Doctor–patient differences in risk and time preferences: A field experiment

Abstract

We conduct a framed field experiment among patients and doctors to test whether the two groups have similar risk and time preferences. We elicit risk and time preferences using multiple price list tests and their adaptations to the healthcare context. Risk and time preferences are compared in terms of switching points in the tests and the structurally estimated behavioural parameters. We find that doctors and patients significantly differ in their time preferences: doctors discount future outcomes less heavily than patients. We find no evidence that doctors and patients systematically differ in their risk preferences in the healthcare domain.

Risk attitudes in medical decisions for others: An experimental approach

“…But whether doctors are really so risk averse in practice is a question that needs further exploration. Galizzi, Miraldo, Stavropoulou, and van Der Pol (2016) conducted a frame field experiment that found mild but significant risk aversion among doctors in the health domain, and a more pronounced risk aversion in the monetary domain. Moreover, the question of whether risk preferences that are elicited in laboratories using small populations and relatively low stakes can be transferred to medical contexts is a contentious issue in the experimental economics literature (Charness et al, 2013) and reflects a wider debate concerning the validity of lab experiments for explaining behaviors in natural settings (see Galizzi and Wiesen, in press;Kessler & Vesterlund, 2015).…”

“…However, it has been scarcely applied in the health domain, where standard gamble methods are the most commonly used methods to elicit preferences towards health risks. The exceptions are Galizzi, Miraldo, and Stavropoulou () and Galizzi, Miraldo, Stavropoulou, and M. van Der Pol (), which adapt the questions of Holt and Laury () to the health domain in order to analyze whether patients' risk preferences differ across the health and financial domains and whether risk preferences differ between patients and their doctors. The Holt & Laury method presents a structured menu of binary choices, where lotteries are compared in pairs, thereby providing interval estimates of risk aversion.…”

Rabin's paradox for health outcomes

“…To our knowledge, our study is one of the first finding risk aversion for moderate individual health outcomes, 6 with another example being Breyer and Fuchs (1982) who consider gambles over days with a 2-hr headache. Risk aversion for larger individual health outcomes, for example, in the range of 0.5 to 20 years of life, is observed frequently (e.g., Galizzi, Miraldo, Stavropoulou, & van der Pol, 2016, Attema, Brouwer, & L'Haridon, 2013, Attema, Brouwer, L'Haridon, & Pinto, 2016, Van Der Pol & Ruggeri, 2008, Oliver, 2018, albeit these studies used a different methodology (i.e., certainty equivalences). For societal outcomes, studies have, for example, found risk aversion for life years (Eraker & Sox, 1981) or lives (Kemel & Paraschiv, 2018).…”

health econometrics

“…These methods typically use hypothetical trade-offs. Galizzi, Miraldo, Stavropoulou, and Van der Pol (2016), for example, compare time preferences for hypothetical healthcare outcomes between doctors and their matched patients. Coller and Williams (1999) seems to be the first incentive-compatible study to assess time preferences for monetary outcomes with real monetary payments.…”