[Skip to Content]
[Skip to Content Landing]
Views 4,840
Citations 0
Editor's Note
June 2017

The Value of Using Registries to Evaluate Randomized Clinical Trial Study Populations

Author Affiliations
  • 1Department of Medicine, University of California, San Francisco
  • 2JAMA Internal Medicine, Chicago, Illinois
JAMA Intern Med. 2017;177(6):889. doi:10.1001/jamainternmed.2017.0761

Randomized clinical trials (RCTs) are considered the gold standard of medical research and influence clinical practice, but they do not address all relevant clinical questions.1 The study designs used in RCTs have many benefits, and are clearly the best way to limit bias and confounding. However, they may only apply to the study population at issue and cannot be assumed to translate to other patient populations.

In the accompanying study, Maddox et al2 highlight the importance of comparing a real-world population to an RCT’s study population. They used data from the American College of Cardiology Practice Innovation and Clinical Excellence (PINNACLE) ambulatory cardiology practice registry to compare the current proportion of real-world patients with acute coronary syndrome (ACS) who would have qualified for the IMPROVE-IT trial. They found that the RCT would only have applied to one-third of current patients with ACS because registry patients were older and had more medical comorbidities. Therefore, the RCT’s impact on clinical practice is limited.

This issue is not isolated to the IMPROVE-IT trial. The study2 highlights the need for data registries to complement the findings of RCTs. The United States lags behind many other countries, such as the Scandinavian countries, who track their population’s health information to continuously improve care. The study population in an RCT is a carefully selected group of individuals, who are often younger, healthier, and less diverse than most patients. Registries in cardiology, oncology, and surgery, such as the National Cardiology Data Registry (NCDR), the Surveillance, Epidemiology, and End Results Program (SEER), and the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) can add real-world data to those of large RCTs. Newer registries can be an invaluable tool in postmarketing surveillance of new drugs, such as the US Food and Drug Administration’s Sentinel Initiative and the National Evaluation System for Health Technology (NEST) for devices.3 These efforts will all be increasingly important in helping clinicians place RCTs in the context of a larger and more diverse patient population and in helping researchers design additional RCTs.

Back to top
Article Information

Corresponding Author: Adam J. Schoenfeld, MD, Department of Medicine, University of California, San Francisco, 2440 Fillmore St, Apt 3, San Francisco, CA 94115 (adam.schoenfeld@ucsf.edu).

Conflict of Interest Disclosures: None reported.

References
1.
Schulz  KF, Altman  DG, Moher  D; CONSORT Group.  CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials.  BMJ. 2010;340:c332.PubMedGoogle ScholarCrossref
2.
Maddox  TM, Tang  F, Downs  JR,  et al.  Applicability of the IMPROVE-IT Trial to current patients with acute coronary syndrome: an NCDR research to practice project  [published online April 24, 2017].  JAMA Intern Med. doi:10.1001/jamainternmed.2017.0754Google Scholar
3.
Shuren  J, Califf  RM.  Need for a national evaluation system for health technology.  JAMA. 2016;316(11):1153-1154. doi:10.1001/jama.2016.8708PubMedGoogle ScholarCrossref
×