:13:06,310 --> 00:13:12,310
adipose and muscle tissues and correlate accretion
rates with nutritional interventions. However,

144
00:13:12,310 --> 00:13:17,230
this method is yet to be fully validated in
the pre-term infant population.

145
00:13:17,230 --> 00:13:22,000
Air displacement plethysmography is a commonly
used tool in infant body composition analysis.

146
00:13:22,000 --> 00:13:27,910
ADP has the benefit of being noninvasive fast
and does not require subjects to be sedated

147
00:13:27,910 --> 00:13:33,370
or expose them to radiation. Body composition
growth charts based on ADP measurements allow

148
00:13:33,370 --> 00:13:38,050
for improved understanding and assessment
of changes in body composition in infants.

149
00:13:38,050 --> 00:13:42,510
While two machines exist so that this assessment
is feasible in infants, children, and adults,

150
00:13:42,510 --> 00:13:49,690
it is not been validated for children between
six and 24 months. Body composition changes

151
00:13:49,690 --> 00:13:54,769
are used in malnutrition, assessment, and
diagnosis. Currently, this objective global

152
00:13:54,769 --> 00:13:59,440
nutrition assessment utilizes a nutrition-focused
physical examination to determine subcutaneous

153
00:13:59,440 --> 00:14:05,170
fat and muscle loss to help identify malnutrition
in classified level of severity. An excerpt

154
00:14:05,170 --> 00:14:10,660
of the assessment is shown here. In a user
reliability of the SGNA has been fair

155
00:14:10,660 --> 00:14:14,839
in published research, highlighting both the
need for trained professionals to perform

156
00:14:14,839 --> 00:14:19,560
this assessment, as well as the opportunity
for utilization of subjective body composition

157
00:14:19,560 --> 00:14:24,630
markers and malnutrition assessment.
Registered dietitians are uniquely qualified

158
00:14:24,630 --> 00:14:29,160
clinical professionals to both perform this
assessment and create a clinical nutrition

159
00:14:29,160 --> 00:14:36,080
intervention based on its findings. To summarize
the challenges of implementing body composition

160
00:14:36,080 --> 00:14:41,130
assessment in the clinical setting, first,
the cost of initial materials ranges between

161
00:14:41,130 --> 00:14:48,320
approaches. Second, feasibility of implementation
is driven by the setting, for example, inpatient

162
00:14:48,320 --> 00:14:52,540
or outpatient, ICU versus stable patients.
Next, performing these measurements requires

163
00:14:52,540 --> 00:14:58,480
adequate staffing and training to ensure consistent
quality in the measurements performed. Finally,

164
00:14:58,480 --> 00:15:04,500
answering, so what? What can we do with this
information when we obtain it? Body composition

165
00:15:04,500 --> 00:15:08,410
assessment in infants and children is uniquely
challenging due to the dynamic growth that

166
00:15:08,410 --> 00:15:13,329
occurs during these phases. While many methods
of body composition assessment exist, there

167
00:15:13,329 --> 00:15:19,160
is not one perfect measurement that can currently
be used for all patients in all settings.

168
00:15:19,160 --> 00:15:24,899
Research priorities in this area are as follows.
First, to validate body composition, methods,

169
00:15:24,899 --> 00:15:29,149
and growth standards as there is a need for
global cooperation and standardized data collection

170
00:15:29,149 --> 00:15:34,730
processes and validation techniques. Second,
to connect body composition changes with clinical

171
00:15:34,730 --> 00:15:39,279
outcomes and interventions and premature infants.
Research has focused on the relationship of

172
00:15:39,279 --> 00:15:44,380
fat-free mass accretion with neurodevelopmental
outcomes, the relationship of long-term metabolic

173
00:15:44,380 --> 00:15:49,449
outcomes with early life body composition
changes and growth has not been elucidated.

174
00:15:49,449 --> 00:15:55,160
Determining clinical interventions to impact
these long-term outcomes is needed. And finally,

175
00:15:55,160 --> 00:16:00,019
to connect body composition changes throughout
life phases as to better understand the complexity

176
00:16:00,019 --> 00:16:05,540
of body composition development throughout
the human lifecycle. Thank you so much for

177
00:16:05,540 --> 00:16:06,540
your time.

178
00:16:06,540 --> 00:16:14,410
DR. LEANNE REDMAN: Hi, everybody. I'm Dr. Leanne
Redman, coming to you from Baton Rouge, Louisiana,

179
00:16:14,410 --> 00:16:21,010
at the Pennington Biomedical Research Center.
I would like to thank the planning committee

180
00:16:21,010 --> 00:16:25,670
for the invitation to speak today. So, over
the next few minutes, I'm gonna be talking

181
00:16:25,670 --> 00:16:29,959
to you about the assessment of energy needs
in pediatric populations, as well as some

182
00:16:29,959 --> 00:16:36,721
advances and future directions that can apply
to this field. So, as we know, malnutrition

183
00:16:36,721 --> 00:16:43,410
or over and undernutrition reflects the perturbation
to energy balance. And energy balance is simply

184
00:16:43,410 --> 00:16:50,690
the relationship between energy intake and
energy expenditure. There are three unique

185
00:16:50,690 --> 00:16:59,090
states of energy imbalance. The first one
being weight stability, or when the energy

186
00:16:59,090 --> 00:17:05,770
intake or caloric intake is equivalent to
the caloric or energy expenditure and the

187
00:17:05,770 --> 00:17:16,169
body energy stores are stable. The second
state of energy imbalance is a state of weight

188
00:17:16,169 --> 00:17:22,640
gain or body energy store gain, and this occurs
when the daily energy intake is sustained

189
00:17:22,640 --> 00:17:31,799
at a much higher level than daily energy expenditure,
and body energy stores are increased and weight

190
00:17:31,799 --> 00:17:33,690
is gained.
And the final state of energy imbalance is

191
00:17:33,690 --> 00:17:40,020
a state of weight loss. And this occurs over
time when the daily energy intake is sustained

192
00:17:40,020 --> 00:17:48,390
at a level which is less than the daily energy
expenditure. Body energy stores, as a result,

193
00:17:48,390 --> 00:17:58,290
decrease to mobilize energy, and weight is
lost. Of course, behind that very simplistic

194
00:17:58,290 --> 00:18:06,860
view of energy balance is the far more complex
physiology of metabolism and dietary intake.

195
00:18:06,860 --> 00:18:12,159
So, on the energy intake side of this equation,
of course, we need to consider the quality

196
00:18:12,159 --> 00:18:18,630
of the diet, because not only does each macronutrient
have different impact on overall total daily

197
00:18:18,630 --> 00:18:25,280
energy intake. Macronutrients also have independent
effects on body energy stores, so whether

198
00:18:25,280 --> 00:18:32,460
or not calories would be preferentially restored
as fat or as lean tissues. And we also know

199
00:18:32,460 --> 00:18:38,039
that macronutrients play a role in total body
water and they impact the energy expenditures

200
00:18:38,039 --> 00:18:45,909
as well. So, similarly, on the energy expenditure
side of the equation, there are a number of

201
00:18:45,909 --> 00:18:53,120
different ways that people partition energy.
So, the first is and the largest component

202
00:18:53,120 --> 00:19:01,340
of someone's daily energy need is their basal
metabolic rate or their resting metabolic

203
00:19:01,340 --> 00:19:07,110
rate is also termed that this accounts for
about 60% of daily energy needs. But in addition

204
00:19:07,110 --> 00:19:13,070
to these, people are also expending calories
through the digestion, processing, and absorption

205
00:19:13,070 --> 00:19:18,010
of nutrients. Of course, physical activity,
although this is the most variable component

206
00:19:18,010 --> 00:19:23,460
of daily energy expenditures and it's divided
into non-exercise expenditures or activities

207
00:19:23,460 --> 00:19:30,530
of daily living as well as structured physical
activity. And of course, we have energy expenditures

208
00:19:30,530 --> 00:19:38,940
related to thermogenesis, which can be adaptations
to perturbations in the diet, changes in the

209
00:19:38,940 --> 00:19:46,159
thermo environment. And in addition to this,
in children, we also have an energy cost associated

210
00:19:46,159 --> 00:19:55,910
with growth. There are a number of different
tools that we can use clinically to evaluate

211
00:19:55,910 --> 00:20:05,210
energy needs in people and in pediatric populations.
So I've summarized a suite of measures that

212
00:20:05,210 --> 00:20:12,070
are available to us in terms of their level
of accuracy, their burden to patients, as

213
00:20:12,070 --> 00:20:21,880
well as to the clinics, the relative cost
of conducting these types of procedures, and

214
00:20:21,880 --> 00:20:27,760
therefore their current utility or dissemination,
particularly for malnutrition and studies

215
00:20:27,760 --> 00:20:33,240
in pediatric populations.
I'm not gonna go through all of these, but

216
00:20:33,240 --> 00:20:40,380
just to let you know that there are sort of
two gold standards here, two gold standard

217
00:20:40,380 --> 00:20:44,840
methods for evaluating energy needs and changes
in energy balance in people. The first would

218
00:20:44,840 --> 00:20:51,070
be doubly labeled water. It can be a tool
that's used in both inpatient as well as outpatient

219
00:20:51,070 --> 00:20:57,280
settings. It's considered the gold standard
because the accuracy is between three and

220
00:20:57,280 --> 00:21:06,000
5% for these tests. It's burdensome because
these measures occur over a period of about

221
00:21:06,000 --> 00:21:12,289
ten days. It measures isotope enrichment during
that time period. So, therefore, that requires

222
00:21:12,289 --> 00:21:17,799
specialized equipment that's not available
in all places. The average cost of one of

223
00:21:17,799 --> 00:21:23,110
these tests ranges depending on someone's
body weight, but it could be in a pediatric

224
00:21:23,110 --> 00:21:33,380
population as low as, say, $300 for an infant
all the way up to $700 or more in an older-aged

225
00:21:33,380 --> 00:21:40,940
adolescent. Of course, we can also directly
assess energy needs clinically in a different

226
00:21:40,940 --> 00:21:47,100
way, and that would be to measure basal metabolic
rate. Already told you that was the largest

227
00:21:47,100 --> 00:21:51,900
component of someone's energy expenditure.
The second largest component is physical activity.

228
00:21:51,900 --> 00:21:55,860
So, by doing a 30-minute assessment of someone's
basal metabolic rate in the clinic, we can

229
00:21:55,860 --> 00:22:01,940
multiply that by an activity factor to derive
a direct measurement of energy needs. This

230
00:22:01,940 --> 00:22:07,090
is not as good as using, say, the doubly labeled
water, but it comes with less burden, it's

231
00:22:07,090 --> 00:22:13,000
a shorter test. The accuracy is still pretty
good and cost is not too extreme for clinics.

232
00:22:13,000 --> 00:22:18,799
So, you'll find that this type of assessment
is routinely used in a larger number of studies,

233
00:22:18,799 --> 00:22:25,950
not only for children but also for adults.
We can, of course, ask people to help us understand

234
00:22:25,950 --> 00:22:32,200
their dietary energy needs and their current
dietary intake using different types of recall

235
00:22:32,200 --> 00:22:39,040
tools. This is obviously prone to quite a
lot of bias. It's been studied a lot. It's

236
00:22:39,040 --> 00:22:42,620
generally not recommended, but these tools
still bring a lot of value for understanding

237
00:22:42,620 --> 00:22:44,809
where those calories are coming from.

238
00:22:44,809 --> 00:22:51,320
But finally, and thank goodness on
the back of a large number of doubly labeled

239
00:22:51,320 --> 00:22:57,070
water studies, as well as studies using resting
metabolic rate to evaluate energy needs, we

240
00:22:57,070 --> 00:23:03,960
now have a large number of prediction equations
that have a fairly moderate and in some cases

241
00:23:03,960 --> 00:23:10,360
high level of precision for estimating energy
needs. These tools are low burden, low cost

242
00:23:10,360 --> 00:23:16,630
and have been used quite widely, especially
in clinical practice, but also in research.

243
00:23:16,630 --> 00:23:26,250
So just to give you an example, here are some
of the prediction equations that are published

244
00:23:26,250 --> 00:23:31,643
in the dietary reference intakes from 2005.
That was the last time that the DRI for Energy

245
00:23:31,643 --> 00:23:41,150
were evaluated and published. But the National
Academy of Medicine has convened a new panel,

246
00:23:41,150 --> 00:23:46,340
and so these equations are currently being
re-evaluated and we would expect a new publication

247
00:23:46,340 --> 00:23:50,820
coming next year. But nevertheless, this is
just to give you an example, this is an example

248
00:23:50,820 --> 00:23:58,550
of one doubly labeled water study that was
done at Baylor College of Medicine. This was

249
00:23:58,550 --> 00:24:01,760
in infants from birth through to two years
of life.

250
00:24:01,760 --> 00:24:05,950
And it's these types of data that we use to
inform these equations to estimate energy

251
00:24:05,950 --> 00:24:13,460
requirements in children back in the 2005
DRI report, just as shown in this example,

252
00:24:13,460 --> 00:24:18,330
these equations are specific to different
age groups of children, which is very appropriate

253
00:24:18,330 --> 00:24:22,700
because we have different rates of growth
and different rates of weight gain that are

254
00:24:22,700 --> 00:24:28,720
needed in different pediatric cohorts. So
you can see that the babies are divided here

255
00:24:28,720 --> 00:24:35,020
into a couple of different smaller cohorts
by age. Once you get into older children,

256
00:24:35,020 --> 00:24:39,860
that's just two cohorts. There's these two
equations here for boys and girls now between

257
00:24:39,860 --> 00:24:45,730
three and eight years and here on the next
slide is the next set of equations, which

258
00:24:45,730 --> 00:24:55,200
would be for children age from nine to 18
years and again sex-specific for boys and

259
00:24:55,200 --> 00:24:58,500
girls. Just as I mentioned before, we have
to account for physical activity here in these

260
00:24:58,500 --> 00:25:02,970
estimates, because these estimates are estimating
basal metabolic rate multiplied by the physical

261
00:25:02,970 --> 00:25:08,400
activity factor, but widespread use because
in clinical practice all we need is the individual's

262
00:25:08,400 --> 00:25:15,610
age, the individual's weight and height, in
order to derive these estimates.

263
00:25:15,610 --> 00:25:23,799
But the field has advanced now. So now in
the context of weight management, and I'd

264
00:25:23,799 --> 00:25:31,260
say more specifically in adults, various groups,
led by those such as Dr. Diana Thomas from

265
00:25:31,260 --> 00:25:37,870
West Point or Dr. Kevin Hall from NIH,as shown
here, have developed now what they call dynamic

266
00:25:37,870 --> 00:25:45,670
energy balance models or models that can not
only predict someone's energy needs or energy

267
00:25:45,670 --> 00:25:52,370
requirements, but also how a change to energy
balance can impact the rate of energy that's

268
00:25:52,370 --> 00:26:01,059
being petitioned into fat-free mass and fat
mass that is shown here on this slide. This

269
00:26:01,059 --> 00:26:08,610
is an example of an energy balance model that
has been developed by Kevin Hall's group in

270
00:26:08,610 --> 00:26:14,860
children. These equations also now take into
account the energy cost of growth. So the

271
00:26:14,860 --> 00:26:20,120
little p here is indicating partitioning
of energy. So in these equations, you can

272
00:26:20,120 --> 00:26:26,760
estimate how differences is in energy from
intake and expenditure and growth over time

273
00:26:26,760 --> 00:26:33,149
can influence the amount of energy that then
would be partitioned into the fat-free mass,

274
00:26:33,149 --> 00:26:41,080
into the fat mass over time.
So this is really exciting for the field because

275
00:26:41,080 --> 00:26:47,919
Kevin Hall and others were able to leverage
these large datasets of doubly labeled water

276
00:26:47,919 --> 00:26:53,720
as well as body composition from various studies
in children to not only develop these models

277
00:26:53,720 --> 00:26:58,330
but to validate them as well. And just to
give you an idea as to the consideration of

278
00:26:58,330 --> 00:27:04,830
the model parameters that are behind the scenes
so, behind that piece of the model that determined

279
00:27:04,830 --> 00:27:11,350
energy partitioning into fat-free mass or
of fat mass. You can see here in this chart

280
00:27:11,350 --> 00:27:16,840
that these models are very, very sophisticated.
They're taking into account growth, not only

281
00:27:16,840 --> 00:27:23,760
overall growth but the growth that is occurring
at different organs as well as tissues. They're

282
00:27:23,760 --> 00:27:30,919
also being able to take into account the metabolic
rates of these organs and tissues, like I

283
00:27:30,919 --> 00:27:37,280
mentioned, as well as the impact of different
levels of physical activity. So I want to

284
00:27:37,280 --> 00:27:42,330
show you the utility of these dynamic energy
balance models in action and really to emphasize

285
00:27:42,330 --> 00:27:44,059
to you the value that this can bring to the
field.

286
00:27:44,059 --> 00:27:51,630
So I called Kevin one day and I said, "Kevin,
I have a problem. I've been working on some

287
00:27:51,630 --> 00:28:00,130
data using infant formula, and we have learned
in our studies that about...that people are

288
00:28:00,130 --> 00:28:06,640
over dispensing formula by about 11% per bottle
that's being made. So if I would extrapolate

289
00:28:06,640 --> 00:28:10,769
this over the total number of servings provided
to babies in a day over by the number of days

290
00:28:10,769 --> 00:28:20,281
in a week, and now I have an estimate of overfeeding.
I want to know what this is going to do to

291
00:28:20,281 --> 00:28:25,511
a baby's body weight over time. If that was
a baby boy and a baby girl." And he's like,

292
00:28:25,511 --> 00:28:32,970
"No problem, we can put this into our infant
growth model." So, we worked on this problem

293
00:28:32,970 --> 00:28:39,490
together. And so, we overlaid here this is
the CDC growth chart for a baby boy (blue) and

294
00:28:39,490 --> 00:28:46,429
a baby girl, it's pink. And assuming that the
baby girl and baby boy is born at the 50th

295
00:28:46,429 --> 00:28:53,450
percentile and they are overfed this number
of calories per week, Kevin's model was able

296
00:28:53,450 --> 00:28:59,250
to predict for us how the body weight would
change, how that person would grow from birth

297
00:28:59,250 --> 00:29:02,690
to six months, which was the period of our
question.

298
00:29:02,690 --> 00:29:09,040
So you can see here in this example, the baby
boy who starts off with, you know, a weight

299
00:29:09,040 --> 00:29:12,730
and length right at the 50th percentile all
over feeding to this degree, based on this

300
00:29:12,730 --> 00:29:19,230
dynamic model, showed us that this baby boy
would have a weight Z score that would be

301
00:29:19,230 --> 00:29:25,039
above the 70 percentile fall and very similar
for the baby girl. The other thing that I

302
00:29:25,039 --> 00:29:33,020
wanted to know was where is this energy being
partitioned. That is then the next exciting

303
00:29:33,020 --> 00:29:37,870
part of Kevin's model. So it's kind of being
cut off here by my video. I apologize for

304
00:29:37,870 --> 00:29:44,720
that. But we were able to then model what
the changes would be in this hypothetical

305
00:29:44,720 --> 00:29:49,700
boy and girl body fat stores as a result of
the overfeeding. So you would see this would

306
00:29:49,700 --> 00:29:56,970
be a baby girl that would be fed according
to a regular goal, which would be to not have

307
00:29:56,970 --> 00:30:03,525
overfeeding. This would be the body fat mass
prediction in the baby girl that had the 11%

308
00:30:03,525 --> 00:30:09,176
overfeeding and the same results here for
the baby boy. So I wanted to show you as an

309
00:30:09,176 --> 00:30:15,000
example of how you could utilize these models
in future studies in this area.

310
00:30:15,000 --> 00:30:20,664
And to give you now a flavor for a proof of
concept with designing an intervention based

311
00:30:20,664 --> 00:30:26,030
on these models. Similar to Kevin Hall, Diana
Thomas she developed a mathematical model

312
00:30:26,030 --> 00:30:31,306
that would predict the amount of calories
a woman would need to eat in each trimester

313
00:30:31,306 --> 00:30:37,809
of pregnancy to gain the recommended amount
of weight that is predetermined by the Institute

314
00:30:37,809 --> 00:30:44,619
of Medicine guidelines. And those are BMI
specific for people in pregnancy. So I brought

315
00:30:44,619 --> 00:30:50,260
this problem to Diana and she created this
model and then we created this applet or online

316
00:30:50,260 --> 00:30:57,100
tool where you can come to the tool. I can
enter in a patient's age, her height, and

317
00:30:57,100 --> 00:31:02,370
her weight. And the model will estimate for
me the number of calories that this person

318
00:31:02,370 --> 00:31:08,510
would need to eat in each trimester of pregnancy
in order to gain weight within this Green

319
00:31:08,510 --> 00:31:14,320
Zone. What's really neat about an applet like
this is, as I mentioned, we can have specific

320
00:31:14,320 --> 00:31:20,360
energy intake targets at different points
in time. We can also as a function of time,

321
00:31:20,360 --> 00:31:26,779
then clinically plot changes to the person's
weight over time and see how well the person's

322
00:31:26,779 --> 00:31:34,809
doing to adhering to these calorie goals based
on the estimated changes in body weight.

323
00:31:34,809 --> 00:31:41,260
We utilize this tool to drive a weight gain
intervention. And this is again, an example

324
00:31:41,260 --> 00:31:45,830
thats in pregnancy. It's a little different,
but it can apply here into interventions say

325
00:31:45,830 --> 00:31:53,669
refeeding in a failure to thrive type scenario.
So in our study, we enrolled overweight and

326
00:31:53,669 --> 00:31:59,309
obese pregnant women. We used this dynamic
gestational weight gain model to provide women

327
00:31:59,309 --> 00:32:05,289
with calorie goal estimates for each trimester
of their pregnancy. The women received these

328
00:32:05,289 --> 00:32:10,389
calorie goals, and they were coached to understand
that their body energy stores or their weight

329
00:32:10,389 --> 00:32:16,620
here in this case was really a reflection
of whether they were hitting this daily energy

330
00:32:16,620 --> 00:32:20,761
intake target. Of course, just like Kevin's
model for the kids, this gestational weight

331
00:32:20,761 --> 00:32:25,760
gain model also took into account maternal
physical activity. So over the course of the

332
00:32:25,760 --> 00:32:29,950
pregnancy, all the women were doing were charting
their weight. They knew if their weight was

333
00:32:29,950 --> 00:32:36,200
in this Green Zone that they were hitting
their dietary intake target. On the back end

334
00:32:36,200 --> 00:32:40,190
we programmed this into an app.
Clinicians could see the weights coming in.

335
00:32:40,190 --> 00:32:45,859
Health coaches were seeing the weights coming
in. If the weight was out of the target zone,

336
00:32:45,859 --> 00:32:50,940
we were able then to provide targeted and
personalized feedback to the woman to help

337
00:32:50,940 --> 00:32:57,750
tailor her dietary intake goals, as well as
her dietary intake in general, to have the

338
00:32:57,750 --> 00:33:04,120
weight falling within this weight gain zone.
So I'm overlaying now the example here of

339
00:33:04,120 --> 00:33:09,600
failure to thrive. Wouldn't it be awesome
to be able to partner with people like Diana

340
00:33:09,600 --> 00:33:15,500
and Kevin to derive not only static estimates
of energy needs for some of these critical

341
00:33:15,500 --> 00:33:22,120
pediatric problems that we're facing with
malnutrition value to thrive as well as obesity,

342
00:33:22,120 --> 00:33:28,810
we could design interventions that not only
provide the energy requirement, but model

343
00:33:28,810 --> 00:33:34,540
for us the changes in these people's fat mass,
fat-free mass and growth over time, and provide

344
00:33:34,540 --> 00:33:42,740
coaching in real-time in outpatients and deliver
interventions. So I'm going to wrap up by

345
00:33:42,740 --> 00:33:51,250
saying energy balance is dynamic. For decades,
the field of nutrition has relied on energy

346
00:33:51,250 --> 00:33:56,269
requirement estimations from static equations
and from the combination of different methods.

347
00:33:56,269 --> 00:34:01,390
But these new dynamic models of energy balance
have improved our ability to prescribe precise

348
00:34:01,390 --> 00:34:08,940
changes in calorie intake and to learn what
are the expected changes not only to a person's

349
00:34:08,940 --> 00:34:15,859
body weight but also to their body composition.
In pediatric populations, models include important

350
00:34:15,859 --> 00:34:21,220
energy partitioning for growth, as well as
the resulting changes in energy partitioning

351
00:34:21,220 --> 00:34:26,770
into the body's energy stores. I've showed
you some examples that we can rely on these

352
00:34:26,770 --> 00:34:33,600
models to guide treatment in certain clinical
conditions such as pregnancy. And I think

353
00:34:33,600 --> 00:34:38,910
that we could use models to guide treatment
in conditions of malnutrition as well as with

354
00:34:38,910 --> 00:34:44,679
childhood obesity and that we can create similar
e-health tools to deliver timely advice to

355
00:34:44,679 --> 00:34:51,839
patients and customized interventions. So
I would like to give a shameless plug for

356
00:34:51,839 --> 00:34:57,230
Pennington and my lab. I love to mentor. We
are looking for postdocs.

357
00:34:57,230 --> 00:35:04,480
DR. VIJAY SRINIVASAN: Hello, I am Vijay Srinivasan,
and it's my pleasure to be with you today.

358
00:35:04,480 --> 00:35:09,390
My presentation today will discuss the concepts
of defining and measuring outcomes from pediatric

359
00:35:09,390 --> 00:35:17,190
malnutrition. I have listed my disclosures
here and do not have any conflicts of interest

360
00:35:17,190 --> 00:35:24,030
relevant to this presentation. Though it may
be tempting to think of discrete and measurable

361
00:35:24,030 --> 00:35:29,540
outcomes directly attributable to malnutrition
in early life, we are typically faced with

362
00:35:29,540 --> 00:35:34,710
the complexities of several built-in risk
factors and etiologies that are often intertwined

363
00:35:34,710 --> 00:35:40,930
in the cause and effect pathway as highlighted
in this figure here. Most of these risk factors

364
00:35:40,930 --> 00:35:45,900
are often part of the malnutrition paradigm,
a key point to be considered when we design

365
00:35:45,900 --> 00:35:53,370
studies and interventions aimed at combating
malnutrition in early life. Further, as these

366
00:35:53,370 --> 00:36:00,859
risks accumulate, early child development
is increasingly compromised. Over 5 million

367
00:36:00,859 --> 00:36:04,450
children under the age of five die each year
with these deaths occurring disproportionately

368
00:36:04,450 --> 00:36:09,450
in low-income and middle-income countries.
For many vulnerable children, the risk factors

369
00:36:09,450 --> 00:36:15,020
of malnutrition, illness, lack of access to
care, poor social support systems, and poverty

370
00:36:15,020 --> 00:36:19,970
contribute to an unstable health trajectory
with repeated episodes of illness interspersed

371
00:36:19,970 --> 00:36:24,730
with short periods of incomplete recovery
highlighted here. In many regions of the world,

372
00:36:24,730 --> 00:36:30,420
acutely ill children may not even access health
care during episodes of illness. Of those

373
00:36:30,420 --> 00:36:35,619
that do studies have shown that children who
are hospitalized have a dramatically higher

374
00:36:35,619 --> 00:36:40,640
risk of death, both during the hospitalization
as well as in the months following discharge

375
00:36:40,640 --> 00:36:48,310
than their community peers. Even when guidelines
for treatment and clinical follow-up are closely

376
00:36:48,310 --> 00:36:54,960
adhered to. Thus contact with the health care
system resulting in hospitalization serves

377
00:36:54,960 --> 00:37:00,380
both as an indicator of vulnerability, as
well as a time point where children are readily

378
00:37:00,380 --> 00:37:06,700
accessible for intervention. Traditionally,
pediatric malnutrition is defined as an imbalance

379
00:37:06,700 --> 00:37:11,210
between nutrient requirements and intake that
results in cumulative deficits of energy,

380
00:37:11,210 --> 00:37:16,490
protein or micronutrients that may negatively
affect growth, development, and other relevant

381
00:37:16,490 --> 00:37:20,770
outcomes.
In 2013, a new classification scheme was developed

382
00:37:20,770 --> 00:37:27,730
by Aspen, led by Dr. Mehta, and consists
of the five domains of chronicity etiology,

383
00:37:27,730 --> 00:37:33,230
mechanisms of nutrient imbalance, severity
of malnutrition, and outcomes. This more granular

384
00:37:33,230 --> 00:37:38,880
and detailed classification scheme is an important
step forward to better characterizing chronicity

385
00:37:38,880 --> 00:37:47,450
and mechanism while laying a better framework
to define and measure outcomes. There are

386
00:37:47,450 --> 00:37:53,339
several key outcomes impacted by nutrient
imbalances in addition to traditional anthropometric

387
00:37:53,339 --> 00:38:00,859
parameters and biomarkers other outcomes affected
by malnutrition include changes in the metabolome,

388
00:38:00,859 --> 00:38:08,410
mortality, organ dysfunction, loss of skin
integrity, greater ICU dependency with hospital

389
00:38:08,410 --> 00:38:14,670
resource utilization. Immune dysregulation
of acquired infections. Alterations in muscle

390
00:38:14,670 --> 00:38:20,800
and bone function as well as cognition and
development. All of these measurable outcomes

391
00:38:20,800 --> 00:38:26,260
can ultimately influence the future trajectory
of health, status, development, education

392
00:38:26,260 --> 00:38:30,910
and socioeconomic status of the individual
and the community as a whole.

393
00:38:30,910 --> 00:38:35,319
Defining outcomes more clearly within the
appropriate framework is therefore valuable

394
00:38:35,319 --> 00:38:42,560
to help us advance greater evidence-based
nutrition practices. I am now going into a

395
00:38:42,560 --> 00:38:47,570
little bit of detail in some of these outcomes,
which I think are relevant for us to understand.

396
00:38:47,570 --> 00:38:53,250
Childhood sarcopenia has only been recently
identified within the context of reduced skeletal

397
00:38:53,250 --> 00:38:59,060
muscle mass. The similarities between Sarcopenia
and malnutrition definitions include the overlapping

398
00:38:59,060 --> 00:39:05,319
principles related to depleted lean body mass
alterations in muscle functionality and suboptimal

399
00:39:05,319 --> 00:39:10,970
nutrient intake leading to nutrition deficiencies.
In children, this also translates to growth

400
00:39:10,970 --> 00:39:17,319
failure with a potential neurodevelopmental
delay in growth and fine motor development

401
00:39:17,319 --> 00:39:23,630
and cognition. There is currently no gold
standard tool to assess motor function impairment

402
00:39:23,630 --> 00:39:28,460
in children being assessed for sarcopenia.
It is challenging to assess muscle function

403
00:39:28,460 --> 00:39:32,760
in a standardized way with infants and young
children due to several factors, including

404
00:39:32,760 --> 00:39:38,380
the development of postural control, coordination,
core stability, and the ability to perform

405
00:39:38,380 --> 00:39:43,349
purposeful in isolated movements.
In older children that is school-aged children

406
00:39:43,349 --> 00:39:49,130
or adolescents handgrip test and a six-minute
walk test are well documented to determine

407
00:39:49,130 --> 00:39:55,940
upper body strength and functional exercise
performance respectively. Cognition and development

408
00:39:55,940 --> 00:40:02,390
is another very vital area in which it is
challenging to measure outcomes. The continuum

409
00:40:02,390 --> 00:40:06,930
from the prenatal stages through the early
to middle childhood years represents a critical

410
00:40:06,930 --> 00:40:11,450
period for child neurodevelopment. Nutritional
deficiencies, infection, and inflammation

411
00:40:11,450 --> 00:40:17,190
are major contributors to impaired child neurodevelopment
in these years. Malnutrition as highlighted

412
00:40:17,190 --> 00:40:24,099
below can result from lack of both macro and
micronutrients. Stunting in early childhood

413
00:40:24,099 --> 00:40:28,820
is associated with poorer cognitive development
and academic performance in later childhood.

