Despite ongoing public health interventions to prevent obesity, the prevalence of obesity is still increasing in the United States. Obesity is also the strongest modifiable risk factor for diabetes. Non-Hispanic African American adults are most susceptible to obesity than Whites. There are also racial disparities in dietary factors that contribute to high energy intake. Compared with whites, African Americans tend to have consume more saturated fat and salt; fewer in fruits, vegetables, and whole grains. While current policies may successfully reduce unhealthy diet behaviors and therefore reduce weight, few of these policy studies conducted racial/ethnic disparities analyses. We applied causal mediation analysis to investigate what the racial disparities in obesity would be if we were able to intervene and eliminate racial disparities regarding diet. For example, how much racial disparities in obesity would be eliminated if soda consumption among Blacks equals Whites. Current studies on obesity disparities are based on regression or Oaxaca­Blinder decomposition framework. These studies at best provide explanatory roles of diet in racial disparities in obesity, instead of a causal interpretation. We applied causal mediation analysis to a longitudinal cohort. The cohort provides a clear temporal ordering of covariate assessment, diet mediators, and weight gain. Strategies to influence the intake of healthy food and amount of energy consumed include a focus on changing nutrition environment. One approach is a 1-cent-per-ounce SSB excise tax. Literature has shown that the soda tax increased SSB retail prices for consumers, and reduced SSB purchase as well as intake. However, current studies using real world data did not evaluate soda tax’s effect on weight and weight-related health outcomes. We took the advantage that four California cities have implemented a 1-cent-per-ounce soda tax since 2015 and applied quasi-experiment design. California also provides a large and racially diverse population to evaluate the health impact of soda tax. Examining multiple cities provides a richer and more complete understanding of how the taxes affect consumption across a range of settings. There has been ongoing debate on implementing soda tax at the state or federal level. What remains unclear is the size of the effect that soda tax has on health and health care costs in the long term. We conducted a cost-effectiveness analysis of soda tax to inform policy decision. Our evaluation of the 1-cent-per-ounce soda tax brought to attention a potential problem for the design of similar policies against unhealthy food. That is, people switch to other high-calorie drinks such as fruit juice and milk as substitutions. Due to extra calorie intake, people’s weight may even increase after soda tax, although it’s still beneficial that people are choosing healthier beverages. Future study is needed to investigate substitution patterns.

Background: Studying risk factors of rare outcomes can be difficult as single studies tend to have few exposed cases, resulting in rather wide confidence intervals. Therefore, combination of multiple studies is usually necessary to draw a conclusion. Pooling of individual patient data (IPD) is considered the gold standard due to increased statistical power through large sample sizes and other advantages. However, present pooling is limited to studies of the same designs, which does not make full use of existing data.

Objectives: To generalize pooling to studies of different designs (including cohort study, case control study, nested case control study and matched case control study), by incorporating simulation of the association of extremely low frequency magnetic field (ELF-MF) and childhood leukemia.

Method: I first simulated large cohort and case control samples based on parameters extracted from both the literature and existing large cohort and case control datasets, which included ELF-MF exposure prevalence, childhood leukemia incidence rate, prevalence of confounders including age, gender, race and SES. Then I combined these simulated data using three different methods: two stage meta-analysis, one stage pooling and two stage meta-analysis with pooling.

Results: Estimates from three synthesis methods were close to the causal estimate and there was no obvious trend of overestimation or underestimation. One stage pooling seemed to have the worst efficiency with the widest 95%CI but the difference was not significant.

Conclusion: The performance of three synthesis methods in the study was not certain. Further simulations with varying parameters and possible mathematical derivations are needed to assess why and when these methods lead to different effect estimates.

Keywords: Individual patient data (IPD), meta-analysis, pooling, simulation with R, Extremely Low Frequency Magnetic Field (ELF-MF), childhood leukemia

Background

Methods

Results

Conclusions