The Potential of Value-Based Insurance Design

A study published last week in the New England Journal of Medicine analyzed one area of value-based insurance design (VBID) that has been commonly researched: the effect of waived or lowered copayments for interventions known to improve health outcomes. Choudhry and co-investigators of the Post-Myocardial Infarction Free Rx Event and Economic Evaluation (MI FREEE) trial (http://www.nejm.org/doi/full/10.1056/NEJMsa1107913?query=featured_home#t=articleResults) tested whether removing copayments for statins, beta-blockers, angiotensin-converting enzyme inhibitors, or angiotensin-receptor blockers in patients who had suffered a myocardial infarction lowered the risk of secondary, post-infarction events. Each of these products has been shown independently to prevent, when taken appropriately, secondary coronary events. This trial tested a study group who had the copayment eliminated for these agents post-infarction against a control group whose usual copayments were unchanged, and were based on existing formulary structures in their health plan (tier 1 for generics, tier 2 for preferred branded products, and tier 3 for nonpreferred brands).

The primary study outcome was a composite of the first readmission for a major vascular event (fatal or nonfatal acute myocardial infarction, unstable angina, stroke, or congestive heart failure) or coronary revascularization (coronary bypass, stenting, or angioplasty). Pre-specified secondary clinical outcomes included the rate of total major vascular events or revascularization.

The study group comprised 2,845 patients, and the control group included 3,010 patients. There were no significant demographic differences between the groups. In the control group, rates of adherence were 35.9% for ACE inhibitors or ARBs compared with 41.5% in the group without any copays. Similarly, those in the usual copayment group had a 45.0% rate of adherence for beta-blockers compared with those without copays, who adhered to beta-blocker therapy at a rate of 49.4%. The differences were 49.0% vs. 55.2% for statins, respectively, and 38.9% vs. 44.3% for all three medication classes. These do not sound like big differences, but they were statistically significant at P < .001 for all comparisons. As a result of the large patient samples, relatively small differences in adherence rates can result in statistical significant findings.

Owing to the adherence gains from waiving copayments, the health outcomes were somewhat improved. A total of 562 fatal or nonfatal vascular events or revascularizations were registered in the usual coverage group compared with 493 patients in the group without copayments (hazard ratio, 0.93; P = .21).

More notable reductions were found in the incidence of secondary outcome events: individual outcomes that composed the composite primary outcome. For example, the rate of stroke was lowered by 31% (hazard ratio, 0.69; P = .03) when copayments were eliminated, and the hazard ratio for the first major vascular event was reduced by 14% (hazard ratio, 0.86; P = .03) in the group without copays.

This VBID intervention showed a limited difference in health spending between the two groups ($66,008 in the full-coverage group vs. $71,778 in the usual-coverage group, P = .68), despite higher pharmaceutical costs in the former.

The investigators pointed out that “despite the improvements in adherence that we observed, overall adherence remained low—less than half of patients in the full-coverage group were fully adherent to their prescribed therapies.”

It seems that even these relatively slight improvements in adherence through the use of VBID resulted in gains in health outcomes. It is well known that adherence to drugs, such as statins, drops off over time in patients with asymptomatic disorders. The outcome of this study suggests that factors other than cost need to be considered for VBID to reach its full potential.