Discover What Works
The focus in medicine has been to improve outcomes and control costs through the application of evidence-based medicine. The idea is that we can improve outcomes if everyone would just follow the clinical guidelines for an individual disease state. This has become the primary focus of pay for performance and the Joint Commission’s mandates on quality indicators, such as early antibiotics in pneumonia care.
This is sound thinking, until you realize that the vast majority of the decisions we have no definitive answer. For example, BMJ Clinical Evidence estimates that of the 2,500 treatments used for a variety of clinical problems, only 36% of them were deemed “beneficial” or “likely to be beneficial,” while 46% were of unknown benefit.3 In the absence of evidence, we are forced to use our clinical judgment, a surprisingly scary proposition when affordable, high-quality care is the goal.
This clinical judgment is what policymakers refer to as discretionary decision-making. The problem is that there is great variability in what experienced, prudent physicians judge to be appropriate. Recently, 1,275 physicians were asked about their decision-making around clinical scenarios with variable levels of evidential support. When asked when they would recommend a routine follow-up visit for a patient with well-controlled hypertension, 47% of physicians in high-resource-use areas (e.g., Miami), compared with 9% of physicians in low-resource-use areas (e.g., Portland), would recommend followup within three months compared with after three months. Management of an elderly clinic patient with new-onset chest pain was met with similar levels of variability in cardiac testing, referral to cardiology, and admission to the hospital between the higher- and lower-resource areas.
Those results contrasted with the consistent use across all spending groups for relatively proven modalities, such as mammographic screening in patients 50 to 70 years old.4 When the definitive answer isn’t available, we are left to use our clinical judgment, which often results in overuse of resources without benefit, and possibly harm.
We need to augment our focus on adherence to current guidelines with the discovery necessary to develop future guidelines. This requires that we move toward better information about what works and what doesn’t.
Enter the Obama administration’s commitment of significant resources to NIH research and comparative effectiveness research. The former allows for continued discovery of new technologies, while the latter informs clinicians about which technologies work the best for a particular clinical disease state.
Coordinating Fragmented Care
As hospitalists, our daily to-do lists are riddled with the consequences of fragmented care. We spend hours trying to track down primary-care physicians, finding test results from outside facilities, and coordinating complex care with multiple providers across multiple continuums. This results in inefficient and costly hospitalist systems, repetition of expensive tests, and overall worse patient outcomes. Thus, the stimulus bill’s push will be to build information technology (IT) infrastructure, such as electronic medical records, with the goal of making safer, more efficient systems that improve outcomes while cutting costs.
While the devil will be in the details—and with medical IT, the details can be daunting—I think this is a wise investment in our future.
We must get healthcare costs under control and improve the quality of care. There are myriad thoughts and proposals on how to accomplish this. The new administration is betting that discovery and technology will provide the answer to what to do with an acute-chest-pain patient and, in turn, the brewing healthcare crisis in America. That doesn’t sound like healthcare rationing to me, Dad. TH
Dr. Glasheen is associate professor of medicine at the University of Colorado Denver, where he serves as director of hospital medicine and the hospitalist training program, and as associate program director of the Internal Medicine Residency Program.