414
00:40:28,820 --> 00:40:35,430
Even mild but persistent malnutrition in early
life that is during the first two years negatively

415
00:40:35,430 --> 00:40:41,119
influences reasoning, visuospatial functions,
IQ, language development, attention, learning,

416
00:40:41,119 --> 00:40:46,560
and academic achievement.
The majority of studies which have investigated

417
00:40:46,560 --> 00:40:52,490
the association between nutrition and cognitive
Development have focused on individual micronutrients,

418
00:40:52,490 --> 00:40:59,400
including omega-three fatty acids, vitamin
B12, folic acid, zinc, iron, and iodine. The

419
00:40:59,400 --> 00:41:03,329
evidence is more consistent from observational
studies which suggest that these micronutrients

420
00:41:03,329 --> 00:41:08,860
play an important role in the cognitive development
of children. However, the results from intervention

421
00:41:08,860 --> 00:41:13,820
trials of simple nutrients are often inconsistent
and inconclusive. We need better controlled

422
00:41:13,820 --> 00:41:19,670
and more adequately powered studies. Additionally,
children living in poor countries may encounter

423
00:41:19,670 --> 00:41:25,329
more multiple micronutrient deficiencies as
opposed to children living in richer countries

424
00:41:25,329 --> 00:41:29,470
who are reasonably well nourished and where
a small deficiency in one nutrient may not

425
00:41:29,470 --> 00:41:35,440
result in measurable long-term change in cognitive
outcomes due to compensation over time. These

426
00:41:35,440 --> 00:41:40,079
are important considerations for us to keep
in mind because nutrients do not act alone.

427
00:41:40,079 --> 00:41:45,650
Rather, they are often synergistic in mechanism
and in other contexts may even have antagonistic

428
00:41:45,650 --> 00:41:51,420
effects with each other.
There are several gaps that we need to account

429
00:41:51,420 --> 00:41:55,410
for as we continue to explore how to define
and measure outcomes related to malnutrition.

430
00:41:55,410 --> 00:42:01,570
For example, how does timing, severity, frequency,
and duration of nutrition deprivation impact

431
00:42:01,570 --> 00:42:07,130
outcomes? What are the right scales, tools,
and surveys to measure outcomes? What should

432
00:42:07,130 --> 00:42:12,480
be the approach to account for confounders
that are often found in lockstep with nutrition

433
00:42:12,480 --> 00:42:19,260
and impact outcomes in very interrelated ways?
The development of a common data element framework

434
00:42:19,260 --> 00:42:23,240
to create a common language among pediatric
nutrition researchers is an important first

435
00:42:23,240 --> 00:42:29,200
step to establish precision, comparison, and
reproducibility of findings. Additionally,

436
00:42:29,200 --> 00:42:33,980
investing in registries and epistries supported
by informatics applications may provide a

437
00:42:33,980 --> 00:42:38,190
sound footing to study both short-term and
long-term outcomes from nutrition evaluation

438
00:42:38,190 --> 00:42:44,720
and interventions. I would like to conclude
by emphasizing that early adversities associated

439
00:42:44,720 --> 00:42:49,000
with severe poverty and chronic undernutrition
contribute to the loss of development potential

440
00:42:49,000 --> 00:42:54,240
and set children onto life course trajectories
associated with negative health, educational,

441
00:42:54,240 --> 00:43:00,510
psychological, and economic consequences that
extended to many subsequent generations.

442
00:43:00,510 --> 00:43:04,210
With the recognition that the foundations
of adult health, wellness, economic capacity,

443
00:43:04,210 --> 00:43:09,330
and well-being begin with experiences from
conception to three years of age, the W.H.O.

444
00:43:09,330 --> 00:43:14,460
has endorsed the promotion of child development
during the first three years as a key strategy

445
00:43:14,460 --> 00:43:21,270
for the success of United Nations Sustainable
Development Goals. Thank you for your time

446
00:43:21,270 --> 00:43:25,134
and attention. I look forward to our discussion
to follow later.

447
00:43:25,134 --> 00:43:35,780
DR. DEBORAH FRANK: [Inaudible] present...this information which comes from
the research group, our Children's Health

448
00:43:35,780 --> 00:43:44,240
Watch, in which I am involved to such a distinguished
audience. The information is available online

449
00:43:44,240 --> 00:43:49,600
at the website I will show you at the end
of the talk. The aims of the presentation

450
00:43:49,600 --> 00:43:55,329
are to summarize data regarding energy insecurity,
housing instability, and cash transfers on

451
00:43:55,329 --> 00:44:00,780
low birth weight, underweight in childhood,
and anemia. I want to also summarize what

452
00:44:00,780 --> 00:44:08,640
little we know about the impact of policy
interventions in these domains on these indicators.

453
00:44:08,640 --> 00:44:18,620
And to suggest some avenues for future research.
Economic hardship acts upon children in the

454
00:44:18,620 --> 00:44:27,329
first 2000 days of life through many mechanisms,
including food insecurity, housing instability,

455
00:44:27,329 --> 00:44:36,819
energy insecurity. These act reciprocally,
each one exacerbating the other and converge

456
00:44:36,819 --> 00:44:45,920
on the body of the baby. Most at risk are
minoritized families, those with young children,

457
00:44:45,920 --> 00:44:51,480
immigrant members, or children with special
health care needs. They are most likely to

458
00:44:51,480 --> 00:44:57,870
have difficulty in meeting their needs for
housing and household energy.

459
00:44:57,870 --> 00:45:03,470
I started out as a food insecurity and underweight
researcher and was not interested at all in

460
00:45:03,470 --> 00:45:12,490
these other dimensions until I showed this
graph to the staff of the local food bank

461
00:45:12,490 --> 00:45:18,660
because I couldn't interpret it in our emergency
room for the three months after the coldest

462
00:45:18,660 --> 00:45:28,200
months, the rates of children whose weight-for-age was less than
the fifth percentile increased by 36%. I was

463
00:45:28,200 --> 00:45:37,480
puzzled, but they weren't. They said that's
heat or eat. Then, this observation from

464
00:45:37,480 --> 00:45:42,840
the front lines compelled my colleagues to
develop empirically a survey instrument to

465
00:45:42,840 --> 00:45:48,830
identify levels of energy insecurity... Realizing
from our own experience that inability to

466
00:45:48,830 --> 00:45:54,680
keep houses and food cold can be as dangerous
as inability to keep people warm. Moderately

467
00:45:54,680 --> 00:46:01,630
energy insecure, meant receiving a threatened
utility shut off in the last year and severe

468
00:46:01,630 --> 00:46:07,589
energy insecurity refer to actual utility
shut off at least one day with no energy for

469
00:46:07,589 --> 00:46:18,160
heating or cooling or use of a cooking stove
as a heating source in the last year. We found

470
00:46:18,160 --> 00:46:27,440
that there was a dose relationship between
these levels of energy and security and food

471
00:46:27,440 --> 00:46:33,000
insecurity particularly at the child level,
which has been shown in other studies to predict

472
00:46:33,000 --> 00:46:43,069
growth faltering anemia.
Whether efforts to mitigate energy insecurity

473
00:46:43,069 --> 00:46:50,849
were effective, as indicated by our finding
that children in families who participate

474
00:46:50,849 --> 00:46:56,089
in the Low Income Home Energy Assistance Program,
were less likely to have a weight-for-age less

475
00:46:56,089 --> 00:47:02,369
than the fifth percentile and had a higher
continuous weight for a Z score. Housing is

476
00:47:02,369 --> 00:47:14,030
harder to measure, homelessness is a well
established threat to child nutrition, but

477
00:47:14,030 --> 00:47:21,440
there are many other dimensions of housing
instability, including moving two or more

478
00:47:21,440 --> 00:47:26,920
times in the past year, being behind on paying
rent or mortgage or being threatened or actually

479
00:47:26,920 --> 00:47:35,070
evicted. Housing instability dimensions and
nutritional indicators in children vary, as

480
00:47:35,070 --> 00:47:40,530
might be expected homelessness or its recent
history, was related to low birth weight,

481
00:47:40,530 --> 00:47:46,940
anemia, iron deficiency and referral for
growth faltering. Multiple moves is tied to

482
00:47:46,940 --> 00:47:52,030
low birth weight and lower average weight
for age and eviction are threatened or actual

483
00:47:52,030 --> 00:48:01,390
pregnancy is tied to low birth weight. Housing
subsidies, have some impact within families

484
00:48:01,390 --> 00:48:08,220
that are already food insecure among children
from 2,764 food insecure families.

485
00:48:08,220 --> 00:48:15,600
Those without housing subsidies compared
to those with showed an adjusted odds ratio

486
00:48:15,600 --> 00:48:23,330
of weight for age greater than two standard
deviations below the mean. This is all that

487
00:48:23,330 --> 00:48:38,900
we knew about targeted policy interventions.
We then asked whether non-targeted subsidies

488
00:48:38,900 --> 00:48:45,579
had any discernible impact on the risk of
early undernutrition. The first was the Earned

489
00:48:45,579 --> 00:48:52,660
Income Tax Credit, which is uniform at the
federal level with various...in terms of state

490
00:48:52,660 --> 00:49:01,829
generosity, and it was found that the level
of EITC available to families per state was

491
00:49:01,829 --> 00:49:07,780
reflected in the reduction of low birth weight
across all racial ethnic groups. Whereas EITC

492
00:49:07,780 --> 00:49:17,210
requires earned income and is given only once
a year. During the COVID pandemic, the advanced

493
00:49:17,210 --> 00:49:25,599
child tax credit was implemented which provided
a monthly subsidy regardless of prior earnings

494
00:49:25,599 --> 00:49:36,380
to families by the age and number of children.
The census household pulse data which is obtained

495
00:49:36,380 --> 00:49:46,190
online found that when the CTC began in July
2021, it was associated for the next few months

496
00:49:46,190 --> 00:49:51,080
with a 26% reduction in food insufficiency,
which is more severe than food insecurity

497
00:49:51,080 --> 00:49:59,650
among households with children.
The CTC monthly payments expired and we have

498
00:49:59,650 --> 00:50:05,270
now been able to document a 25% increase in
food insufficiency among households with children

499
00:50:05,270 --> 00:50:18,760
through July 2022. The census household pulse
survey has no clinical correlates available.

500
00:50:18,760 --> 00:50:24,500
We have some take home messages. Energy and
housing and security are correlated with food

501
00:50:24,500 --> 00:50:30,690
insecurity often at the child food insecurity
level, which in turn is a precursor of indicators

502
00:50:30,690 --> 00:50:37,190
of malnutrition. Even after controlling for
food insecurity, energy and housing insecurity

503
00:50:37,190 --> 00:50:41,960
pose incremental risk to young children's
growth and health. Most data is available

504
00:50:41,960 --> 00:50:51,890
for low birth weight and there's real sparse
postneonatal data, low income energy assistance

505
00:50:51,890 --> 00:50:56,580
and housing subsidies are associated with
improved weight for age, and young children

506
00:50:56,580 --> 00:51:02,950
and cash transfers associated with produce
low birth weight. Cash transfers are also

507
00:51:02,950 --> 00:51:08,650
associated with decreased food insufficiency,
which increases on transfer searches continued.

508
00:51:08,650 --> 00:51:13,059
We need to inform our future research with
guidance from those who have lived experience

509
00:51:13,059 --> 00:51:23,040
both quantitative and qualitative.
We should focus on sentinel groups. We need

510
00:51:23,040 --> 00:51:28,619
clinical correlation to national surveys,
monitoring nutritional status to identify

511
00:51:28,619 --> 00:51:35,050
the impact in the first 2,000 days of fluctuating
environmental and political conditions such

512
00:51:35,050 --> 00:51:42,569
as heat waves, or anti immigrant rhetoric
accelerating evictions. There are natural

513
00:51:42,569 --> 00:51:48,270
experiments which exacerbate or mitigate hardships,
simultaneously in the context of each other

514
00:51:48,270 --> 00:51:58,780
rather than isolation. There will be new electronic
health records innovations which mandate in

515
00:51:58,780 --> 00:52:06,390
documentation of housing and food insecurity
in adult inpatient populations but no other

516
00:52:06,390 --> 00:52:13,510
I would suggest using our research to prioritize
policy relevance. Prenatal and child malnutrition

517
00:52:13,510 --> 00:52:18,060
correlates and material hardships other than
food security are understudied. There are

518
00:52:18,060 --> 00:52:24,819
measurement tools but not uniform. Early electronic
health record measures of hardship should

519
00:52:24,819 --> 00:52:31,230
be expanded to include obstetric and pediatric
populations. Important for planning intervention

520
00:52:31,230 --> 00:52:36,410
to disentangle which effects are mediated
by food insecurity and nutrition benefits

521
00:52:36,410 --> 00:52:39,670
which are identified incremental to those
factors.

522
00:52:39,670 --> 00:52:47,760
And we need to identify on clinical indicators
of child malnutrition policies that broadly

523
00:52:47,760 --> 00:52:52,559
support or reduce family incomes versus those
which are targeted to a specific hardship.

524
00:52:52,559 --> 00:52:57,910
Remember, the public policies and economic
conditions are written on the bodies and brains

525
00:52:57,910 --> 00:53:03,180
of babies. And please feel free to be in touch
with me or go to the website for more information.

526
00:53:03,180 --> 00:53:04,180
Thank you.

527
00:53:04,180 --> 00:53:11,460
DR. LAURIE NOMMSEN-RIVERS: Good day listeners.
In this brief talk, I will provide an overview

528
00:53:11,460 --> 00:53:18,960
of excess weight loss in the context of the
exclusively breastfeeding newborn. Today's

529
00:53:18,960 --> 00:53:24,651
talk will focus on the scope of the problem,
limitations of current knowledge and research

530
00:53:24,651 --> 00:53:32,160
priorities across the translational research
continuum. Exclusive breastfeeding is recommended

531
00:53:32,160 --> 00:53:39,390
by all major public health organizations for
the first six months of life. While most US

532
00:53:39,390 --> 00:53:45,450
infants start off breastfeeding, the drop
off is steep and exclusive breastfeeding rates

533
00:53:45,450 --> 00:53:51,980
are quite low. One of the two healthy people
2030 goals for breastfeeding is to increase

534
00:53:51,980 --> 00:54:01,420
the rate of exclusive breastfeeding at six
months, from a baseline of 24.9% to 42.4%.

535
00:54:01,420 --> 00:54:10,230
Thus, promotion of exclusive breastfeeding
is a public health priority. It is universally

536
00:54:10,230 --> 00:54:15,330
recognized that newborns lose weight in the
first few days of life for exclusively breastfeeding

537
00:54:15,330 --> 00:54:21,530
newborns initially infant feeding volume is
very small. However, sometime in the first

538
00:54:21,530 --> 00:54:28,910
few days postpartum milk production ramps
up, we see a parallel increase in infant weight.

539
00:54:28,910 --> 00:54:35,059
Note that typically infants are still on the
downward trajectory with their weight at the

540
00:54:35,059 --> 00:54:41,230
time of the birth hospital discharge. Despite
weight loss being a normal adjustment to extra

541
00:54:41,230 --> 00:54:48,570
uterine life, sometimes weight loss is excessive
or never rebounds with potentially life threatening

542
00:54:48,570 --> 00:54:55,339
sequela. Thus, there is a fine balance between
avoiding unnecessary supplementation and avoiding

543
00:54:55,339 --> 00:55:01,710
readmission and health consequences due to
breastfeeding failure. You may be thinking

544
00:55:01,710 --> 00:55:07,230
why not just give all newborns some formula
as a bridge to avoid excess weight loss? There

545
00:55:07,230 --> 00:55:13,410
are several reasons for avoiding formula supplement,
one of which is the clearly established negative

546
00:55:13,410 --> 00:55:20,329
impact it has on sustaining breastfeeding.
In the study shown here, we found the adjusted

547
00:55:20,329 --> 00:55:26,770
risk of stopping breastfeeding altogether
to be over three fold higher if formula was

548
00:55:26,770 --> 00:55:35,030
used during the birth hospitalization. And
these results were in mothers with intention

549
00:55:35,030 --> 00:55:41,900
to exclusively breastfeed. Nonetheless, there
are serious risks to the newborn when weight

550
00:55:41,900 --> 00:55:47,390
loss is excessive, including complications
such as hypoglycemia, jaundice, dehydration

551
00:55:47,390 --> 00:55:51,990
and hypernatremia.
And if left undetected excess weight loss

552
00:55:51,990 --> 00:56:00,869
can lead to seizures, renal failure, or hypovolemic
shock. When it comes to newborn weight loss,

553
00:56:00,869 --> 00:56:07,520
unfortunately U.S. babies are the biggest losers.
In two separate cohort studies with hundreds

554
00:56:07,520 --> 00:56:13,089
of newborns, we observed greater newborn weight
loss as compared to the growth velocity data

555
00:56:13,089 --> 00:56:20,329
from the WHO growth reference study. The median
WHO baby is already surpassing birth weight

556
00:56:20,329 --> 00:56:28,319
at day seven. U.S. babies on average have gained
three to five ounces less weight in the first

557
00:56:28,319 --> 00:56:36,170
week of life, with 16% of babies in Davis,
California, and 18% of babies in Sacramento,

558
00:56:36,170 --> 00:56:39,690
California losing 10% or more of their birth
weight. And what does it mean to lose 10%

559
00:56:39,690 --> 00:56:50,000
or more birth weight? In this systematic review,
the authors compiled the admission weight

560
00:56:50,000 --> 00:56:56,150
and serum sodium concentration of newborns
readmitted for breastfeeding associated hypernatremia.

561
00:56:56,150 --> 00:57:04,369
All had dangerously elevated serum sodium.
96% have weight loss greater than or equal

562
00:57:04,369 --> 00:57:11,799
to 10% of birth weight. Most clinicians see
weight loss of 10% or greater as a genuine

563
00:57:11,799 --> 00:57:15,730
concern and to be avoided.
Throughout this talk when I refer to excess

564
00:57:15,730 --> 00:57:22,410
weight loss, I'm using the cut off of 10%
weight loss or more. The newborn weight loss

565
00:57:22,410 --> 00:57:29,099
tool or Newt was developed to improve tracking
of newborn weight trajectories based on their

566
00:57:29,099 --> 00:57:36,000
birth hospitalization rates. The Newt can
be used to characterize current weight loss

567
00:57:36,000 --> 00:57:42,490
relative to a large database of U.S. newborns
weighed during their birth hospitalization.

568
00:57:42,490 --> 00:57:49,730
That context is important. The orange line
represents the 90th percentile, which for

569
00:57:49,730 --> 00:57:55,349
vaginally born infants is consistent with
more than 10% weight loss, suggesting an at

570
00:57:55,349 --> 00:58:03,039
risk trajectory. Even though the Newt is useful
for characterizing current weight loss relative

571
00:58:03,039 --> 00:58:08,690
to other U.S. newborns, the data are entirely
based on one to two weights per infant during

572
00:58:08,690 --> 00:58:14,520
their birth hospitalization. And this raises
the question as to whether the Newt is valid

573
00:58:14,520 --> 00:58:20,570
as a predictive tool. Thus, we use data from
the Davis site of the WHO growth reference

574
00:58:20,570 --> 00:58:26,650
study to test the validity of the Newt in
predicting excess weight loss once discharged

575
00:58:26,650 --> 00:58:30,859
to home.
Unfortunately, we found the Newt to be a very

576
00:58:30,859 --> 00:58:37,099
poor predictor of excess weight loss. Only
21% of true cases of excess weight loss that

577
00:58:37,099 --> 00:58:43,849
were detected on home visits were in the at
risk range based on the Newt trajectory from

578
00:58:43,849 --> 00:58:50,020
the hospital weights. Plus, in a recent randomized
trial using the Newt as a clinical aid in

579
00:58:50,020 --> 00:58:55,720
the decision to supplement, there was no difference
in hospital readmissions being about three

580
00:58:55,720 --> 00:59:03,220
and a half percent of exclusively breastfed
infants readmitted in both groups. You may

581
00:59:03,220 --> 00:59:08,430
be wondering why birth hospitalization weights
are a poor predictor of post discharge weights.

582
00:59:08,430 --> 00:59:15,760
Winnicott once said, "There is no such thing
as a baby. If you set out to describe a baby,

583
00:59:15,760 --> 00:59:22,359
you will find you're describing a baby and
someone." In other words, we need to view weight

584
00:59:22,359 --> 00:59:30,830
loss risk factors in the context of the breastfeeding
DYAD both maternal risk factors and infant

585
00:59:30,830 --> 00:59:37,589
risk factors. While the first few days of
weight loss are related to infant factors,

586
00:59:37,589 --> 00:59:43,220
once discharged to home maternal factors play
a big role in newborn weight.

587
00:59:43,220 --> 00:59:48,130
For example, based on the data site of the
WHO growth reference study, the risk of excess

588
00:59:48,130 --> 00:59:54,440
weight loss was nearly seven fold greater
when the mother experienced delayed onset

589
00:59:54,440 --> 01:00:00,750
of lactogenesis of more than 72 hours from
birth. We saw a similar pattern in our Sacramento

590
01:00:00,750 --> 01:00:10,680
cohort a first time mother infant DYAD. Unfortunately,
delayed lactogenesis is surprisingly common

591
01:00:10,680 --> 01:00:17,950
in the U.S. in the Sacramento cohort, which
was entirely comprised of first time mothers

592
01:00:17,950 --> 01:00:24,359
44% experienced delayed lactogenesis. Keep
in mind that common doesn't necessarily mean

593
01:00:24,359 --> 01:00:30,250
normal for our species. When we look across
cultures, we see tremendous variation in the

594
01:00:30,250 --> 01:00:35,470
timing of lactogenesis where some cultures
commonly experiencing lactogenesis within

595
01:00:35,470 --> 01:00:42,280
the first 24 hours of birth. Many factors
such as hospital routines and maternal obesity

596
01:00:42,280 --> 01:00:49,859
negatively influenced the timing of lactogenesis
in the U.S. The bottom line is we have a situation

597
01:00:49,859 --> 01:00:55,049
in the U.S. where exclusive breastfeeding is
the public health goal. But we are operating

598
01:00:55,049 --> 01:01:02,300
in an environment where there are many risk
factors for delayed Lactogenesis and lactation

599
01:01:02,300 --> 01:01:05,380
problems.
Situation is magnified by our very weak safety

600
01:01:05,380 --> 01:01:13,670
net for supporting the breastfeeding DYAD
once they are discharged to home. To summarize,

601
01:01:13,670 --> 01:01:18,190
I would like to apply a translational research
framework that was developed for science and

602
01:01:18,190 --> 01:01:24,170
human lactation and infant feeding. I led
a workgroup in the development of this framework

603
01:01:24,170 --> 01:01:33,010
as part of the NIH begin project. Our report
is currently under review. When it comes to

604
01:01:33,010 --> 01:01:37,930
access newborn weight loss, research needs
span the translational research continuum.

605
01:01:37,930 --> 01:01:46,329
At the T1 discovery stage, there is a need
to elucidate the underlying drivers of delayed

606
01:01:46,329 --> 01:01:53,200
lactogenesis and insufficient milk production.
At the T2 human health implication stage,

607
01:01:53,200 --> 01:01:59,750
we need to better characterize newborn weight
change patterns, and their relation to both

608
01:01:59,750 --> 01:02:08,130
healthy and unhealthy clinical outcomes across
diverse groups. At the T3 clinical implication

609
01:02:08,130 --> 01:02:14,930
stage, we need to develop effective interventions
to support timely onset of lactogenesis, abundant

610
01:02:14,930 --> 01:02:18,530
milk production, and healthy newborn weight
gain.

611
01:02:18,530 --> 01:02:25,180
We also need to develop and validate culturally
appropriate screening tools to identify DYADs

612
01:02:25,180 --> 01:02:34,109
at high priority for close post discharge
follow up. At the T4 implementation stage,

613
01:02:34,109 --> 01:02:40,680
we need to develop implementation strategies
that ensure equity and access to optimal breastfeeding

614
01:02:40,680 --> 01:02:47,789
support during the birth hospitalization,
and post discharge. There is also a need for

615
01:02:47,789 --> 01:02:53,930
educational interventions targeting health
care providers, and also interventions targeting

616
01:02:53,930 --> 01:03:03,240
education to parents. At the T5 impact stage,
we need surveillance of weight change patterns,

617
01:03:03,240 --> 01:03:10,920
readmission rates, and severity of hypernatremia
in exclusively breastfeeding newborns. And

618
01:03:10,920 --> 01:03:18,510
importantly, it is essential to monitor disparities
in access to support for exclusive breastfeeding

619
01:03:18,510 --> 01:03:27,910
both during and after the birth hospitalization,
and to monitor disparities in efforts at prevention

620
01:03:27,910 --> 01:03:35,849
and treatment of excess newborn weight loss.
So thank you for your time today and I leave

621
01:03:35,849 --> 01:03:50,789
you with a list of key references that were
the foundation for this talk.

622
01:03:50,789 --> 01:04:01,940
DR. NILESH MEHTA: Hello, welcome back to the Q&A session.
Thank you so much to all our wonderful speakers.

623
01:04:01,940 --> 01:04:09,790
This was indeed a very informative session.
And it has already stimulated a lot of questions.

624
01:04:09,790 --> 01:04:15,299
To everyone participating, please consider
sending in your questions using the box on

625
01:04:15,299 --> 01:04:24,140
your screen. We'll try to get to as many as
we can. Let's get this off. I have 

626
01:04:24,140 --> 01:04:31,170
[inaudible], Dr. Ashley Vargas, in addition to all
the speakers here. And I'd like to start with

627
01:04:31,170 --> 01:04:37,891
our first speaker, Dr. Barr, Stephanie Barr,
this question is for you. Could you comment

628
01:04:37,891 --> 01:04:44,270
on the inherent limitations of predictive
equations for populations related to race,

629
01:04:44,270 --> 01:04:50,289
age, sex and secular trends over time? And
how these are revalidated over time? Are there

630
01:04:50,289 --> 01:04:55,260
any strategies to get these data at regular
intervals to revalidate?

631
01:04:55,260 --> 01:05:00,880
DR. STEPHANIE BARR: I think that's a great
question. And as indicated, there are inherent

632
01:05:00,880 --> 01:05:06,260
limitations of these predictive equations
with the factors that were mentioned. And

633
01:05:06,260 --> 01:05:13,010
Sharon Assadi and group in I believe New Zealand
had a really nice paper that is referenced

634
01:05:13,010 --> 01:05:19,520
in the slides for when you all can take a
closer look at them that explored this specifically.

635
01:05:19,520 --> 01:05:25,230
And I think it's important to know that, as
referenced, there are equations that exist

636
01:05:25,230 --> 01:05:30,250
and I think they could be useful tools in
specific clinical settings, but they're definitely

637
01:05:30,250 --> 01:05:36,420
not perfect. In terms of strategies on how
to move forward, I believe that what needs

638
01:05:36,420 --> 01:05:41,140
to be done in order to going off of what Laurie
was discussing with moving things through

639
01:05:41,140 --> 01:05:45,809
the translational research spectrum, there
needs to be continued greater collaboration

640
01:05:45,809 --> 01:05:51,140
outside as little individual research bubbles
to get a better understanding of typical growth

641
01:05:51,140 --> 01:05:55,020
patterns at a population level.

642
01:05:55,020 --> 01:06:04,820
DR. NILESH MEHTA: Thank you. Yeah, it's absolutely true.
On a similar reign, Dr. Redman next question

643
01:06:04,820 --> 01:06:12,460
is for you. This is from Laurie Bishop from
Boston, how might the energy balance model

644
01:06:12,460 --> 01:06:18,500
be adjusted to predict energy requirements
in children undergoing surgery or who are

645
01:06:18,500 --> 01:06:19,569
hospitalized for acute illness?

646
01:06:19,569 --> 01:06:25,420
DR. LEANNE REDMAN: Yeah, that's an excellent question.
So and it's an important distinction because

647
01:06:25,420 --> 01:06:33,210
the dynamic models that I shared from Kevin
Hall, rely mostly on information coming from

648
01:06:33,210 --> 01:06:40,039
healthy children. But there are a large number
of studies and data available, say from CHOP,

649
01:06:40,039 --> 01:06:46,190
right, Virginia Stallings has done a lot of
work on assessing energy needs in critically

650
01:06:46,190 --> 01:06:52,559
ill babies and young children. Those types
of data can be used to make very specific

651
01:06:52,559 --> 01:06:59,829
models for specific sets of medical conditions
and patient populations. And to the point

652
01:06:59,829 --> 01:07:04,770
of another question, which mentioned that
there are fairly substantial changes that

653
01:07:04,770 --> 01:07:09,780
occur over time in a critically ill child's
resting metabolic rate, which would need to

654
01:07:09,780 --> 01:07:15,770
be taken into account over time. So as long
as the models are developed and validated

655
01:07:15,770 --> 01:07:21,570
on populations that were intending to target,
they should be able to do a good job and they

656
01:07:21,570 --> 01:07:29,620
should definitely be able to be detested in
inpatient type situations as well as outpatient

657
01:07:29,620 --> 01:07:34,700
settings in the future.

658
01:07:34,700 --> 01:07:41,070
DR. NILESH MEHTA: Thank you, Dr. Redman. The next question
is for Dr. Srinivasan. Dr. Srinivasan, what

659
01:07:41,070 --> 01:07:47,440
nutrients are nutrient combinations would
you prioritize, to study for their impact

660
01:07:47,440 --> 01:07:50,619
on neurodevelopmental outcomes in pediatrics?

661
01:07:50,619 --> 01:07:57,029
DR. VIJAY SRINIVASAN: Thank you for that question.
Gosh, I wish I could narrow it down. There

662
01:07:57,029 --> 01:08:03,309
are so many of interest and choices here.
But I think if I were to sort of think about

663
01:08:03,309 --> 01:08:09,279
what are some of the key nutrients that impact
in that continuum that we are focused on right

664
01:08:09,279 --> 01:08:15,812
now, definitely the essential fatty acids,
omega three fatty acids, still a big area

665
01:08:15,812 --> 01:08:23,049
of interest in terms of how they impact neurocognition
and development. And in addition, you know,

666
01:08:23,049 --> 01:08:28,500
iron is, as in my opinion, still an important
nutrient that is there are deficiencies worldwide

667
01:08:28,500 --> 01:08:35,120
and I think it is such an important yet often
overlooked in some ways nutrient. In addition

668
01:08:35,120 --> 01:08:41,540
to the B vitamins as well. I would also add
zinc, which is again another important component

669
01:08:41,540 --> 01:08:42,790
in this and...

670
01:08:42,790 --> 01:08:47,651
and finally, in addition to just
the nutrients, it's even just, you know, looking

671
01:08:47,651 --> 01:08:53,029
at the dietary composition, the quality of
food and, you know, the timing of food, I

672
01:08:53,029 --> 01:08:57,750
mean, there's data emerging, you know, the
intervals at which one eats food in these

673
01:08:57,750 --> 01:09:05,940
early childhood days beyond infancy can impact.
So definitely more questions than answers,

674
01:09:05,940 --> 01:09:12,779
but these are some of the key nutrients that
I would like to sort of recommend that be studied.

