Controversy has surrounded the latest publication of the Diagnostic and Statistical Manual (DSM) of mental health disorders, in part because of concerns that the guideline pathologizes many behaviors that some people might consider normal, theoretically increasing the opportunity to prescribe pharmaceuticals for non-pathological behavior. But beyond the field of psychiatry, there are increasing concerns that “medicalization” may be doing more harm than good for patients (especially where tests and therapies have marginal benefit but potentially great risks), and may be influenced by profit motives and desires to define disease so expansively as to intrude on normal living to a stifling degree. A recent comprehensive study of medical panels’ decisions about expanding disease definitions shed some light on this debate, and revealed some concerning findings…
In 2009, the Institute of Medicine published a landmark report entitled “Conflict of interest in medical research, education, and practice”. In the report, the IOM found “widespread relationships with industry have created significant risks that individual and institutional financial interests may unduly influence professionals’ judgments” in medicine. Subsequent recommendations from the IOM revealed that a major area of concern was the writing of medical guidelines that physicians around the country (and, indeed, around the world) typically use to decide when to screen patients for diseases, how to define a disease, and when and how to treat such diseases. The influence of these guidelines on day-to-day practice cannot be understated: the guidelines are the subject of testing among medical students and residents, periodic examination among providers to obtain board certification; and the subject of numerous initiatives intended to reduce morbidity and mortality under the premise that we can detect diseases earlier, treat them more effectively, and reduce long-term disability and/or death.
But the IOM found so many conflicts among the developers of these guidelines that it suggested “whenever possible” that guideline-developing panels either “should not have” conflicts of interest, or that only a minority of members should have conflicts and that chairs of the committees should be free of conflicts. The problem was that accepting fees and contracts from pharmaceutical and device companies had become so common that few senior researchers and experts in sub-specialties were available to write these guidelines if conflicts excluded participation; hence, “disclosure” of conflicts became the mantra, even though such disclosure has been shown to be insufficient to prevent unconscious biases such as the desire to prescribe a drug more aggressively than would be objectively justified based on a risk-benefit ratio.
This month, a peer-reviewed article published in PLoS Medicine provided a fascinating glimpse into how the internal politics of disease-defining panels remains a subject of serious conflicts of interest. The study was a cross-sectional evaluation analyzing major medical guidelines published between 2000 and April 2013. The researchers, who themselves were not on these panels, identified which panels recently made decisions about definitions or diagnostic thresholds for conditions that are common in the US. The researchers then catalogued the various conflicts of interest including industry ties among the panel members that were disclosed, and correlated these to the changes in disease definitions proposed by the panels.
Among publications in which expert panels proposed changing disease definitions and/or diagnostic criteria for fourteen common conditions, panels proposed changes widened the disease definition for ten of the fourteen diseases, and narrowed it for one disease. Only six of the panels mentioned possible harms of the proposed changes; based on a systematic analytic framework, the researchers were able to find that none of the panels appeared to rigorously assess the downsides of expanding definitions.
Of the fifteen panels involved in the guideline development, twelve included members who disclosed financial ties to multiple companies, with 75% of panel members having such ties. The median number of ties to the industry was seven companies per person on the panel. The most common ties were to companies marketing drugs for the disease being considered by that panel. Sadly, there was no notable difference between the ties among panels that released their guidelines before the 2011 IOM recommendations, and those that released their recommendations after the IOM recommendations.
The article highlights the pervasiveness of conflicts of interest among medicine’s top brass. It also parallels our recent analysis in which we also found numerous similar conflicts of interest and industry ties in international aid organizations and charity groups that decide major global health campaigns. Overall, this suggests a continuing culture in which private for-profit interests meld with public health decision-making in ways that may expand the reach of medicine, but not clearly in the interest of patients.