Background: In US colleges, alcohol and tobacco pose substantial health risks but little is known about their co-occurrence, limiting development and use of appropriately targeted prevention efforts.
Methods: Data from the 2001 HSPH College Alcohol Study (student n = 10,924; college n = 120) were analyzed using multivariate logistic regression. Co-occurring smoking/drinking risks were examined in aggregate, by gender, and by age of onset of regular drinking and smoking across 10 measures of drinking style, three categories of transitional drinking behavior, and by treatment behavior.
Results: Over 98% of current smokers drink; a fraction (<1%) of lifetime and 7% of past year alcohol abstainers smoke. Across drinking pattern measures, 44-59% of drinkers smoke. Co-occurrence risks are greatest among youth reporting high total alcohol consumption (adjusted OR 4.21, p < 0.0001), drinking problems (adjusted OR 3.31, p < 0.0001) or symptoms of diagnosable alcohol abuse (adjusted OR 3.02, p < 0.0001). Risks are sharply elevated among students who self-medicate for drinking problems and do not seek/receive help (adjusted OR 1.87, p < 0.0001), and those reporting symptoms of diagnosable alcohol abuse who may or may not seek/receive help (adjusted OR 2.67, p < 0.01 and 2.96, p < 0.01, respectively). Females and students reporting early adolescent onset of regular smoking are at greatest risk.
Conclusions: Virtually all college smokers drink, and smoking and drinking are powerfully interrelated across drinking measures. Efficiency argues for jointly targeting both behaviors and developing interventions tailored to the special vulnerabilities of young women, students experiencing difficulties adjusting to college, and students reporting early onset of regular smoking.