675
01:09:12,779 --> 01:09:20,480
DR. NILESH MEHTA: Thanks, Dr. Srinivasan. The next question
is for Dr. Frank. Dr. Frank, you gave a terrific

676
01:09:20,480 --> 01:09:26,009
overview of the social determinants and food
insecurity and how they relate to malnutrition.

677
01:09:26,009 --> 01:09:31,980
Could you say a little more about any research
or literature out there on the effects of

678
01:09:31,980 --> 01:09:38,859
local, state, or federal policies and health
outcomes in chronically ill children? We have

679
01:09:38,859 --> 01:09:47,219
DR. DEBORAH FRANK: We have looked at that in our cohort, which is a exclusively
low-income pediatric cohort seeking health

680
01:09:47,219 --> 01:09:52,600
care of kids under the age of four. We found
that kids with special health care needs are

681
01:09:52,600 --> 01:10:00,150
at higher risk of food insecurity and housing
insecurity, families, and that SSI in both cases

682
01:10:00,150 --> 01:10:07,060
has a (INAUDIBLE), which means the child is
sicker than your obvious kid with special

683
01:10:07,060 --> 01:10:13,130
health care needs and is very hard to get.
But those kids are less likely to have food

684
01:10:13,130 --> 01:10:18,830
insecurity at the child level while they still
have it...the family still has food insecurity

685
01:10:18,830 --> 01:10:25,179
compared to kids without special health care
needs and are also less likely to be housing

686
01:10:25,179 --> 01:10:33,830
insecure. I haven't looked at...It's a good
question about energy insecurity. States vary

687
01:10:33,830 --> 01:10:40,210
in the degree to which doctors can prevent
families from having their power cut off if

688
01:10:40,210 --> 01:10:46,100
they have a special health care-need child
in the house. And that would be a very good

689
01:10:46,100 --> 01:10:56,210
thing to look at. But that is clearly...many
of the outcomes we look at like for age, hospitalization,

690
01:10:56,210 --> 01:11:03,220
and so on are already increased from the special
health care needs. And that so you have a

691
01:11:03,220 --> 01:11:10,300
different baseline when you're saying, OK,
so what are the health effects?

692
01:11:10,300 --> 01:11:17,489
DR. NILESH MEHTA: Thank you. Thank you for that. Dr. Nommsen-Rivers, have
you seen any clever examples of research,

693
01:11:17,489 --> 01:11:23,699
any good research designs that are able to
address questions across the translational

694
01:11:23,699 --> 01:11:29,561
spectrum, especially in respect to the risk
of malnutrition in infants?

695
01:11:29,561 --> 01:11:41,295
[pause]

696
01:11:41,295 --> 01:12:05,429
You might be muted. [pause] I don't think we can hear
you. There seems to be some trouble with your

697
01:12:05,429 --> 01:12:16,040
audio. We'll come back to you. Let's jump
to...along the themes of some of these ideas

698
01:12:16,040 --> 01:12:21,580
for research for the future, there are some
questions related to what technologies out

699
01:12:21,580 --> 01:12:26,719
there might facilitate some of this work.
Dr. Srinivasan, you talked about outcomes.

700
01:12:26,719 --> 01:12:32,949
Traditionally, particularly the long-term
outcomes have relied on follow-up clinics

701
01:12:32,949 --> 01:12:40,460
and therefore difficult to capture in a large
country with referral work and mobility of

702
01:12:40,460 --> 01:12:45,580
populations. Are there any cognitive functional
outcomes or other outcomes in children that

703
01:12:45,580 --> 01:12:49,100
are validated for remote data gathering that
might facilitate getting these at regular intervals?

704
01:12:49,100 --> 01:12:56,000
DR. VIJAY SRINIVASAN: Yeah, that's a great question.
And I think you've really, you know, documented

705
01:12:56,000 --> 01:13:01,409
or identified this really key area that's
a problem because most studies start off strong,

706
01:13:01,409 --> 01:13:04,760
and then follow-up becomes challenging in
the long term, which is actually probably

707
01:13:04,760 --> 01:13:10,360
the most useful information that we all need,
particularly in terms of developmental and

708
01:13:10,360 --> 01:13:15,440
neurocognitive outcomes.
There are definitely tools that are available

709
01:13:15,440 --> 01:13:21,420
at different stages of life. For example,
in infants and toddlers, you could use the

710
01:13:21,420 --> 01:13:26,110
Bayley scales, and then, you know, in older
children as well, you know, you could use

711
01:13:26,110 --> 01:13:32,650
the Wechsler scales. Parent and self-reported
measures such as the ages and stages questionnaire

712
01:13:32,650 --> 01:13:37,030
as well as the Vineland adaptive behavior
scales have been used in clinical trials.

713
01:13:37,030 --> 01:13:43,700
The important key points in this to keep in
mind are one, we need trained data gatherers.

714
01:13:43,700 --> 01:13:49,591
Two, we need to also make sure parents have
access to be able to do this. And again, that

715
01:13:49,591 --> 01:13:54,520
sort of can create a source of bias because
many of these parents that are accessible

716
01:13:54,520 --> 01:14:01,570
to do remote follow-up may also be the ones
that, you know, have preferentially have Internet

717
01:14:01,570 --> 01:14:08,389
and or cell phone connectivity or other sorts
of, you know, features that make them different

718
01:14:08,389 --> 01:14:12,070
from the population of interest that may not
have access to such things. So one has to

719
01:14:12,070 --> 01:14:18,370
be careful in how we sort of attempt to do
follow-up. I will conclude by saying that

720
01:14:18,370 --> 01:14:22,820
another novel method of trying to do follow-up
is the use of remote vehicles.

721
01:14:22,820 --> 01:14:30,170
If one were to bring the lab or one were to
bring imaging or other, you know, the analysts

722
01:14:30,170 --> 01:14:34,730
sort of say to the patient's home, that is
certainly something that is also being explored

723
01:14:34,730 --> 01:14:42,000
in remote communities and poses a great, you
know, sort of opportunity to track more long term data.

724
01:14:42,000 --> 01:14:50,630
DR. NILESH MEHTA: That's a great idea. I know that
some of our participants I know Lori Bechard

725
01:14:50,630 --> 01:14:55,450
in particular, has participated in remote
measurement of body composition and energy

726
01:14:55,450 --> 01:15:00,770
assessments in patients who are technology
dependent. I also appreciate what you said

727
01:15:00,770 --> 01:15:07,929
about being careful when you employ some of
these technologies in terms of excluding a

728
01:15:07,929 --> 01:15:13,699
certain percent of the population that may
not have access to them. That's a very valid

729
01:15:13,699 --> 01:15:21,250
and extremely important piece that the institutional
review boards, the IRBs will definitely question

730
01:15:21,250 --> 01:15:27,720
look at for these research. Thank you for
that. We might have Dr. Nommsen-Rivers' audio

731
01:15:27,720 --> 01:15:32,610
back. And if you do, we'll get back to the
question of are there any examples of good

732
01:15:32,610 --> 01:15:38,790
translational research in malnutrition in
infants.

733
01:15:38,790 --> 01:15:42,445
DR. LAURIE NOMMSEN-RIVERS: Well, yes. Can you hear me now? Does it sound
like I'm back? 

734
01:15:42,445 --> 01:15:43,813
DR. NILESH MEHTA: Yes, you're back. 

735
01:15:43,813 --> 01:15:52,560
DR. LAURIE NOMMSEN-RIVERS: Yay, OK. The answer was to refresh my presentation for some reason.
Well, I think the clever examples are we're,

736
01:15:52,560 --> 01:16:00,870
again, looking at the diet. If we take the
newborn excess weight loss tool, for example,

737
01:16:00,870 --> 01:16:09,110
the Newt all of the inputs on that are infant
focused. It's the type of birth, the sex,

738
01:16:09,110 --> 01:16:15,030
age of the infant, and without...I don't know
if everyone else is hearing rumbling, I'm

739
01:16:15,030 --> 01:16:25,600
hearing some odd audio rumbling. And so I
think where we need to go with future research

740
01:16:25,600 --> 01:16:35,120
in the realm of preventing excess weight loss
in exclusively breastfed newborns is that

741
01:16:35,120 --> 01:16:39,969
we are collecting data on the DYAD. You know,
a lot of times we are mining the electronic

742
01:16:39,969 --> 01:16:46,889
medical record for our research productivity
and how can we create future electronic medical

743
01:16:46,889 --> 01:16:51,930
records that link the mother's record with
the infant so that we can holistically look

744
01:16:51,930 --> 01:16:59,940
at risk factors and protective factors and
not just look at the infant individually.

745
01:16:59,940 --> 01:17:03,159
And, you know, within this translational research
space, I also think it's important that we

746
01:17:03,159 --> 01:17:09,699
look at where was there a concern of insufficient
breastmilk intake.

747
01:17:09,699 --> 01:17:18,600
But these babies aren't showing up as readmissions
because of the mother not able to get any

748
01:17:18,600 --> 01:17:22,350
help with breastfeeding and turning to infant
formula and stopping breastfeeding. So the

749
01:17:22,350 --> 01:17:28,000
ones that end up readmitted really just represent
the tip of the iceberg and they tend to represent

750
01:17:28,000 --> 01:17:37,239
disproportionately white, upper-middle-class
families. And whereas, other families are

751
01:17:37,239 --> 01:17:43,250
much more quick to introduce formula, they
don't have, you know, their context is very

752
01:17:43,250 --> 01:17:50,639
different when it comes to how they're approaching
this. So looking at socioeconomic differences

753
01:17:50,639 --> 01:17:55,510
and how we're approaching insufficient breast
milk intake is going to be important too.

754
01:17:55,510 --> 01:18:02,889
DR. NILESH MEHTA: Thank you, Dr. Nommsen-Rivers. Thank you for
that. You mentioned electronic health records.

755
01:18:02,889 --> 01:18:07,770
I just want to put in a plug for our participants
who are listening. After the break when we

756
01:18:07,770 --> 01:18:13,390
finish this Q&A, we come back for the second
session and it's exciting one. In Session

757
01:18:13,390 --> 01:18:18,210
10, we will have Documenting Nutrition in
EHR, the Electronic Health Records by Robert

758
01:18:18,210 --> 01:18:26,940
Grundmeier and then followed by a pretty
comprehensive on guideline development and

759
01:18:26,940 --> 01:18:31,851
research design by Liam McKeever.
So, do consider joining for that. But let's

760
01:18:31,851 --> 01:18:37,570
get back to the questions. This question from
Dr. Vargas is for all of you, and it addresses

761
01:18:37,570 --> 01:18:44,260
one of the common conundrums for research
related to micronutrients in particular. Blood

762
01:18:44,260 --> 01:18:50,411
volume is a big limitation in evaluating micronutrient
deficiencies, especially in young kids. And

763
01:18:50,411 --> 01:18:56,020
there are limitations that are not necessarily
a limit in adults. Are there any novel developments

764
01:18:56,020 --> 01:19:03,110
in technology that you have seen in your areas
that are promising, where you can limit the

765
01:19:03,110 --> 01:19:13,762
burden of bloodletting in malnutrition risk
assessment in kids?

766
01:19:13,762 --> 01:19:22,760
MS. STEPHANIE MERLINO BARR: I'm happy to kick us off. [crosstalk] So as I mentioned, I'm
a registered dietitian in NICU. So this is

767
01:19:22,760 --> 01:19:27,760
my life. And I wouldn't say that it's necessarily
a novel development, but I'll definitely put

768
01:19:27,760 --> 01:19:35,199
in a plug for my profession where from a nutrition
intervention perspective, having experts in

769
01:19:35,199 --> 01:19:41,350
the field actively participating in multidisciplinary
care is critical. And then in terms of actually

770
01:19:41,350 --> 01:19:47,630
assessing malnutrition.
I mentioned this quite briefly in my presentation

771
01:19:47,630 --> 01:19:51,670
about utilizing skills that registered dietitians
and nutritionists have like the nutrition-focused

772
01:19:51,670 --> 01:19:59,200
physical examination, which specifically we
have training on identifying signs of micronutrient

773
01:19:59,200 --> 01:20:04,190
deficiencies. And this physical examination
are one tool that is certainly less invasive

774
01:20:04,190 --> 01:20:11,390
than a blood draw, which may not be feasible
for my 600 grandbabies upstairs.

775
01:20:11,390 --> 01:20:18,200
DR. NILESH MEHTA: Thank you. Anyone else want to jump in? 

776
01:20:18,200 --> 01:20:32,420
DR. LAURIE NOMMSEN-RIVERS: Well, I'll jump
in from a macroscopic perspective. This is not my area of expertise at all, but I am
dipping my toe into the machine learning space.

777
01:20:32,420 --> 01:20:39,600
And absolutely machine learning needs to be
paired with experts as Stephanie is talking

778
01:20:39,600 --> 01:20:42,889
about. You know, you want to pair with the
nutrition expert so you can understand what's

779
01:20:42,889 --> 01:20:48,080
going on. But I think we have a lot of opportunity
with looking at clusters of features that

780
01:20:48,080 --> 01:20:58,040
may help us to understand in a more macroscopic
way how, you know, what might be some effect

781
01:20:58,040 --> 01:21:06,600
of ratios that we wouldn't think of from just,
you know, plugging away one at a time and looking at that.

782
01:21:06,600 --> 01:21:12,420
DR. NILESH MEHTA: Yeah. I endorse these views as well. I agree.

783
01:21:12,420 --> 01:21:17,510
I think the context is also important, as
you rightly said, you know, what is the setting?

784
01:21:17,510 --> 01:21:23,550
Is it a critically ill child? Is it a child
on parenteral nutrition? Is this, you know,

785
01:21:23,550 --> 01:21:28,290
some other unique disease, unique therapies
that might interfere with metabolism? And

786
01:21:28,290 --> 01:21:36,336
I think if one were to utilize the context
appropriately, I completely agree. I think one can be more targeted and specific in the
approach. 

787
01:21:36,336 --> 01:21:40,860
DR. LEANNE REDMAN: And I will say that the ways in

788
01:21:40,860 --> 01:21:46,220
which we've gotten around some of this in
our research studies is by using dried blood

789
01:21:46,220 --> 01:21:52,660
spots. So just with a simple fingerstick or
heel stick, you can collect dry blood on the

790
01:21:52,660 --> 01:21:58,889
filter paper. And then now there are a large
number of assays that are available using

791
01:21:58,889 --> 01:22:04,350
this technology. So I'm certain that many
of the micronutrients could be evaluated in

792
01:22:04,350 --> 01:22:10,150
this way. So it's brought a low burden assessment
with a lot of possibilities into clinical research.

793
01:22:10,150 --> 01:22:17,450
DR. NILESH MEHTA: That's a great point. Of course,
we haven't even alluded to the fact that micronutrient

794
01:22:17,450 --> 01:22:23,370
levels in blood, especially in my population,
hospitalized, sick children are so unreliable

795
01:22:23,370 --> 01:22:30,870
and the need to use functional measures of
micronutrient deficiency and it becomes even

796
01:22:30,870 --> 01:22:35,430
more relevant in that population.
But thank you. These are great thoughts. There

797
01:22:35,430 --> 01:22:40,560
is a question from Dr. Stevens from Columbia
University for Dr. Redman, and this is

798
01:22:40,560 --> 01:22:45,601
an interesting one related to energy balance.
Dr. Redman, in sick children, are there

799
01:22:45,601 --> 01:22:53,239
prospective randomized data to validate that
feeding based on some measure of assessing

800
01:22:53,239 --> 01:23:00,760
energy requirement like predictive equation
has better outcomes? And can you randomize

801
01:23:00,760 --> 01:23:06,830
other data that children would do better to
randomized to minimizing caloric deficit or to overfeeding for growth? 

802
01:23:06,830 --> 01:23:13,140
DR. LEANNE REDMAN: Yeah, I don't know
about these studies, but I know with the dietitian

803
01:23:13,140 --> 01:23:18,139
in the room that these predictive equations
are used all the time, at least as the starting

804
01:23:18,139 --> 01:23:25,280
point to define feeding plans for critically
ill babies. Then the energy balance part of

805
01:23:25,280 --> 01:23:29,970
it, right? Weight is often tracked in the
NICU over time. And the dietitians use the

806
01:23:29,970 --> 01:23:35,100
weight then to titrate the calories up or
down to get the desired result that they're

807
01:23:35,100 --> 01:23:41,060
after. I don't know the extent to which these
types of data have been published.

808
01:23:41,060 --> 01:23:45,400
And I don't know whether or not these things
have been done in a randomized controlled

809
01:23:45,400 --> 01:23:54,659
fashion, but we certainly should be pulling
the EMR data, which we'll hear in the next segment to be able to look at some of this.
Yeah. 

810
01:23:54,659 --> 01:23:58,090
DR. DEBORAH FRANK: This is Deborah Frank. As a clinician,

811
01:23:58,090 --> 01:24:06,770
there is a real urgency to get kids somehow
repleted for their immune function. And that

812
01:24:06,770 --> 01:24:17,810
versus the long-term risks of too rapid catch-up
growth is a real dilemma, particularly because

813
01:24:17,810 --> 01:24:21,680
these kids go up and down in terms of their
infectious burden and then their weight loss and we start over and so on.

814
01:24:21,680 --> 01:24:31,020
DR. NILESH MEHTA: Great point.
Yes. Well, while we are on the theme of any

815
01:24:31,020 --> 01:24:37,350
future technologies to assist some of this
research, Dr. Redman, on the topic of

816
01:24:37,350 --> 01:24:42,891
energy assessment, particularly the role,
the burden of physical activity, have you

817
01:24:42,891 --> 01:24:50,330
seen any data on accelerometers? I've heard
a lot about that application to accurately

818
01:24:50,330 --> 01:24:58,290
try to gather data on physical activity. But
is the technology now reliable? 

819
01:24:58,290 --> 01:25:03,770
DR. LEANNE REDMAN: This is an excellent question because so many households
now have different wearable devices.

820
01:25:03,770 --> 01:25:09,389
And one thing that we don't want parents to
do is to put this on the wrist of their child

821
01:25:09,389 --> 01:25:14,380
and try to assess for themselves, you know,
the amount of calories that the device is

822
01:25:14,380 --> 01:25:21,360
saying and equate that to a truth. What we
know from accelerometry research, right, these

823
01:25:21,360 --> 01:25:27,840
are research-grade devices. They're very good
at measuring steps, very good at estimating

824
01:25:27,840 --> 01:25:33,659
the number of minutes per day that you are
being physically active, but they do a very

825
01:25:33,659 --> 01:25:40,600
poor job in estimating physical activity,
energy expenditure, or estimating total number

826
01:25:40,600 --> 01:25:45,210
of calories burned throughout the day. So
in our studies, actually, we provide people

827
01:25:45,210 --> 01:25:50,760
a pedometer if we want to be coaching on physical
activity because we only want them to ever

828
01:25:50,760 --> 01:25:59,010
see step information and not to try to give
themself any pleasure in knowing that they

829
01:25:59,010 --> 01:26:03,449
have burned 2,000 calories today. Like if
you go and do a workout class and reward themselves

830
01:26:03,449 --> 01:26:07,489
with something, especially for our people
on like a weight loss thing, but these can

831
01:26:07,489 --> 01:26:11,480
be really detrimental to what you're trying
to do in clinical care.

832
01:26:11,480 --> 01:26:16,560
So to answer the question, they're good for
what they're designed to do, which is just

833
01:26:16,560 --> 01:26:21,040
to track the number of minutes we're being
physically active. They're very good for the

834
01:26:21,040 --> 01:26:25,940
steps. Research grade ones are much better
for us to understand more granular things

835
01:26:25,940 --> 01:26:34,949
about physical activity, but not energy expenditure.
DR. NILESH MEHTA: Thank you. I believe there is an explosion

836
01:26:34,949 --> 01:26:40,090
of these devices, so there is a tendency to
try to use them. So hopefully in the near

837
01:26:40,090 --> 01:26:47,060
future, some good validation studies will
help. The next question is for Stephanie Barr.

838
01:26:47,060 --> 01:26:55,260
You gave a wonderful overview of various methodologies.
This question from Lori Bechard, what body

839
01:26:55,260 --> 01:27:00,710
composition methodology would you prioritize
for validation studies, particularly which

840
01:27:00,710 --> 01:27:05,060
one has the promise to best capture acute
changes in the clinical setting for kids?

841
01:27:05,060 --> 01:27:10,460
MS. STEPHANIE MERLINO BARR: Such a great question, and I'll answer with
the caveat that my perspective is really in

842
01:27:10,460 --> 01:27:15,920
the NICU population. So my answer for that
population is different from what you might

843
01:27:15,920 --> 01:27:21,540
say in an outpatient perspective or even in
an older pediatric population.

844
01:27:21,540 --> 01:27:26,860
In my research, I'm specifically interested
in using air displacement plethysmography

845
01:27:26,860 --> 01:27:31,830
and utilizing that machine in a clinical setting,
which we have been doing here at my hospital,

846
01:27:31,830 --> 01:27:37,350
MetroHealth, which is the county hospital
in Cleveland, Ohio for the past several years,

847
01:27:37,350 --> 01:27:44,230
have been collaborating with the group at
Cincinnati Children's who is also implementing

848
01:27:44,230 --> 01:27:52,719
this device in a clinical mode and looking
for collaborators externally as well. So if

849
01:27:52,719 --> 01:27:57,820
you're out there and you have an ADP machine,
please do reach out. I'd love to know who

850
01:27:57,820 --> 01:28:03,250
you are and what you're doing with it. There
are, of course, pretty big limitations with

851
01:28:03,250 --> 01:28:08,000
the specific device that is on the market
for the NICU population in that you need to

852
01:28:08,000 --> 01:28:14,640
be clinically stable in order to be measured
in it. So, a secondary goal is looking at

853
01:28:14,640 --> 01:28:21,400
what other measurements or modes of assessment
can we use for infants with bronchopulmonary

854
01:28:21,400 --> 01:28:26,119
dysplasia, who are really some of the kiddos
we're most interested in knowing what body

855
01:28:26,119 --> 01:28:31,720
composition development is like.
So using things like we're trying mid-upper

856
01:28:31,720 --> 01:28:35,435
arm circumferences, we have some flank skin fold
measurements that are being done in our unit.

857
01:28:35,435 --> 01:28:40,770
I know the group in Minnesota has looked at
ultrasonography for body composition assessment.

858
01:28:40,770 --> 01:28:45,179
I think all of these are interesting, but
still with very big question marks on how

859
01:28:45,179 --> 01:28:53,530
we can use these and specifically for validation
purposes. 

860
01:28:53,530 --> 01:28:59,080
DR. NILESH MEHTA: Thanks so much. That's very helpful.I know we are coming close to the end of this
session. I could squeeze in one very quick

861
01:28:59,080 --> 01:29:05,659
question for Dr. Srinivasan, and you mentioned
registries epistry based and I know some other

862
01:29:05,659 --> 01:29:11,830
research areas trauma have done well with
this and national registries in certain countries

863
01:29:11,830 --> 01:29:17,000
in Europe have worked amazing for a large
country like U.S. mobile populations larger

864
01:29:17,000 --> 01:29:25,060
capture. Any thoughts on how one would develop
a registry for malnutrition? 

865
01:29:25,060 --> 01:29:30,670
DR. VIJAY SRINIVASAN: What a great question. Thank you. I think it's a tough
question. And I think COVID gave us many examples

866
01:29:30,670 --> 01:29:36,969
of what to do well and what not to do well.
And I think we may be helped by those instances.

867
01:29:36,969 --> 01:29:44,179
I think key challenges with registries, while
they are very useful to sort of have population-based

868
01:29:44,179 --> 01:29:50,320
data, is the disparate populations across
the 50 different states. In some ways, individual

869
01:29:50,320 --> 01:29:54,210
states might be able to take the lead and,
you know, develop their local sort of systems

870
01:29:54,210 --> 01:30:01,650
that works better for their citizens. And
it's tough because one has to contend with

871
01:30:01,650 --> 01:30:06,600
transparency. I think there is always the
fear of patient-level privacy and data concerns.

872
01:30:06,600 --> 01:30:11,449
There is always the fear of how much does
government get involved. On the other hand,

873
01:30:11,449 --> 01:30:17,690
if the data is readily available to the participants
and it can inform their future, that might

874
01:30:17,690 --> 01:30:22,200
actually be appealing. So it's about leveraging
the data that's there and not letting it be

875
01:30:22,200 --> 01:30:26,730
trapped, but actually making it available
perhaps in real-time or near real-time. And

876
01:30:26,730 --> 01:30:33,429
these are some of the challenges that we hopefully
can, you know, I think, overcome or at least

877
01:30:33,429 --> 01:30:38,020
work towards. But to conclude, I would really
like to say, in the moment, I think using

878
01:30:38,020 --> 01:30:43,619
scientific networks through the NIH, Aspen
and other large societies and organizations

879
01:30:43,619 --> 01:30:50,050
may be the way to go to still use the trust
that we have to build these networks.

880
01:30:50,050 --> 01:30:57,070
Well said. Well said. Thank you for that.
Looks like it's one...past 1:30. That brings

881
01:30:57,070 --> 01:31:03,500
us to the end of session nine. My sincere
thanks to all our expert speakers. Thank you

882
01:31:03,500 --> 01:31:08,150
for a very stimulating discussion. Thank you
to all the participants for joining, particularly

883
01:31:08,150 --> 01:31:13,520
all the questions. Keep them coming. At this
stage, we take a 15-minute break. And as I

884
01:31:13,520 --> 01:31:19,680
mentioned earlier, we come back at 1:45, which
is Eastern Standard Time. So 15 minutes from

885
01:31:19,680 --> 01:31:24,820
now for session 10. And session 10 promises
to be equally engaging, exciting. We will

886
01:31:24,820 --> 01:31:32,211
have electronic health records. We'll have
a discussion on guidelines and study design.

887
01:31:32,211 --> 01:31:38,140
And then in the Q&A stage session, we'll focus
on study design considerations. And we'll

888
01:31:38,140 --> 01:31:45,369
be joined in addition to the speakers by our
panelists (INAUDIBLE), Sherry Kirk and Julianne

889
01:31:45,369 --> 01:31:51,330
Patterson. So enjoy the 15-minute break and
I hope to see all of you back for session

890
01:31:51,330 --> 01:31:53,119
ten. Bye-bye.

891
01:31:53,119 --> 01:31:59,600
ASHLEY VARGAS: Hello and welcome everyone
to session ten on Early Life Malnutrition

892
01:31:59,600 --> 01:32:04,659
Moving Towards a Unified Approach. I'm Dr.
Ashley Vargas. I'm a program director at NICHD.

893
01:32:04,659 --> 01:32:10,770
And I'll provide the brief introduction for
this session. I'm joined by Dr. Nilesh Mehta

894
01:32:10,770 --> 01:32:16,290
and Dr. Gail Cresci. Dr. Mehta will serve as
your primary moderator for the end of this

895
01:32:16,290 --> 01:32:21,740
session. I'm really delighted to bring to
you all today some wonderful talks. We start

896
01:32:21,740 --> 01:32:27,310
with Bob Grundmeier of Children's Hospital
of Philadelphia, where he will focus on documentation

897
01:32:27,310 --> 01:32:33,550
of nutrition and the electronic health records.
Followed by Dr. Liam McKeever from Aspen who

898
01:32:33,550 --> 01:32:39,580
will focus on the support needed for better
pediatric malnutrition guidelines. After this,

899
01:32:39,580 --> 01:32:46,080
we'll pause for questions briefly then move
into a panel session. This panel session features

900
01:32:46,080 --> 01:32:52,560
Dr. Wendy Sue Swanson from SpoonfulOne and
Stanford. Dr. Shelley Kirk from Cincinnati

901
01:32:52,560 --> 01:32:57,790
Children's Hospital Medical Center, and Dr.
Julie Ann Patterson from Northern Illinois

902
01:32:57,790 --> 01:33:03,690
University. All three of these individuals
come from slightly different areas not directly

903
01:33:03,690 --> 01:33:08,030
involved in childhood malnutrition research.
But what they do bring to this panel that

904
01:33:08,030 --> 01:33:15,630
we're very excited about is experienced doing
some new novel applied methods for clinical

905
01:33:15,630 --> 01:33:20,850
trials, data collection, registries, etc.
Maybe some new tools that we can use in the

906
01:33:20,850 --> 01:33:25,540
malnutrition space to help move this research
forward. So that panel session will be moderated

907
01:33:25,540 --> 01:33:31,480
by Dr. Mehta. During the session, please remember
to share all your questions for the presenters

908
01:33:31,480 --> 01:33:37,290
and panelists in the chat box. And we'll try
to address as many as we can during the panel

909
01:33:37,290 --> 01:33:43,150
discussion. Without further ado, we'll move
into the presentations. Thank you.

910
01:33:43,150 --> 01:33:53,560
ROBERT W. GRUNDMEIER: Hi, my name is Bob Grundmeier
and I am happy to be here today for this opportunity

911
01:33:53,560 --> 01:33:59,530
to talk about documenting nutrition in the
electronic health record. My focus for this

912
01:33:59,530 --> 01:34:05,050
talk will primarily be on the primary care
pediatrician's perspective on collecting nutrition

913
01:34:05,050 --> 01:34:10,830
information in the electronic health record.
Over the course of my remarks, I will discuss

914
01:34:10,830 --> 01:34:15,320
information that is readily available in the
electronic health record. I will talk about

915
01:34:15,320 --> 01:34:22,770
other information that is available, but might
require some extra effort to be able to extract.

916
01:34:22,770 --> 01:34:26,699
And I will conclude with some thoughts on
how we might improve data collection regarding

917
01:34:26,699 --> 01:34:33,670
nutritional status as part of routine care
in the future. One of the most useful types

918
01:34:33,670 --> 01:34:39,469
of information that is almost universally
available in the electronic health record

919
01:34:39,469 --> 01:34:45,909
is information about child growth. But unfortunately,
as you might imagine, there are some inaccuracies

920
01:34:45,909 --> 01:34:53,480
in measurements that as part of routine care,
and some adjustments must be made to account

921
01:34:53,480 --> 01:34:57,619
for that.
What's shown on this figure is an example

922
01:34:57,619 --> 01:35:07,440
of a sample of about 100 growth charts overlaid
on top of each other, showing how there can

923
01:35:07,440 --> 01:35:12,489
often be implausible measurements. We see
the measurements sort of jumping up and down

924
01:35:12,489 --> 01:35:21,179
unexpectedly and implausibly. This slide shows
the same information after it has been cleaned

925
01:35:21,179 --> 01:35:25,510
using an algorithm that I had the pleasure
to be involved with the development of with

926
01:35:25,510 --> 01:35:32,210
my colleague Dr. Daymont. The citation is shown
there for your reference. And in this slide

927
01:35:32,210 --> 01:35:38,610
we see that there is implausible measurements
that were in the routinely collected data

928
01:35:38,610 --> 01:35:45,830
have been eliminated. Flipping back and forth
between the screens, we can see how nicely

929
01:35:45,830 --> 01:35:50,270
the algorithm gets rid of the implausible
measurements while retaining those measurements

930
01:35:50,270 --> 01:35:56,290
that are likely reflective of the child's
growth. Laboratory data is another kind of

931
01:35:56,290 --> 01:36:01,620
information that is easy to extract from the
electronic health record. And in some cases,

932
01:36:01,620 --> 01:36:06,980
we have useful lab data that is collected
on virtually all children depending on their

933
01:36:06,980 --> 01:36:09,960
age.
For example, there are recommendations to

934
01:36:09,960 --> 01:36:17,389
collect hemoglobin or complete blood counts
at particular ages, such as in infancy and

935
01:36:17,389 --> 01:36:23,600
in the second year of life, and typically
an additional check for anemia in adolescence,

936
01:36:23,600 --> 01:36:31,230
especially amongst girls of menstruating age,
there's also guidance to check cholesterol

937
01:36:31,230 --> 01:36:38,350
at particular ages. And this is also routinely
captured typically in the tween years as well

938
01:36:38,350 --> 01:36:44,290
as shortly before transition to adulthood.
Some other labs are periodically collected

939
01:36:44,290 --> 01:36:50,489
in particular situations that might be useful
from a nutrition standpoint. For example,

940
01:36:50,489 --> 01:36:56,530
vitamin D, glucose, and liver enzymes are
routinely collected for children who may have

941
01:36:56,530 --> 01:37:02,090
overweight status. And even less frequently,
but also useful, there are labs that may be

942
01:37:02,090 --> 01:37:07,130
collected specifically among children with
poor growth. This can include comprehensive

943
01:37:07,130 --> 01:37:12,920
metabolic panels, which typically includes
nutritional measures such as albumin. There

944
01:37:12,920 --> 01:37:20,580
are inflammatory measures that we often collect
such as C-reactive protein or erythrocyte

945
01:37:20,580 --> 01:37:24,930
sedimentation rate as well as occasionally
thyroid status.

