Unpacking the Relationship Between PTSD and Cigarette Use: The Moderating Role of Household Income

Background: Prior research has linked current posttraumatic stress disorder (PTSD) to increased cigarette smoking. However, existing work has not yet examined factors which may moderate this relationship, despite the critical implications of such knowledge for intervention development. We know that SES is also related to both PTSD and cigarette use, but little is known about how these factors might influence the relationship between PTSD and smoking. We predicted that this relationship is likely to vary as a function of income and propose that this variation may signal differences in preferred coping strategies across income levels.

Methods: The sample consisted of nearly 1500 adult Detroit residents who participated in a NIDA R01-funded study designed to assess exposure to trauma, PTSD, depression symptoms, cigarette and alcohol use. Participants were chosen from a probability sample of households within the city limits of Detroit followed by randomly selecting one adult from each household. A weighted 7.1% of the sample met cluster criteria for PTSD in the past month. Self report of household income was measured using a 7-level multiple choice question (M=3.75, SE=.06; $25,000 mean income) and cigarette smoking consisted of the total number of cigarettes smoked in the past month (M=8.36 packs/month, SE=.47). Due to the skewed nature of the dependent variable, cigarette use, data were analyzed using negative binomial regression. All analyses, conducted with STATA 10.0, were weighted and corrected the standard errors to account for the complex sampling design.

Results: First, we examined effects of PTSD and income on cigarette use in the past month, controlling for race, gender, age, and education. Meeting criteria for a PTSD diagnosis in the past month was associated with increased smoking in our overall sample (p<.01). We also observed a main effect of income (p<.01) and a marginal effect of race (p=.10), such that lower-income and non-African American race were associated with increased smoking. Including a PTSD and income interaction in the model revealed that the relationship between PTSD and smoking is moderated by income such that as one meets PTSD criteria, higher-income people smoke more and lower-income people smoke less (p<.05). There was no significant interaction between PTSD and race.

Conclusions and Implications: We expand upon prior work documenting a link between PTSD and increased cigarette use by demonstrating that this relationship is moderated by income such that only higher-income individuals increase smoking when they develop PTSD. It may be that smoking is a particularly valued mechanism for coping with PTSD symptoms among those with higher income (who are less likely to smoke otherwise). Interestingly, plotting the interaction reveals a steep decline in cigarette use by lower-income individuals with PTSD. To better tailor smoking cessation interventions, further research is needed to understand the differential smoking patterns of individuals with PTSD across the SES spectrum. Whereas treating PTSD symptoms and teaching alternative coping strategies may effectively decrease smoking among higher-income individuals, additional methodologies are likely to be needed for lower-income individuals who may be less likely to use cigarettes purely for coping with PTSD. 4-30-2010-->