The Challenge of Evaluating State Heat Standards: A Cautionary Tale on Data Limitations in Occupational Safety Research

Abstract

As climate change intensifies heat exposure, occupational heat illness has emerged as a growing public health concern. Five U.S. states have adopted comprehensive occupational heat standards, yet no multi-state causal evaluation exists. This paper documents the fundamental data barriers that prevent credible evaluation of these policies using publicly available data. State-level heat-related fatality counts are suppressed by the Bureau of Labor Statistics due to small cell sizes, forcing researchers to impute state outcomes from national totals. I demonstrate that this imputation approach—using fixed state shares applied to national death counts—mechanically prevents identification of state-specific policy effects, since treated states' imputed outcomes cannot diverge from control states by construction. Difference-in-differences estimates using this imputed outcome are uninformative about policy effectiveness, regardless of statistical significance. These findings highlight a critical gap in occupational safety data infrastructure: credible evaluation of state workplace policies requires access to restricted-use microdata or alternative administrative records (workers' compensation claims, emergency department visits) that are not systematically available to researchers. Until better data infrastructure exists, policymakers must rely on the physiological evidence base and single-state evaluations rather than multi-state econometric studies.