Behaving irrationally: High-deductible health plans

In theory, high-deductible health plans should create incentives for rational spending in healthcare. In practice, they often do not.

I was interested in Professor Chandra’s comments at the end of class this week about the incentives created by high-deductible health plans. It would seem that having a higher upfront, out-of-pocket fee would create incentives for patients to be selective about how to spend that money, in theory encouraging them to research care options, switch to low cost and high quality providers, and be more discriminant in the care they seek. However, studies have shown that patients often don’t respond rationally to these incentives. Some reduce all care indiscriminately. Some have worse adherence to medication regimens (particularly asymptomatic conditions like high blood pressure or high cholesterol), and one study showed minimal switching to less expensive generic medications.  The article points out that a rational consumer 1) behaves rationally, 2) knows his/her preferences, and 3) has all relevant facts about what he/she plans to purchase. There are many reasons to explain why these don’t apply to a consumer of healthcare today.

Regarding the first criterion, experiments in behavioral economics have shown that people often don’t behave rationally. The article notes that people are highly loss averse (they experience twice the dissatisfaction in the face of loss compared to the satisfaction they experience when they gain something). So the idea of having to spend money on high deductibles is a significant psychological barrier that can impede pursuit of care that the patient knows he/she may benefit from.

Regarding knowing one’s preferences and having the relevant facts to make a decision, information asymmetry is a big issue. Medical literature is often complex and full of jargon that can be difficult for a lay-person to understand and then weigh pros and cons for himself/herself. Even with the help of sophisticated decision aids that attempt to simply concepts and focus on outcomes patients care about, patients often feel most comfortable making medical decisions jointly with or deferring decisions altogether to their doctors who have the expertise. The problem is that in a fee-for-service world, physicians are often subject to a number of conscious and unconscious biases toward more care. In addition, patients often have limited information on relative costs of procedures. Furthermore, individual outcomes for each purchase are never guaranteed (taking a statin may not prevent a heart attack). These gaps in information and uncertainties about outcomes make it even harder for patients to make rational decisions.

After identifying barriers to rational decision making, the article makes several suggestions for enabling better decisions. Health savings accounts can reduce feelings of loss when patients have to pay high deductibles, given that money has already been set aside for healthcare spending. There are increasing efforts to increase price transparency throughout medicine, though these are admittedly still in progress. Finally, shifting away from fee-for-service can reduce provider incentives to prescribe more care (that a patient may indiscriminately reduce across the board because he/she doesn’t have information to choose among the recommendations) and align patient and physician incentives more closely.

8 thoughts on “Behaving irrationally: High-deductible health plans

  1. I think this is such an interesting issue within healthcare–how do we design the optimal plan to create the right incentives for patients/consumers?

    My main concern with high deductible plans is that patients often don’t seek out the preventative care that they need (unless those consultations are free, which many are now by law under Obamacare). I think your piece does a great job of laying out some of the behavioral economics reasons why, such as loss aversion. In addition, I think a key issue can be that patients are likely to estimate their level of risk for negative health outcomes to be lower than it actually is. We like to think of ourselves as relatively healthy, and if we believe there’s a low likelihood of the doctor finding anything that will improve our health outcomes (and save us money) down the road, we choose not to visit the doctor.

    Patient education around the costs of procedures is definitely a pre-requisite for high deductible plans improving incentives, but so is education around the necessity of preventive care.

  2. Such an interesting topic! Can you please share the article link? I completely agree that the theories from behavioral economics can have powerful implications on how we think about designing healthcare incentives. Sam, great point on preventative care under-utilization. In addition to poor risk estimates, I think we are seeing the behavioral economics theory of “present bias” at work. Present bias is another concept that describes the tendency for individuals to value gains or losses in the present more than the future. Therefore, to the average human, the inconvenience of scheduling and going to an appointment holds more weight than the vague concept of a health implication some time in the future. Making consultations convenient to set up and attend, in addition to making losses felt early-on (e.g. monetary penalties for not gong to yearly check-ups), could do the trick. These steps are more likely to move the needle on preventative care usage then seemingly-rational incentives of price pressures and increasing access to information.

    1. Sorry forgot the article link! http://www.gfoa.org/sites/default/files/GFR021628.pdf

  3. Health deductibles are such an interesting topic. As mentioned above, people are notoriously bad at estimating risk. This is particularly an issue in chronic disease conditions where the impact of the disease is often silent until an adverse outcome occurs. This makes it difficult for consumers to evaluate the value of their care so they do not seek medical attention as they are deterred by paying a high deductible. As noted in your post, this issue is further complicated by the lack of information about pricing in the healthcare market. One problem is that most doctors have no idea of the cost of individual tests or treatment plans. Even if they do know the cost of the drug or test, the degree of coverage dependent on what insurance a patient has changes this. Thus I completely agree that shifting away from fee for service is necessary. This will not only incentivize better utilization of the healthcare dollar but also promote the need for transparency around cost.

  4. Great post on a very interesting topic thanks for sharing. Borrowing from Public Reporting, Consumerism, and Patient Empowerment (Huckman and Kelley, 2016) published in the NEJM, the confluence of several recent trends in health care, including 1) increases in cost-sharing (deductibles and copayments) and 2) public reporting of quality data etc. are part of a larger movement of health care consumerism that is purported to help facilitate shared-decision and reduce waste etc. Unfortunately as you point out higher cost-sharing is leading to reduced utilization that may not actually be associated with better (i.e. more prudent) decision making on the part of consumers. Poor understanding of not only of quality information but also of patient’s individual health needs is pervasive and payers are responding to decades of wasteful spending that is largely a result of the perverse incentives inherent in fee-for-service payment models by shifting these costs (and the risk) onto patients. I think that as mobile app health solutions (as extensions of primary care) continue to proliferate, assuming there is better integration with EMR systems moving forward, patients will become more informed and involved in their care, which hopefully help preserve value as the as providers and payers continue to concern themselves with costs.

  5. Great article, Nicole. Thank you for sharing. This is an incredibly important issue since it is exactly what is going on in the U.S. healthcare market right now. Deductible payments have skyrocketed over the past 5 years, yet there has not been much of a decrease in patients seeing care providers. With no apparent correlation, it begs the question will the ACA be successful? Patients are still demanding care and the costs keep tallying. With no curtailment in frivolous patient services, payers and providers are still on the line. Moreover, providers will most likely not get paid on a fee-for-service basis. This means they will be seeing patients for non-serious circumstances and not get paid for their time. It seems that since patients are not rational actors when it comes to their health, the system may need some adjustments in order to be financially viable and sustainable.

  6. Thank you, Nicole! You note that it’s hard for patients have limited information on the relative costs of alternative treatment plans, and I would argue that doctors don’t know enough about these relative costs either! When prescribing a medication, a physician has little information on the actual cost and out-of-pocket cost a patient will incur. How about the relative cost between two different specialists? Nope. How about the relative cost between procedures at different locations? Nope. The list goes on and on. Doctors can play an important in being a rational steward of resources, but they, like patients, need more information to do so effectively.

  7. Great topic to address. For me, the main feeling that this brings up is a potential disproportionate negative impact on those people with low education, and likely less income. The healthcare system is complex to navigate, and I feel that asking consumers to take a large personal financial responsibility for this – could have adverse effects on a weaker segment of the population. There are certain areas, whether products or services, where the government or a semi-private body needs to take an active role in order to instrument people to make optimal decisions. To me, healthcare is such an area.

Leave a Reply