Does the deductible influence the demand for healthcare at the end of the year?

How do individuals respond to price changes? This question is central to economics. According to classical theory, all market participants act "rationally," considering all aspects of price in their purchasing decisions. This means individuals are forward-looking and account for future price developments in their present-day choices. Whether this is actually the case can be studied through the demand for healthcare services. Due to deductibles and co-payments, the “short-term” and “long-term” prices for healthcare services can differ, as cost-sharing resets to zero at the start of a new year. If a patient has exceeded their cost-sharing limit during the current year, the “December price” for a medical service is significantly lower than the “January price”. Chronically ill patients who reach their deductible every year experience this “price jump”, but in the long term, the price remains exactly the same. They save not a single franc by moving services forward to take advantage of the lower December price. However, if we observe that patients change their behavior despite no long-term savings, it suggests that cost-sharing leads to unintended behavioral distortions.

To determine whether individuals adjust their healthcare demand (irrationally) based on short-term price changes, we focus on patients with chronic illnesses who require regular medications: depression (antidepressants), type I diabetes (insulin), type II diabetes (antidiabetics), high blood pressure (antihypertensives), or high cholesterol (lipid-lowering drugs). These medications incur high healthcare costs, leading affected patients to often choose the lowest deductible. Due to these high costs, they typically reach their deductible early in the year. While they also experience the sharp increase in cost-sharing at the start of each year, the long-term price remains unchanged across the year. Medications are particularly suitable for analysis because chronically ill patients have a consistent need for them, and medications are also storable. The timing of their purchase is therefore manageable, allowing us to closely observe potential price responses.

If patients with chronic illnesses respond to the price jump at the turn of the year, they would purchase more medications before the new year than afterward. To investigate this, we use daily data on the amount of defined daily dosage (DDD) purchased on a given day. This data structure allows us to perform a regression discontinuity analysis, estimating the difference in the average number of DDDs purchased at the end of the year compared to the beginning of the next year. Since we observe the same patients before and after the turn of the year, and nothing other than the cost-sharing changes, we can identify the causal effect of the price change on the amount of DDD purchased. Our results are clear: At the end of the year, when patients can obtain their medications (almost) “free",  they purchase significantly more medications than after the turn of the year, when they must pay most of the costs themselves. The difference varies by medication group, ranging from 0.6 DDD per day for antidepressants to 1.1 DDD per day for antidiabetics. Scaled up to a month, this corresponds to an average of about 30 DDDs, essentially a one-month supply. In other words, on average, one additional month's supply is purchased in December compared to January. Over the entire year, an average of 2.2 DDDs are purchased per day. Our estimated effect of 1.1 DDD corresponds to approximately 50% of that total—indicating a substantial response to the short-term price change.

Between Christmas and New Year, access to healthcare institutions is limited due to the holidays. Patients may therefore purchase more medications as a precaution to ensure they are adequately supplied during the holiday period. If we do not account for this “holiday effect”, we risk overestimating the effect of the price change. To test this, we apply the same method to the summer months, where a similar holiday effect might be expected. Indeed, we find evidence of such an effect during the summer. However, when applied to the year-end period, at least one-third of the price effect remains. This means that while holidays and festive periods do play a role, the short-term price change at the turn of the year remains a significant factor driving the increased demand for medications in December.

Another potential factor is the so-called “January slump”. Patients might buy their medications in December to avoid expenses in January when disposable income might be limited. To investigate this, we compare patients who are more likely to face financial difficulties with those for whom this is less likely. One way to do this is by comparing individuals who receive premium subsidies with those who do not, as these groups differ significantly in income. However, patients receiving premium subsidies and those not receiving them show a comparable price response at the turn of the year. From this, we conclude that the “January slump” does not influence the observed patient behavior.

Our results show that people respond to the resetting of cost-sharing at the beginning of each year. Short-term price changes influence their behavior—even though this does not save them any out-of-pocket costs in the long term. While they act with foresight in the short term, they seem to struggle with fully grasping the complexity of the longer-term implications. From the perspective of health insurers and in terms of overall costs, these “Christmas purchases” are not fundamentally problematic, as the saved cost-sharing simply shifts to the following year for patients. However, our findings provide further evidence that the design of cost-sharing mechanisms can lead to irrational and potentially even health-compromising behavior. One option could be to exempt chronic medications from the deductible entirely and instead implement a constant cost-sharing structure. Additionally, patients should be better informed so that they can understand the rules of health insurance and make well-informed decisions.