Monday, October 24, 2016

Underperforming big ideas in diabetes and breast cancer

- Kenny Lin, MD, MPH

Management of type 2 diabetes and screening for breast cancer make up a large portion of most family physicians' practices, including my own. Care and prevention for these patients is based on straightforward underlying theories of disease causation and behavior. Patients with type 2 diabetes have high blood glucose levels; treatment involves normalizing blood glucose through lifestyle modification and medication. Small, nonpalpable breast cancers eventually become large, symptomatic tumors. Smaller tumors are more likely to be curable, so undergoing regular screening mammography is preferable to not doing so.

But what if these underlying theories are wrong?

In a recent editorial in JAMA, Drs. Michael Joyner, Nigel Paneth, and John Ioannidis explored how the "big idea" or narrative that investments in genetics and information technology will lead to a revolution in health care has captured a large share of biomedical research funding and journal publications. They then illustrated how this big idea has "underperformed," as central assumptions of precision/personalized medicine have not been borne out in studies and tens of billions of dollars invested into electronic health records since 2009 have not made patient care measurably better or patient data more accessible to researchers.

Is tight glycemic control for patients with type 2 diabetes mellitus an underperforming clinical big idea? In an analysis in Circulation: Cardiovascular Quality and Outcomes, Drs. Rene Rodriguez-Gutierrez and Victor Montori compared clinical policy statements and practice guidelines for patients with type 2 diabetes between 2006 and 2015 with evidence from randomized controlled trials. Despite little or no evidence that tight glycemic control (hemoglobin A1c <6.5 or 7.0%) improves microvascular or macrovascular outcomes compared to less strict hemoglobin A1c goals, the majority of guidelines continued to endorse tight control for one or both of those outcomes. (In contrast, AFP editorials and articles have long asserted that "Physicians should not let well-intentioned but misguided concern for glucose levels distract them from attending to other interventions that more profoundly affect mortality [in patients with type 2 diabetes]: smoking cessation, blood pressure control, metformin therapy, and lipid reduction.")

And do small breast tumors detected by mammograms become large, lethal ones? Sometimes, but not as often as most patients and physicians think, according to an observational study in the New England Journal of Medicine that concluded: "Women [with tumors detected on mammography] were more likely to have breast cancer that was overdiagnosed than to have earlier detection of a tumor that was destined to become large." This study also concluded that most of the reduction in breast cancer mortality over the past 40 years could be attributed to improved systemic therapy rather than earlier tumor detection. In an AFP editorial on counseling women about breast cancer screening, Dr. Mark Ebell and I discussed the benefits and harms of mammography in younger women and noted that for every additional breast cancer death prevented by starting at age 40, two women will be overdiagnosed with (and overtreated for) breast tumors that never would have become clinically apparent.