Evaluating the Health Impacts of Food and Beverage Taxes.

Several jurisdictions are now imposing taxes on food and beverages to prevent obesity (and related conditions). Existing evidence concerning their effects comes largely from simulation studies and trials in closed settings, both of which have limitations. Rigorous evaluation of actual taxes may provide richer evidence with greater external validity to support policy making. This article describes existing evaluation studies and outlines an implicit underlying theoretical framework for how taxes are expected to affect health. It then explores three important issues for future studies: selection of an appropriate evaluative perspective (comparing realist and biomedical experimental paradigms); approaches to causal inference; and the challenge of a low signal-to-noise ratio. We argue that evaluation should be informed by a realist perspective as well as making appropriate use of established empirical quasi-experimental approaches to testing causal effects. This should be underpinned by a theoretical framework that acknowledges complexity and the potential diversity of impacts.