NIH Collaboratory Distributed Research Network (DRN)
Millions of people. Strong collaborations. Privacy first.
Leadership: Richard Platt and Lesley Curtis
Project Manager: Sarah Malek
The NIH Collaboratory Distributed Research Network (DRN) enables investigators funded by the NIH and other not-for-profit sponsors to collaborate with investigators based in health plans that participate in the FDA’s Sentinel System. The DRN is especially useful for supporting multisite research programs.
The DRN fully leverages the FDA Sentinel System’s data, methods, tools, and querying infrastructure. It can also directly contact providers and health plan members to collect new information or in support of randomized clinical trials. Read more about the Sentinel System.
Examples of Capabilities
Observational Studies of Comparative Effectiveness and Safety
With support from the FDA, investigators from DRN research partners evaluated the safety and effectiveness of dipeptidyl peptidase-4 inhibitors compared with other oral antihyperglycemic agents (Toh et al, 2018[a]). The investigators conducted a population-based, retrospective, new-user cohort study to examine associations between the use of saxagliptin and sitagliptin and the risk of hospitalization for heart failure in 18 health insurance plans and health systems in the FDA’s Mini-Sentinel program. The study did not find a higher risk of hospitalization for heart failure among new users of saxagliptin or sitagliptin compared with other antihyperglycemic agents.
Investigators also used the DRN to conduct a prospective safety surveillance study that included 7 sequential assessments comparing the myocardial infarction risk of saxagliptin with sitagliptin, pioglitazone, second-generation sulfonylureas, and long-acting insulin (Toh et al, 2018[b]). The analysis used disease risk score stratification and propensity score matching to adjust for potential confounders.
Prospective Data Collection
Patient-reported data can be collected from health plan members using, for example, the FDA MyStudies App, a public domain software application developed on behalf of the FDA. This smartphone app facilitates input of real-world data directly by patients and can be linked securely to the patient’s electronic health data in support of traditional clinical trials, pragmatic trials, observational studies, and registries. To demonstrate its use, the Kaiser Permanente Washington DRN research partner used the FDA MyStudies App in a pilot of a descriptive study that collected exposures and health care outcomes from 64 pregnant women. This experience can be leveraged to scale across the NIH Collaboratory DRN to prospectively collect patient-generated outcomes from recruited participants to supplement data.
Randomized Clinical Trials
The NIH Collaboratory DRN research partners can conduct multicenter randomized trials involving direct outreach to clinicians and health plan members. An example is IMPACT-AFib, a project funded by FDA-Catalyst. The study is an 80,000-person individually randomized trial using the FDA Sentinel System to identify and conduct a direct-mail intervention to health plan members with atrial fibrillation who are at high risk for stroke. The intervention encourages these patients to consider oral anticoagulant treatment. The patients’ clinicians are also contacted as part of the study.
Studies of Live Birth Pregnancies and Pregnancy Outcomes
The DRN has linked over 4 million pairs of mothers and infants, making it possible to study birth outcomes of maternal health and antenatal exposures. Read more about the Mother-Infant Linkage Table in the Sentinel Common Data Model.
A Distributed Data and Research Network
The NIH Collaboratory DRN research partners have access to administrative claims data and, in some cases, linked clinical data. Also, because each research partner has direct identifiers and a relationship with potential participants, the DRN enables researchers to conduct prospective longitudinal observational studies.
The DRN uses a distributed analysis approach that enables authorized researchers to send queries to the DRN research partners who hold data. Queries take the form of program code that the DRN research partner can execute on a pre-existing, extensively curated dataset transformed into the Sentinel Common Data Model. The querying capability of the network reduces the need to share confidential or proprietary data. The DRN research partner can return the query result, typically aggregated (count) data, rather than the dataset itself. This form of remote querying reduces legal, regulatory, privacy, proprietary, and technical barriers associated with data sharing for research.
The DRN uses the PopMedNet software application and complies with standards for distributed querying supported by the Query Health Initiative of the Office of the National Coordinator for Health Information Technology and adopted by other research collaborations, including Sentinel and PCORnet.
