In his Editorial, Dr. Klepser considers the effectiveness and the future of value-based insurance design.
The idea makes a lot of sense. If there is something you want people to use more of and price is a barrier, you should lower the price. Furthermore, most people would agree that if you can increase the use of evidence-based, cost-effective healthcare, you should improve care and may even be able to reduce the overall cost of healthcare by eliminating expensive downstream services and procedures. Such is the idea behind value-based insurance design (VBID)—reduce patient cost sharing on services proven to be a good value for the money spent as a way of increasing utilization and improving adherence.1 To date, however, there is limited evidence about the effects of VBID as it relates to medication adherence.
In this issue of The American Journal of Managed Care, Sedjo and Cox examine a reduction in patient cost sharing on patient adherence to 3-hydroxy-3-methyl-glutaryl-CoA reductase inhibitor (statin) therapy.2 The authors are to be commended for their efforts in looking at an important and timely issue. They use the patent expiration of Zocor (simvastatin) and subsequent switching of patients from the brand name product to the generic version along with its lower copayments as a natural experiment. They compared medication possession ratios (MPRs) for the patients moved to the generic medication with patients taking other brand name statins. Although they showed a statistically significant improvement in adherence after the reduction in copays, the actual improvements were modest. The modest change in adherence raises a number of questions. First, was this natural experiment truly consistent with VBID? Second, and more importantly, if it was, what does this mean for the effectiveness of VBID?
To determine whether the current study should be considered an examination of a VBID, there are several issues to consider. Encouraging the utilization of generic medications by lowering patient cost sharing for generic medications is consistent with VBID. In this study, however, there was no systematic change in the generic cost sharing relative to branded products. Some patients faced no reductions in cost sharing, while it was relatively modest for others. In addition, the authors recognize that this natural experiment did not include any patient outreach, education, or promotion about the decreased cost sharing; therefore, there was no clear message that the generic product was a better value for patients than a competing brand name statin. The only patients who would be aware of the “higher value” generic would be those who were already using simvastatin. Finally, this study included a mandatory switch to a generic medication in addition to a reduction in copayments.
Although the safety and effectiveness of generic medications continues to be promoted as identical to the branded products, there are still patients who do not view them as direct substitutes for their brand name counterparts. The results of this study hint at that effect, since patients with no reduction in cost sharing had a decrease in MPR of 3.22%, which is larger than the decrease in the control group (2.02%). Therefore, it is difficult to assess how patient behavior would have changed if there had only been a reduction in cost sharing. Although this study definitely looked at elements consistent with a VBID, the lack of promotion of the benefit change and the mandatory substitution of the generic product suggest that this natural experiment should not be viewed as indicative of broader programs. That said, the results seen in this study are not inconsistent with other studies looking at VBID. This leads to the bigger questions raised by this study: (1) Why are the results so modest? (2) What does it mean for the future of VBID? Although the answers are complex and vary for every medication or service, there are 2 possible explanations: they do not address the value patients place on healthcare, nor do they account for the role price plays in signaling quality.
Many variables influence patients’ decisions to use a particular service. With VBID, the cost barrier is reduced for services deemed to be of high value by someone other than the patient. Other important factors that influence patients’ decisions to use a given product or service are not directly influenced by VBID. Most notably, there is no accounting for the value patients place on a given service. For example, if patients appreciate the long-term value of statin therapy, they likely are already adherent. Under that scenario, the effect of a VBID is to subsidize behavior that was already taking place. Conversely, if asymptomatic patients do not notice a direct improvement in their functional health status from taking a statin, they may assign little or no value to their statin therapy. If that is the case, patients may not be willing to take a medication even if it were free. This is especially true when people have already decided not to use a product or service because of its price. Consumer behavior literature suggests that people will find reasons to support their initial decisions to explain why they did not utilize a service. If this is the case, patients are less likely to modify their behavior based solely on a reduction in price.
It is generally accepted that healthcare is unique in that people do not usually make the decision about the products or services they need, nor do they fully understand the value of the product or service. At its best, VBID has an opportunity to influence both the product or service selected and a patient’s valuation of it. Guided by evidence-based medicine and using price as a signal to both providers and patients, VBID has the potential to encourage provider use of high-value products and services. Likewise, an educated patient would recognize the cue that lower copays mean higher value services. Unfortunately, this is contrary to most goods and services in our society. Basic economic theory suggests that the more people value something, the more they are willing to pay for it and the more they are willing to consume at any given price. For this reason, price is seen as an indicator of quality, with higher prices being equated to more valuable products and services. From the patient’s perspective, VBID turns this notion upside down—the more value (clinical benefit) a product or service provides, the lower the cost to patients, which they may translate to mean less valuable (or effective).
Does any of this mean that VBID is a bad idea or that researchers should not continue to look at natural experiments? No, but it does indicate that there is more to VBID than lowering patient cost sharing. This is especially true when the lowered cost sharing is also a shift to a generic medication.
Address correspondence to: Donald G. Klepser, PhD, MBA, Department of Pharmacy Practice, University of Nebraska Medical Center, 986045 Nebraska Medical Center, Omaha, NE 68198-6045. E-mail: firstname.lastname@example.org.
Funding Source: None reported.
Author Disclosure: The author reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
1. Fendrick AM, Chernew ME. Value-based insurance design: aligning incentives to bridge the divide between quality improvement and cost containment. Am J Manag Care. 2006;12(1):1-7.2. Sedjo RL, Cox ER. Lowering copayments: impact of simvastatin patent expiration on patient adherence. Am J Manag Care. 2008;14(12):813-818.