1 00:00:12,645 --> 00:00:14,481 >> Jon McKeeby: My name is Jon McKeeby. 2 00:00:14,481 --> 00:00:17,350 I'm the clinical information officer of the Clinical Center 3 00:00:17,350 --> 00:00:19,219 of the National Institutes of Health. 4 00:00:19,886 --> 00:00:21,688 Today, we'll talk about the Clinical 5 00:00:21,688 --> 00:00:25,258 Data Interchange Standards Consortium, CDISC. 6 00:00:29,629 --> 00:00:32,399 We will first discuss the basics. 7 00:00:36,035 --> 00:00:40,039 From the principles and practice of clinical research book, 8 00:00:40,740 --> 00:00:46,880 CDISC is reviewed as the standards for clinical research. 9 00:00:47,380 --> 00:00:50,683 Standards exist for organizing clinical research data 10 00:00:50,683 --> 00:00:53,186 to enable sharing and aggregation. 11 00:00:54,454 --> 00:00:57,590 CDISC was created as a standard 12 00:00:57,590 --> 00:01:00,493 for submitting regulatory information to the FDA. 13 00:01:01,528 --> 00:01:04,264 CDISC with FDA, HL7, 14 00:01:04,264 --> 00:01:07,867 NCI combined efforts with the Biomedical Research 15 00:01:07,867 --> 00:01:12,939 Integrated Domain Group, BRIDG, to operationalize CDISC. 16 00:01:14,841 --> 00:01:18,611 CDISC is a standards development organization 17 00:01:19,546 --> 00:01:23,049 whose focus is medical research data linked with healthcare. 18 00:01:23,817 --> 00:01:26,853 It is to enable information system interoperability 19 00:01:26,853 --> 00:01:30,323 to improve medical research and related areas of healthcare. 20 00:01:31,157 --> 00:01:33,560 The mission of CDISC is to develop 21 00:01:33,560 --> 00:01:36,963 and support global platform independent data standards, 22 00:01:36,963 --> 00:01:40,667 to improve medical research and related areas of healthcare. 23 00:01:40,667 --> 00:01:45,839 It is to ensure interoperability between research and healthcare. 24 00:01:45,839 --> 00:01:50,577 It is to streamline the workflow from EHRs clinicians 25 00:01:50,577 --> 00:01:52,178 who conduct research. 26 00:01:54,881 --> 00:01:58,284 The benefits are reduced time to set up research, 27 00:01:58,284 --> 00:02:00,153 traceability from the EHR 28 00:02:00,153 --> 00:02:02,489 to the clinical research database, 29 00:02:02,489 --> 00:02:05,425 improve data quality, improve data sharing 30 00:02:05,425 --> 00:02:08,528 and interoperability across research organizations, 31 00:02:09,028 --> 00:02:12,398 reduce costs for setup and overhead of data analysis 32 00:02:12,398 --> 00:02:16,069 for the length of study, and increase predictability 33 00:02:16,069 --> 00:02:19,405 by creating repeatable and streamlined process. 34 00:02:20,874 --> 00:02:22,342 In another conversation, 35 00:02:22,342 --> 00:02:27,080 we talked about common data elements. 36 00:02:28,014 --> 00:02:31,317 CDISC provides standards, common data elements, 37 00:02:31,317 --> 00:02:32,919 in a common data model. 38 00:02:34,420 --> 00:02:38,858 CDISC importance is research studies can be initiated faster. 39 00:02:38,858 --> 00:02:41,661 Gartner reported in 2007 40 00:02:41,661 --> 00:02:44,564 that it reduced setup time by 70 to 90 percent. 41 00:02:45,765 --> 00:02:48,067 It also allows research studies 42 00:02:48,067 --> 00:02:51,571 to significantly have fewer resources, 43 00:02:51,571 --> 00:02:53,206 both in time and personnel. 44 00:02:54,407 --> 00:02:57,911 It ensures the data quality and integrity 45 00:02:57,911 --> 00:03:00,980 that's higher than without having standards. 46 00:03:01,581 --> 00:03:05,885 It improves communication across the teams for research, 47 00:03:05,885 --> 00:03:09,055 across organizations, across stakeholders. 48 00:03:09,656 --> 00:03:12,258 It ensures data sharing and interoperability 49 00:03:12,258 --> 00:03:14,794 between systems and organizations. 50 00:03:14,794 --> 00:03:18,064 And it provides high quality reporting to the FDA, 51 00:03:18,064 --> 00:03:22,368 clinicaltrials.gov, and other regulatory organizations. 52 00:03:27,106 --> 00:03:30,743 The guiding principles of CDISC are to develop standards 53 00:03:30,743 --> 00:03:33,446 of the highest quality to allow all researchers 54 00:03:33,446 --> 00:03:36,482 to leverage and share information from individuals 55 00:03:36,482 --> 00:03:38,084 and studies around the world. 