Applications of behavioural economics to health: three studies in health decision making and behaviour

Abstract

To maximise the health of society within a limited budget, decision makers in public health systems need to decide which health treatments to fund and how best to support engagement by patients with treatments. Current practice is heavily influenced by rational choice theory. In this thesis we apply an alternative behavioural economics perspective to inform decisions on which treatments to fund and how to support engagement with services. Decisions on which treatments to fund are often informed by the expected gains in patients’ quality-adjusted life years (QALYs). QALYs are derived from objective mortality data weighted by appraisals made by members of the general population of the likely impact on well-being associated with health states (the Q in QALYs). Concerns have been raised about the way in which the quality component of QALYs is calculated, leading to calls for ways to obtain experience informed general population appraisals. In Chapter 2 we test the effect on general population preferences (N = 155) of being informed of patients’ mean ratings of their health state and whether the direction of an effect depends on people’s prior beliefs of patients’ mean rating of the health states. We find that when the mean ratings given by patients are higher (lower) than expected, participants in the intervention group provide significantly higher (lower) valuations than participants in the control group. In Chapter 3 we examine whether people (N = 1259) selectively underestimate the well-being consequences of moderate anxiety / depression as compared to other dimensions of health, and we test if being informed of actual changes in well-being associated with health states changes appraisals of their relative undesirability. We find that people provided with information on the consequences of health states for life satisfaction or for day affect report a higher preference for avoiding living with moderate anxiety / depression. Both Chapters show that informing people of these summary measures before they appraise health states is a feasible way to obtain experience informed preferences and that experience informed preferences differ to those obtained using the current method. Non-attendance for hospital appointments is a problem. One way to increase attendance is to improve the accuracy of waiting lists by writing to patients to check if a procedure is still required. However, the did not return (DNR) rate to such letters is substantial. In Chapter 4 we test (N = 2855) whether the DNR rate is reduced by introducing nudges to validation letters. We find that the redesigned validation letter reduced DNRs, by 4.73 percentage points or 19.73%. Taken together these studies show the importance of applying a behavioural economics perspective to inform decisions on how to maximise the health of society.