946
01:37:24,930 --> 01:37:32,869
But these are not collected routinely, only
in particular cases where there are concerns

947
01:37:32,869 --> 01:37:39,639
about poor growth. Now, I would like to turn
our attention to some types of information

948
01:37:39,639 --> 01:37:44,200
that may be particularly useful, but are a
little more challenging to use because the

949
01:37:44,200 --> 01:37:49,989
information is often in free text. Pediatricians
collect information about child nutrition

950
01:37:49,989 --> 01:37:55,699
status at basically all preventive health
care visits, but they do so in a wide variety

951
01:37:55,699 --> 01:38:03,020
of different ways. I searched through the
templates that were available in an electronic

952
01:38:03,020 --> 01:38:07,739
health record vendor product that is shared
across many health systems and identified

953
01:38:07,739 --> 01:38:15,510
that there were 307 different structured templates,
meaning sort of checkbox based templates that

954
01:38:15,510 --> 01:38:21,619
are used across 19 different pediatric health
systems. Looking within my own health system

955
01:38:21,619 --> 01:38:26,840
at Children's Hospital of Philadelphia, I
identified that we have 25 different structured

956
01:38:26,840 --> 01:38:33,179
templates which allow you to pick responses
from lists, for example. And in addition,

957
01:38:33,179 --> 01:38:38,900
there are 617 at least free text documentation
templates.

958
01:38:38,900 --> 01:38:43,320
This does not include the nutrition assessments
that might be embedded in the context of other

959
01:38:43,320 --> 01:38:49,000
visits such as preventive health care or other
visit templates. Here is an example of one

960
01:38:49,000 --> 01:38:55,950
such template at CHOP. This one is related
to breastfeeding and generally allows clinicians

961
01:38:55,950 --> 01:39:02,619
to type in whatever they want and allows flexibility.
In this particular screen, the three asterisks

962
01:39:02,619 --> 01:39:08,950
indicate a field where the clinician can really
type in whatever they want and they can also

963
01:39:08,950 --> 01:39:12,890
delete whatever they want. So you may not
get complete information from a template such

964
01:39:12,890 --> 01:39:18,020
as this, but it does remind the clinician
about particular types of information that

965
01:39:18,020 --> 01:39:23,020
you might want to collect regarding an infant
who is breastfeeding. And with some effort,

966
01:39:23,020 --> 01:39:26,489
one can actually get this information back
out of the electronic health record, even

967
01:39:26,489 --> 01:39:32,920
though it is actually stored as text. Here's
another example of a template that is used

968
01:39:32,920 --> 01:39:36,960
as part of a healthy weight assessment that
has a little bit more structure to it and

969
01:39:36,960 --> 01:39:42,119
also a little bit more detail.
In this case, the clinicians still have flexibility

970
01:39:42,119 --> 01:39:46,580
to ignore certain prompts. They don't have
to respond to all of these, but it serves

971
01:39:46,580 --> 01:39:52,850
the double purpose of reminding clinicians
about the topics that are important to discuss,

972
01:39:52,850 --> 01:40:00,000
as well as providing some structured response
options as shown in the yes/no response options

973
01:40:00,000 --> 01:40:05,160
in the highlighted in yellow. So it gives
you a lot of sort of yes/no indicators for

974
01:40:05,160 --> 01:40:14,599
different categories of information and can
be very useful and easy to extract information

975
01:40:14,599 --> 01:40:22,179
from. This slide shows examples of documentation
from three different providers all who cared

976
01:40:22,179 --> 01:40:28,510
for the same patient at different points in
time. The first example, which happens to

977
01:40:28,510 --> 01:40:34,739
be a note that I wrote, was at the newborn
visit where I attempted to the best of my

978
01:40:34,739 --> 01:40:44,260
ability to quantify the formula intake for
this particular child. The next note was somewhat

979
01:40:44,260 --> 01:40:50,850
later at a follow-up visit by one of my colleagues.
The final note is actually from a specialist

980
01:40:50,850 --> 01:40:56,780
who we had consulted regarding this child's
poor growth status, who again attempted to

981
01:40:56,780 --> 01:41:00,139
quantify information.
As you can see, we're all generally collecting

982
01:41:00,139 --> 01:41:06,639
the same kinds of information, including the
caloric strength of the formula that's being

983
01:41:06,639 --> 01:41:12,770
consumed in the case of the specialist. Although
it may not have the degree of precision that

984
01:41:12,770 --> 01:41:19,920
might be desired for research purposes. Here
is an example of some narrative documentation

985
01:41:19,920 --> 01:41:27,300
related to the nutritional status of an adolescent
patient. As you can see in this note, the

986
01:41:27,300 --> 01:41:33,690
clinician captured a 24-hour diet recall with
some attempts at quantifying specifics about

987
01:41:33,690 --> 01:41:39,210
the oral intake, such as the number of pieces
of pizza and the number of chicken wings that

988
01:41:39,210 --> 01:41:45,640
were consumed. At the bottom of the screen
as the assessment and plan for this same child.

989
01:41:45,640 --> 01:41:52,360
As you can see, the clinician has recorded
an assessment of the malnutrition status with

990
01:41:52,360 --> 01:41:58,070
some quantified information, as well as some
thoughts about the potential causes in the

991
01:41:58,070 --> 01:42:05,530
context of this child's medical complexity.
On this screen, we see an example of self-reported

992
01:42:05,530 --> 01:42:10,320
nutrition status that is collected as part
of a pre visit questionnaire.

993
01:42:10,320 --> 01:42:16,219
I think this is a promising mechanism for
improving the collection of quantifiable data

994
01:42:16,219 --> 01:42:20,739
related to nutrition status going forward.
This is currently in a pilot project phase

995
01:42:20,739 --> 01:42:26,929
in our organization, but has been found to
be very acceptable to our patient families

996
01:42:26,929 --> 01:42:31,739
and is also very much appreciated by the clinicians
who feel that these types of information are

997
01:42:31,739 --> 01:42:38,920
useful to collect. Finally, I would like to
share some concluding thoughts. I think there

998
01:42:38,920 --> 01:42:44,620
are definitely many opportunities for improving
the current status quo of how nutritional

999
01:42:44,620 --> 01:42:50,510
information is collected as part of routine
care, which could be very helpful for observational

1000
01:42:50,510 --> 01:42:58,360
research studies related to nutritional status.
And clinical teams really are willing to collect

1001
01:42:58,360 --> 01:43:05,369
information about nutrition status, but they
need to know how to do so, for example, by

1002
01:43:05,369 --> 01:43:10,070
providing them with templates or guidance
on what questions to ask, which need to be

1003
01:43:10,070 --> 01:43:16,670
incorporated into clinical workflows. And
it's also necessary that clinicians feel that

1004
01:43:16,670 --> 01:43:19,139
the information is being collected is actually
important.

1005
01:43:19,139 --> 01:43:25,510
And from the clinician perspective, that means
how might this information be used to improve

1006
01:43:25,510 --> 01:43:32,961
health outcomes in some fashion? I hope this
whirlwind tour of a few aspects of how nutrition

1007
01:43:32,961 --> 01:43:37,840
is assessed in routine primary care has been
somewhat thought provoking, and I look forward

1008
01:43:37,840 --> 01:43:45,400
to our discussion on how we might improve
the documentation of nutritional status in

1009
01:43:45,400 --> 01:43:50,100
the future.

1010
01:43:50,100 --> 01:43:53,250
LEAH MCKEEVER: Hi, I'm Leah McKeever. And
I am the methodologist and director for the

1011
01:43:53,250 --> 01:43:57,389
ASPEN Clinical Guidelines. I'm gonna talk
to you today about what I think it would take

1012
01:43:57,389 --> 01:44:03,010
to create a good pediatric malnutrition guideline,
though everything I'm about to say applies

1013
01:44:03,010 --> 01:44:08,630
equally to adult malnutrition, and I have
no conflicts of interest to disclose. Now,

1014
01:44:08,630 --> 01:44:13,150
to help you understand the issues in this
field, I need to first expose some very pervasive

1015
01:44:13,150 --> 01:44:18,980
myths in our field. People seem to think that
all study designs can answer all questions

1016
01:44:18,980 --> 01:44:23,410
with the exception that some study designs
are more rigorous than others. The truth is

1017
01:44:23,410 --> 01:44:28,570
actually far more complex than that. The reality
is that all study designs have assumptions

1018
01:44:28,570 --> 01:44:34,090
that must be met in order for them to be informative
at all. And many of the research questions

1019
01:44:34,090 --> 01:44:38,940
in nutrition violate these assumptions in
a way that renders these study estimates either

1020
01:44:38,940 --> 01:44:44,730
completely invalid or misinformative. For
this reason, I'm very careful how I set up

1021
01:44:44,730 --> 01:44:50,000
study design inclusion criteria on a guideline,
we usually start by looking for randomized

1022
01:44:50,000 --> 01:44:53,949
control trials, because these studies will
provide unconfounded estimates.

1023
01:44:53,949 --> 01:45:00,150
When RCTs are not available, we will look
for quasi-experimental designs. Now, we may

1024
01:45:00,150 --> 01:45:06,580
drop down into observational prospective cohorts,
but only if there are no known unmitigated

1025
01:45:06,580 --> 01:45:11,580
confounders for a research question. And we're
gonna describe that in a second 'cause, you

1026
01:45:11,580 --> 01:45:18,639
know, usually there are unmitigated confounders.
So this system right here limits us to only

1027
01:45:18,639 --> 01:45:23,410
using studies where the data was collected
for research purposes, and that utilize a

1028
01:45:23,410 --> 01:45:29,179
study design that can actually answer the
question at hand. So what is the problem here?

1029
01:45:29,179 --> 01:45:34,260
Why is confounding such an issue? What's at
risk? Well, in all studies, we're looking

1030
01:45:34,260 --> 01:45:38,139
at the relationship between two variables.
In this case, let's look at the relationship

1031
01:45:38,139 --> 01:45:44,199
between nutrition and clinical outcome. We
know that there are people who are more severely

1032
01:45:44,199 --> 01:45:50,991
ill and that more severely ill people will
have worse clinical outcomes. If that disease

1033
01:45:50,991 --> 01:45:57,219
acuity is also associated with nutrition exposure,
then an observational estimate is not going

1034
01:45:57,219 --> 01:46:00,469
to be valid.
In this case, the estimate that you get for

1035
01:46:00,469 --> 01:46:05,989
the nutrition outcome relationship will be
confounded. Another word for that is actually

1036
01:46:05,989 --> 01:46:12,369
blended with the relationship between disease
acuity and outcome, and we won't be able to

1037
01:46:12,369 --> 01:46:18,310
separate them, OK? So the relationship between
nutrition and outcome will be blended in the

1038
01:46:18,310 --> 01:46:23,119
estimate with the relationship between disease
acuity and outcome, and therefore you will

1039
01:46:23,119 --> 01:46:29,199
have an invalid estimate. That resulting estimate
is not just an issue of it lacking in rigor,

1040
01:46:29,199 --> 01:46:35,850
it is literally misinformative. This is the
problem for us because disease acuity is inseparable

1041
01:46:35,850 --> 01:46:43,900
from malnutrition status. There's a reason
our patients are malnourished. An RCT or quasi-experimental

1042
01:46:43,900 --> 01:46:48,730
study bypasses this issue. On the left, you
see an observational study and you can see

1043
01:46:48,730 --> 01:46:54,430
that almost every aspect of a patient's care
relates to their disease acuity, including

1044
01:46:54,430 --> 01:46:59,030
nutrition. So there's tons of confounding
here. In an RCT, which is what we see on the

1045
01:46:59,030 --> 01:47:04,429
right, all those variables still connect with
outcome, but none of them connect with nutrition

1046
01:47:04,429 --> 01:47:08,380
exposure because nutrition exposure is randomly
allocated.

1047
01:47:08,380 --> 01:47:13,739
So we now have a clear signal accounting for
the relationship of interest. I'm sure there

1048
01:47:13,739 --> 01:47:19,409
are many that will say, yes, but we adjust
for illness severity through things like,

1049
01:47:19,409 --> 01:47:24,409
you know, for adults we use Apache II Score,
we have illness severity indices. The problem

1050
01:47:24,409 --> 01:47:29,760
is that illness severity is not a hard endpoint
that you can cleanly capture. These indices

1051
01:47:29,760 --> 01:47:34,989
are a constellation of surrogate exposures
that just predict mortality. Compare that

1052
01:47:34,989 --> 01:47:40,719
to adjusting for age or sex, right? Two hard
endpoints that can completely mitigate the

1053
01:47:40,719 --> 01:47:46,920
confounding through adjustment. With illness
severity indices, you may dampen some of

1054
01:47:46,920 --> 01:47:52,090
the confounding but it's still going to leave
plenty of residual confounding to directionally

1055
01:47:52,090 --> 01:47:54,070
push on that estimate.

1056
01:47:54,070 --> 01:47:58,860
DR. LIAM MCKEEVER: In these cases, what this means is
that the observational research simply cannot

1057
01:47:58,860 --> 01:48:04,590
answer the question. It's estimates are not
just uninformative, they are actually misinformative,

1058
01:48:04,590 --> 01:48:11,510
if there is any known unmitigated confounding.
So what can we do? Well, I think your field

1059
01:48:11,510 --> 01:48:16,699
has two overarching questions that need to
be answered. One concerns construct validity

1060
01:48:16,699 --> 01:48:22,159
of your tools. We need to know that your tools
are measuring what we say they're measuring.

1061
01:48:22,159 --> 01:48:28,800
Second, we need to know whether intervening
on the results of those measurements leads

1062
01:48:28,800 --> 01:48:34,980
to better clinical outcomes. So let's tackle
the first one. So we need to make sure that

1063
01:48:34,980 --> 01:48:40,679
our tools are properly validated. This can
even be done retrospectively against a gold

1064
01:48:40,679 --> 01:48:45,659
standard measure of body composition. So the
current gold standard is Radiographic Imaging.

1065
01:48:45,659 --> 01:48:50,040
And at this point someone usually says, but
if you do that, you're gonna be limiting your

1066
01:48:50,040 --> 01:48:55,630
population to cancer patients because they're
the only ones with Radiographic Imaging. I

1067
01:48:55,630 --> 01:48:59,290
don't think that's actually true.
We can do an abdominal pelvic CT scan for

1068
01:48:59,290 --> 01:49:04,580
any undiagnosed abdominal pain, but let's
just go ahead and say that's true. Fine, do

1069
01:49:04,580 --> 01:49:10,750
that. Show us that your tool is properly validated
in cancer patients. If it validates well,

1070
01:49:10,750 --> 01:49:15,360
and it also validates well with other less
invasive tools, then the use of those less

1071
01:49:15,360 --> 01:49:20,730
invasive tools in other populations will become
more meaningful because we've seen the tool

1072
01:49:20,730 --> 01:49:27,090
properly validated against a gold standard
in cancer patients. Another common comment

1073
01:49:27,090 --> 01:49:32,869
that I get at this point is to say that malnutrition
is more than just body composition. And I

1074
01:49:32,869 --> 01:49:37,900
think usually they're talking about the fact
that we're looking at risk. My question for

1075
01:49:37,900 --> 01:49:42,760
you is, is it really? I mean, you're a container
of carbohydrates, proteins, fats, vitamins

1076
01:49:42,760 --> 01:49:47,369
and minerals. From a research perspective,
the most stable of these is protein status.

1077
01:49:47,369 --> 01:49:52,130
So that's our gold standard for body comp.
And if you're malnourished, that should be

1078
01:49:52,130 --> 01:49:58,190
detectable as a loss in protein status. If
you're instead at risk for future malnutrition,

1079
01:49:58,190 --> 01:50:02,409
that should be detectable in the future as
a loss in protein status.

1080
01:50:02,409 --> 01:50:09,130
So we do need to properly validate our tools
wherever we're able. We then need unconfounded

1081
01:50:09,130 --> 01:50:15,719
study designs to tell us that intervening
upon malnutrition findings alters outcomes.

1082
01:50:15,719 --> 01:50:21,150
In many cases, this can be done through a
randomized controlled trial, and the common

1083
01:50:21,150 --> 01:50:25,460
objection at this point is, yeah, but that's
not ethical. We cannot randomize someone to

1084
01:50:25,460 --> 01:50:31,179
become malnourished. Well, you don't have
to. Instead, what you'll do is you'll randomize

1085
01:50:31,179 --> 01:50:35,739
to nutrition exposure and then you can look
at how the nutrition outcome relationship

1086
01:50:35,739 --> 01:50:42,390
differs between malnourished and normal nourished
patients. In epidemiology, we call that effect

1087
01:50:42,390 --> 01:50:47,750
modification, and in statistics, it shows
up as an interaction term in your model. In

1088
01:50:47,750 --> 01:50:53,810
populations where standard care fall short,
we need RCTs. So if standard care is falling

1089
01:50:53,810 --> 01:51:00,239
short of estimated needs, then we need RCTs
that compare an intensive nutrition model

1090
01:51:00,239 --> 01:51:06,239
to standard care. This can come about through
enteral feeding algorithms or supplemental

1091
01:51:06,239 --> 01:51:13,119
PN. No one is getting randomized to malnourishment
here, so the IRB should be happy.

1092
01:51:13,119 --> 01:51:18,429
In your study, you will have both malnourished
and normal nourished patients and you're then

1093
01:51:18,429 --> 01:51:25,969
going to see if malnutrition status modifies
the effect of nutrition on clinical outcome.

1094
01:51:25,969 --> 01:51:31,390
Again, this is a simple interaction term in
your model that allows you to see if malnourished

1095
01:51:31,390 --> 01:51:39,060
patients respond differently to nutrition
than normal nourished patients. Now, in some

1096
01:51:39,060 --> 01:51:44,139
populations, we may have put the cart before
the horse unfortunately, and started intervening

1097
01:51:44,139 --> 01:51:50,010
with intensive nutrition without the necessary
supporting data. In this case, you may not

1098
01:51:50,010 --> 01:51:55,270
be able to ethically withhold nutrition. So
in this case we'll have to use quasi-experimental

1099
01:51:55,270 --> 01:52:00,130
study designs. So what you'll do is you'll
go back in the medical records to the moment

1100
01:52:00,130 --> 01:52:06,320
where you began intervening, and you will
compare mortality data from before that intervention

1101
01:52:06,320 --> 01:52:12,449
to mortality data after you began intervening.
Now you'll have to account for patient differences

1102
01:52:12,449 --> 01:52:19,139
between those two time points. So you're gonna
obtain a robust collection of ancillary variables.

1103
01:52:19,139 --> 01:52:23,151
This may be a good place to use propensity
score adjustment, but this would be a way

1104
01:52:23,151 --> 01:52:30,679
to address this problem if intervention is
not an option. So this is what it's gonna

1105
01:52:30,679 --> 01:52:37,020
take to demonstrate efficacy. Most PICO questions
on a clinical guideline would likely not be

1106
01:52:37,020 --> 01:52:42,150
able to include the vast field of available
observational research that we currently have

1107
01:52:42,150 --> 01:52:47,870
because again, its measurement is flawed and
its estimates are gonna be misleading. This

1108
01:52:47,870 --> 01:52:53,650
is not better than expert opinion. I know
we think that, but it isn't. Observational

1109
01:52:53,650 --> 01:52:58,420
research, if it's confounded, is not better
than nothing. It is actually misleading. So

1110
01:52:58,420 --> 01:53:03,949
it's worse than nothing. And so in this case,
we would forego that and instead do expert

1111
01:53:03,949 --> 01:53:09,489
opinion. This is what we would need for most
PICO questions on a malnutrition guideline

1112
01:53:09,489 --> 01:53:14,480
until we get the proper research. Now this
is not a major problem, we can do a guideline

1113
01:53:14,480 --> 01:53:19,790
that's all expert opinion, but we can only
do so many guidelines at once. And an expert

1114
01:53:19,790 --> 01:53:24,730
opinion guideline is not always a major priority
compared to guidelines that have actual research

1115
01:53:24,730 --> 01:53:28,719
behind them.
So until there is proper research in our field

1116
01:53:28,719 --> 01:53:34,930
of malnutrition, we may be better served by
position papers. But I very much look forward

1117
01:53:34,930 --> 01:53:51,469
to discussing this further in our panels,
and I thank you for listening. Thank you for

1118
01:53:51,469 --> 01:53:59,710
these fantastic, very stimulating discussion
presentations. I'm excited to welcome you

1119
01:53:59,710 --> 01:54:08,489
all back to the Q&A session, and I'm also
excited to welcome our panelists. If participants

1120
01:54:08,489 --> 01:54:15,929
want to still submit questions, please feel
free to do so. You'll be able to use the box

1121
01:54:15,929 --> 01:54:20,640
and I will try to get to them even during
the panel discussion. So thank you again for

1122
01:54:20,640 --> 01:54:28,500
all those who have put in questions. Let's
get started. We have questions. Let's start

1123
01:54:28,500 --> 01:54:35,480
with Dr. Grundmeier. What a fantastic overview
of a topic which is both exciting but also

1124
01:54:35,480 --> 01:54:43,760
pain point for many people on the front lines.
You've outlined... This question is for Dr.

1125
01:54:43,760 --> 01:54:47,780
Grundmeier. You outline many great features
within the EMR that can be used to explore

1126
01:54:47,780 --> 01:54:53,369
nutritional status. Have you had experience
with using this information to create alerts

1127
01:54:53,369 --> 01:54:58,070
for patients who might be at risk for malnutrition
or to trigger a nutrition intervention or

1128
01:54:58,070 --> 01:54:59,070
evaluation?

1129
01:54:59,070 --> 01:55:04,969
DR. GRUNDMEIER: Yeah. Thank you. That's actually
a great question. Hopefully I'm audible or

1130
01:55:04,969 --> 01:55:13,150
somebody will let me know if I'm not. So the
short answer to that is no, I have not yet

1131
01:55:13,150 --> 01:55:18,270
been involved in a project to do any kind
of alerting related to nutritional status

1132
01:55:18,270 --> 01:55:25,739
beyond what you would consider to be the creation
of passive content where you work through

1133
01:55:25,739 --> 01:55:31,230
a set of questions. And then there's maybe
some reminders in text that a clinician can

1134
01:55:31,230 --> 01:55:37,099
interpret it to decide, for example, whether
a referral to our nutritional experts or to

1135
01:55:37,099 --> 01:55:42,580
our healthy weight program would be indicated
based on the information. But it's not an

1136
01:55:42,580 --> 01:55:48,190
automated assessment. It's up to the doctor
to think about it. I'm actually now working

1137
01:55:48,190 --> 01:55:54,090
with a collaborator who's just starting a
project who's very interested in nutritional

1138
01:55:54,090 --> 01:56:04,410
status as evidence of child maltreatment,
as a specific sub domain. And as part of that,

1139
01:56:04,410 --> 01:56:11,070
we are actually going to be making extensive
use of the growth patterns. And I anticipate

1140
01:56:11,070 --> 01:56:20,050
that those rules we might craft to provide
some automated alerting will be useful for

1141
01:56:20,050 --> 01:56:23,660
nutritional assessment in general.
I think it's a tremendous area of opportunity,

1142
01:56:23,660 --> 01:56:28,469
especially when we're asking people to enter
a lot of information. You really want to give

1143
01:56:28,469 --> 01:56:33,010
them some value back and that is, I think,
been a missed opportunity thus far.

1144
01:56:33,010 --> 01:56:38,639
DR. NILESH MEHTA: Thank you. That's very exciting.
We have a lot more questions related to EHR.

1145
01:56:38,639 --> 01:56:45,120
Let me jump to Dr. McKeever. This is a question
from Andrea Garber from University of California,

1146
01:56:45,120 --> 01:56:50,070
San Francisco. Excellent talk. What are your
thoughts on balancing the time and cost of

1147
01:56:50,070 --> 01:56:55,800
an RCT with the allure of doing retrospective
studies using data that can be drawn easily

1148
01:56:55,800 --> 01:57:04,389
out of the EHR? Liam, you're muted. I can't
hear you.

1149
01:57:04,389 --> 01:57:14,520
DR. MCKEEVER: Sorry. That was a great question.
RCTs, yes, they're very difficult. I run them

1150
01:57:14,520 --> 01:57:19,760
myself. They make your whole life a nightmare
till they're done. But at the end of the day,

1151
01:57:19,760 --> 01:57:23,730
it just comes down to you. If you want an
answer to your question, you have to ask the

1152
01:57:23,730 --> 01:57:28,760
question in a way that it can be answered,
right? You need a study design that can answer

1153
01:57:28,760 --> 01:57:36,520
it. So for an electronic health record, there
are many questions that could probably be

1154
01:57:36,520 --> 01:57:42,130
answered in an observational way, and it would
just be a matter of maybe there are confounders

1155
01:57:42,130 --> 01:57:48,849
we aren't considering. I'm not saying that
observational is bad, OK? I'm saying that

1156
01:57:48,849 --> 01:57:55,869
if you know there's a known confounder, if
you know it and you know that you can't fully

1157
01:57:55,869 --> 01:58:00,810
mitigate that confounding, and because you're
a scientist, you are starting on the assumption

1158
01:58:00,810 --> 01:58:09,489
of the null hypothesis being true. Then you
cannot take that estimate and draw causal

1159
01:58:09,489 --> 01:58:13,909
inference from it. You just can't. No one
wants to hear it, I get it. I never hear anyone

1160
01:58:13,909 --> 01:58:23,000
else say it. But it's a mathematical reality.
So this is where you would have to do either

1161
01:58:23,000 --> 01:58:28,530
a randomized controlled trial or, again, you
could do a quasi-experimental design. And

1162
01:58:28,530 --> 01:58:34,119
there are sometimes ways at going back into
retrospective studies and making the quasi-experimental

1163
01:58:34,119 --> 01:58:42,199
study out of that. The whole idea is you want
the exposure to be not a product of any aspect

1164
01:58:42,199 --> 01:58:47,579
of the person's care. So some people do that
by saying, OK, well, we know that this physician

1165
01:58:47,579 --> 01:58:53,719
always tends to dose more and this physician
tends to dose less. So we'll compare his patients

1166
01:58:53,719 --> 01:59:00,119
and that is a way to kind of remove some of
that confounding influence. You have to get

1167
01:59:00,119 --> 01:59:05,119
creative when you go quasi-experimental, and
it's not as good as in RCT, but it does get

1168
01:59:05,119 --> 01:59:10,260
around that fundamental issue that says it
doesn't really matter what your finding is,

1169
01:59:10,260 --> 01:59:12,520
because you're finding it's just measuring
illness severity.

1170
01:59:12,520 --> 01:59:18,610
DR. NILESH MEHTA: Well said. Yeah, we'll have more
discussion around this when our panelists

1171
01:59:18,610 --> 01:59:26,130
join us in terms of study design. But well
said. Dr. Grundmeier, question from Vijay

1172
01:59:26,130 --> 01:59:31,550
Srinivasan from University of Pennsylvania,
not too far from you. Can we leverage informatics

1173
01:59:31,550 --> 01:59:36,489
to develop pipelines to automatically extract
nutrition data from EMR and to provide information

1174
01:59:36,489 --> 01:59:42,270
on adequacy, especially around interventions
in real time?

1175
01:59:42,270 --> 01:59:51,260
DR. GRUNDMEIER: Well, yes, we can with sufficient
resources build, really improve on the current

1176
01:59:51,260 --> 01:59:56,659
state. I think one of the barriers right now,
which I alluded to in my remarks, is so much

1177
01:59:56,659 --> 02:00:04,650
of the richness of what at least in my settings
we're trying to capture currently end up in

1178
02:00:04,650 --> 02:00:09,400
free text documentation, which is always a
little challenging to build pipelines around.

1179
02:00:09,400 --> 02:00:14,380
But of course it is possible. It just takes
some work. And there's, I think, an unanswered

1180
02:00:14,380 --> 02:00:20,710
question of, is the work better spent trying
to build the pipeline to deal with the data

1181
02:00:20,710 --> 02:00:26,699
you have versus redesigning the way you're
collecting the data in a way that is relevant

1182
02:00:26,699 --> 02:00:32,389
for clinical practice, gets the research data
quality data you want. And one of the things

1183
02:00:32,389 --> 02:00:39,940
I think is very promising is patients are
really willing to collect data for us potentially

1184
02:00:39,940 --> 02:00:45,590
in ways that are more valid than one pediatrician
versus another, asking the questions that

1185
02:00:45,590 --> 02:00:53,849
they learned to ask 20 years ago and probably
don't ask the same every time. So certainly

1186
02:00:53,849 --> 02:00:58,420
good opportunity to build pipelines.
Real time is always a little bit of a challenge.

1187
02:00:58,420 --> 02:01:04,489
I mean, soon enough is probably quite possible
in terms of, like for example, involved in

1188
02:01:04,489 --> 02:01:12,170
registries where we put, well, nutrition information
is in it, but not to a high degree of granularity.

1189
02:01:12,170 --> 02:01:17,490
But it's pulled out every month and cleaned
and there's a review process to make sure

1190
02:01:17,490 --> 02:01:23,889
it's OK. And that's often good enough for
research purposes if you can get back the

1191
02:01:23,889 --> 02:01:29,190
past the issues of bias and confounding of
course. But real time is a little more challenging.

1192
02:01:29,190 --> 02:01:35,370
If you want to do an alert based on something
that was just entered in the chart, that is

1193
02:01:35,370 --> 02:01:41,320
also possible, but that is a higher barrier
and people do that. But it is much more costly

1194
02:01:41,320 --> 02:01:45,630
to do that level of real time. I hope that
was answering the question. I might have used

1195
02:01:45,630 --> 02:01:49,030
that to talk about what I wanted to talk about
rather than your question.

1196
02:01:49,030 --> 02:01:54,530
DR. NILESH MEHTA: No, that's great. No, thank you for
that. This is great. I think there are more

1197
02:01:54,530 --> 02:01:59,090
questions coming in, but this would be a great
time to welcome our panelists, Wendy Sue Swanson,

1198
02:01:59,090 --> 02:02:04,900
Cernich, Julie Patterson. And well, you guys
are coming on board. A very quick question.

