High Deductible Plans and Diabetes Treatment

One reaction of employers to higher health plan costs has been to impose more of the burden of those costs on employees, both in higher premium contributions and more service cost-sharing, with higher deductibles, copays and coinsurance.  The most popular “innovation” over the last decade has been significantly higher deductibles.  One concern about this greater cost-sharing burden is that it may deter patients from getting necessary care.  A study in the Journal of the American Medical Association Network Open examines whether people with Type 2 diabetes are more or less likely to stay on recommended medications if they are in a high-deductible plan.   (JAMA Article)   The study used a matched cohort design and data from a large commercial claims set.  About 1500 patients with Type 2 diabetes were in both the high-deductible and non-high-deductible arms of the study and their medication use for 2013 and 2014 was analyzed.  The authors further studied effects on use of generic versus branded drugs.  Theoretically patients would be less likely to abandon generic medications, as they typically have lower cost and cost-sharing.  All the patients were in a traditional plan in 2013 and half stayed in one for 2014 and half enrolled in an HDPD for 2014.  The groups were closely matched on age, sex and health status scores.  The primary outcome was refill rates.  Looking at all anti-hyperglycemic medications, the difference between the two groups was not significant in regard to refill rates or discontinuation of therapy.  But for branded drugs, 20.5% of patients in a high-deductible plan did not get refills versus 14% in the non-HD plan.  For both groups, failure to refill was much higher for branded medications than for generic ones.  But the rate of discontinuance of branded therapy was similar between the two groups.  The simplest explanation is that when patients have greater cost-sharing, they switch to cheaper therapies, such as generics.  The authors suggest that this could be problematic if a brand drug is better for a particular patient than a generic one, but they offer no explanation at all for why this would be the case, and they don’t even have any evidence that in fact patients whose doctors thought they should be on a branded medication were among those who either discontinued therapy or switched to a generic.  I think the authors are trying to create a problem that doesn’t exist, on the face of their own evidence.  In fact, the study suggests that at least in regard to one form of Type 2 diabetes treatment, the high-deductible plans are working exactly as intended.  People covered by them don’t have any higher rate of discontinuance of therapy, but are switching to cheaper generics.  So what’s the problem?