Electronic records, which have become increasingly prevalent in recent years, represent a valuable repository of medical data. There are over 300 million people in the United States, most of whom are receiving some sort of medical care. If anonymized medical records were made available to researchers, these e-records could be leveraged to improve care at low cost.
In today’s quest to answer questions about medicine and human health, the large-scale, controlled clinical trial is central. New drugs, surgical techniques, non-surgical interventions and medical devices are typically tested in such studies, which require the creation of control and test groups, controlling for confounding factors such as age and lifestyle, and the tracking of patients.
While these studies are essential for testing new drugs and interventions, not every medical question can or should be tested using a clinical trial, given the human and monetary resources that are required to conduct such studies.
Physicians and researchers already have other accepted ways of advancing medical knowledge.
Observational studies and chart reviews typically analyze groups of patients based on the condition displayed or the intervention used, and are often performed when large randomized studies are infeasible. They can yield important results even without the controls needed for clinical trials.
Physicians author case reports as a means of sharing their experiences with especially instructive cases. These reports are regarded as valuable parts of the medical literature and are a standard part of peer-reviewed medical journals. The medical field recognizes and accepts knowledge gained through observational studies, chart reviews and case reports in addition to controlled clinical studies.
With the use of e-records, the scope of chart reviews and the sharing of case reports are dramatically expanded by many orders of magnitude. This represents a unique opportunity to leverage vast amounts of newly available data to explore many medical questions.
Recently, the drug Vioxx was recalled due to side effects causing higher rates of heart attack than comparable drugs. The recall took place one month after results of a study using medical records from 1.4 million people were reported. This recall is one example of how collecting and using data from actual patients can lead to medical advances.
Another example of data gathering that has led to significant recent advances is the ongoing Framingham Heart Study. This longitudinal study has tracked over 10,000 individuals from three consecutive generations, monitoring their physical health and lifestyle choices, in order to learn about cardiovascular disease. That information has been used by researchers to make many advances, ranging from genetics to the role of social networks.
Many other studies have been based upon survey and medical reporting data collected by the CDC. The data that could be made available dwarfs current resources.
As previously noted, medical care data is logged for many millions of people in the United States. We can leverage the sheer volume of available data, combined with new pattern-recognition methods and theoretical advances in data analysis, to increase our knowledge in ways beyond the practical reach of other methods.
At the very least, analyzing these data could reveal previously unknown connections–for example, between a particular medication, a bit of medical history, and a seemingly unrelated disease–that would provide clues as to what questions should be pursued in formal, controlled studies.
Also, since each person’s medical records may cover many years, we can learn about long term effects much more easily and cost-effectively by analyzing these available data than by conducting longitudinal studies on a particular therapy. Thus, we can use these data to discover long-term effects that may otherwise not be detected at all.
Leveraging the availability of care data to increase our knowledge can’t and shouldn’t replace controlled studies or physician experience. But it can be a powerful and cost-effective tool, allowing us to utilize huge amounts of information and new methods of analysis to increase our medical knowledge, improving our ability to treat patients and take care of ourselves.
Data—be they from study results or individual experiences—are the raw material that we use to build our medical knowledge. Gaining access to such a huge volume of new information about human health is like inventing a microscope that can see objects that are much smaller, or a telescope that can see much farther away. This new data can lead to many new discoveries.
Analyses of these data would be an important addition to the medical research toolkit, augmenting traditional research methods.
The governmental agencies that oversee various aspects of healthcare practice and research should work with medical organizations to make electronic medical records available in an anonymous, analyzable form. They should encourage use of this vast, important resource to propel our knowledge of medicine forward.
Coming Next...Step VIII: Promote "First Day" Celebrations.
For Further Reading
1. R. Rubin, How did Vioxx debacle happen? USA Today (10/12/2004).
2. D. J. Graham, D. Campen, R. Hui, M. Spence, C. Cheetham, G. Levy, S. Shoor, W. A. Ray, Risk of acute myocardial infarction and sudden cardiac death in patients treated with cyclo-oxygenase 2 selective and non-selective non-steroidal anti-inﬂammatory drugs: nested case-control study, Lancet 365, 475–81 (2005).
