1. State LGBTQ policies and binge drinking among sexual minority youth in the US: a multilevel analysis
- Author
-
Gabriel L. Schwartz, Linglin Huang, Yung-Shin Chien, and Ichiro Kawachi
- Subjects
Health (social science) ,Social Psychology ,Epidemiology ,Multilevel model ,Binge drinking ,Odds ratio ,Youth Risk Behavior Survey ,030227 psychiatry ,Sexual minority ,03 medical and health sciences ,Psychiatry and Mental health ,0302 clinical medicine ,Transgender ,030212 general & internal medicine ,Lesbian ,Psychology ,Demography ,Adolescent health - Abstract
Though higher rates of binge drinking have been reported among sexual minority (lesbian, gay, and bisexual) youths compared to their heterosexual peers in the United States, questions remain about which specific structural factors drive these inequities. We test whether state LGBTQ (lesbian, gay, bisexual, transgender, queer/questioning) policy climate is associated with youth binge drinking, as well as whether that association is unique to sexual minority youth. We use LGBTQ policy scores compiled by the Movement Advancement Project in 2017 and analyze 2017 Youth Risk Behavior Survey data (126,432 youths living in 24 states). We fit multi-level logistic regression models to estimate odds ratios for binge drinking with state-level LGBTQ policy scores. More progressive state-level overall LGBTQ policies were associated with lower odds of binge drinking among sexual minority youths (interaction term OR = 0.990, CI 0.984–0.996, p = 0.0009), but not among heterosexuals (OR = 1.001, CI 0.992, 1.009, p = 0.881). For sexual minorities, living in the most LGBTQ-affirming state was associated with a probability of binge drinking 27% lower than that of those living in the least LGBTQ-affirming state. In sensitivity analyses, policies specific to LGBTQ youth (opposed to general LGBTQ people) were even more strongly related to sexual minorities’ binge drinking (interaction term: state-level youth policy score × individual-level sexual identity—of OR = 0.949, CI 0.927–0.971, p = 0.00001). In sex-stratified models, associations with overall LGBTQ policy scores were significant only for girls (OR = 0.986, CI 0.979–0.993, p
- Published
- 2021