1199
02:02:04,900 --> 02:02:10,770
Back to Dr. McKeever. This is from David Suarez.
Is validation against a gold standard? Validation

1200
02:02:10,770 --> 02:02:15,070
of the gold standard doesn't actually measure
the outcome in question. For example, nutrition

1201
02:02:15,070 --> 02:02:18,400
screening tools predict mortality but not
response to nourishment.

1202
02:02:18,400 --> 02:02:25,340
DR MCKEEVER: So that's a problem in my eyes.
So if this assumption, we make an assumption

1203
02:02:25,340 --> 02:02:33,170
when we use a tool like that. And that assumption
is that intervening with nutrition will...

1204
02:02:33,170 --> 02:02:38,010
Intervening upon the finding of that screening
with nutrition will alter outcomes in some

1205
02:02:38,010 --> 02:02:43,140
way. I've seen that with the NUTRIC Score,
for example. There's all these papers out

1206
02:02:43,140 --> 02:02:52,440
there that show that NUTRIC Score is related
to outcome. And from that they say, wow, so

1207
02:02:52,440 --> 02:02:58,560
we really need to provide better nutrition
based on this NUTRIC Score. But again, if

1208
02:02:58,560 --> 02:03:04,150
the NUTRIC Score is only measuring illness
severity, of course it's related to outcome.

1209
02:03:04,150 --> 02:03:09,639
There's not a single nutritional parameter
on the score. So to properly look at the NUTRIC

1210
02:03:09,639 --> 02:03:15,570
Score, you would have to have a randomized
controlled trial that looked at whether intervening

1211
02:03:15,570 --> 02:03:21,800
upon that score with nutrition impacted clinical
outcomes. And without that, I don't see its

1212
02:03:21,800 --> 02:03:22,800
utility.

1213
02:03:22,800 --> 02:03:31,480
DR. NILESH MEHTA: Thanks, Liam. With our panelists
on board, this is a question for all three

1214
02:03:31,480 --> 02:03:38,700
of you. And we can go around the panel to
ask. In this setting of malnutrition gaps

1215
02:03:38,700 --> 02:03:45,650
and research identification which is our task,
can you discuss some of your study designs,

1216
02:03:45,650 --> 02:03:49,250
particularly those that could be applied to
explore pediatric malnutrition from the lens

1217
02:03:49,250 --> 02:03:55,880
of prevention, identification or diagnosis.
Essentially something in your repertoire or

1218
02:03:55,880 --> 02:04:02,610
past or current experience that we might learn
from? We can go around the panel for this

1219
02:04:02,610 --> 02:04:12,889
one. Just jump in. Thank you so much for that
question. Oh. That's OK. Go ahead. So my research

1220
02:04:12,889 --> 02:04:17,940
actually emerged as a result of the Joint
Commission identifying a standardised definition

1221
02:04:17,940 --> 02:04:22,489
of exclusive breastfeeding. And I think what
we've heard in this conference is the need

1222
02:04:22,489 --> 02:04:27,820
for standardized definitions. And so an entity
such as the Joint Commission working in collaboration

1223
02:04:27,820 --> 02:04:33,270
with them to develop these definitions and
then track them, allows you the opportunity

1224
02:04:33,270 --> 02:04:38,360
to look at outcomes across health care systems
using a nationwide dataset.

1225
02:04:38,360 --> 02:04:43,440
So that's exactly what my research kind of
centered on, was looking at exclusive breastfeeding

1226
02:04:43,440 --> 02:04:48,889
outcomes using the Joint Commission definition
in baby-friendly and non-baby-friendly hospitals

1227
02:04:48,889 --> 02:04:53,340
while looking like controlling for demographic
characteristics as well, which I obtained

1228
02:04:53,340 --> 02:04:58,300
from the American Communities Survey. So I
feel like, you know, sometimes it requires

1229
02:04:58,300 --> 02:05:07,470
merging several data sets in order to answer
those research questions. How easily accessible

1230
02:05:07,470 --> 02:05:14,550
were these data sets that you relied on or
that you accessed? I think that it's good

1231
02:05:14,550 --> 02:05:22,070
to have friends in good spaces, and that would
be some of my collaborators at the University

1232
02:05:22,070 --> 02:05:26,790
of Wisconsin-Madison as an example. The University
of Wisconsin-Madison developed the neighborhood

1233
02:05:26,790 --> 02:05:32,570
atlas, which was another indicator that I
added to my most recent study, which was exploring

1234
02:05:32,570 --> 02:05:36,850
the area deprivation index. And again, in
this conference, we've learned a lot about

1235
02:05:36,850 --> 02:05:42,350
the relationship between where a person lives
and the need to provide context around where

1236
02:05:42,350 --> 02:05:45,409
a person lives.
And so the area deprivation index is that

1237
02:05:45,409 --> 02:05:50,860
tool. All you need is the person's address,
and then it provides you with a number that

1238
02:05:50,860 --> 02:05:56,791
gives you a level of deprivation. So I applied
this to my most recent dataset, and then I

1239
02:05:56,791 --> 02:06:03,920
looked for access to baby-friendly hospitals
in areas that were highly deprived versus

1240
02:06:03,920 --> 02:06:08,849
areas that were low deprivation and then looked
at exclusive breastfeeding rates across those

1241
02:06:08,849 --> 02:06:14,159
area deprivation indices. So I feel like as
we talk about risk factors, understanding

1242
02:06:14,159 --> 02:06:19,010
where a person lives and what they have access
to, the area deprivation index is a readily

1243
02:06:19,010 --> 02:06:24,920
available tool. It includes 17 variables from
the American Community Survey. It collapses

1244
02:06:24,920 --> 02:06:31,380
those into one very great number that you
can use to interpret the resources in a person's

1245
02:06:31,380 --> 02:06:37,159
community. And it can be a great screening
tool, I believe, as well as a research tool

1246
02:06:37,159 --> 02:06:45,270
to look at outcomes. Thank you so much. That's
very helpful. Let's move along the panel.

1247
02:06:45,270 --> 02:06:54,110
Who wants to go next? Oh, I'm fine with going
next. So I have been involved with POWER,

1248
02:06:54,110 --> 02:06:58,320
the Pediatric Obesity Weight Evaluation Registry
for the past eight years.

1249
02:06:58,320 --> 02:07:06,390
And this has been a...the intent was to have
national participation of comprehensive multicomponent

1250
02:07:06,390 --> 02:07:12,580
pediatric weight management programs. And
to get people to sign on, this was something

1251
02:07:12,580 --> 02:07:17,989
that comes as a cost. And so we do at this
point, over the eight years we've been doing

1252
02:07:17,989 --> 02:07:24,659
this, we have not been able to get outside
funding to support programs, to either systematically

1253
02:07:24,659 --> 02:07:31,250
be collecting data, to, you know, have the
resources on board. In which to do that, we

1254
02:07:31,250 --> 02:07:39,980
have options for both manual data entry as
well as electronic upload that requires IT

1255
02:07:39,980 --> 02:07:46,250
specialists at the sites to help do that. We
also tried very hard to standardize the data

1256
02:07:46,250 --> 02:07:51,710
collection, make sure we're getting data elements
in some kind of systematic way. So when I

1257
02:07:51,710 --> 02:07:57,559
listened to research design, when I'm just
trying to put the infrastructure together

1258
02:07:57,559 --> 02:08:05,320
to get at outcomes, which is programs that
are set up in a certain way, who's doing better

1259
02:08:05,320 --> 02:08:10,500
based on either the resources they have available,
the staffing, the training, there are so many

1260
02:08:10,500 --> 02:08:14,679
outstanding questions.
And the research that we publish so far of

1261
02:08:14,679 --> 02:08:20,740
the data elements that we have, have yet to
point to something either in program design

1262
02:08:20,740 --> 02:08:28,260
or even patient characteristics that say these
do better given this very generic way of providing

1263
02:08:28,260 --> 02:08:33,739
multicomponent pediatric weight management
services. So even though eight years we've

1264
02:08:33,739 --> 02:08:38,329
spent on this, we're sort of still in what
I would refer to as the infancy and trying

1265
02:08:38,329 --> 02:08:46,600
to discern this, we've yet to establish a
consensus on an outcome that we see as successful.

1266
02:08:46,600 --> 02:08:52,119
What measures should we be pointing to that
says with pediatric obesity, you have a successful

1267
02:08:52,119 --> 02:08:57,719
outcome? We had a previous presenter that
says in the adult world, we have markers of

1268
02:08:57,719 --> 02:09:05,679
know when you have success. But with pediatrics,
that doesn't exist. We've yet to have a consensus

1269
02:09:05,679 --> 02:09:14,210
on either the outcome measure or the outcome
itself to know we are doing better. (LAUGHS)

1270
02:09:14,210 --> 02:09:19,929
In terms of the next step that would be, if
you had a magic wand, where would you want

1271
02:09:19,929 --> 02:09:25,901
to see help? I know you mentioned external
funding, of course, for something on a large

1272
02:09:25,901 --> 02:09:31,929
basis like this.
Anything else that you would prioritize as

1273
02:09:31,929 --> 02:09:38,450
a hurdle that might help you jump to the next
level? I think if sites had access to be able

1274
02:09:38,450 --> 02:09:46,210
to do the electronic upload of the data of
interest, so you remove that as a barrier.

1275
02:09:46,210 --> 02:09:53,619
If we use more validated screening tools and
have the electronic medical record, have that

1276
02:09:53,619 --> 02:09:58,909
in place, or whether it is established templates
that people who are participating in this

1277
02:09:58,909 --> 02:10:07,070
are using that reliably. But this all requires
from an infrastructure standpoint, rather

1278
02:10:07,070 --> 02:10:12,739
than just designing the study, we need to
expand. We were at a point where we have 42

1279
02:10:12,739 --> 02:10:18,219
sites participating at one point in time.
They can't sustain the participation. We have

1280
02:10:18,219 --> 02:10:25,119
our own data coordinating center at Cincinnati
Children's that comes as a cost. But we use

1281
02:10:25,119 --> 02:10:30,590
what's called the rave data system that's
very high quality, and does the sort of thing

1282
02:10:30,590 --> 02:10:36,760
that you heard the first speaker talk about
making sure that we remove data that is not

1283
02:10:36,760 --> 02:10:42,070
reliable, that is outside limits. So we're
able to do quality checks.

1284
02:10:42,070 --> 02:10:48,429
But I still think that we can do far better
and getting a representative registry that

1285
02:10:48,429 --> 02:10:57,139
doesn't have the barrier of cost to participate.
Thank you. Let's go to Wendy Sue Swanson.

1286
02:10:57,139 --> 02:11:12,460
Your comments on the same question. I'm sorry.
I can't hear.

1287
02:11:12,460 --> 02:11:29,389
WENDY SUE SWANSON: Oh, I'm sorry. Did you
call me? Sorry, yeah. (CROSSTALK) Yeah, sorry.

1288
02:11:29,389 --> 02:11:30,880
I'm Dr. Wendy Sue Swanson.

1289
02:11:30,880 --> 02:11:32,369
DR. NILESH MEHTA: Do jump in. Yeah.

1290
02:11:32,369 --> 02:11:37,760
WENDY SUE SWANSON: I was, you know, I think
I have a very different take. So I was at

1291
02:11:37,760 --> 02:11:43,119
Seattle Children's Hospital for about 10 years
doing digital health and innovation. And now,

1292
02:11:43,119 --> 02:11:47,940
in part, because of funding challenges, I'm
now both at Stanford in digital health. And

1293
02:11:47,940 --> 02:11:53,909
also with SpoonfulOne, which is an outside-funded
venture capital-funded company working on

1294
02:11:53,909 --> 02:11:58,400
diet diversity in early life, early allergen
and exposure. But one thing I wanted to just

1295
02:11:58,400 --> 02:12:02,610
bring up is, you know, I'm sitting very intentionally
in the kitchen, in that, you know, so many

1296
02:12:02,610 --> 02:12:06,639
of these challenges for families, we don't
solve at clinic, and we don't solve them in the

1297
02:12:06,639 --> 02:12:11,119
hospital. They're solved ultimately in the
kitchen. And, you know, I started my work

1298
02:12:11,119 --> 02:12:15,770
in digital health as a mom blogger for 10
years really translating health and science

1299
02:12:15,770 --> 02:12:20,300
information, because I wanted families to
reach that. So when we're thinking about clinical

1300
02:12:20,300 --> 02:12:25,550
trial design, of course, the more convenient
we can create in any design, and the more

1301
02:12:25,550 --> 02:12:31,463
digital in my mind, we can scale. So one thing
you can look up, but the INTENT Study, so

1302
02:12:31,463 --> 02:12:36,309
it's I-N-T-E-N-T study.com, you'll see a lot
of this.

1303
02:12:36,309 --> 02:12:41,790
But we designed a trial where we were able
to recruit 1,700 infants between the age of

1304
02:12:41,790 --> 02:12:46,310
four and six months in less than 12 months.
And we did that through multiple different

1305
02:12:46,310 --> 02:12:51,260
channels through experimentation. It's a full
end-to-end digital trial, meaning that families

1306
02:12:51,260 --> 02:12:55,849
never go to clinic for outcomes, we rely on
patient-reported outcomes for primary and

1307
02:12:55,849 --> 02:13:01,409
secondary endpoints. And we also, you know,
have a pretty long. It's a randomized, controlled

1308
02:13:01,409 --> 02:13:06,570
prospective trial design, it'll run 18 months.
So it's pretty long. Families, though, however,

1309
02:13:06,570 --> 02:13:12,000
could learn about it online and social channels.
We also tested through Duke Clinical Research

1310
02:13:12,000 --> 02:13:16,460
Institute, entire, you know, swaths of any
child who aged in at four months of age through

1311
02:13:16,460 --> 02:13:21,510
my chart, getting an invitation to learn about
the trial. And then we use, you know, Facebook,

1312
02:13:21,510 --> 02:13:26,000
Instagram, traditional social media channels.
And we also use BabyCenter. And what's so

1313
02:13:26,000 --> 02:13:30,360
extraordinary about BabyCenter? Is it's where
moms and dads already are. So when I would

1314
02:13:30,360 --> 02:13:37,159
go to recruit a group of four-month-old infants,
I could go to 165,000 moms to four-month-olds

1315
02:13:37,159 --> 02:13:42,099
right then, describe the trial design and,
you know, let them learn about their eligibility.

1316
02:13:42,099 --> 02:13:47,180
They fill out eligibility online. When they
met eligibility, they download an app. After

1317
02:13:47,180 --> 02:13:52,080
filling out informed consent, they're immediately
randomized into either the intervention arm

1318
02:13:52,080 --> 02:13:56,590
or the control. And then they began the trial
through the app that has internal and external

1319
02:13:56,590 --> 02:14:02,270
intrinsic, you know, validation and gamification,
and also allows families to input information

1320
02:14:02,270 --> 02:14:07,820
there. So, you know, I think the goal here
in some ways of joining this panel is to think

1321
02:14:07,820 --> 02:14:13,140
really critically about populations. We got
children from all 50 states. 30% of our study

1322
02:14:13,140 --> 02:14:20,480
population had by the [inaudible], inventory, a diagnosis
of atopic dermatitis, and we're able to capture,

1323
02:14:20,480 --> 02:14:24,230
you know, 1,700 infants that we can now follow
in that time. And this was through the pandemic,

1324
02:14:24,230 --> 02:14:28,489
that we were able to recruit them. And so
my hope in some ways in joining this is just

1325
02:14:28,489 --> 02:14:32,469
to, you know, we did a number of different
trials at Seattle Children's that involved

1326
02:14:32,469 --> 02:14:40,590
user-centered design, family-centered care
methods that bring patients families and caregivers

1327
02:14:40,590 --> 02:14:45,619
to the table in the design of how you actually
create technology, how you create the study

1328
02:14:45,619 --> 02:14:50,550
design and how you put it in so that, you
know, that convenient model is really a part

1329
02:14:50,550 --> 02:14:53,469
of how we get there.
And I think, you know, what was interesting

1330
02:14:53,469 --> 02:14:57,900
about this trial and our recruitment, we fully
recruited back in April. So we're in the midst

1331
02:14:57,900 --> 02:15:03,780
of data collection at this time. And like,
you know, some of the presenters have talked

1332
02:15:03,780 --> 02:15:06,340
about, that's kind of what, you know, what
I live and breathe at this point until we're

1333
02:15:06,340 --> 02:15:11,020
done with the trial in about a year. But,
you know, when you have an infancy, of course,

1334
02:15:11,020 --> 02:15:14,530
during these critical periods of development,
and these critical periods of iron intake

1335
02:15:14,530 --> 02:15:19,170
and all the other issues and critical periods
of early breastfeeding, for example, there's

1336
02:15:19,170 --> 02:15:23,860
such small windows of eligibility and enrollment.
That to scale and to get the numbers that

1337
02:15:23,860 --> 02:15:28,619
we often need, in this case, we're looking
at outcomes for food allergy diagnoses, which

1338
02:15:28,619 --> 02:15:32,699
because of the low rates of prevalence in
early life, we required large trial designs

1339
02:15:32,699 --> 02:15:36,389
just to really think critically about getting
these studies into Pearson's hands through

1340
02:15:36,389 --> 02:15:41,300
mobile apps, really thinking critically about
where to moms and dads live online.

1341
02:15:41,300 --> 02:15:46,941
And how do we go find them, educate them,
you know, garner their informed consent, and

1342
02:15:46,941 --> 02:15:51,050
involve them in research moving forward, using
social tactics like videos and education,

1343
02:15:51,050 --> 02:15:55,960
nudges, notifications, text messages, etc.
A part of the way they live their entire day,

1344
02:15:55,960 --> 02:15:58,270
bring research into that same model.

1345
02:15:58,270 --> 02:16:05,540
DR. NILESH MEHTA: That is really very exciting. And
congratulations on all your success. We discussed

1346
02:16:05,540 --> 02:16:12,059
this briefly in our previous session. One
of the discussion points we had, Dr. Srinivasan

1347
02:16:12,059 --> 02:16:18,599
from Pennsylvania alluded to that is the concept
that using social media platforms or digital

1348
02:16:18,599 --> 02:16:27,559
tools for data gathering in studies may limit
the study to only populations that have access

1349
02:16:27,559 --> 02:16:32,420
to it. And there are plenty of examples that
people don't have access to an app or computer

1350
02:16:32,420 --> 02:16:35,710
or a phone. Any comments on that? How do you
overcome that?

1351
02:16:35,710 --> 02:16:39,910
DR. WENDY SUE SWANSON: Yeah, thanks for your question.
I think over the last 10, 15 years that I've

1352
02:16:39,910 --> 02:16:44,349
been in digital health, it's always the first
question because we worry about marginalized

1353
02:16:44,349 --> 02:16:48,809
populations. You know, the act that was just
signed-in in this last year on broadband Internet

1354
02:16:48,809 --> 02:16:54,469
access, 30 more million Americans that Biden
signed in, for example, are collapsing the

1355
02:16:54,469 --> 02:16:58,770
gaps that people have towards broadband access.
And in reality, if you start to look at the

1356
02:16:58,770 --> 02:17:03,469
Pew Internet data, that we ultimately know,
increasingly, you know, north of even 80%

1357
02:17:03,469 --> 02:17:08,859
of 12-year-olds have access to a smartphone,
and increasingly, that it's not a reason not

1358
02:17:08,859 --> 02:17:13,160
to get involved using these digital tools
and tactics. But to your specific question

1359
02:17:13,160 --> 02:17:17,679
about social media, I want to bring that up.
Data collection in social media is fraught,

1360
02:17:17,679 --> 02:17:21,880
you know, without question. You know, when
we did recruitment, for example, using Facebook

1361
02:17:21,880 --> 02:17:27,050
and using Facebook or meta now owned Instagram,
we actually had a lot of challenges with hackers

1362
02:17:27,050 --> 02:17:30,359
from all over the world.
But we had to rechange our patient identification

1363
02:17:30,359 --> 02:17:35,149
and verification process at the beginning
using IP addresses from exactly where people

1364
02:17:35,149 --> 02:17:41,260
are. The goal in some ways I would suggest
is to, from a recruitment standpoint and uptake,

1365
02:17:41,260 --> 02:17:45,950
is to find people where they live, where they
think, where they learn, and where they share,

1366
02:17:45,950 --> 02:17:49,400
and bring them into secure environment. So
in partnership with Duke Clinical Research

1367
02:17:49,400 --> 02:17:54,359
Institute, we used another app called Pattern
Health that had been used in multiple other

1368
02:17:54,359 --> 02:17:59,570
clinical research trials, never fully intended
virtual. But we use it after it had been verified,

1369
02:17:59,570 --> 02:18:03,200
it had been encrypted, it followed all the
recommend, and then brought patients and families

1370
02:18:03,200 --> 02:18:08,530
into that environment for data access and
for data gathering in that way. Because I think,

1371
02:18:08,530 --> 02:18:12,910
you know, I'm increasingly, I mean, I was
one of the very first physicians on Twitter

1372
02:18:12,910 --> 02:18:17,670
and Facebook, one of the first bloggers in
this country for a pediatric hospital. But

1373
02:18:17,670 --> 02:18:23,660
I think that I'm increasingly skeptical and
concerned about the kind of ownership of these

1374
02:18:23,660 --> 02:18:30,310
social channels and their lack of care and
concern for both privacy, human rights, public

1375
02:18:30,310 --> 02:18:34,770
health and education.
So I think we go and get people and we bring

1376
02:18:34,770 --> 02:18:39,929
them into a place where we can really protect
them and serve them in ways that we would like.

1377
02:18:39,929 --> 02:18:44,380
DR. NILESH MEHTA: That's a great point. Dr. Grundmeier,
Julie Patterson had this question too, relate

1378
02:18:44,380 --> 02:18:51,540
it to some other presentations in this workshop,
particularly tools such as photographs, the

1379
02:18:51,540 --> 02:18:57,560
mobile food record, and I know other companies
that have invested in it. Do you see opportunities

1380
02:18:57,560 --> 02:19:02,570
for these type of apps on patient or caregiver's
phone that can be integrated into the EMR?

1381
02:19:02,570 --> 02:19:09,370
We've certainly seen photographs being included
for other purposes in the EMR.

1382
02:19:09,370 --> 02:19:15,319
DR. ROBERT GRUNDMEIER: Yeah. Question for me? That's
great. (CROSSTALK) One of my favorite topics

1383
02:19:15,319 --> 02:19:21,450
right now. And it's actually related to the
notion that families will provide us more

1384
02:19:21,450 --> 02:19:27,530
accurate information than we're able to collect
in the office in our limited time. And I think

1385
02:19:27,530 --> 02:19:33,330
the variety of apps that are available for
collecting data sets such as the example of

1386
02:19:33,330 --> 02:19:40,580
taking photographs to quantify nutritional
intake, which is, I think, really exciting

1387
02:19:40,580 --> 02:19:47,050
in terms of possibility of making it, providing
new ways for families to easily capture information.

1388
02:19:47,050 --> 02:19:51,780
I mean, the number of times I've asked a family
to just keep track of nutritional intake for

1389
02:19:51,780 --> 02:19:57,430
three days or something like that. And it
seems to be a big barrier for people to do

1390
02:19:57,430 --> 02:20:04,410
that the old-fashioned way. Whereas taking
pictures is great. The frustration I have

1391
02:20:04,410 --> 02:20:09,740
is that there are always people making these
wonderful apps. And it seems like almost nobody

1392
02:20:09,740 --> 02:20:16,500
is thinking about how that information could
be incorporated into routine clinical care.

1393
02:20:16,500 --> 02:20:23,340
And that, I think, is an incredible
gap and opportunity that I really hope future

1394
02:20:23,340 --> 02:20:28,890
research efforts would address. I could go
on and on, but it has to be done with great

1395
02:20:28,890 --> 02:20:33,660
care because clinicians don't want to be deluged
with information that they don't know what

1396
02:20:33,660 --> 02:20:40,620
to do with. But with thoughtful approaches,
these new types of data capture could both

1397
02:20:40,620 --> 02:20:46,880
be useful for research, patient engagement,
and self-management, as well as helping to

1398
02:20:46,880 --> 02:20:52,860
inform the clinical team so that they can
do their part to navigate the family towards better outcomes. 

1399
02:20:52,860 --> 02:21:01,610
DR. NILESH MEHTA: Thank you. Thank you so much.
Yeah, it's an area that many people are watching.

1400
02:21:01,610 --> 02:21:06,979
There is, of course, patient confidentiality
in addition to extracting these in an objective

1401
02:21:06,979 --> 02:21:14,040
manner to provide information. A question
for the panelists: at this point, yesterday,

1402
02:21:14,040 --> 02:21:19,890
around this time, we talked about disease-related
malnutrition. We alluded to a specific group

1403
02:21:19,890 --> 02:21:30,569
of diseases. And we found that many of these
researchers found that these unique situations

1404
02:21:30,569 --> 02:21:37,430
as a result of being pigeonholed into their
unique areas in order to explore the possibility

1405
02:21:37,430 --> 02:21:43,280
of pushing the field towards multiple disease
condition associated malnutrition areas at

1406
02:21:43,280 --> 02:21:52,450
the same time, do you have any opportunities
that you see how you can combine some of these

1407
02:21:52,450 --> 02:21:57,550
disparate conditions?
And I don't know if that means registries

1408
02:21:57,550 --> 02:22:04,560
that house many or trials that incorporate
distinct disease-specific malnutrition. And this is a question for anyone on the panel.

1409
02:22:04,560 --> 02:22:12,433
DR. SHELLEY KIRK: 
I'd like to address that if I can. I don't know if I'm still on mute or you can hear
me.

1410
02:22:12,433 --> 02:22:14,460
 DR. NILESH MEHTA: No, you're good. We can hear you. 

1411
02:22:14,460 --> 02:22:20,960
DR. SHELLEY KIRK: OK great. One of the things we captured in our
registry is that for each patient encounter,

1412
02:22:20,960 --> 02:22:27,930
we also list medical diagnoses. So it's not
just about obesity. So we're capturing whether

1413
02:22:27,930 --> 02:22:36,330
it's dyslipidemia or we're capturing NASH
or any of the other comorbidities of obesity.

1414
02:22:36,330 --> 02:22:41,770
We have an opportunity to look at that. And
so what we've recently added to our registry

1415
02:22:41,770 --> 02:22:48,479
are liver health measures. And this may be
the first time that will be able, in combination

1416
02:22:48,479 --> 02:22:56,780
with obesity, be looking at NASH and NAFLD
and see of the additional measures that it's

1417
02:22:56,780 --> 02:23:02,561
not just about change in liver enzymes. There's
a lot more going on to investigate. So this

1418
02:23:02,561 --> 02:23:09,030
may actually offer the first time having a
registry that's looking at liver health and obesity combined.

1419
02:23:09,030 --> 02:23:20,790
DR. NILESH MEHTA: 
Thank you. Anyone else? Any other thoughts?

1420
02:23:20,790 --> 02:23:29,850
A question for Dr. McKeever, in addition to
the cost of doing an RCT, many RCTs in recent

1421
02:23:29,850 --> 02:23:36,450
times because of the nature of an RCT and
the need to conform to the experimental condition

1422
02:23:36,450 --> 02:23:42,680
find themselves lacking in external validity.
So, for example, many RCTs have had patient

1423
02:23:42,680 --> 02:23:48,150
eligibility criteria such that a very small
fraction of my practice actually caters to

1424
02:23:48,150 --> 02:23:53,830
their trial eligibility. And therefore, when
I go back to my practice, I barely find external

1425
02:23:53,830 --> 02:23:57,710
validity. Any comments on that, Liam in terms
of our RCTs?

1426
02:23:57,710 --> 02:24:02,510
DR. LIAM MCKEEVER: It's a...it's a really important consideration,
right? The consideration of who does this

1427
02:24:02,510 --> 02:24:08,730
generalize to is the most important question
you can ask anytime you read a study. So here's

1428
02:24:08,730 --> 02:24:15,390
the thing. It depends on your population, right?
So like I did research in adult septic patients,

1429
02:24:15,390 --> 02:24:22,859
that's an incredibly heterogeneous population.
If I were going to just do that pragmatically

1430
02:24:22,859 --> 02:24:30,970
or just, you know, a "come all", then you really
need a fairly large sample size to be able

1431
02:24:30,970 --> 02:24:37,470
to deal with that heterogeneity. And a lot
of RCTs are not in a position to do that because

1432
02:24:37,470 --> 02:24:43,189
they're expensive, they're time-consuming.
And so I think the kind of... I think the

1433
02:24:43,189 --> 02:24:50,050
line of questioning has to go like this, in
my opinion. I think, you have to first say, can we

1434
02:24:50,050 --> 02:24:57,010
demonstrate the efficacy of this in any population?
Clean it up. And if you can demonstrate, yes,

1435
02:24:57,010 --> 02:25:02,470
with people with acute pancreatitis, this
is happening, this intervention is working.

1436
02:25:02,470 --> 02:25:07,750
It might not be working for these people over
here, but ask yourself what would actually

1437
02:25:07,750 --> 02:25:11,810
be the value of a population estimate anyways,
right?

1438
02:25:11,810 --> 02:25:17,450
This idea that everybody needs the same thing
is not true. One thing that was brought up

1439
02:25:17,450 --> 02:25:22,390
before we started this and I'm so sorry I
can't see the names of all the panelists and

1440
02:25:22,390 --> 02:25:26,890
I don't know which one brought it up, but
she was talking about like, maybe we should

1441
02:25:26,890 --> 02:25:32,620
be collecting data on genetics. And wouldn't
that be great if we could do that? And there

1442
02:25:32,620 --> 02:25:37,310
are people who do that like the (UNKNOWN)
collects data on septic patients, actually

1443
02:25:37,310 --> 02:25:44,080
has a gene chip, and she's got a huge database.
This, you know, we need to start thinking

1444
02:25:44,080 --> 02:25:50,790
about like what people is this intervention
going to work for? And if we keep sticking,

1445
02:25:50,790 --> 02:25:55,840
you know, I run the guidelines, but I'm the
first to say a guideline is a population estimate.

1446
02:25:55,840 --> 02:26:00,040
I'm not sure how useful that is. It might
be good for some and bad for others. And I

1447
02:26:00,040 --> 02:26:04,850
think the null findings we often see is that
some people are benefiting from it and some

1448
02:26:04,850 --> 02:26:11,630
people are not. And it's all equaling out
in the middle. So personally, I don't have

1449
02:26:11,630 --> 02:26:14,880
a problem with this idea that you need to
get specific.

1450
02:26:14,880 --> 02:26:20,950
We just need that many different population
studies happening. We need robust funding

1451
02:26:20,950 --> 02:26:25,720
and robust studies. Otherwise, I'm not sure
that our pragmatic answers are going to give

1452
02:26:25,720 --> 02:26:30,050
us anything more than a population estimate,
and I don't know how useful that is. Yeah.

1453
02:26:30,050 --> 02:26:38,270
DR. SHELLEY KIRK:  I'd just like to add that POWER, our
registry has just added genetic testing as

1454
02:26:38,270 --> 02:26:43,550
an additional data element going forward.
And the reason why we're looking at in-depth

1455
02:26:43,550 --> 02:26:49,250
liver health or genetics is that we're not
getting the answers we're looking for in terms

1456
02:26:49,250 --> 02:26:53,610
of figuring out who's going to be successful
in our interventions or not, because we're

1457
02:26:53,610 --> 02:27:00,770
not collecting the information that might
uncover what it is that we're missing. That

1458
02:27:00,770 --> 02:27:06,140
is the advantage of the registry that has
established that once you hit a roadblock,

1459
02:27:06,140 --> 02:27:11,300
you can change that. You can do it differently.
So I appreciate you bringing that up because

1460
02:27:11,300 --> 02:27:14,270
we're poised to get that kind of data going
forward.