3. Framingham Heart Study: A Project of National Heart, Lung, and Blood Institute and Boston University (1948-present).
4. C. J. O'Donnell, K. Lindpaintner, M. G. Larson, V. S. Rao, J. M. Ordovas, E. J. Schaefer, R. H. Myers, D. Levy, Evidence for association and genetic linkage of the angiotensin-converting enzyme locus with hypertension and blood pressure in men but not women in the Framingham Heart Study, Circulation 97, 1766-1772 (1998).
5. D. Levy, A. L. DeStefano, M. G. Larson, C. J. O’Donnell, R, P. Lifton, H. Gavras, L. A. Cupples, R. H. Myers, Evidence for a gene influencing blood pressure on chromosome 17: genome scan linkage results for longitudinal blood pressure phenotypes in subjects from the Framingham Heart Study, Hypertension 36, 477 (2000).
6. N. A. Christakis, J. H. Fowler, The spread of obesity in a large social network over 32 years, New England Journal of Medicine 357, 370-379 (2007).
7. Centers for Disease Control and Prevention, Data & Statistics.
8. Centers for Disease Control and Prevention, National Center for Health Statistics.
9. Centers for Disease Control and Prevention, WONDER online databases.
10. University of Virginia School of Medicine, Clinical Data Repository (CDR).
11. Wake Forest University School of Medicine, Clinical Data Repository (CDR).
12. R. Schoenberg, C. Safran, Internet based repository of medical records that retains patient confidentiality, British Medical Journal 321, 1199-1203 (2000).
13. G. Narcisi, Clinical data repositories: boosting patient care & research for a data-intensive future, CMIO 9-10 (April 2010).
14. E. R. Weitzman, L. Kaci, K. D. Mandl, Sharing medical data for health research: the early personal health record experience. Journal of Medical Internet Research 12, e14 (2010).
15. A. Schwartz, C. Pappas, L. J. Sandlow, Data repositories for medical education research: issues and recommendations, Academic Medicine 85, 837-843 (2010).
16. M. Jenicek, Clinical case reporting in evidence-based medicine, 2nd edition (Hodder Arnold Publication, London, 2001).
17. M. Kidd, C. Hubbard, Introducing Journal of Medical Case Reports, Journal of Medical Case Reports 1, 1 (2007).
18. M. L. Richardson, F. S. Chew, Radiology Case Reports: a new peer-reviewed, open-access journal specializing in case reports, Radiology Case Reports 1, 1-3 (2006).
19. K. Benson, A. J. Hartz, A comparison of observational studies and randomized, controlled trials, New England Journal of Medicine 342, 1878-1886 (2000).
20. N. Black, Why we need observational studies to evaluate the effectiveness of health care, British Medical Journal 312, 1215 (1996).
21. R. W. Haley, D. R. Schaberg, D. K. Mcclish, D. Quade, K. B. Crossley, D. H. Culver, W. M. Morgan, J. E. Mcgowan, Jr., R. H. Shachtman, The accuracy of retrospective chart review in measuring nosocomial infection rates: results of validation studies in pilot hospitals, American Journal of Epidemiology 111, 516-533 (1980).
22. C. G. Victora, J. Habicht, J. Bryce, Evidence-based public health: moving beyond randomized trials, American Journal of Public Health 94, 400-405 (2004).
23. D. C. Des Jarlais, C. Lyles, N. Crepaz, the TREND Group, Improving the reporting quality of nonrandomized evaluations of behavioral and public health interventions: the TREND statement, American Journal of Public Health 94, 361-366 (2004).
24. D. G. Altman, K. F. Schulz, D. Moher, M. Egger, F. Davidoff, D. Elbourne, P. C. Gøtzsche, T. Lang, The revised CONSORT statement for reporting randomized trials: explanation and elaboration, Annals of Internal Medicine 134, 663-694 (2001).
25. J. J. Allison, T. C. Wall, C. M. Spettell, J. Calhoun, C. A. Fargason, R. Kobylinski, R. Farmer, C. I. Kiefe, The art and science of chart review, Joint Commission Journal of Quality Improvement 26, 115-36 (2000).
Writing and Editing Credits
Yaneer Bar-Yam with Shlomiya Bar-Yam, Karla Z. Bertrand, and Nancy Cohen
Page 1: "The Electronic Medical Record" by the Agency for Healthcare Research and Quality
Page 2: "Drug Capsule" © iStockphoto / Anthony Coulie
Page 3: "Information Flow Diagram" by Alexander S. Gard-Murray and Yaneer Bar-Yam
Alexander S. Gard-Murray