main set of analyses, we extended the observation period from 1990–2007 to 1990–2010; four additional states (Arizona, Michigan, New Jersey, and New Mexico) and the District of Columbia passed medical marijuana policies during this time (Lynne-Landsman et al., 2013). Initial models were based on the same covariates as the comparison analyses, while subsequent models included individual-level demographic covariates (age, sex, race/ethnicity and educational attainment) and several additional state-level covariates (citizen political orientation, per-capita mental health spending, percentage of uninsured adults, cigarette excise taxes and a smoke-free air policy score).