1461
02:27:14,270 --> 02:27:21,029
DR. LIAM MCKEEVER: I'll tell you, my research showed in
a very small sample and it has to be reproduced,

1462
02:27:21,029 --> 02:27:29,380
but it showed that 40% of the adult ICU population
has a genetic snip and the genetic allele

1463
02:27:29,380 --> 02:27:36,590
status such that those people will create
more oxidative stress from nutrition versus

1464
02:27:36,590 --> 02:27:42,970
people who don't. So it's like for those people
they might need a completely different intervention,

1465
02:27:42,970 --> 02:27:47,149
a completely different dosing, and you can't
capture that any other way. Right. I think

1466
02:27:47,149 --> 02:27:49,580
you're on the cutting edge if you're doing
that. That's great.

1467
02:27:49,580 --> 02:27:56,430
DR. SHELLEY KIRK: Thanks. DR. NILESH MEHTA: These are great comments
alluding to what both of you said. If you

1468
02:27:56,430 --> 02:28:03,010
have identified a relevant patient population,
whether it's the adult with sepsis or children

1469
02:28:03,010 --> 02:28:07,350
of a certain age who are going to undergo
some kind of nutritional intervention, like

1470
02:28:07,350 --> 02:28:14,330
the POWER registry or others that we've heard
of. One thing that's growing in some areas

1471
02:28:14,330 --> 02:28:19,970
is the concept of platform trials, wherein
you then move from one intervention to another

1472
02:28:19,970 --> 02:28:26,410
but utilize the resources that you've invested
into developing the ability to gather data,

1473
02:28:26,410 --> 02:28:33,050
whether you're using EMR or whether you're
using research personnel. Liam, first you and

1474
02:28:33,050 --> 02:28:39,060
then anyone else after that, jump in platform
trials gathering strength. Are you seeing

1475
02:28:39,060 --> 02:28:41,040
any emerging?

1476
02:28:41,040 --> 02:28:48,609
DR. LIAM MCKEEVER:  I'm not aware of anything. I'm not 100%
sure, though, that I followed the whole question.

1477
02:28:48,609 --> 02:28:54,080
DR. NILESH MEHTA: Yeah, it essentially is a population of interest
and then the intervention changes. So you

1478
02:28:54,080 --> 02:28:58,979
finish one succinct trial and then you can
actually go for the next intervention in that

1479
02:28:58,979 --> 02:29:03,439
patient population. So all your resources
that you've invested in developing the infrastructure

1480
02:29:03,439 --> 02:29:09,160
continue to be used and the enrolment and
everything. So you are...it's a force multiplier.

1481
02:29:09,160 --> 02:29:14,979
And the adult critical care group, the Canadian
Trials Group, have good examples of that.

1482
02:29:14,979 --> 02:29:20,710
And I know there are efforts in certain parts
of NIH as well that they've seen proposals

1483
02:29:20,710 --> 02:29:21,710
like that.

1484
02:29:21,710 --> 02:29:27,189
DR. LIAM MCKEEVER: I think it's a great idea. It's a
wonderful idea. Go ahead. 

1485
02:29:27,189 --> 02:29:34,720
DR. SHELLEY KIRK: One thing that we also have established in power are these ongoing
reports and what we're able to report on selected

1486
02:29:34,720 --> 02:29:40,260
outcomes that we're looking to expand on those
reports is having six-month cohorts. So if

1487
02:29:40,260 --> 02:29:45,560
a program wanted to try out something new,
they can be looking at their outcome data

1488
02:29:45,560 --> 02:29:52,330
for a particular six-month period and compare
that to a previous stretch of time. And again

1489
02:29:52,330 --> 02:29:57,510
it would be over six months and only include
those that were enrolled during those different

1490
02:29:57,510 --> 02:30:04,560
cohort periods. So we have an infrastructure
here to answer quite a bit. We just need more

1491
02:30:04,560 --> 02:30:11,160
resources in which to make sure we have a
representative national population that has

1492
02:30:11,160 --> 02:30:16,370
the diversity. That's the other thing that
a registry is going to be able to do is capture

1493
02:30:16,370 --> 02:30:23,180
populations both from a geographic standpoint
as well as a race-ethnicity. And I think then

1494
02:30:23,180 --> 02:30:29,140
our outcomes are going to be more applicable
when we can't account for those differences

1495
02:30:29,140 --> 02:30:37,680
in the patient demographics.
DR. NILESH MEHTA: Thank you. Yeah, that's a great point. Dr. Grundmeier, 

1496
02:30:37,680 --> 02:30:43,380
we're talking about resources and
the ability to harness what's available. EMR,

1497
02:30:43,380 --> 02:30:50,290
of course, will remain in the spotlight, and
frankly, a lot more to be desired. Where does

1498
02:30:50,290 --> 02:30:56,670
natural language processing and the ability
to use it and then gather information that

1499
02:30:56,670 --> 02:31:02,740
can provide both clinical decision support,
but also research data? 

1500
02:31:02,740 --> 02:31:07,270
DR. ROBERT GRUNDMEIER: Yeah. Actually, I was going to make a comment about one of the
things I've been thinking about is listening

1501
02:31:07,270 --> 02:31:14,960
to the conversation is whether the EHR can
be a part of the mechanism by which randomized

1502
02:31:14,960 --> 02:31:21,890
clinical trials could be completed more efficiently
and in more diverse, you know, a broader variety

1503
02:31:21,890 --> 02:31:27,830
of populations. To your question about NLP,
things are getting much better. There's a

1504
02:31:27,830 --> 02:31:37,050
lot more tools and ways to develop NLP pipelines.
I am less aware of whether it's been extensively

1505
02:31:37,050 --> 02:31:43,380
applied to the nutrition challenge, but I
think we've all had the experience of NLP

1506
02:31:43,380 --> 02:31:49,000
helping us in our everyday lives in different
ways, whether it's Siri or Alexa.

1507
02:31:49,000 --> 02:31:52,880
So it's really close. I can remember when
I started my career 20 years ago, we were

1508
02:31:52,880 --> 02:31:58,490
just waiting for automated voice recognition
to be a thing we could do and it was really

1509
02:31:58,490 --> 02:32:03,000
close for a long time. And now it's, you know,
now it works. So I think we're right at that

1510
02:32:03,000 --> 02:32:10,029
cusp with NLP where increasingly you can get
actually accurate, reliable information through

1511
02:32:10,029 --> 02:32:16,439
pipelines that are more or less off the shelf.
We're not quite there yet, though. It still

1512
02:32:16,439 --> 02:32:21,800
takes a lot of tinkering and a little bit
of research and development efforts. So but

1513
02:32:21,800 --> 02:32:27,990
I think that's going to be and that's what
I say now is I don't worry so much about enforcing

1514
02:32:27,990 --> 02:32:33,279
a structure on the data collection process,
but just try to get people to give you the

1515
02:32:33,279 --> 02:32:39,500
information that you need in some fashion
in the text, and then we can get it back out

1516
02:32:39,500 --> 02:32:46,810
afterwards because that tends to be more acceptable
to people than being forced to check boxes.

1517
02:32:46,810 --> 02:32:52,229
Doctors don't like to do that. That's not
the way we're trained. So yeah, I think that's

1518
02:32:52,229 --> 02:32:58,399
the way going forward will be to allow free
text information to be collected and develop

1519
02:32:58,399 --> 02:33:05,040
better ways to make efficient use of it.
DR. NILESH MEHTA: Thank you. Thank you. We have time for one

1520
02:33:05,040 --> 02:33:07,109
quick comment from Julie Patterson. Julie,
please jump in.

1521
02:33:07,109 --> 02:33:11,890
DR. JULIE PATTERSON: Yes. So thank you so much.
I just wanted to kind of share one of the

1522
02:33:11,890 --> 02:33:16,070
things that I'm hearing people say is sometimes
barriers and facilitators to implementing

1523
02:33:16,070 --> 02:33:21,170
change. And so I want to kind of put in a
plug for dissemination, implementation, science

1524
02:33:21,170 --> 02:33:27,300
frameworks. There's a really great resource
dissemination-implementation.org, which has

1525
02:33:27,300 --> 02:33:33,400
a kind of a tool that you can use. A lot of
times it's hard to identify which dissemination/implementation

1526
02:33:33,400 --> 02:33:37,340
framework you would want to use, and it helps
you to identify what stage of the process

1527
02:33:37,340 --> 02:33:42,950
you're in. Are you looking at developing an
intervention or implementing evidence-based

1528
02:33:42,950 --> 02:33:49,970
practices so that there's a really good opportunity
for implementation science to be a strategy

1529
02:33:49,970 --> 02:33:53,970
for overcoming a lot of the barriers that
we've heard of in this conference in terms

1530
02:33:53,970 --> 02:33:59,170
of identifying and diagnosing patients with
malnutrition. And then finally, just to put

1531
02:33:59,170 --> 02:34:04,510
in a plug for understanding where your organization
is at. There's a really great organization

1532
02:34:04,510 --> 02:34:09,310
change management tool that can assess kind
of the readiness for change and the readiness

1533
02:34:09,310 --> 02:34:14,770
to develop these or to implement these interventions.
And I think that we see better success when

1534
02:34:14,770 --> 02:34:22,120
we have all parties engaged. And so there's
a great tool that I can put in the resources

1535
02:34:22,120 --> 02:34:25,830
information that will be available, I believe,
after the conference, along with the links

1536
02:34:25,830 --> 02:34:30,540
for these implementation science frameworks
to help people think about research, study

1537
02:34:30,540 --> 02:34:32,569
designs, and strategies.

1538
02:34:32,569 --> 02:34:37,470
DR. NILESH MEHTA: That's wonderful. Thank you for doing
that. Thank you for your comment. To all our

1539
02:34:37,470 --> 02:34:41,950
participants, we are almost at the end of
this session. I hope that you will now spend

1540
02:34:41,950 --> 02:34:50,330
the next hour until 3:45 and visit the poster
hall. There are some fantastic abstracts,

1541
02:34:50,330 --> 02:34:55,350
posters displaying work from all around the
world. The poster session is now going to

1542
02:34:55,350 --> 02:35:02,149
be open for the early life track two. At 3:45
right after the poster session, we come back

1543
02:35:02,149 --> 02:35:08,160
and this is the last and a very exciting session
on early life Malnutrition conclusions. I

1544
02:35:08,160 --> 02:35:13,061
will be joined in moderating that session
by my colleague, Dr. Todd Rice, and my colleague,

1545
02:35:13,061 --> 02:35:20,250
Dr. David Seres. In addition, we have a very
esteemed panel to help us discuss some of

1546
02:35:20,250 --> 02:35:28,290
the future avenues for this area. We have
Professor Christopher Duggan from Boston Children's.

1547
02:35:28,290 --> 02:35:33,580
William Evans from University of California,
Berkeley, Liam McKeever joins us back. David

1548
02:35:33,580 --> 02:35:38,890
Suskind from Seattle Children's. Tamara Hannon
from Indiana University and Vijay Srinivasan

1549
02:35:38,890 --> 02:35:44,330
from Children's Hospital of Philadelphia.
So I hope you will join us. At this stage,

1550
02:35:44,330 --> 02:35:50,620
I just want to express my deepest thanks to
our two speakers today, Bob Grundmeier,

1551
02:35:50,620 --> 02:35:56,770
and opening up the EHR campaign for us. It
will stay on our minds. And Liam McKeever,

1552
02:35:56,770 --> 02:36:02,760
always challenging us with the study, design,
and research, and we really appreciate that.

1553
02:36:02,760 --> 02:36:07,661
And huge thanks to our panelists. Julie Patterson,
Shelley Kirk, and Wendy Sue Swanson. Thank

1554
02:36:07,661 --> 02:36:12,621
you for your insightful comments. Thank you
all for joining us. See you at the poster

1555
02:36:12,621 --> 02:36:18,290
session. And then at 3:45 for the concluding
session of this workshop. Bye.

1556
02:36:18,290 --> 02:36:22,560
DR. NILESH MEHTA: Hello, everyone. Welcome to this most exciting

1557
02:36:22,560 --> 02:36:31,391
and sadly, the last session of our five day
workshop. This is session 11. This session

1558
02:36:31,391 --> 02:36:38,270
is Early Life Malnutrition Conclusions where
we bring together lessons learned, dreams

1559
02:36:38,270 --> 02:36:47,290
dreamt, expert panelists from previous sessions,
and then hopefully with input from our participants,

1560
02:36:47,290 --> 02:36:53,970
try and discuss the future of malnutrition
research in early life. I'm delighted to be

1561
02:36:53,970 --> 02:36:58,939
joined in this session as co-moderators by Dr.
Todd Rice from Vanderbilt University Medical

1562
02:36:58,939 --> 02:37:06,189
Center, and David Seres from Columbia University
Irving Medical Center. Although I will be

1563
02:37:06,189 --> 02:37:12,471
speaking to begin with, they will jump in
intermittently as well. In addition, we have

1564
02:37:12,471 --> 02:37:18,130
a very exciting panel. Our heartfelt thanks
to this esteemed panel for their generous

1565
02:37:18,130 --> 02:37:23,410
time and for spending the next hour with us.
We have Christopher Duggan from Boston Children's

1566
02:37:23,410 --> 02:37:30,540
Hospital, William Evans from UC Berkeley,
Liam McKeever from the American Society for

1567
02:37:30,540 --> 02:37:37,569
Parenteral and Interim Nutrition, David Suskind
from Seattle Children's Hospital, Tamara Hannon

1568
02:37:37,569 --> 02:37:43,390
from Indiana University School of Medicine,
and Vijay Srinivasan from the Children's Hospital

1569
02:37:43,390 --> 02:37:49,670
of Philadelphia.
Welcome to all of you panelists and my co-moderators.

1570
02:37:49,670 --> 02:37:54,790
As I mentioned, this session goes on until
as long as we want, but hopefully we'll shut

1571
02:37:54,790 --> 02:38:01,720
it down at the end of an hour if we do get
there. After that, I will close the session

1572
02:38:01,720 --> 02:38:08,720
today and please do hang on until the end
because I have the poster awards for the adult

1573
02:38:08,720 --> 02:38:14,700
and pediatric posters that will be announced
right at the end of this session. So having

1574
02:38:14,700 --> 02:38:21,431
said that, I am once again going to thank
our NIH Scientific Leadership Christopher

1575
02:38:21,431 --> 02:38:28,600
Lynch, Charlotte Pratt and Ashley Vargas.
Their foresight into thinking this and creating

1576
02:38:28,600 --> 02:38:34,060
this platform for such a lively discussion
on five days is tremendously appreciated.

1577
02:38:34,060 --> 02:38:42,760
It has been a great opportunity for us to
identify gaps and think about future studies.

1578
02:38:42,760 --> 02:38:46,680
This particular session is important because
we bring together panelists and speakers from

1579
02:38:46,680 --> 02:38:52,020
previous session, and what we would appreciate
is for the panelists to provide their own

1580
02:38:52,020 --> 02:39:00,050
thoughts. They would use their past experience,
their expertise and their view of the sessions

1581
02:39:00,050 --> 02:39:04,350
that they've had the chance to attend in the
last five days.

1582
02:39:04,350 --> 02:39:10,080
And based on that, I am going to ask them
in the beginning to start diving into what

1583
02:39:10,080 --> 02:39:15,910
they think would be a classic, a pivotal moonshot
study. What would be the moonshot study that

1584
02:39:15,910 --> 02:39:24,330
they would want to see that promises improving
outcomes in pediatric malnutrition or prevents

1585
02:39:24,330 --> 02:39:29,279
pediatric malnutrition and thereby improve
outcomes; which was the basic premise of our

1586
02:39:29,279 --> 02:39:36,540
workshop. So their ideas on research opportunities
and gaps, how to close some of the gaps that

1587
02:39:36,540 --> 02:39:42,010
were identified in the workshop so far would
be also very helpful. So I look forward to

1588
02:39:42,010 --> 02:39:51,081
a semi-structured but hopefully a, a generous
and exciting engaging discussion. So let's

1589
02:39:51,081 --> 02:39:55,710
kick this off. Once again for those of you
who are participating, if you look at the

1590
02:39:55,710 --> 02:40:00,491
box on your screen for questions, we would
love for you to send in your questions, your

1591
02:40:00,491 --> 02:40:04,960
thoughts, your comments along the lines of
anything you heard in the workshop specifically

1592
02:40:04,960 --> 02:40:11,760
related to the future studies that you would
like to see. At this point, I would let my

1593
02:40:11,760 --> 02:40:18,149
co-moderators Todd and David say hello to
everybody.

1594
02:40:18,149 --> 02:40:24,830
Alright, David waved and Todd waved, thank
you, welcome. Let's kick this off. I think

1595
02:40:24,830 --> 02:40:29,880
I'm going to start as I promised, asking each
of the panelists to spend some time to kick

1596
02:40:29,880 --> 02:40:36,510
us off in relation to what they think would
be a moonshot study that they would like to

1597
02:40:36,510 --> 02:40:45,140
see, a pivotal study, a classic study. And
I will start this with, let's start with David

1598
02:40:45,140 --> 02:40:46,350
Suskind.

1599
02:40:46,350 --> 02:40:55,640
DR. DAVID SUSKIND: Yeah, there are so many studies
that I think are needed in the world of nutrition

1600
02:40:55,640 --> 02:41:04,760
and pediatrics that would be pivotal, and
it's hard to say one, but my focus of interest

1601
02:41:04,760 --> 02:41:11,990
is inflammatory bowel disease. And I think
as many of us know, this is a condition that

1602
02:41:11,990 --> 02:41:20,510
has increased rather rapidly over the last
50 plus years. And nutrition and diet are

1603
02:41:20,510 --> 02:41:32,880
integrated into the IBD paradigm because of
the microbiome and the effect of diet on the

1604
02:41:32,880 --> 02:41:44,010
overall health. And so the moonshot study
that I would love to see done would actually

1605
02:41:44,010 --> 02:41:55,550
be multiple, but would be looking at how diet
impacts disease and disease development, looking

1606
02:41:55,550 --> 02:42:04,240
at not only something called exclusive enteral
nutrition which we know works amazingly well

1607
02:42:04,240 --> 02:42:13,540
for inflammatory bowel disease, not only improving
nutritional outcomes but also actually healing

1608
02:42:13,540 --> 02:42:23,600
the disease itself. And then looking at that
in terms of how a whole foods diet impacts

1609
02:42:23,600 --> 02:42:34,390
outcomes. And I think as we all know, there
is not enough resources, there are not enough

1610
02:42:34,390 --> 02:42:41,340
researchers looking at the impact of nutrition
in these types of conditions.

1611
02:42:41,340 --> 02:42:49,010
And so really providing those resources so
we could do those prospective interventional

1612
02:42:49,010 --> 02:42:58,899
nutritional studies in inflammatory bowel
disease itself. And that would be my desire

1613
02:42:58,899 --> 02:43:06,229
because there are so many different dietary
interventions that could impact disease, but

1614
02:43:06,229 --> 02:43:14,439
they're just not being done as of yet because
the resources are currently not there for us.

1615
02:43:14,439 --> 02:43:20,220
DR. NILESH MEHTA: Thanks, David. Very helpful. David,
can I ask you what outcome would be your most

1616
02:43:20,220 --> 02:43:26,790
priority or what couple of outcomes would
you be most interested in in such a study?

1617
02:43:26,790 --> 02:43:33,890
DR. DAVID SUSKIND: So definitely look at the nutritional
improvement in this population. There have

1618
02:43:33,890 --> 02:43:42,210
been two interventions in inflammatory bowel
disease that have been shown to improve nutritional

1619
02:43:42,210 --> 02:43:48,350
outcomes in this population. One is exclusive
of enteral nutrition, the other are fairly

1620
02:43:48,350 --> 02:43:58,850
strong immunosuppressants biologics, anti-TNFs
specifically. And so we want to look at that,

1621
02:43:58,850 --> 02:44:08,930
but we would also wanna look at disease
activity. We'd like to look at the microbiome

1622
02:44:08,930 --> 02:44:19,350
and the metabolome which we believe are instigators
of the disease itself because by looking at

1623
02:44:19,350 --> 02:44:25,870
the nutritional intervention, we actually
are getting closer to looking at what the

1624
02:44:25,870 --> 02:44:35,109
etiopathogenesis of inflammatory bowel disease
is and getting our understanding of why this

1625
02:44:35,109 --> 02:44:48,340
occurs in patients. We still don't have 100%
understanding of why IBD occurs, but we do

1626
02:44:48,340 --> 02:44:57,050
know that an intervention that does not use
immunosuppressants is this nutritional therapy.

1627
02:44:57,050 --> 02:45:05,189
And so we can actually understand the etiopathogenesis
by studying exclusive enteral nutrition and

1628
02:45:05,189 --> 02:45:12,470
other dietary barriers for this population.
So it would be multifocal in terms of the

1629
02:45:12,470 --> 02:45:18,439
outcomes we would want to look at, but definitely
would want to look at the presumed trigger of

1630
02:45:18,439 --> 02:45:19,439
the disease itself.

1631
02:45:19,439 --> 02:45:27,250
SPEAKER: So can I just, thank you and ask
you to think even bigger than this because

1632
02:45:27,250 --> 02:45:33,100
if your nutritional status improves but you
don't survive, I don't really care about that.

1633
02:45:33,100 --> 02:45:40,390
So beyond these things that are very integral
and very important even if just as markers

1634
02:45:40,390 --> 02:45:48,140
for severity of illness, what beyond that
would you like us to shoot for in terms of

1635
02:45:48,140 --> 02:45:56,310
the best outcomes for kids? Cognitive function
growth, quality of life, activity levels,

1636
02:45:56,310 --> 02:46:00,960
broken bones in their 70s, really think, think
bigly here.

1637
02:46:00,960 --> 02:46:07,180
DR. DAVID SUSKIND: No, I agree a thousand fold.
I mean, we really need to take a holistic

1638
02:46:07,180 --> 02:46:18,140
approach to our evaluation and what outcomes
that we really want to. From a clinician standpoint,

1639
02:46:18,140 --> 02:46:29,380
outside of my actual research, really, I need
to look at quality of life, looking at nutritional

1640
02:46:29,380 --> 02:46:41,391
parameters, including growth, looking at development,
these are all integral for our understanding.

1641
02:46:41,391 --> 02:46:47,510
And I think one of the difficult aspects of
doing nutritional research is that there are

1642
02:46:47,510 --> 02:46:56,580
so many endpoints that we really want to examine
and trying to get a platform that would incorporate

1643
02:46:56,580 --> 02:47:04,220
all of those. One of the things that I've
been working on of late is using digital technology

1644
02:47:04,220 --> 02:47:13,430
to really understand the whole world with
the whole patient in terms of psychosocial,

1645
02:47:13,430 --> 02:47:22,130
in terms of disease activity, in terms of
integrating that biomarkers that we look for

1646
02:47:22,130 --> 02:47:31,330
in both nutritional outcomes and disease activity
and trying to integrate that into platform.

1647
02:47:31,330 --> 02:47:39,450
But you're right. I mean, we do need to look
at this as a kind of a holistic evaluation

1648
02:47:39,450 --> 02:47:46,290
of the patient.
So looking at all the parameters that we can.

1649
02:47:46,290 --> 02:47:51,200
DR. NILESH MEHTA: Thanks, David. Let's go to Chris
Duggan, Chris.

1650
02:47:51,200 --> 02:47:56,800
DR. CHRISTOPHER DUGGAN: Thanks Nilesh, thanks
the organizers of this really exciting conference.

1651
02:47:56,800 --> 02:47:58,760
Is there an echo or is that to me?

1652
02:47:58,760 --> 02:48:02,420
SPEAKER: Everybody should maybe mute.

1653
02:48:02,420 --> 02:48:08,220
DR. CHRISTOPHER DUGGAN: And whenever I hear the
term moonshot, of course, as a Massachusetts

1654
02:48:08,220 --> 02:48:12,720
native, I'm reminded of John F. Kennedy's
approach, right? And anyone who's flown into

1655
02:48:12,720 --> 02:48:18,359
Logan Terminal E will see the quote. And as
we all know, he said we're going to go to

1656
02:48:18,359 --> 02:48:24,630
the moon, not because it's easy, because it's
hard. And indeed when you think about what

1657
02:48:24,630 --> 02:48:30,640
a moonshot should be, it should be hard. And
in the field of pediatric nutrition, I think

1658
02:48:30,640 --> 02:48:36,800
one of the things that makes me think what
a moonshot would be would be looking very

1659
02:48:36,800 --> 02:48:42,390
critically at what the important populations
are at risk of severe nutritional problems

1660
02:48:42,390 --> 02:48:48,950
or and as maybe we'll talk about later I view
it, I think this has to be thought of globally,

1661
02:48:48,950 --> 02:48:55,050
not just in one set of developed economies,
but perhaps more broadly. But suffice it to

1662
02:48:55,050 --> 02:49:01,350
say we do know that children who have...are
subject to inflammatory or infectious stresses

1663
02:49:01,350 --> 02:49:09,439
are especially susceptible to malnutrition.
And certainly, we know in our country the

1664
02:49:09,439 --> 02:49:14,110
burden of prematurity and congenital anomalies,
specifically congenital heart disease, those

1665
02:49:14,110 --> 02:49:20,470
are two populations that we really have plenty
of patients with those conditions that it

1666
02:49:20,470 --> 02:49:27,840
shouldn't be hard with enough resources to
design a multicenter interventional trial

1667
02:49:27,840 --> 02:49:33,390
in which the sample size would be large enough
so that significant conclusions can be made

1668
02:49:33,390 --> 02:49:37,939
on the interventions.
And as if to kind of illustrate the fact that

1669
02:49:37,939 --> 02:49:42,510
we're in such a sad state of affairs now,
you only have to think about going up into

1670
02:49:42,510 --> 02:49:47,340
your local NICU and realizing that ten years
ago, we had one form of intravenous fat and

1671
02:49:47,340 --> 02:49:52,430
now we have three forms of intravenous fat,
and no one can tell us whether one is superior

1672
02:49:52,430 --> 02:49:58,240
than the other. They've been brought to market
with short term studies, no long term outcomes,

1673
02:49:58,240 --> 02:50:03,710
no long term growth outcomes, there are few,
I should say, and that's just a great example

1674
02:50:03,710 --> 02:50:12,670
of an opportunity lost. So that would be my
moonshot. Pick a population of children at

1675
02:50:12,670 --> 02:50:18,630
risk of significant malnutrition. And obviously,
maybe people here aren't so interested in

1676
02:50:18,630 --> 02:50:24,700
the effects of nutrition in other countries
but clearly children with pneumonia, diarrhea,

1677
02:50:24,700 --> 02:50:30,090
malaria and others are at risk of significant
malnutrition globally. And indeed, studies

1678
02:50:30,090 --> 02:50:36,060
are ongoing which I'm a part of, one to investigate
in interventional, multi-country way what

1679
02:50:36,060 --> 02:50:40,470
approach makes the best sense. But here in
the United States, it seems to me the burden

1680
02:50:40,470 --> 02:50:45,630
of pediatric nutrition has to focus on premature
infants and children with anomalies such as

1681
02:50:45,630 --> 02:50:46,950
congenital heart disease.

1682
02:50:46,950 --> 02:50:54,609
SPEAKER: Just a comment. [Cross talk - Thanks, Chris.] There
have been speakers talking about sort of you

1683
02:50:54,609 --> 02:51:00,649
know, there is no lack of food insecurity
in the U.S. And while this conference was sort

1684
02:51:00,649 --> 02:51:09,450
of more about malnutrition in sick people,
there is you can't ignore, I think, that continuum

1685
02:51:09,450 --> 02:51:16,130
of how they enter into their illness and how
well nourished or not nourished they are to

1686
02:51:16,130 --> 02:51:27,270
begin with. So, yeah, that's good stuff. Thanks.
DR. TODD RICE: Chris, do you think that we know or have some

1687
02:51:27,270 --> 02:51:33,640
idea of which patients in those populations
are the patients that we need to aggressively

1688
02:51:33,640 --> 02:51:40,760
intervene in? And potentially, is the intervention
different for different subgroups in those populations? 

1689
02:51:40,760 --> 02:51:46,160
DR. CHRISTOPHER DUGGAN: Yeah, it could well be. And again,
I'm not a neonatologist, but it seems to me that

1690
02:51:46,160 --> 02:51:49,149
those children... well, certainly the cardiologists
will tell you that the children in single

1691
02:51:49,149 --> 02:51:55,410
ventricle physiology would be the ones at
highest risk of poor growth in neurodevelopment

1692
02:51:55,410 --> 02:52:02,410
over the first years of life. And in the NICU
patients, obviously, those children with extreme

1693
02:52:02,410 --> 02:52:07,200
prematurity, those children with either congenital
anomalies or severe bronchopulmonary dysplasia

1694
02:52:07,200 --> 02:52:12,430
would be highly likely to suffer from nutritional
problems as well.

1695
02:52:12,430 --> 02:52:18,640
But here is where and that question really
illustrates an important point. And David's

1696
02:52:18,640 --> 02:52:23,500
point about, about IBD illustrates this too,
in my mind, which is that if we're designing

1697
02:52:23,500 --> 02:52:27,739
nutrition study but don't take into consideration
that new biological therapies are going to

1698
02:52:27,739 --> 02:52:32,100
be rolled out for IBD and don't take that
into consideration or don't think about new

1699
02:52:32,100 --> 02:52:37,010
surgical therapies for single ventricle physiology,
then we'll be really scuppered. You know,

1700
02:52:37,010 --> 02:52:41,470
we really need to develop these and design
these studies with people who don't think

1701
02:52:41,470 --> 02:52:47,830
of themselves as nutrition scientists, but
those who are content experts as well. Yeah, great point.

1702
02:52:47,830 --> 02:52:53,680
DR. WILLIAM EVANS: And I'll be talking a little
bit about our study in premature infants,

1703
02:52:53,680 --> 02:52:59,431
but it might be a good time just... So I said,
we published an article called D-3 Creatine

1704
02:52:59,431 --> 02:53:05,149
Dilution for the Noninvasive Measurement of
Skeletal Muscle Mass and Premature Infants.

1705
02:53:05,149 --> 02:53:19,489
We had about 90 infants from a birth weight
of 550g to 4.2kg, and we measured them while

1706
02:53:19,489 --> 02:53:27,370
they were in the NICU and 20 of our patients
weighed less than 1,000g at the start of the

1707
02:53:27,370 --> 02:53:30,770
study.
So I'll be speaking a lot about that. But

1708
02:53:30,770 --> 02:53:38,560
it was my first foray into pediatrics and
especially into neonatology. And it actually

1709
02:53:38,560 --> 02:53:47,149
it turns out it's pretty easy to measure lean
body mass and muscle mass by taking urine

1710
02:53:47,149 --> 02:53:56,600
from a diaper. DR. NILESH MEHTA: Thank you, Bill. That's very
timely. Bill, in fact, I'll jump to you so

1711
02:53:56,600 --> 02:54:01,319
that we can have some continuity right after
this. But one of the things that Chris brings

1712
02:54:01,319 --> 02:54:07,760
up and related to Todd's question and David's
question, a David Seres question before during

1713
02:54:07,760 --> 02:54:14,740
the day. When you talk about doing studies
in a specific population because you believe

1714
02:54:14,740 --> 02:54:21,569
that that's where the burden of the disease
and therefore the potential for the intervention

1715
02:54:21,569 --> 02:54:27,580
being effective or efficacious or having a
signal is more that lends itself to a certain

1716
02:54:27,580 --> 02:54:33,100
study design; and you can actually, with a
multi-center, get large numbers. But the concept

1717
02:54:33,100 --> 02:54:37,640
in recent years has been one of enrichment
where whether do you have tools that will

1718
02:54:37,640 --> 02:54:43,739
help you identify which ones, even among those
are more likely to benefit from the intervention?

