Accessibility statement

Policy impact evaluation

This topic explores the use of both ex-post econometric approaches to the evaluation of public health and health system interventions, based on observational evidence, as opposed to evidence from randomised controlled trials.

The key to these approaches is identifying a source of variation in treatment that is independent of other factors that influence outcomes. Often natural experiments are used to support estimation approaches such as instrumental variables, regression discontinuity and difference-in-differences.

In recent years there has been a heavy emphasis on keeping parametric assumptions to a minimum and the literature has favoured methods such as matching, nonparametric regression, and control function approaches and on making inferences that are robust to functional form and distributional assumptions.

CHE's work in this area has thus far focused predominantly on applications in high income countries. Recent work on low- and middle-income countries includes econometric analysis of the impact of health insurance reforms on health outcomes. Other work comprises policy impact evaluation of sugar-sweetened beverage taxes in middle-income countries.

Research staff involved: Marc Suhrcke (Global Health team); Nigel Rice (HEDG team).

Example publications:

  • R Nakamura, M Suhrcke, R Pechey, M Morciano, M Roland, TM Marteau. Impact on alcohol purchasing of a ban on multi‐buy promotions: a quasi‐experimental evaluation comparing Scotland with England and Wales. Addiction 109 (4), 558-567. Download from Wiley
  • Nakamura R, Coombes E, Suhrcke M. Do Economic Incentives Promote Physical Activity? Evaluating the Impact of the London Congestion Charge on Active Commuting and Health (March 27, 2014). SSRN Working Paper. Download from Social Science Research Network
  • Jones AM, Laporte A, Rice N, Zucchelli E. Do smoking bans have an impact on active smoking? A model with evidence from a policy experiment in the UK. Health Economics, 2015, 24(2), 175–192. Download from Wiley