Liana Jacobi

Study protocol: combining experimental methods, econometrics and simulation modelling to determine price elasticities for studying food taxes and subsidies (The Price ExaM Study)

BackgroundThere is a need for accurate and precise food price elasticities (PE, change in consumer demand in response to change in price) to better inform policy on health-related food taxes and subsidies.Methods/DesignThe Price Experiment and Modelling (Price ExaM) study aims to: I) derive accurate and precise food PE values; II) quantify the impact of price changes on quantity and quality of discrete food group purchases and; III) model the potential health and disease impacts of a range of food taxes and subsidies. To achieve this, we will use a novel method that includes a randomised Virtual Supermarket experiment and econometric methods. Findings will be applied in simulation models to estimate population health impact (quality-adjusted life-years [QALYs]) using a multi-state life-table model. The study will consist of four sequential steps:1.We generate 5000 price sets with random price variation for all 1412 Virtual Supermarket food and beverage products. Then we add systematic price variation for foods to simulate five taxes and subsidies: a fruit and vegetable subsidy and taxes on sugar, saturated fat, salt, and sugar-sweetened beverages.2.Using an experimental design, 1000 adult New Zealand shoppers complete five household grocery shops in the Virtual Supermarket where they are randomly assigned to one of the 5000 price sets each time.3.Output data (i.e., multiple observations of price configurations and purchased amounts) are used as inputs to econometric models (using Bayesian methods) to estimate accurate PE values.4.A disease simulation model will be run with the new PE values as inputs to estimate QALYs gained and health costs saved for the five policy interventions.DiscussionThe Price ExaM study has the potential to enhance public health and economic disciplines by introducing internationally novel scientific methods to estimate accurate and precise food PE values. These values will be used to model the potential health and disease impacts of various food pricing policy options. Findings will inform policy on health-related food taxes and subsidies.Trial registrationAustralian New Zealand Clinical Trials Registry ACTRN12616000122459 (registered 3 February 2016).

Causal effects from panel data in randomized experiments with partial compliance

We present Bayesian models for finding the longitudinal causal effects of a randomized two-arm training program when compliance with the randomized assignment is less than perfect in the training arm (but perfect in the non-training arm) for reasons that are potentially correlated with the outcomes. We deal with the latter confounding problem under the principal stratification framework of Sommer and Zeger (1991) and Frangakis and Rubin (1999), and others. Building on the Bayesian contributions of Imbens and Rubin (1997), Hirano et al. (2000), Yau and Little (2001) and in particular Chib (2007) and Chib and Jacobi (2007, 2008), we construct rich models of the potential outcome sequences (with and without random effects), show how informative priors can be reasonably formulated, and present tuned computational approaches for summarizing the posterior distribution. We also discuss the computation of the marginal likelihood for comparing various versions of our models. We find the causal effects of the observed intake from the predictive distribution of each potential outcome for compliers. These are calculated from the output of our estimation procedures. We illustrate the techniques and ideas with data from the 1994 JOBS II trial that was set up to test the efficacy of a job training program on subsequent mental health outcomes.