56 00:03:39,185 --> 00:03:41,921 To facilitate the ability for implementers 57 00:03:42,822 --> 00:03:45,391 to effectively structure and analyze data 58 00:03:45,391 --> 00:03:48,194 so it's easily interpreted, understood, 59 00:03:48,194 --> 00:03:50,730 and navigated by regulatory reviewers; 60 00:03:51,965 --> 00:03:54,434 ensure the standards are developed in a manner 61 00:03:54,434 --> 00:03:57,337 that emphasize content, structure, and quality, 62 00:03:57,937 --> 00:04:01,240 transcending implementation strategy and platform; 63 00:04:02,308 --> 00:04:05,678 provide a global, multidisciplinary 64 00:04:05,678 --> 00:04:07,614 cross-functional community of members, 65 00:04:07,614 --> 00:04:09,682 volunteers, and stakeholders 66 00:04:09,682 --> 00:04:11,684 from across the research spectrum 67 00:04:11,684 --> 00:04:13,786 to develop consensus-based standards; 68 00:04:14,821 --> 00:04:16,255 collaborate and partner 69 00:04:16,255 --> 00:04:19,292 with fellow thought leaders and organizations 70 00:04:19,292 --> 00:04:22,295 on key initiatives to foster efforts 71 00:04:22,295 --> 00:04:24,263 to advance standards and semantics; 72 00:04:25,264 --> 00:04:27,734 and accomplish this without promoting 73 00:04:27,734 --> 00:04:30,403 any individual vendor or organization. 74 00:04:31,104 --> 00:04:35,675 Create a set of standards that is agnostic to any vendor. 75 00:04:40,313 --> 00:04:41,914 There's been many milestones. 76 00:04:41,914 --> 00:04:45,718 It started in 1997 as a volunteer group. 77 00:04:46,719 --> 00:04:49,922 It then became more formal in 2000. 78 00:04:50,823 --> 00:04:54,093 Created as an independent nonprofit organization. 79 00:04:55,261 --> 00:04:58,297 Global participation started in 2001. 80 00:04:58,865 --> 00:05:01,701 And then the various standards have been created 81 00:05:01,701 --> 00:05:04,537 over time from 2001 to current. 82 00:05:05,371 --> 00:05:11,844 This is just a list of the major standards 83 00:05:11,844 --> 00:05:13,680 that were created by CDISC. 84 00:05:17,517 --> 00:05:21,154 In 2017, there was a paper that was created 85 00:05:21,154 --> 00:05:26,426 to review the importance of data standards in clinical research. 86 00:05:26,926 --> 00:05:28,761 And they identified the importance 87 00:05:28,761 --> 00:05:30,630 to promote interoperability. 88 00:05:31,464 --> 00:05:34,967 And in the paper, they specifically mentioned CDISC. 89 00:05:34,967 --> 00:05:38,204 And you should use data standards that are common, 90 00:05:38,805 --> 00:05:40,773 known and accepted. 91 00:05:41,340 --> 00:05:43,242 And that the data standards from CDISC 92 00:05:43,242 --> 00:05:45,812 should be considered as the best starting point 93 00:05:45,812 --> 00:05:48,448 for defining and coding data 94 00:05:48,448 --> 00:05:50,750 and metadata in a consistent way. 95 00:05:51,317 --> 00:05:55,455 They identified that CDISC is an operational data model 96 00:05:55,455 --> 00:05:57,657 that provides a good starting point. 97 00:05:58,157 --> 00:06:01,761 It provides metadata that provides one mechanism 98 00:06:01,761 --> 00:06:04,731 to remove ambiguity. 99 00:06:04,731 --> 00:06:09,001 And then it has the advantage of ensuring data traceability 100 00:06:09,001 --> 00:06:12,739 and providence of data through the research process. 101 00:06:15,641 --> 00:06:19,278 Ways CDISC is used. It's an exchange format. 102 00:06:19,278 --> 00:06:23,750 It's used to exchange data utilizing XML format. 103 00:06:24,283 --> 00:06:27,553 We see that in its use to provide protocol data 104 00:06:27,553 --> 00:06:32,358 to clinicaltrials.gov as well as pharmaceutical data to the FDA. 105 00:06:33,893 --> 00:06:36,496 CDISC provides a controlled terminology 106 00:06:37,563 --> 00:06:38,865 common data elements. 107 00:06:38,865 --> 00:06:43,236 They create a common set of terms across therapeutic areas. 108 00:06:43,870 --> 00:06:47,540 And then as a reference model, a physical database. 109 00:06:47,540 --> 00:06:52,211 It's used as a common data model to establish semantic and syntax 110 00:06:52,211 --> 00:06:54,781 to share with stakeholders across studies.