The old saying, “Penny-wise and dollar-foolish” comes to mind when you read the new NBER study that shows increases in copayments and/or coinsurance for prescription drugs cause people to stop taking their drugs, and as a result, die at significantly higher rates.
A domino effect
The mortality increase was unexpectedly high. A 34 percent increase in cost – or an average of $10.40 per drug – was enough to trigger a death spiral that took 32.7 percent more lives. “I never thought we would get a mortality effect of this size,” said Amitabh Chandra, one of the study’s authors. A big driver of higher mortality was the cutback of relatively low-cost, but life-saving drugs such as statins (preventing heart attacks) and antihypertensives (for blood pressure/ stroke).
These findings challenge the idea that cost sharing practices such as copays and coinsurance cause people to make wiser health choices because they have “skin in the game.”
These findings challenge the idea that cost sharing practices such as copays and coinsurance cause people to make wiser health choices because they have “skin in the game.” This principle was established decades ago when the RAND Health Insurance Experiment concluded that cost sharing results in less utilization of health services without negative impacts on health.
Not so, according to the NBER study, which tracked monthly prescription drug fills among a sample of over 358,000 healthy Medicare patients. An additional sample of 127,000 dual Medicare/Medicaid patients was also tracked to provide predictive behavior of patient spending when very little cost sharing was required.
Is this penny-wise behavior, or something else?
Why would a patient choose to cut back or totally eliminate their prescription drugs when their out-of-pocket costs go up? Wouldn’t they value their lives more than the $10.40 per medication cost increase?
“We find that small increases in cost cause patients to cut back on drugs with large benefits, ultimately causing their death.”
Current cost sharing principles assume patients are rational. For example, if prescription drug copays/coinsurances are increased, they would first cut back on less valuable medications. The NBER study did not find such rational behavior. “We find that small increases in cost cause patients to cut back on drugs with large benefits, ultimately causing their death,” the authors — Amitabh Chandra, Evan Flack, and Ziad Obermeyer — wrote. “Cutbacks are widespread, but most striking are those seen in patients with the greatest treatable health risks, in whom they are likely to be particularly destructive.”
The study makes a correlation that mortality rates rise as the number of dropped drugs rise, both in absolute numbers within a population and for individuals taking multiple medications. When faced with price increases, 18 percent more patients choose to fill no drugs, regardless of how many drugs they had been on previously, or their health risks.
Choice Fatigue
The many factors that go into an individual’s cost/benefit calculus are generally highly idiosyncratic and may change over time. The study posits that “choice fatigue” may be occurring precisely because there are too many choices and decisions to be made, with too little information to base those choices upon. For example, should the patient discontinue the most expensive drug(s), or the drug(s) that treats their least serious health condition? Might they consider dropping the drug that was added most recently to their regimen, or the drug that they’ve been taking the longest, but not feeling a lot of benefit from?
The idea of doing any cost/benefit analysis may be daunting, and the easiest choice is to simply drop them all. The study authors point out that, “in several real-world and experimental settings, having more choices pushes decision makers towards simpler options, even when they are more risky.”
… shouldn’t those designing the prescription plan be the ones to wrestle with the penny-wise problem? Specifically, what are the costs – both fiscal and ethical – of raising coinsurance under certain conditions, or of having copayments/coinsurance at all?
This begs the question: shouldn’t those designing the prescription plan be the ones to wrestle with the penny-wise problem? Specifically, what are the costs – both fiscal and ethical – of raising coinsurance under certain conditions, or of having copayments/coinsurance at all?
Planning for best outcomes
The economic argument for prescription drug copays/coinsurance stops making sense when it causes patients to make irrational decisions that end up costing their lives and/or more expensive medical care. Indeed, the authors make the argument that “improving the design of prescription drug insurance offers policy makers the opportunity to purchase large gains in health at extremely low cost per life-year.” Optimally, a plan should be designed to help patients make the best decisions for good health outcomes. Ever-present monetary constraints of course have a moderating effect on optimal plan design, but a cost/benefit analysis that leaves out the ethical repercussions of plan design choices isn’t living up to its purpose.
The economic argument for prescription drug copays/coinsurance stops making sense when it causes patients to make irrational decisions that end up costing their lives and/or more expensive medical care.
One suggestion the study offers for improving plan design is to require a value-based approach to setting coinsurance rates. Proven treatments for life-saving conditions – antihypertensives, for example – would have zero (or even negative) copays. Drugs that treat other conditions would have a standard coinsurance. It’s wisely pointed out that this approach should be taken on a regulatory level, so all insurance plans comply with the same standards. Otherwise, lower premium plans that don’t offer zero copays for life-saving drugs would counteract the intended benefit of removing obstacles to those drugs.
Other regulatory level changes could create cost sharing limits on employer plans, which account for the largest number of health insurance plans in the United States. A good model, according to study author Chandra, is the government’s Medicaid program. Out-of-pocket obligations for those who qualify for Medicaid are generally quite small. First-dollar coverage for the sickest among us offers great value for maintaining a healthy population, no matter if you are getting health coverage from a government program or employer sponsored plan.
Hopefully, the clear conclusions drawn from the NBER study on the impact of cost sharing for prescription drugs will act as a bugle reveille to everyone from health plan sponsors to health policy makers. We need to wake up, assemble and evaluate how to create effective pathways to optimal health outcomes. Change is never easy and rarely happens with lightning speed, but the conversations about how we reform healthcare in the U.S. are occurring with renewed intensity and a sense of possibility. The time for being penny-wise and dollar-foolish should be over.
