Game Theory in the Health Care Industry
The United States spends significantly more on health care than any other developed country. It’s not like Americans have particularly poor health compared to other populations. If anything, the U.S. demographics and health behaviors suggest that we should have lower health costs. Furthermore, compared to other developed nations, the United States records relatively low numbers of hospital admissions, physician office visits, and drug prescriptions. But of course, there are a number of factors that contribute to the explanation of this complex story, including the fact that the U.S. doesn’t ration expensive medical technology like some other countries such as the United Kingdom and that multiple insurers drive up administrative costs. Moreover, with a greater GDP than many other developed countries, we simply have the ability to spend more on health care.
Game theory can be applied to health care to help explain why players defect from a set of strategies that is seemingly the best outcome. Players are simply acting in their own self-interest. Using Presh Talwalkar’s example from the blog linked below, suppose we have two players: the doctor and a patient who struggles with obesity and has elevated blood pressure and cholesterol. The doctor’s two strategies are either to prescribe medicine or provide advice on potential lifestyle changes (ie: exercise, diet). Prescribing drugs only takes 5 minutes while giving lifestyle advice requires more of a time investment on the doctor’s part. And lifestyle advice is considered a more effective treatment because behavioral factors are the most significant determinants of overall health. As for the patient, he can either comply with the doctor’s directions or be non-compliant, and want a second opinion from a different doctor.
The best outcome would be for both the patient and the doctor to cooperate (doctor provides lifestyle advice and patient complies), however, at the self-interested individual level, the choice to cooperate is irrational. The players in this type of game make their decision based on how they think the other player is going to act, making their best choice in response. From the patient’s point of view, his dominant strategy is non-compliance. Regardless of how the doctor acts, the patient will always prefer to be non-compliant. If the doctor prescribes ineffective drugs, then the patient will simply ignore the advice, but if the doctor actually gives good lifestyle advice, then the patient will want a second opinion to confirm. As for the doctor, he is better off quickly treating the patient by prescribing drugs since the patient isn’t going to comply anyways. So, in the end, the doctor prescribes drugs and the patient is non-compliant to whatever strategy the doctor employs.
The prisoner’s dilemma occurs in the health care industry because doctors and patients don’t have any incentive to cooperate. While there are movements toward value-based care, because the current medical system still functions under fee-for-service where physicians are paid per treatment rather than for the health outcomes of their patients, doctors are poorly compensated for their efforts if they choose to invest their time and suggest ways in which the patients can lead healthier lifestyles.
Fortunately, the same game theory that is used to explain non-cooperative behavior in the health care industry can also be used to tackle the rising health care costs. Game theory can be applied to medical spending and pricing in order to encourage hospitals, physicians, and insurance companies to cooperate. Through simulations that identify different conditions, game theory helps to reason how players will behave in a game given the particular conditions/assumptions. And in the health care industry, our main interest is identifying the conditions under which parties (players) cooperatively interact to arrive at a Nash Equilibrium that is collectively in the best interest of everyone involved.
Sources:
https://www.insidescience.org/news/game-theory-tackles-rising-health-care-costs
http://www.ncbi.nlm.nih.gov/pubmed/23329381