1719
02:54:43,739 --> 02:54:49,310
And, you know, Todd and David in their world
have addressed that with the new trick score

1720
02:54:49,310 --> 02:54:56,310
and similar ones. But it could be something
as simple as marker of underlying nutritional

1721
02:54:56,310 --> 02:55:02,149
status. But any anyone want to comment on
how to find that population where you are

1722
02:55:02,149 --> 02:55:08,189
more than likely for the intervention to benefit
because we now have a graveyard full of negative

1723
02:55:08,189 --> 02:55:12,200
randomized controlled trials where we imagine
everybody in my hospital or everybody in this

1724
02:55:12,200 --> 02:55:16,819
population should benefit from a therapy and
they don't. And therefore the summative result

1725
02:55:16,819 --> 02:55:25,690
is negative. DR. TODD RICE: You know, I think in the adult
world, we're starting to get a little bit

1726
02:55:25,690 --> 02:55:31,100
of an understanding of there are patients
that are and I think David Seres may have

1727
02:55:31,100 --> 02:55:37,319
coined this term, have refractory malnutrition
which means that they are truly malnourished,

1728
02:55:37,319 --> 02:55:42,130
but they also are malnourished to the point
that we can't intervene, at least in their

1729
02:55:42,130 --> 02:55:46,590
current state. And there's a really nice article
not in critically ill patients which is what

1730
02:55:46,590 --> 02:55:51,729
I do, but in hospitalized patients otherwise,
that if you have high inflammatory markers

1731
02:55:51,729 --> 02:55:55,899
in the adult world, that it actually doesn't
improve your outcomes.

1732
02:55:55,899 --> 02:56:01,550
If we provide any meet malnourished criteria,
if we provide nourishment. But if you have

1733
02:56:01,550 --> 02:56:07,069
less inflammatory, high inflammatory markers,
then, additional nutritional support may improve

1734
02:56:07,069 --> 02:56:12,529
your outcomes. And so the question I think
I always have is we, we treat lots of groups

1735
02:56:12,529 --> 02:56:18,640
of patients as the same, but there may be
patients that the nutritional intervention

1736
02:56:18,640 --> 02:56:24,221
has to be part of a different treatment plan,
i.e. you need an anti-inflammatory and a nutritional

1737
02:56:24,221 --> 02:56:29,140
intervention. Otherwise, the patient's so
inflamed that they're just not going to be able

1738
02:56:29,140 --> 02:56:34,390
to utilize the nutritional support that we
give them. 

1739
02:56:34,390 --> 02:56:42,340
DR. WILLIAM EVANS: I think that's a really important question. I'm also part of the International
Cachexia Society looking at changes in body

1740
02:56:42,340 --> 02:56:47,920
composition, particularly muscle mass in ill
people and the patient population where you

1741
02:56:47,920 --> 02:56:55,650
get the greatest loss of muscle of any patient
population are older people in the ICU. And

1742
02:56:55,650 --> 02:57:02,420
part of the problem that is not so well recognized
is that they all have some degree of enteral

1743
02:57:02,420 --> 02:57:07,210
feeding intolerance.
And what that is, is actually related to is

1744
02:57:07,210 --> 02:57:13,939
unclear. But almost certainly the raging inflammation
that they have. So treating with potential,

1745
02:57:13,939 --> 02:57:22,360
you know, anti-inflammatory enteral feeding
formulas might be appropriate. But as I just

1746
02:57:22,360 --> 02:57:28,300
heard, someone said, you know, we don't really
know how well they work. They're marketed as

1747
02:57:28,300 --> 02:57:33,729
anti inflammatory enteral feeding formulas,
but the degree to which they actually affect

1748
02:57:33,729 --> 02:57:43,479
inflammation is really unknown. SPEAKER: I would also
point to a long littered path of anti-inflammatory

1749
02:57:43,479 --> 02:57:52,380
attempts at anti-inflammatory therapies in
critically ill patients that have proven either

1750
02:57:52,380 --> 02:57:58,990
no effect or be harmful. Some of us are old
enough to remember growth hormone in the ICU

1751
02:57:58,990 --> 02:58:08,410
or, you know, so that, you know, obviously
these would have to be tested in a randomized setting.

1752
02:58:08,410 --> 02:58:17,310
DR. WILLIAM EVANS: I think that Chris brought up the
area perhaps of maybe the greatest impact

1753
02:58:17,310 --> 02:58:24,180
because the nutritional status of neonates
in a NICU is literally all over the place.

1754
02:58:24,180 --> 02:58:29,060
And what is the best marker of nutrition status
and maybe even more importantly, what is the

1755
02:58:29,060 --> 02:58:37,060
best indicator of healthy or healthy growth?
SPEAKER: I'll come back to this when it's my turn. But

1756
02:58:37,060 --> 02:58:46,069
just an answer. Somebody was talking about
the effort trial which is where this inflammation

1757
02:58:46,069 --> 02:58:54,010
data came from. The answer to how to go find
it really, I think we should go back to the

1758
02:58:54,010 --> 02:58:59,721
basics and get simple because in the effort
trial, they used a very simple thing that

1759
02:58:59,721 --> 02:59:06,910
mostly focused on weight and intake. And then
I think one other thing and, and that predicted

1760
02:59:06,910 --> 02:59:13,550
response to nourishment. And so I think that
that the problem is that we've gotten way

1761
02:59:13,550 --> 02:59:21,490
ahead of ourselves on this. That malnutrition
means something, it's certainly predicts poor

1762
02:59:21,490 --> 02:59:27,120
outcomes, but it doesn't predict who gets
response to nourishment and that I think it's

1763
02:59:27,120 --> 02:59:33,470
prompted me to actually request in many venues
that we call it something else. I have proposed

1764
02:59:33,470 --> 02:59:40,540
Matilda; that hasn't gone over well. But the
problem is that if somebody has malnutrition,

1765
02:59:40,540 --> 02:59:48,989
to say that they're malnourished it may not
be accurate because they may be perfectly

1766
02:59:48,989 --> 02:59:53,870
well nourished, it's just that they're wasting
from inflammation or they're wasting from

1767
02:59:53,870 --> 02:59:56,390
disuse or they're wasting from underlying
disease. So I'll talk more when it's my turn. 

1768
02:59:56,390 --> 03:00:04,590
DR. WILLIAM EVANS: Well,
there were some really famous studies done

1769
03:00:04,590 --> 03:00:12,760
in the late 70s, early 80s in Boston using
TPN to feed the critically ill patients. You

1770
03:00:12,760 --> 03:00:17,830
can maintain their body weight, I mean through
TPN but their muscle mass continues to decline.

1771
03:00:17,830 --> 03:00:23,920
DR. CHRISTOPHER DUGGAN: It was all fat. And they were getting 4,000 calories a day
intravenously. And that was the proof of principle.

1772
03:00:23,920 --> 03:00:31,230
Shouldn't be called hyperpigmentation. Right.
DR. NILESH MEHTA: Great point. I think (UNKNOWN) was now in

1773
03:00:31,230 --> 03:00:37,960
Boston but wasn't there before talked about
CRP as a marker of switching to ableism and

1774
03:00:37,960 --> 03:00:43,689
don't bother until it starts dropping. And
this was in the 80s and in the surgical world,

1775
03:00:43,689 --> 03:00:47,250
people talked about inflammation. And then,
of course, Gordon Jensen in the adult section

1776
03:00:47,250 --> 03:00:52,729
reminded us of what he's been saying. So this
is spot on. Bill, why don't you jump in, where

1777
03:00:52,729 --> 03:00:58,000
are you going? You talked a little bit about
the deuterated creatinine levels and your

1778
03:00:58,000 --> 03:01:03,460
efforts to now have a solid marker of muscle
mass. But where do you want to see that go in

1779
03:01:03,460 --> 03:01:10,229
terms of a moonshot study?
DR. WILLIAM EVANS: Well, I can say right now that we have ten

1780
03:01:10,229 --> 03:01:22,260
NIH funded grants using this method in adults.
And yet I think where it really needs to be

1781
03:01:22,260 --> 03:01:28,600
used is in infants and children because the
body composition data for infants and children

1782
03:01:28,600 --> 03:01:39,140
is woefully inadequate. So we've just published
a paper in boys with Duchenne dystrophy using

1783
03:01:39,140 --> 03:01:44,930
this method. And what this method does without
if you want to know the details, you can go back

1784
03:01:44,930 --> 03:01:52,460
and look at the PowerPoint presentation I've
submitted. It measures functional muscle mass

1785
03:01:52,460 --> 03:01:58,290
because creatine and creatine phosphate is
co-located with the contractual components

1786
03:01:58,290 --> 03:02:05,640
of muscle. So it's undiluted by water, it's
undiluted by fat, it's undiluted by fibrotic

1787
03:02:05,640 --> 03:02:14,471
tissue. And then and we compared boys with
DMD to healthy controls, healthy boys. And

1788
03:02:14,471 --> 03:02:21,500
we have some teenage boys and one 15 year old
non ambulant boy had less than 5% of his body

1789
03:02:21,500 --> 03:02:30,130
weight as muscle. So this is the first kind
of very simple. When you talk about simple,

1790
03:02:30,130 --> 03:02:35,640
you know, measuring the muscle mass that's
been widely available and the two studies

1791
03:02:35,640 --> 03:02:43,510
that I would be most interested in pediatrics.
One is, in talking with my NICU cohorts is

1792
03:02:43,510 --> 03:02:53,720
very low birth weight babies tend to develop
neurocognitive problems by the time they're

1793
03:02:53,720 --> 03:03:02,350
two. And the indicators of neurocognitive
status are not great. And the measures of

1794
03:03:02,350 --> 03:03:11,170
growth or healthy growth are not great and
the ability to supply extra high quality protein

1795
03:03:11,170 --> 03:03:17,351
for these babies while they're in the NICU
is also not great because you have to obviously

1796
03:03:17,351 --> 03:03:23,359
give it along with breast milk. So that's
one. I would be really interested in kind

1797
03:03:23,359 --> 03:03:28,680
of understanding the role of changes in muscle
mass because we can measure muscle mass every

1798
03:03:28,680 --> 03:03:34,770
two weeks if we want. You can measure it however
you want. It's completely noninvasive and

1799
03:03:34,770 --> 03:03:41,279
it's relationship to neurocognitive changes
because I maintain that establishing muscle

1800
03:03:41,279 --> 03:03:47,930
mass which is a heavily myelinated tissue
is absolutely related to the changes that

1801
03:03:47,930 --> 03:03:54,140
occur in the central nervous system. And then
secondly, and probably the big moonshot, we

1802
03:03:54,140 --> 03:04:01,770
know that body mass index of body weight in
infants and children is highly associated

1803
03:04:01,770 --> 03:04:08,989
with long term outcomes.
Obesity being one and chronic diseases being

1804
03:04:08,989 --> 03:04:14,120
two and what component of body mass index
is most important? We've assumed that it's

1805
03:04:14,120 --> 03:04:20,590
fat. I maintain that that's probably not the
case, that maybe muscle. Muscle is an extremely

1806
03:04:20,590 --> 03:04:30,060
high energy requiring tissue. It is the reason
why the basal metabolic rate is different

1807
03:04:30,060 --> 03:04:36,960
between boys and girls and men and women,
and changes with age and changes with growth.

1808
03:04:36,960 --> 03:04:42,160
So I would be really interested in knowing
what the body composition and I would predict

1809
03:04:42,160 --> 03:04:52,010
muscle is associated with these longer term
outcomes related to obesity and and the sequelea

1810
03:04:52,010 --> 03:04:59,319
of events following that. (CROSSTALK)
DR. NILESH MEHTA: How easy is this to do in living subjects, your technique?

1811
03:04:59,319 --> 03:05:05,680
DR. WILLIAM EVANS: We just finished a study in
rural Bangladesh using the method, so it's

1812
03:05:05,680 --> 03:05:12,899
easy. SPEAKER: How about in patients with advanced
renal failure? DR. WILLIAM EVANS: Yeah, as long as they can produce

1813
03:05:12,899 --> 03:05:19,800
some urine, it's fine. We look at the enrichment
of creatinine, so the method uses a very small

1814
03:05:19,800 --> 03:05:26,100
dose of deuterated creatine either encapsulated
or for children we dissolve it in a liquid

1815
03:05:26,100 --> 03:05:29,350
and we can dissolve it in a little small amount
of heavy water.

1816
03:05:29,350 --> 03:05:37,140
And we can get from that lean body mass, fat
mass, muscle mass from a simple sip of water,

1817
03:05:37,140 --> 03:05:42,760
and then we take a saliva and urine sample
a little bit later on. And as long as someone

1818
03:05:42,760 --> 03:05:46,980
produces some urine, we're looking at the
enrichment of creatine and not the amount

1819
03:05:46,980 --> 03:05:52,370
of creatine that's excreted. So as I said,
we're doing studies in some really sick people

1820
03:05:52,370 --> 03:05:57,660
right now. SPEAKER: And just be curious to see if you
could do it out of the dialysis or something?

1821
03:05:57,660 --> 03:06:04,580
DR. WILLIAM EVANS: Well, it would be a little tricky to be honest.
That would be one of the populations. You

1822
03:06:04,580 --> 03:06:11,970
have to take it from the blood sample rather
than the dialysis. But that's... it's cool,

1823
03:06:11,970 --> 03:06:21,040
DR. NILESH MEHTA: Thanks Bill. Let’s keep moving. DR. WILLIAM EVANS: We've outdated against a body MRI. DR. LIAM MCKEEVER: OK, wonderful. DR. TAMARA HANNON: Well, this discussion

1824
03:06:21,040 --> 03:06:29,000
is... Do you want me to go ahead? DR. NILESH MEHTA: Yes, please,
yeah.

1825
03:06:29,000 --> 03:06:37,060
DR. TAMARA HANNON: So I'm a pediatric endocrinologist,
so I often see these kids that are fed in

1826
03:06:37,060 --> 03:06:44,180
the NICU when they have metabolic problems
later in life. So I completely agree that

1827
03:06:44,180 --> 03:06:58,160
measures of body fat and growth in the newborn
period are, are not what we should rely unnecessarily.

1828
03:06:58,160 --> 03:07:05,740
I don't know what they are, but to indicate
the best health. In fact, if you have to overfeed

1829
03:07:05,740 --> 03:07:12,410
a person for them to store fat as adipose,
that's probably not metabolically normal.

1830
03:07:12,410 --> 03:07:20,239
Like that's not biologically normal. It's
biologically normal for humans to store fat

1831
03:07:20,239 --> 03:07:29,020
without being overfed. So I spend a lot of
my time thinking about where to intervene

1832
03:07:29,020 --> 03:07:37,610
in the cycle of diabetes which you might not
think of as a condition of malnourishment.

1833
03:07:37,610 --> 03:07:48,120
However, you know, it can be considered malnourishment
in the way that most people with diabetes

1834
03:07:48,120 --> 03:07:57,689
are not getting a diet, a nourishing diet,
a diet that's relatively high in vitamins

1835
03:07:57,689 --> 03:08:05,850
and minerals in relationship to the calories.
But what I would say from my moonshot study,

1836
03:08:05,850 --> 03:08:14,760
here I go. This is out there.
But you all know what WIC is? WIC is this

1837
03:08:14,760 --> 03:08:21,729
special supplemental nutrition program for
women, infants and children. And WIC serves

1838
03:08:21,729 --> 03:08:30,930
more than 50% of the infants in the U.S. It
is a huge program. And we know that babies

1839
03:08:30,930 --> 03:08:39,260
and young women, women of childbearing age
who get WIC services have better health outcomes.

1840
03:08:39,260 --> 03:08:46,050
We know they have lower rates of obesity.
We know that they're better nourished. It's

1841
03:08:46,050 --> 03:08:57,149
very burdensome to sign up for WIC. So the
first thing I would propose is an implementation

1842
03:08:57,149 --> 03:09:06,170
study to allow all nutritionally at risk women,
even if they're not pregnant, even pre pregnancy,

1843
03:09:06,170 --> 03:09:15,319
why shouldn't an 18 year old who's going to become
pregnant at age 21 already be on WIC if we

1844
03:09:15,319 --> 03:09:23,189
want to prevent bad outcomes? So the implementation
of an easier way for young women of child

1845
03:09:23,189 --> 03:09:31,470
bearing age to access a evidence-based program
that is managed by the states in many different

1846
03:09:31,470 --> 03:09:36,890
WIC offices. So implementing an easier way
to get on this and serving women who are at

1847
03:09:36,890 --> 03:09:45,130
high risk this lends itself to a multitude
of studies of the biology of what happens

1848
03:09:45,130 --> 03:09:53,960
when you try to nourish women prior to a pregnancy
or during a pregnancy, and what happens to

1849
03:09:53,960 --> 03:09:58,750
them after pregnancy and what happens to their
infants.

1850
03:09:58,750 --> 03:10:07,700
So that's my Moonshot, billions of dollars study
design. Now, of course, so that we could we

1851
03:10:07,700 --> 03:10:18,210
could suggest anything that we want and not
be limited to practicalities. But what if

1852
03:10:18,210 --> 03:10:26,510
you did make WIC available to anyone, any
woman who was of childbearing age? Not everyone

1853
03:10:26,510 --> 03:10:33,200
would sign up for WIC, right? What if you
made it really easy? Would still half the

1854
03:10:33,200 --> 03:10:38,970
people sign up for WIC, would even half the
people sign up for WIC? If we know it works

1855
03:10:38,970 --> 03:10:48,130
and we know it lends itself to better outcomes
and we know who signs up for WIC, and we can

1856
03:10:48,130 --> 03:10:54,720
study what happens to those women and what
happens to their babies and what they eat?

1857
03:10:54,720 --> 03:11:02,820
What they eat so we can study actually what's
distributed to them. That seems about as close

1858
03:11:02,820 --> 03:11:07,319
as what I'd want to do for the rest of my career
as anything.

1859
03:11:07,319 --> 03:11:16,931
DR. NILESH MEHTA: That's a great one. Tamara you're addressing
both the mother and child almost at a family

1860
03:11:16,931 --> 03:11:23,399
level intervention and an intervention which
at a system level is far out, far reaching

1861
03:11:23,399 --> 03:11:30,210
and implementation is well said, well-meaning
interventions such as this still reach only

1862
03:11:30,210 --> 03:11:41,350
a certain fraction. I like it, it's a very,
it's not easy as you said, but it's far reaching

1863
03:11:41,350 --> 03:11:52,360
and hopefully there will be support to address
that. Thank you. We'll come back. We'll have

1864
03:11:52,360 --> 03:11:57,530
more. It's Vijay's turn. Vijay, you want to
jump in and give us your moonshot study?

1865
03:11:57,530 --> 03:12:03,000
DR. VIJAY SRINIVASAN: Oh, sure. Thank you. This
is an absolutely wonderful discussion to be

1866
03:12:03,000 --> 03:12:10,620
a part of, and I am just so excited and stimulated
by all the discussions we're having. In fact,

1867
03:12:10,620 --> 03:12:14,970
Moonshot is interesting because I'm just in
the middle of our towards the end of finishing

1868
03:12:14,970 --> 03:12:21,510
my fourth or fifth read of Apollo 13, which
I always sort of am inspired by as probably

1869
03:12:21,510 --> 03:12:27,989
the most successful failure and a good metaphor
for a lot of what we do in our research and

1870
03:12:27,989 --> 03:12:33,930
enterprise. So many...lots of hard work, some
successes, but a good deal of failure. But

1871
03:12:33,930 --> 03:12:38,600
the more important message that I'm always
reminded by that book when I read it every

1872
03:12:38,600 --> 03:12:45,020
time is how human ingenuity overcame adversity
through the constant sort of adaptation and

1873
03:12:45,020 --> 03:12:50,689
sort of tinkering and constant course corrections
throughout the whole journey to bring the

1874
03:12:50,689 --> 03:12:55,740
astronauts back home safely. So using that
metaphor, keeping in mind that I'm a pediatric

1875
03:12:55,740 --> 03:13:03,640
intensivist and I deal with the full continuum
from acute illness all the way through survivorship,

1876
03:13:03,640 --> 03:13:09,410
through chronic technology dependence, which
is now my latest sort of place of focus.

1877
03:13:09,410 --> 03:13:16,900
It is easy or hard, depending on how you look
at it. And I am struck by the fact that I

1878
03:13:16,900 --> 03:13:21,420
deal with noise. There is noise from critical
illness, there's noise from interventions,

1879
03:13:21,420 --> 03:13:26,670
there's noise from all the toxicities. And
I'm trying to do an intervention that will

1880
03:13:26,670 --> 03:13:32,210
give me a signal that's big enough. And usually,
I feel because it is so noisy, the intervention

1881
03:13:32,210 --> 03:13:39,100
has to be a massive thump to really generate
that signal. And perhaps 50 years ago our

1882
03:13:39,100 --> 03:13:44,300
science and care was not as good and we had
better signals from smaller interventions

1883
03:13:44,300 --> 03:13:50,390
because we were sort of coming up with bigger
game changers. So in that difficult context,

1884
03:13:50,390 --> 03:13:55,830
I in fact, I'm sort of as I increasingly think
I kind of am thinking a little against conventional

1885
03:13:55,830 --> 03:13:59,350
wisdom in that should we be thinking about
one intervention or should we be thinking

1886
03:13:59,350 --> 03:14:05,500
about a bundle of interventions, recognizing
that this is about bringing the human, the

1887
03:14:05,500 --> 03:14:11,040
child of the adult back to health? So in other
words, nutrition coupled with sleep, with

1888
03:14:11,040 --> 03:14:18,140
physical rehab, with aspects of nurturing,
and how does one implement all that?

1889
03:14:18,140 --> 03:14:24,739
Put it together and it doesn't stop at the
point of the inpatient, as Tamara just said,

1890
03:14:24,739 --> 03:14:29,590
you have to continue it in the outpatient
setting and you have to continue the process.

1891
03:14:29,590 --> 03:14:36,029
So to me, as I think of this design, how do
I sort of bring all these elements of the

1892
03:14:36,029 --> 03:14:43,010
bundle together? How can I make sure that
practitioners are able to collect data, measure

1893
03:14:43,010 --> 03:14:50,270
what they're supposed to do accurately and
comparably outcomes such as body function,

1894
03:14:50,270 --> 03:14:56,260
composition, muscle mass, the metabolome,
the microbiome, neurocognition, development,

1895
03:14:56,260 --> 03:15:01,411
and more importantly, to continue this into
the families when they go back home. And how

1896
03:15:01,411 --> 03:15:07,109
do they sort of maintain and continue to improve
on their trajectory, recognizing that many

1897
03:15:07,109 --> 03:15:12,300
of the kids that come to the pediatric ICU
are repeat offenders, they often have chronic

1898
03:15:12,300 --> 03:15:17,260
comorbidities, whether they're NICU graduates,
whether they are congenital cardiac malformation,

1899
03:15:17,260 --> 03:15:22,811
whether they have neurological impairment.
It is a challenge to make sure that they can

1900
03:15:22,811 --> 03:15:27,010
sort of reach their full potential, which
is sort of it goes back to, again, my earlier

1901
03:15:27,010 --> 03:15:39,470
presentation about the sustainable goals and
what does it mean for us as an individual

1902
03:15:39,470 --> 03:15:40,470
and a population.

1903
03:15:40,470 --> 03:15:45,020
DR. NILESH MEHTA: Vijay, that was brilliant. I'm
not surprised that in your world, and critical

1904
03:15:45,020 --> 03:15:50,780
care, you have challenged the dogma of the
silver bullet because there isn't one. And

1905
03:15:50,780 --> 03:15:58,350
there are going to be bundled interventions
more and more. Well said, that's a very exciting

1906
03:15:58,350 --> 03:16:01,989
concept. Anybody want to chime in? Have
questions for Vijay?

1907
03:16:01,989 --> 03:16:09,810
DR. DAVID SUSKIND:  Now, I'd like to second that I think
especially in the world that I live in with

1908
03:16:09,810 --> 03:16:20,810
inflammatory bowel disease, it isn't just
one intervention that works to our children.

1909
03:16:20,810 --> 03:16:28,310
We are intervening from a lifestyle standpoint
as well as medication therapy, as well as

1910
03:16:28,310 --> 03:16:37,760
nutritional therapy, and understanding that
one thing by itself may not be that silver

1911
03:16:37,760 --> 03:16:44,310
bullet, but really incorporating individuals
I suppose, and then studying that, so you

1912
03:16:44,310 --> 03:16:52,601
are able to then kind of push the science
for or I think is central to all of the conditions

1913
03:16:52,601 --> 03:16:57,160
that we focus on. So kudos for that.

1914
03:16:57,160 --> 03:17:04,540
DR. NILESH MEHTA: Thank you. You know, we won't,
I won't elaborate on this, but this is very

1915
03:17:04,540 --> 03:17:10,380
similar to the concept of not just bundling
it, but bundling it in relation to the individual's

1916
03:17:10,380 --> 03:17:15,710
needs and the whole concept of precision nutrition
as well. This is a good time to go to Liam.

1917
03:17:15,710 --> 03:17:21,140
Liam, I don't want to put you on the spot,
and please feel free to go as deeper, as little

1918
03:17:21,140 --> 03:17:25,891
as you want. But you've heard of all these
considerations. You've heard of bundling interventions,

1919
03:17:25,891 --> 03:17:33,350
you heard of patient populations that are
likely to be most beneficial. They might be

1920
03:17:33,350 --> 03:17:38,260
tools that might allow you to identify as
someone in your position who gets to look

1921
03:17:38,260 --> 03:17:44,000
at the literature and try to make sense of
it in the form of guiding it, bringing it

1922
03:17:44,000 --> 03:17:49,771
back to the bedside. What kind of study? What
is your moonshot study design? Where does

1923
03:17:49,771 --> 03:17:55,510
this nutritional world going and what would
you like to see from researchers when you

1924
03:17:55,510 --> 03:17:58,660
sit down and assemble the next panel for guidelines
generation?

1925
03:17:58,660 --> 03:18:03,930
DR. LIAM MCKEEVER: I mean, we're always our own
best critic. And one of the criticisms I have

1926
03:18:03,930 --> 03:18:09,350
about the work I do on the guidelines is that
we are providing population estimates, and

1927
03:18:09,350 --> 03:18:15,870
I don't believe a population estimate is a
useful thing. I think that it is a major misleading

1928
03:18:15,870 --> 03:18:22,910
thing that I spend a lot of time working on.
But, so precision nutrition is actually a

1929
03:18:22,910 --> 03:18:28,790
good segway for what I would want to do. I
have two things. I want to demonstrate efficacy,

1930
03:18:28,790 --> 03:18:34,383
but I want to also be able to account for
the fact that we don't...that not everybody

1931
03:18:34,383 --> 03:18:38,899
is going to respond to nutrition in the exact
same way. I think malnutrition is one of the

1932
03:18:38,899 --> 03:18:50,560
starting points for that, right? So I'd like
to see an RCT where we demonstrate that intervening

1933
03:18:50,560 --> 03:18:58,189
upon our nutritional screenings or assessments
actually alters clinical outcome. And to do

1934
03:18:58,189 --> 03:19:02,950
that, you have to be looking at people who
are fed more than people who are randomized

1935
03:19:02,950 --> 03:19:08,670
to be fed less. And that could be standard
care versus supplemental PN or standard care

1936
03:19:08,670 --> 03:19:13,740
versus an enteral algorithm.
I think there are ways to do it that don't

1937
03:19:13,740 --> 03:19:20,180
involve starving someone and then just showing
that you'd have to have four groups. You'd

1938
03:19:20,180 --> 03:19:30,100
have to have people who were designated malnourished.
And then you look at the people who are malnourished

1939
03:19:30,100 --> 03:19:35,899
and high fed, malnourished and low fed and
then not malnourished, high fed, not malnourished,

1940
03:19:35,899 --> 03:19:46,949
low fed. And what you're looking for is does...
basically, does malnutrition modify the effect

1941
03:19:46,949 --> 03:19:52,010
of feeding an outcome? And you would get to
that through on interaction term. So it's

1942
03:19:52,010 --> 03:19:58,430
a very simple thing that we've never done
in RCT that I've seen, at least not in adults.

1943
03:19:58,430 --> 03:20:03,830
But I don't think you guys have that in kids
either. Then the other thing that I would

1944
03:20:03,830 --> 03:20:16,870
add to that is I would be taking genetics
data and I would also want to be able to look

1945
03:20:16,870 --> 03:20:25,430
at it stratified by the various different
illness states. So, as much as someone could

1946
03:20:25,430 --> 03:20:31,149
write one giant grant, I think we are not
served by having one researcher answer all

1947
03:20:31,149 --> 03:20:36,989
those questions. I would rather seek coordination
where there are several different principal

1948
03:20:36,989 --> 03:20:42,760
investigators who are all kind of working
together to answer these questions individually

1949
03:20:42,760 --> 03:20:49,020
at different hospitals with their own sensibilities.
But certain things held in place to make sure

1950
03:20:49,020 --> 03:20:55,010
that when I have to do my job, I can conflate
them all into a forest plot. So the interventions

1951
03:20:55,010 --> 03:21:00,279
were similarly administered, the outcomes
were similarly measured, things like that.

1952
03:21:00,279 --> 03:21:05,340
But I think that would be important too, to
not have it be this thing where forever we're

1953
03:21:05,340 --> 03:21:11,510
all just looking at rehashings of one giant
study that someone was allowed to do.

1954
03:21:11,510 --> 03:21:19,949
DR. CHRISTOPHER DUGGAN: But how do you avoid the problem
with 20 underpowered studies that probably

1955
03:21:19,949 --> 03:21:25,410
aren't ethical to accomplish? The power is
not there...DR. WILLIAM EVANS: Or even harder to find malnutrition.

1956
03:21:25,410 --> 03:21:29,050
DR. LIAM MCKEEVER: You do that with a summary
statistics. So that's what a meta-analysis

1957
03:21:29,050 --> 03:21:30,050
is about, right?

1958
03:21:30,050 --> 03:21:35,680
DR. CHRISTOPHER DUGGAN: ...is not an individual study with the
idea of, hey, someday my study is going to

1959
03:21:35,680 --> 03:21:36,720
be part of a meta-analysis.

1960
03:21:36,720 --> 03:21:42,109
DR. LIAM MCKEEVER: Well, it's not some day. It's
part of an organized effort. Everybody knows

1961
03:21:42,109 --> 03:21:49,800
up front that we've pulled these different
researchers in to do this massive undertaking.

1962
03:21:49,800 --> 03:21:54,199
DR. CHRISTOPHER DUGGAN: And to do it as a multi-test, so you're
arguing for a multicenter trial?

