Steven Katz and Sarah Hawley of the University of Michigan authored a Viewpoint in last week’s Journal of the American Medical Association asking if we are “expecting too much” by suggesting that shared decision making (SDM) has the potential to reduce overtreatment and lower health care costs. They believe the promotion of SDM may be distracting from more effective physician- and institutional-level interventions aimed at addressing costs.
While we whole-heartedly endorse solutions that focus attention on all parts of the health care system. We believe strongly that SDM should continue to be a key element of efforts to tackle the challenges of overtreatment and rising health care costs.
In arguing that SDM doesn’t reduce overtreatment, the authors cite a study that found that when women were more involved in decisions, they were, according to Katz and Hawley, more likely to choose “more extensive treatment,” in this case, mastectomy for early-stage breast cancer.
This perspective reflects a common misunderstanding about SDM. It is not a tool to reduce procedure rates, but a way to help ensure that the patients receive the care they want in situations when they have a choice among options. Lumpectomy followed by radiation and mastectomy are both reasonable choices for many women with early-stage breast cancer, but Clara Lee and colleagues found that doctors were 10 times more likely than breast cancer patients to say that keeping the breast is the top goal when choosing a treatment approach. This mismatch between what informed patients want and what clinicians think they want—called preference misdiagnosis by Al Mulley and others is some of the strongest evidence in favor of shared decision making. Dale Collins Vidal at Dartmouth found that when women were fully informed of their options, a number of them actually preferred mastectomy. In the case of breast cancer, it’s possible that there is actually overuse of the less extensive treatment option. Without thoughtful studies like those done by Collins and others, it is impossible to know the right rate.
SDM is a process that can help address both overuse and underuse, but especially here in the U.S., overuse is likely more common. Several studies have demonstrated the impact of informing and involving patients in decisions where overdiagnosis and overtreatment is believed to occur—including screening for breast and prostate cancer and diagnostic workups for low-risk chest pain. When patients share in decisions about cancer screening, some choose to decline screening—which many experts agree is a reasonable choice. Patients with low-risk chest pain who were informed about their condition and told they could choose to be admitted for observation or to follow up with a clinician the next day understood their condition better and were less likely to want to be admitted to the hospital. When declining screening or choosing a less intensive clinical workup leads to less-intensive treatment, including unnecessary downstream tests and treatments (i.e., overuse), the potential cost savings are meaningful.
Published analyses of cost savings—including those cited by Katz and Hawley—have focused primarily on interventions for which overuse and widespread practice variation is well documented. Interventions such as SDM that involve patients in deciding when those interventions are needed (and wanted) are important components of—not alternatives to—institutional and provider-level interventions to improve the value of health care. Approaches that neglect the important role of patient preferences run the risk of being perceived by patients as rationing or limiting access to care. And while it is true, as the authors say, that many clinicians have not received training in how to facilitate SDM or integrate it into routine workflows, we think this argues for more robust SDM interventions to address these barriers through use of tools, training and implementation strategies.
Addressing the challenges of overuse and rising health care costs requires a multifaceted approach, and while SDM may be part of a solution, it is no silver bullet. Glyn Elwyn and colleagues have also written recently on this topic, arguing that it’s counterproductive and harmful to justify SDM primarily as a cost-saving measure. Informing and involving patients in treatment decisions through SDM has value in and of itself, and is an important component of any solution that aims to put patients—not cost savings—at the center of care.