Participating Research Partners and Lead Investigators
- Aetna (Healthagen LLC)
- Blue Cross Blue Shield of Massachusetts
- Harvard Pilgrim Health Care Institute
- HCA Healthcare
- HealthCore, Inc. Government & Academic Research
- HealthPartners Institute
- Humana, Inc.
- Kaiser Permanente Hawaii, Center for Integrated Health Care Research
- Kaiser Permanente Mid-Atlantic States
- Kaiser Permanente Northern California, Division of Research
- Kaiser Permanente Northwest Center for Health Research
- Kaiser Permanente Washington Health Research Institute
- Marshfield Clinic Research Institute
- Vanderbilt University Medical Center, Department of Health Policy (Tennessee Medicaid data)
Together, these DRN research partners are actively following nearly 50 million individuals and have curated data covering hundreds of millions of person-years. See a snapshot of the Sentinel Distributed Database.
All participating DRN research partners have the academic training and experience needed to serve as coinvestigators in large, multi-institutional, longitudinal observational analyses. Each has demonstrated the ability and obtained the permissions to access and use their respective data.
All DRN research activities are opt-in. Each research partner decides on a project-by-project basis whether to participate in a potential study. The network seeks to build strong and trusting collaborations with external investigators.
The DRN research partners require funding for investigator and analyst effort and other research-related costs. Depending on the circumstances, they perform preparatory-to-research assessments without direct compensation. These preparatory assessments typically describe an eligible population, assess duration of follow-up, and track the rates of outcomes of interest.
For more information, contact firstname.lastname@example.org.
Drs. Jeffrey Brown and Lesley Curtis explain the NIH Collaboratory DRN.
Gewandter JS, Kleckner AS, Marshall JH, et al. Chemotherapy-induced peripheral neuropathy (CIPN) and its treatment: an NIH Collaboratory study of claims data. Support Care Cancer. 2019 Sep 7. [Epub ahead of print] PMID: 31494735.
Panozzo CA, Curtis LH, Marshall J, et al. Incidence of statin use in older adults with and without cardiovascular disease and diabetes mellitus, January 2008-March 2018. PLoS One. 2019;14(12):e0223515. PMID: 31805056.
Agiro A, Sridhar G, Gordon A, Brown J, Haynes K. Antibiotic dispensing following pediatric visits in the US emergency departments and outpatient settings from 2006 to 2016. Pharmacol Res Perspect. 2019;7(5):e00512. PMID: 31467679.
Raman SR, Brown JS, Curtis LH, et al. Cancer screening results and follow-up using routinely collected electronic health data: estimates for breast, colon, and cervical cancer screenings. J Gen Intern Med. 2019;34(3):341-343. PMID: 30350029.
Brown J, Klan J, Murphy S, Swan B, NIH Collaboratory Electronic Health Records Core. NIH Collaboratory Distributed Research Network: PopMedNet-i2b2 Integration Proof of Concept Report. March 25, 2015. (See also PopMedNet-i2b2 Integration Proof of Concept [video].)
Haynes K, Selvam N, Cziraky MJ. Bidirectional data collaborations in distributed research. EGEMS (Wash DC). 2016;4(2):1205. PMID: 27141518.
Klann JG, Buck MD, Brown J, et al. Query Health: standards-based, cross-platform population health surveillance. J Am Med Inform Assoc. 2014;21(4):650-656. PMID: 24699371.
November 14, 2014: Jeffrey Brown, Lesley H. Curtis, Richard Platt. Grand Rounds Presentation: Using the NIH Collaboratory's and PCORnet's Distributed Data Networks for Clinical Trials and Observational Research: A Preview. (See video and slides.)
June 13, 2014: Jeffrey Brown, Lesley Curtis, Richard Platt. Grand Rounds Presentation: The NIH Collaboratory Distributed Research Network: A Privacy Protecting Method for Sharing Research Data Sets. (See video and slides.)