UCSF

Introduction: On April 1, 2017, California Proposition 56 (Prop 56) was implemented, increasing the excise tax on cigarettes by $2/pack. This study compares the association of Prop 56 with smoking prevalence and smoking intensity across racial/ethnic groups, further examining distinctions across income subgroups within each racial/ethnic group.

Methods: The study used pooled cross-sectional data from the 2012-2018 California Behavioral Risk Factor Surveillance System. We examined two outcomes: current smoking prevalence and smoking intensity conditional on current smoking. A two-part econometric model was used to estimate the association of Prop 56 with smoking prevalence and intensity using multiple logistic regression and multiple linear regression, respectively. The two-part model was run separately for all adults (full sample) and each racial/ethnic group. Within each racial/ethnic group, we ran stratified analyses by income subgroups.

Results: The results indicated that Prop 56 was negatively associated with smoking prevalence among full-sample, Hispanic, White, and African American adults; and negatively associated with smoking intensity among full-sample and White smokers. Stratified analyses by race/ethnicity and income showed that Prop 56 was negatively associated with smoking prevalence among low-income full-sample and White adults and among middle-income smokers in the full, Hispanic, White, African American, and Asian samples. Prop 56 was negatively associated with smoking intensity among middle-income Hispanic and high-income White smokers. The association between Prop 56 and smoking intensity was positive among high-income African American smokers.

Conclusion: Prop 56 was associated with a reduction in smoking prevalence across multiple racial/ethnic groups, particularly within the low- and middle-income subgroups.

Policy implications: Our findings indicate that the reduction in smoking prevalence immediately following the implementation of California Proposition 56 tobacco tax increase was significant across a variety racial/ethnic groups, particularly low- and middle-income subgroups. We found differential responses in smoking prevalence across income groups among Whites but not among racial/ethnic minorities. We found no evidence of any significance association between Proposition 56 and smoking intensity among minorities and economically vulnerable populations, except for middle-income Hispanics. Researchers, policymakers, and advocates should consider additional merits of targeted, community-based, non-economic tobacco control interventions in reaching low- and middle-income groups within racial/ethnic minorities.