1963
03:21:54,199 --> 03:21:59,580
DR. LIAM MCKEEVER: I am. But it's not one it's
not just guided by one person. It's literally

1964
03:21:59,580 --> 03:22:08,330
we are splitting things up and we're, so that
here's the thing, right? There are so many

1965
03:22:08,330 --> 03:22:13,220
little things in any one study design that
become problematic. I'd like to diversify

1966
03:22:13,220 --> 03:22:20,310
a little bit. If you put it all in the hands
of one person, then you end up in situations

1967
03:22:20,310 --> 03:22:26,600
where maybe like someone doesn't have enough
stock in what they're doing, that it's OK.

1968
03:22:26,600 --> 03:22:30,210
For example, a lot of times you have this
happening and you're not, you put it in the

1969
03:22:30,210 --> 03:22:36,310
hands of a dietitian to run it and you're
not paying them anything so they don't have

1970
03:22:36,310 --> 03:22:42,810
a horse in the race. Next thing you know,
when they are bringing somebody into the study,

1971
03:22:42,810 --> 03:22:48,540
they're looking at like things like how long
is this person going to be in? How much data

1972
03:22:48,540 --> 03:22:54,120
am I going to have to collect? And if you
don't have somebody there, someone's research

1973
03:22:54,120 --> 03:23:02,330
fellow or somebody who their career is hanging
in the midst on that to be close to the study

1974
03:23:02,330 --> 03:23:07,270
and really monitoring everything, the value
of those smaller studies.

1975
03:23:07,270 --> 03:23:13,540
And I'm not talking super small. I mean, each
of these can be large studies, but the beauty

1976
03:23:13,540 --> 03:23:18,911
of smaller studies is there's somebody there
whose life is hanging in the balance. Somebody

1977
03:23:18,911 --> 03:23:24,859
there who cares whether or not that data is
collected properly, who is watching and monitoring

1978
03:23:24,859 --> 03:23:29,470
how things are going down, calling the nurses
every day to make sure that they understand

1979
03:23:29,470 --> 03:23:37,300
what was given those things. I think when
you let something get too big fall apart,

1980
03:23:37,300 --> 03:23:45,380
I think that there's a benefit to doing that.
But the power is only an issue if that's the

1981
03:23:45,380 --> 03:23:50,819
only study in existence and it's too small.
We can have them all adequately powered. If

1982
03:23:50,819 --> 03:23:56,920
they're in, you know, the truth is, the more
you stratify, the less, you know, the more

1983
03:23:56,920 --> 03:24:02,140
you increase your power to see, in effect,
you decrease the heterogeneity in your sample.

1984
03:24:02,140 --> 03:24:07,330
Whereas instead, if you just look at everybody,
then you need a huge sample size, right? Because

1985
03:24:07,330 --> 03:24:10,310
there's so much heterogeneity in your data.

1986
03:24:10,310 --> 03:24:20,260
DR. NILESH MEHTA: Liam, you mentioned... Liam,
thank you for that. You mentioned stratification,

1987
03:24:20,260 --> 03:24:25,970
Vijay used the word course correction. I'm
going to go to Todd next. Todd, there's a lot

1988
03:24:25,970 --> 03:24:33,220
to learn from centers like yours. Vanderbilt
has been leading in terms of innovative study

1989
03:24:33,220 --> 03:24:37,260
designs, and you've accomplished as a group
quite a lot. Any comments on adaptive trial

1990
03:24:37,260 --> 03:24:43,520
design and where are we with that in terms
of adaptive randomization? Adaptive trials?

1991
03:24:43,520 --> 03:24:52,100
DR. TODD RICE: Yeah. You know, I think there's
promise in adaptive trials, but I think and

1992
03:24:52,100 --> 03:24:55,890
somebody may have said this earlier, but I
think we have to be really cautious of putting

1993
03:24:55,890 --> 03:25:02,729
the cart ahead of the horse. A lot of adaptive
trials, you have to have some fundamental

1994
03:25:02,729 --> 03:25:07,989
understanding of mechanisms and things that
are going on in order to adapt. And, I mean,

1995
03:25:07,989 --> 03:25:17,460
you guys have heard me say for a number of
years that we took for a long time, at least

1996
03:25:17,460 --> 03:25:22,930
in the adult world, this one size fits all
approach to nutritional care, especially in

1997
03:25:22,930 --> 03:25:28,500
the ICU, but in lots of places. And I think
we're now realizing that one size does not

1998
03:25:28,500 --> 03:25:36,040
fit all. And an adaptive design can be great
for those. But we have to understand how the

1999
03:25:36,040 --> 03:25:41,330
people have different sizes and what they
need is different things. I think we'll get

2000
03:25:41,330 --> 03:25:47,310
there, but I think we still have some learning
to go in that regard.

2001
03:25:47,310 --> 03:25:52,580
DR. NILESH MEHTA: Todd what's your moonshot study?
What are you going to dream about.

2002
03:25:52,580 --> 03:25:59,729
DR. TODD RICE: Yeah, I caution people that I'm
not a pediatrician and I do adult medicine,

2003
03:25:59,729 --> 03:26:06,840
but I'm not sure it's different. For me, I
think there are multiple steps in the study.

2004
03:26:06,840 --> 03:26:11,720
I'm a clinical trial by design. I do randomized
trials. And so for me, it's going to be a randomized

2005
03:26:11,720 --> 03:26:16,810
trial. But on the front side, before we put
people into the randomized trial, before we

2006
03:26:16,810 --> 03:26:22,291
put kids into the randomized trial, we need
to understand the differences in their malnutrition

2007
03:26:22,291 --> 03:26:30,290
states. And for me, that then leads to different
interventions for different folks, for different

2008
03:26:30,290 --> 03:26:37,040
kids with different malnutrition states, and
studying that in order to. Improve. An outcome

2009
03:26:37,040 --> 03:26:42,220
that and I don't know what this is in the
pediatric world, but it has to be an outcome

2010
03:26:42,220 --> 03:26:47,279
that clinicians care about, that administrators
care about, and that kids and families care

2011
03:26:47,279 --> 03:26:52,440
about. And we've spent a lot of time in the
adult world with the first two parts of those

2012
03:26:52,440 --> 03:26:58,149
the clinicians care about and the administrators
care about. And we're now understanding that

2013
03:26:58,149 --> 03:27:02,390
we need to focus a lot on what the patients
care about as their outcome.

2014
03:27:02,390 --> 03:27:07,060
And so that's kind of the moonshot. If you
listen to me and I'm going to say it again because

2015
03:27:07,060 --> 03:27:10,430
I actually kind of fell in love with it in
the last three or four days. If you listen

2016
03:27:10,430 --> 03:27:17,700
to me on the adult side, we try and do nutritional
interventions in a short period of time and

2017
03:27:17,700 --> 03:27:26,020
then look at a longer-term outcome and using
the personal trainer concept I would love

2018
03:27:26,020 --> 03:27:31,949
to see a trial that has a personal nutritionist
for every patient, and that personal nutritionist

2019
03:27:31,949 --> 03:27:36,699
identifies the needs of that patient and then
helps them fulfill those needs, not just that

2020
03:27:36,699 --> 03:27:42,739
point in time, but across sort of their illness,
their hospitalization into their outpatient

2021
03:27:42,739 --> 03:27:48,680
setting. And obviously, it's a real moonshot.
And you'd have to have 100 million dietitians

2022
03:27:48,680 --> 03:27:56,320
and you'd have to have lots of money. But Dr. Lynch,
Dr. Pratt and Dr. Vargas all told us we don't

2023
03:27:56,320 --> 03:28:01,720
have any financial constraints. We don't have
any logistical constraints. And so we're going for it.

2024
03:28:01,720 --> 03:28:05,640
DR. TAMARA HANNON: When you went out and... (CROSSTALK)

2025
03:28:05,640 --> 03:28:08,090
DR. NILESH MEHTA: Say that again?

2026
03:28:08,090 --> 03:28:10,060
DR. TODD RICE: Sorry, Tamara.

2027
03:28:10,060 --> 03:28:16,399
DR. TAMARA HANNON: Well, you would...I mean that
would be great, but when I think about what

2028
03:28:16,399 --> 03:28:22,050
my patients want and they need, they want to
be able to pay for their medicine and they

2029
03:28:22,050 --> 03:28:27,220
want to be able to get the food that they need.
They want their kids to have a meal at school

2030
03:28:27,220 --> 03:28:36,160
that they'll eat, they want, I mean...they
want, they can have a coach, a personal nutritionist,

2031
03:28:36,160 --> 03:28:44,790
a trainer, but you actually have to give them
the food.

2032
03:28:44,790 --> 03:28:49,320
DR. TODD RICE: Yeah. Access. Right? Yeah. And
I mean, I...you know, we're dreaming big and

2033
03:28:49,320 --> 03:28:54,960
there are no constraints. So our coach is
in charge of both. All of that, that I put forth and

2034
03:28:54,960 --> 03:28:55,960
access.

2035
03:28:55,960 --> 03:28:59,370
DR. VIJAY SRINIVASAN: Yeah. I think is.

2036
03:28:59,370 --> 03:29:03,680
DR. NILESH MEHTA: And in the modern era, with
technology advances, nutrition code doesn't

2037
03:29:03,680 --> 03:29:10,239
have to be 100 million dietitians. This could
be remotely accessed. And then, as you are

2038
03:29:10,239 --> 03:29:15,300
alluding to Tamara, the intervention would
be how to then get them the individualized

2039
03:29:15,300 --> 03:29:16,300
prescribed nutrition.

2040
03:29:16,300 --> 03:29:22,700
DR. TODD RICE: I mean, I don't know if you know
this, but, Tamara, I have a little bit of

2041
03:29:22,700 --> 03:29:27,970
Indiana backgrounds and the background has
a little bit to do with the impact program

2042
03:29:27,970 --> 03:29:34,109
in Kenya that is actually kind of stationed
from Indiana. And what they actually found

2043
03:29:34,109 --> 03:29:39,859
and it's fascinating and it's worked phenomenal
is, is that to treat their HIV patients in

2044
03:29:39,859 --> 03:29:46,189
Kenya, they actually had to teach the patients
how to essentially farm their own food because

2045
03:29:46,189 --> 03:29:50,760
food became a higher priority than, even though
they were free, the antiretrovirals that were

2046
03:29:50,760 --> 03:29:55,660
being provided them either weren't taken or
didn't work if you didn't have the right nutritional

2047
03:29:55,660 --> 03:30:00,210
support with them. And so they learned this
and they spent an inordinate amount of time

2048
03:30:00,210 --> 03:30:05,330
and resources to teach their patients how
to do farming stuff and how to grow their

2049
03:30:05,330 --> 03:30:09,229
own food and access to their own food, so
much so that they got so good at it that they

2050
03:30:09,229 --> 03:30:13,540
now sell it. And that's an income for them.
In addition what they have as their food.

2051
03:30:13,540 --> 03:30:18,870
But I think that's what you're talking about,
right, is, is that it's not just bringing food

2052
03:30:18,870 --> 03:30:23,500
to people, but it's teaching them how to get
that access, whether it's doing it themselves

2053
03:30:23,500 --> 03:30:26,990
or finding it or having the resources to do
it, etc.

2054
03:30:26,990 --> 03:30:28,520
DR. VIJAY SRINIVASAN: Well, it's interesting...

2055
03:30:28,520 --> 03:30:32,740
DR. TAMARA HANNON: And that's exactly what I'm talking about because
when you talk about whole person health, we're

2056
03:30:32,740 --> 03:30:40,181
talking about this, this small amount of it,
but if you don't have food, you don't have

2057
03:30:40,181 --> 03:30:48,109
health. So it doesn't matter how many specialists
you have on your team. It's exactly what

2058
03:30:48,109 --> 03:30:49,109
I'm talking about.

2059
03:30:49,109 --> 03:30:54,040
DR. WILLIAM EVANS: So I was involved in a in a meeting
at the Gates Foundation and they were talking

2060
03:30:54,040 --> 03:31:00,150
about looking at the effects of increased
dietary protein in malnourished children.

2061
03:31:00,150 --> 03:31:07,210
So they started giving eggs to the children,
but quickly found out the children weren't

2062
03:31:07,210 --> 03:31:13,170
getting the eggs. It was going to the parents
and wasn't going to the children. And so there's

2063
03:31:13,170 --> 03:31:18,109
a lot of infrastructures. So what ended up
happening is that they established egg farms

2064
03:31:18,109 --> 03:31:25,000
so they could provide adequate nutrition and
increased protein to the entire family before

2065
03:31:25,000 --> 03:31:31,939
they could actually start to look at the infants
and children in the study. So there's a huge

2066
03:31:31,939 --> 03:31:38,819
infrastructure. And early on I heard that
there isn't any real consensus definition

2067
03:31:38,819 --> 03:31:44,000
of what malnutrition is. And I know in the
adult world, especially in the geriatrics

2068
03:31:44,000 --> 03:31:49,620
world, I've been to daylong meetings trying
to define malnutrition. And the only criteria

2069
03:31:49,620 --> 03:31:55,930
that everybody agreed to was involuntary weight
loss. And so I think it's it can be a moving

2070
03:31:55,930 --> 03:32:04,760
target in trying to do RCTs in, "malnourished
children" without really specifying what the

2071
03:32:04,760 --> 03:32:09,300
specific nutrient you're interested in and
the component of malnutrition that you're

2072
03:32:09,300 --> 03:32:13,189
most interested in trying to correct.

2073
03:32:13,189 --> 03:32:23,060
DR. VIJAY SRINIVASAN: Yeah, I just you know, (CROSSTALK)
With the... Sorry, Nilesh. Just a quick comment

2074
03:32:23,060 --> 03:32:27,970
about, you know, there's actually a potential
opportunity for a natural experiment that's

2075
03:32:27,970 --> 03:32:32,600
already kind of happened. And I kind of want
to sort of extrapolate from Tamara's point

2076
03:32:32,600 --> 03:32:39,149
about it's about availability and access.
All the information advice we give our families

2077
03:32:39,149 --> 03:32:43,710
will not matter if you don't have access.
And one of the things that we have just done

2078
03:32:43,710 --> 03:32:49,020
is expand the child tax credit. And we know
that has lifted a lot of children out of poverty.

2079
03:32:49,020 --> 03:32:54,199
And I wonder if there is more granular data
to see what the nutritional status also did

2080
03:32:54,199 --> 03:32:59,649
during this period when they had more sort
of access. Because every parent I mean, invariably

2081
03:32:59,649 --> 03:33:03,560
they all want to do the right thing for their
child, which is to put food on the table and

2082
03:33:03,560 --> 03:33:07,970
feed them. And if they had the resources,
they would do so. And I think that's one of

2083
03:33:07,970 --> 03:33:12,820
those things where I'd be very interested
to see if, you know, NIH and other organizations

2084
03:33:12,820 --> 03:33:24,430
can actually tease out some of that data to
see if it made a difference.

2085
03:33:24,430 --> 03:33:26,130
DR. NILESH MEHTA: Great point, Vijay.

2086
03:33:26,130 --> 03:33:36,149
DR. DAVID SUSKIND: And that's a really interesting point.
But I think. Oh. Sorry. But I think to the

2087
03:33:36,149 --> 03:33:40,710
other way of you.... with psychosocial and
lifestyle issues are there and having the

2088
03:33:40,710 --> 03:33:49,239
money to get the proper foods are important
but what are those proper foods and what are

2089
03:33:49,239 --> 03:33:57,640
their purchasing power with that so you know
as we and many of the sessions have talked

2090
03:33:57,640 --> 03:34:05,930
about, you know, malnutrition can be under
nutrition and over nutrition. So we may be

2091
03:34:05,930 --> 03:34:11,930
providing the economic ability to get the
foods, which is an essential component to

2092
03:34:11,930 --> 03:34:20,500
it, but also the educational component. The
knowledge that programs like WIC provide are

2093
03:34:20,500 --> 03:34:31,449
essential to do so. So I think it kind of
underlies the difficulty with knowing any

2094
03:34:31,449 --> 03:34:38,330
intervention and what is the outcome that
we're looking for because it is such a complex

2095
03:34:38,330 --> 03:34:48,170
issue and trying to find that granular outcomes
with a very broad stroke I think can be difficult

2096
03:34:48,170 --> 03:34:52,630
if not done in a very prospective way.

2097
03:34:52,630 --> 03:35:00,750
DR. NILESH MEHTA: Thanks, David. David Seres,
do you want to have the last word?

2098
03:35:00,750 --> 03:35:11,480
DR. DAVID SERES: Sure, I get to say all the above, plus
a couple of things. Ron [inaudible] once wrote:

2099
03:35:11,480 --> 03:35:19,310
The only nutritional outcomes are mortality,
health, economics, and quality of life. And

2100
03:35:19,310 --> 03:35:26,470
I kind of hold to that because one of the
things that really matter and then there are

2101
03:35:26,470 --> 03:35:31,300
things on the way to those things that really
matter too. So we could incorporate those.

2102
03:35:31,300 --> 03:35:37,590
But we really need to agree on what it is
that we're talking about when we use the term

2103
03:35:37,590 --> 03:35:43,370
malnutrition, Are we really talking about
people who are malnourished? And if we are,

2104
03:35:43,370 --> 03:35:49,170
then we need to change our definitions because
they include all of these things that are

2105
03:35:49,170 --> 03:35:55,910
disease-related and disease epiphenomena. If
I were designing a moonshot study, it would

2106
03:35:55,910 --> 03:36:03,350
go as follows: given an uninhibited funding
policy and everything else, that since we

2107
03:36:03,350 --> 03:36:09,000
have become very clear about the fact that
we really don't know when, how much, and what

2108
03:36:09,000 --> 03:36:15,960
to feed and what route to feed people who
are sick. Now, I don't want to...the free hospital

2109
03:36:15,960 --> 03:36:22,310
issues. Let's take those off the table now.
But for the sick patient, we don't know the

2110
03:36:22,310 --> 03:36:27,470
answer to any of those things. Therefore,
no patients should be fed in the hospital

2111
03:36:27,470 --> 03:36:34,060
unless they're part of a randomized controlled
trial. This is my Moonshot because we need

2112
03:36:34,060 --> 03:36:40,010
a huge study. We need an enormous study. And
what we are really dealing with is something

2113
03:36:40,010 --> 03:36:47,069
that truly is experimental and we are experimenting
on every single patient that we treat with

2114
03:36:47,069 --> 03:36:52,510
nourishment. Now, there is no controversy
that if you live long enough and get fed less

2115
03:36:52,510 --> 03:36:58,739
enough, you will die of starvation or the
consequences of the starvation will worsen

2116
03:36:58,739 --> 03:37:04,180
your outcomes. There's no controversy over
that. The controversy is how long do you have

2117
03:37:04,180 --> 03:37:11,430
to starve? What can you do to change that
starvation? Is what we're describing as malnutrition,

2118
03:37:11,430 --> 03:37:16,600
starvation, or is it something else? I think
it's a lot of things put together. And I said

2119
03:37:16,600 --> 03:37:24,060
at the previous Moonshot that I also want
to put all of the cachexia, sarcopenia development,

2120
03:37:24,060 --> 03:37:30,160
and malnutrition, folks in a room and kind
of knock their heads together and say, "Guys,

2121
03:37:30,160 --> 03:37:34,979
aren't you talking about something that's
almost exactly the same with different, slightly

2122
03:37:34,979 --> 03:37:40,390
different inputs?" But all of them have multiple
inputs. You can develop cachexia by wasting

2123
03:37:40,390 --> 03:37:44,720
or you can develop cachexia by starving.
You can develop malnutrition in both of those

2124
03:37:44,720 --> 03:37:52,180
ways. So we have to really start thinking
not as nutrition interventionalists because

2125
03:37:52,180 --> 03:37:56,590
that's a conflict of interest in on all of
our parts. All of us want to intervene with

2126
03:37:56,590 --> 03:38:03,859
nutrition. So as I think Chris mentioned,
we need to pull in people who are content

2127
03:38:03,859 --> 03:38:11,100
experts in the diseases rather than in the
nutrition who are not wanting to look for

2128
03:38:11,100 --> 03:38:18,390
things to do. I've always said that being
an expert is one of the most insidious forms

2129
03:38:18,390 --> 03:38:22,680
of conflict of interest, because if I don't
know what to do or if I have nothing to do,

2130
03:38:22,680 --> 03:38:26,490
then you're not going to come talk to me and
treat me like an expert. And there goes my

2131
03:38:26,490 --> 03:38:31,670
salary and my ego gratification and all of
the fun things I get to do when I go to work

2132
03:38:31,670 --> 03:38:38,561
every day. So I really have a vested interest
in being able to do something. So I've backed

2133
03:38:38,561 --> 03:38:43,600
off on all of my thoughts about modifying
disease. When I first started in practice,

2134
03:38:43,600 --> 03:38:48,510
I thought I was going to put every patient in
the hospital on some form of parenteral support

2135
03:38:48,510 --> 03:38:51,580
because I was going to stamp out hospital
malnutrition.

2136
03:38:51,580 --> 03:38:56,550
I'm serious. That was my goal based on what
I had been taught. And thankfully I had some

2137
03:38:56,550 --> 03:39:03,020
people knock my head around a little bit and say
"Excuse me, there's no data for that." And now what

2138
03:39:03,020 --> 03:39:09,870
I am doing as a practitioner is trying my
best to safely treat starvation. And I don't

2139
03:39:09,870 --> 03:39:16,310
know how to do that, and it drives me nuts.
So we need to start from simple. We need to

2140
03:39:16,310 --> 03:39:23,850
look very simply at the smallest criteria
that we can find that actually predicts response

2141
03:39:23,850 --> 03:39:28,250
to nourishment in whatever group that we're
talking about in every group, every group

2142
03:39:28,250 --> 03:39:35,819
is going to be different. So in the effort,
trial, inflammation clearly predicted a non-response,

2143
03:39:35,819 --> 03:39:41,939
but starvation predicted a response. So, we
need to go back to the very finest basics

2144
03:39:41,939 --> 03:39:47,410
and do the finest granular studies that we
can possibly do where we randomize all of

2145
03:39:47,410 --> 03:39:52,090
these things to be able to chop out what it
is that actually determines whether or not

2146
03:39:52,090 --> 03:39:59,439
somebody is starving, meaning that they have...
that they are becoming malnourished or developing

2147
03:39:59,439 --> 03:40:04,069
malnutrition in the truest sense that I want
it to be.

2148
03:40:04,069 --> 03:40:08,350
And then all of these other things that are
wasting due to inflammation and so forth,

2149
03:40:08,350 --> 03:40:15,239
which have nothing to do with us as nutrition
interventionalists. I think I got at all.

2150
03:40:15,239 --> 03:40:21,300
DR. NILESH MEHTA: David, That was fantastic. You
know, one hour is not enough with such a panel.

2151
03:40:21,300 --> 03:40:29,270
I can't tell you how enjoyable this is. And
I don't know if this is legit. And maybe Ashley

2152
03:40:29,270 --> 03:40:32,680
can tell us and then maybe they can ignore
it. But if you had thoughts that you were

2153
03:40:32,680 --> 03:40:40,100
unable to fit in here, do send an email and
it would be nice to consolidate some of your

2154
03:40:40,100 --> 03:40:45,260
thoughts that we were unable to get to. But
it is time for me to close this out so that

2155
03:40:45,260 --> 03:40:54,560
we have the next 10 minutes to award the poster
winners and then close this workshop. This

2156
03:40:54,560 --> 03:41:02,810
means I have to offer my sincere thanks right
now to this panel. It's been truly...

2157
03:41:02,810 --> 03:41:12,500
DR. DAVID SERES: Ashley said Yes, please. Ashley
said (CROSSTALK). Send emails with your ideas

2158
03:41:12,500 --> 03:41:18,890
and your comments. Hopefully, the organizers
will be getting together to write up proceedings,

2159
03:41:18,890 --> 03:41:25,939
to publish, and these are things that could
be taken into consideration. And I thank my

2160
03:41:25,939 --> 03:41:34,830
co-organizers and this has just been a real
pleasure. I'm sorry Nilesh back to you. 

2161
03:41:34,830 --> 03:41:40,370
DR. NILESH MEHTA: Thank you...thank you to Tamara Hannon,
David Susskind, Bill Evans, Chris Duggan,

2162
03:41:40,370 --> 03:41:47,979
Vijay Srinivasan, Liam McKeever, thank you
to Todd Rice and David Seres. At this point,

2163
03:41:47,979 --> 03:41:54,100
our Session 11 is closed and we will move
to the closing and adjournment of day five.

2164
03:41:54,100 --> 03:41:59,052
Thank you all. Have a great evening. 

2165
03:41:59,052 --> 03:42:17,315
[pause] Alright.

2166
03:42:17,315 --> 03:42:22,939
SPEAKER: Great job man. You’re the best. You really are smooth. (Inaudible). You’re so [inaudible] smooth.

2167
03:42:22,939 --> 03:42:24,830
DR. NILESH MEHTA: Sorry. Oh, thank you.

2168
03:42:24,830 --> 03:42:28,540
SPEAKER: (UNKNOWN) Do we go somewhere
else for the poster sessions?

2169
03:42:28,540 --> 03:42:35,819
DR. NILESH MEHTA: I am looking for some guidance
here. Ashley, do you know if Brett is still

2170
03:42:35,819 --> 03:42:39,510
here? Is he going to close the sessions or
(UNKNOWN) you going to close the session and

2171
03:42:39,510 --> 03:42:49,460
move to the slide for the poster awards? There
you are. Am I live?

2172
03:42:49,460 --> 03:42:56,390
DR. TAMARA HANNON: Yes, you are, I believe.

2173
03:42:56,390 --> 03:43:02,199
DR. NILESH MEHTA: Oh, thank you. Thank you all.
Thank you to the participants. Thank you to

2174
03:43:02,199 --> 03:43:09,040
the many poster presentation submissions that
we had. At this point, I'd like to announce

2175
03:43:09,040 --> 03:43:17,180
our poster winners in the next few slides.
In the next slide, I will just allude to the

2176
03:43:17,180 --> 03:43:23,930
criteria for the poster awardees. These posters,
we received to see posters from around the world.

2177
03:43:23,930 --> 03:43:31,920
This was really fun to review and NIH scientific
committee review these and score them for

2178
03:43:31,920 --> 03:43:39,260
these outlined criteria. The purpose is clearly
stated. The methods and data are present,

2179
03:43:39,260 --> 03:43:45,729
the quality of presented results are evaluated,
appropriateness of the conclusions based on

2180
03:43:45,729 --> 03:43:54,761
the results and the innovation, and the potential
impact of the results. So based on that, let's

2181
03:43:54,761 --> 03:44:01,560
now look at the next slide for the adult poster
winners. It is my pleasure to announce that

2182
03:44:01,560 --> 03:44:08,819
the Adults Track Awards for the posters were
awarded to these three individuals, Nina Rocca

2183
03:44:08,819 --> 03:44:17,709
and her group who presented... who submitted
the work on implementation of a TOC nutrition

2184
03:44:17,709 --> 03:44:23,409
intervention for patients with malnutrition.
Thank you. Congratulations to Nina and her

2185
03:44:23,409 --> 03:44:29,930
group. The next one is Brooke Helleher from
University of New Hampshire. And Brooke and

2186
03:44:29,930 --> 03:44:36,150
her group submitted their work on the Home
Food environment and its association with

2187
03:44:36,150 --> 03:44:43,630
age, food security and obesity, co-morbidities
in Hispanics and Latino populations. Congratulations

2188
03:44:43,630 --> 03:44:52,170
to Brooke Helleher and her group. The next
one is Lynne Hiller from James Valley Veteran

2189
03:44:52,170 --> 03:44:58,500
Hospital, Florida Estimation of 24-Hour Urinary
Creatinine Excretion to Identify Malnutrition

2190
03:44:58,500 --> 03:45:04,210
in critically ill Veterans. Congratulations
to all three winners in the Adult Track Awards

2191
03:45:04,210 --> 03:45:11,870
and thank you for your work. Congratulations
and good luck for your future projects. Next,

2192
03:45:11,870 --> 03:45:18,330
we will go to the Pediatric FAC Award winners
for the posters. On the next slide, the pediatric

2193
03:45:18,330 --> 03:45:25,940
awards go to first Nicole Gilbert from Alberta
Health Sciences, their work on assessing the

2194
03:45:25,940 --> 03:45:32,140
prevalence of malnutrition in pediatric Canadian
pediatric hospitals. They evaluated the application

2195
03:45:32,140 --> 03:45:39,069
of the AND ESPEN definitions.
Congratulations and Nicole and group. And

2196
03:45:39,069 --> 03:45:44,229
then Katherine Bell from Brigham and Women's
Hospital, Harvard Medical School. Their work

2197
03:45:44,229 --> 03:45:49,980
on preterm infant nutritional status comparing
to body composition references. Congratulations

2198
03:45:49,980 --> 03:45:56,399
Katherine Bell and colleagues. And then finally,
Laura Gollins from Texas Children's Hospital

2199
03:45:56,399 --> 03:46:03,061
Baylor, where they presented their work on
mid-upper arm circumference as

2200
03:46:03,061 --> 03:46:07,960
a predictor of body composition in extremely
low birth weight infants. Congratulations

2201
03:46:07,960 --> 03:46:13,359
to all three award winners in the pediatric
or the early life track. And good luck with

2202
03:46:13,359 --> 03:46:22,600
your future projects. Next slide. The Scientific
Consulting Group will contact all the awardees,

2203
03:46:22,600 --> 03:46:26,600
and if you don't hear from them, please reach
out to Mark Dennis in the next two... If you

2204
03:46:26,600 --> 03:46:33,460
don't hear from them in the next two weeks.
Thank you so much. At this point, we are close

2205
03:46:33,460 --> 03:46:40,220
to the end of the hour and it remains simply
for me to give heartfelt thanks to all of

2206
03:46:40,220 --> 03:46:46,150
you for joining over the last two days. And
then on the three days before the week before

2207
03:46:46,150 --> 03:46:53,210
in the Adult Malnutrition workshop, we've
experienced a variety of topics expertly presented

2208
03:46:53,210 --> 03:46:58,649
by some of the leading minds and researchers
in the group, and it allowed us to highlight

2209
03:46:58,649 --> 03:47:02,760
gaps.
I thoroughly enjoyed the last session where

2210
03:47:02,760 --> 03:47:08,649
with the panel, expert panel we talked about
moonshot studies, just as we did in the adult

2211
03:47:08,649 --> 03:47:15,260
session last week. My sincere thanks to the
NIH, the scientific group led by the leadership

2212
03:47:15,260 --> 03:47:23,970
of Christopher Lynch, Charlotte Pratt, and
Ashley Vargas. Your involvement, your dedication,

2213
03:47:23,970 --> 03:47:30,241
and your investment into the future of malnutrition
research is, is really inspiring. And thank

2214
03:47:30,241 --> 03:47:35,260
you for providing us with this platform and
for all of us to join and put our heads together

2215
03:47:35,260 --> 03:47:41,880
on where we go next. The goal of this workshop
was to improve outcomes by preventing malnutrition,

2216
03:47:41,880 --> 03:47:48,439
and we went from an entire spectrum of early
life to adults. And finally, I would say thanks

2217
03:47:48,439 --> 03:47:53,910
on behalf of my co-chairs, Dr. Gail Cresci,
David Seres, and Dr. Todd Rice. It's been

2218
03:47:53,910 --> 03:48:00,830
a pleasure. I wish you all the very best and
have a great evening and good luck in all

2219
03:48:00,830 --> 03:48:04,510
your research endeavors. Goodbye and good
luck.