Skip to main content
eScholarship
Open Access Publications from the University of California

A Team-Based Behavioral Economics Experiment on Smoking Cessation

  • Author(s): White, Justin S.
  • Advisor(s): Dow, William H
  • et al.
Abstract

Tobacco use is a leading cause of death worldwide, yet smoking cessation services are not widely available in many low-resource settings. Popular approaches also fail to help smokers to maintain self-control and motivation. The degree to which peer pressure promotes self-control in team-based health interventions remains largely untested. Moreover, peer pressure and cash incentives have rarely been mobilized in tandem. To this end, we conducted a randomized controlled trial in 42 villages in Thailand to test a novel intervention that combines commitment contracts for smoking cessation with team incentives that activate peer pressure. We randomly assigned 201 participants, 11% of all smokers in the study area, to a control group that received smoking cessation counseling or a treatment group that received counseling plus a commitment contract, team incentives, and text message reminders for smoking cessation. We find that, relative to the control group, the intervention increased biochemically verified smoking abstinence by 25% points at six months (three months post-intervention). Moreover, the intervention cost about $300 per marginal quitter, less than half that of common smoking cessation aids in Thailand. We find evidence that exogenously selected teammates had a large causal effect on each other's outcomes. The team effects are heterogeneous with respect to participants' ex ante quit predictions: the success of less confident smokers increases with a teammate's degree of self-confidence whereas the success of more confident smokers does not change. Further analyses indicate that heterogeneous teams result in higher aggregate quitting than do homogeneous teams. Our team commitment intervention may offer a viable cost-effective alternative to smoking cessation approaches in low-resource settings.

Main Content
Current View