The effect of pharmacological and psychotherapeutic interventions on risk and loss aversion in anxiety

Background:

Anxiety disorders are highly prevalent and have been associated with altered economic decision-making such as preference for certain over risky decisions, despite lower possible monetary outcomes. This aversion to risk is thought to be driven by two key factors: aversion to the uncertainty of the choice, or aversion to the negative outcome itself [2]. Previous research attempting to disentangle these influences suggests that anxiety-related preference for lower value certain options during a digitised gambling task, is likely driven by an aversion to risky decisions even in the absence of potential monetary losses, highlighting risk aversion as a potential target for anxiety treatment interventions [1]. This study aims to replicate patient-control differences in risk averse behaviours and investigate whether these are modulated by first line pharmacological and psychotherapeutic interventions; Escitalopram and cognitive behavioural therapy.

Methods:

At two time-points, 94 healthy controls, and 202 anxiety patients completed a loss-risk aversion gambling task, which independently manipulates loss and risk aversion through the inclusion of gain only (gain/£0) and mixed (gain/loss) gamble trials. Data was collected across two randomised controlled trials, comparing the effect of (a) sub-chronic escitalopram administration (n = 18) versus placebo (n = 21), and (b) cognitive behavioural therapy (n = 65) versus waiting list (n = 44). Prospect theory [2] derived models were fit to the data of each group at each session. Parameter estimates were compared using (a) pairwise t-test to replicate patient-control differences, and (b) linear models to examine the effects of intervention on follow-up estimates adjusting for baseline.

Results:

The results of the patient-control analysis suggest that in a highly powered sample, patients exhibit greater loss aversion relative to controls (t(223) = -3.59, p < .001), but no difference in risk aversion (t(206) = -1.32, p = 0.19). Although no difference in gamble propensity was observed at the behavioural level (F(1, 294) = 1.43, p = .23). The intervention effects analysis revealed a significant treatment group by timepoint interaction on the loss aversion parameter estimate (F(3, 135) = 2.85, p = .04), such that at follow-up, loss aversion was significantly reduced following a period on the waiting list relative to all other interventions (t(135)’s > 2.04, p’s < .04).

Conclusion:

Contrary to previous findings [1], the results of this high-powered study demonstrate that relative to controls, anxiety patients exhibit greater loss aversion but not risk aversion, although this did not translate into a behavioural effect. Counterintuitively, we also found that whilst neither CBT nor sub-chronic escitalopram or placebo administration improve risk or loss aversion, loss aversion may in fact reduce after a period of time on the waiting list. This suggests that both psychological and pharmacological interventions may modulate loss aversion by attenuating a reduction in loss aversion that may otherwise occur over time.

References

1. Charpentier, C. J., Aylward, J., Roiser, J. P., & Robinson, O. J. (2017). Enhanced risk aversion, but not loss aversion, in unmedicated pathological anxiety. Biological psychiatry, 81(12), 1014-1022.
2. Kahneman, D., & Tversky, A. (2013). Prospect theory: An analysis of decision under risk. In Handbook of the fundamentals of financial decision making: Part I (pp. 99-127).