--as it applies specifically to diabetes:
A group from the Pittsburgh VA looked at variability across their system’s outpatient facilities in prescribing two expensive drugs for Type 2 diabetes—the glitazone-type drugs, and long-acting insulin analogues. They selected these categories because a) the best evidence shows no advantage for the vast majority of patients compared to much cheaper drugs; and b) the VA’s own formulary discourages their use.
The authors found that despite the evidence and VA policy, nearly 9 percent of the patients received a glitazone, with an incredibly wide range of 1.5-26 percent among facilities. Of those taking some sort of insulin, about 40 percent were getting the high-priced versions, with again a very wide range of 4 to 71 percent. (If we looked at non-VA practices, these percentages would no doubt be higher.)
Drs. Timothy J. Wilt and Amir Qaseem, also of the VA, commented on the study and speculated on why these results. It is notable that one factor that would normally have loomed large in data from standard community practice was not mentioned here as a possibility—enthusiastic drug marketing—since the VA, among all healthcare settings in the US, may be the most insulated against drug detailing. However, a factor that was identified was specialty guidelines and pay-for-performance criteria that reward trying to reduce the patient’s blood sugar to the lowest possible levels, regardless of what harm that might cause the patient, and regardless of the evidence that reducing sugar does not prevent major diabetic complications.
This in turn drives home the point about “controlling the channels”—that the way drugs are marketed in the US, the people who write the clinical practice guidelines and the pay-for-performance rules end up being brainwashed by the industry “line” on diabetes. This means that even physicians who are safely kept in a cocoon and protected from the wiles of detail people may still end up practicing in a way that’s heavily influenced by Pharma-think. We can run but we can’t hide.
Wilt and Qaseem have what seems a good idea about guidelines and pay-for-performance—it’s not enough to promulgate guidelines that only talk about ideal treatment. When in a case like diabetes, overuse of various drugs has been identified as a specific problem, guideline writers ought to pick criteria for overuse and specifically identify, and penalize, physicians who meet those criteria.
Gellad W, Mor M, Zhao X, et al. Variation in the use of high-cost diabetes mellitus medications in the VA healthcare system. Archives of Internal Medicine 172:1608-9, Nov. 12, 2012.
Wilt TJ, Qaseem A. Implementing high-value, cost-conscious diabetes mellitus care through the use of low-cost medications and less-intensive glycemic control target. Archives of Internal Medicine 172:1610-11, Nov. 12, 2012.