Introduction: Homeless populations present a unique opportunity for smoking cessation interventions, as homeless smokers report both higher rates of smoking and increased likelihood to engage smoking behaviors defined as high-risk, as well as equal interest in quitting smoking when compared to the general population. However, there has been a relative lack of investigation regarding whether housing status impacts smoking cessation treatment delivery or outcomes, and if so, how these are impacted. Gaining further insight as to how treatment delivery and smoking cessation rates differ between housed and unhoused smokers provides important insight to tailor treatment strategies in order to better serve smokers experiencing homelessness. We compared treatment delivery and outcomes between homeless and housed smokers recruited from two safety net hospitals, who were either referred to the state quitline or provided intensive counseling.

Methods: Baseline characteristics and cessation rates were retrospectively analyzed for 1604 hospitalized smokers recruited at two safety-net hospitals in New York City. Participants were randomized to standard therapy (nicotine replacement therapy (NRT), varenicline, bupropion) and either referral to the state quit line (1-2 telephone counseling sessions) or intensive counseling (7 telephone counseling sessions). Results were stratified based on housing status as determined by participant’s responses when queried regarding their housing in the week prior to hospital admission. Baseline characteristics were analyzed using chi-squared tests, while outcome measures, including 30-day abstinence and number of quit attempts, reported at 2-and 6-month follow-up were evaluated using a multivariable model.

Results: In agreement with prior findings, unstably-housed smokers were found to have higher rates of co-morbid substance use and roughly equal rates of interest in quitting smoking compared to stably-housed smokers. Homeless and housed smokers were equally likely to positively rate smoking cessation counseling received in the hospital prior to discharge. Housed smokers randomized to receive intensive counseling were more likely to report the use of NRT at 6-month follow-up compared to peers who received a quitline referral. Homeless smokers in both the intervention and control group were less likely to report utilizing NRT at 6-month follow-up. Unstably-housed smokers in the intensive counseling group were more likely to report high confidence in quitting compared to quitline participants at follow-up; while housed smokers reported roughly equal confidence in quitting despite treatment group. Housed smokers that were randomized to intensive counseling were more likely to report being smoke free for the prior 30 days at both 2-month and 6-month follow-up compared to those in the control group. In contrast, unstably housed smokers reported equal rates of 30-day abstinence at 2-month and 6-month follow-up regardless of randomization.

Conclusions: While the intensive counseling intervention was shown to be more effective in achieving both smoking abstinence and also increasing the number of quit attempts for smokers with stable housing. Unstably housed smokers were not more likely achieve abstinence or engage in quit attempts after receiving the intensive counseling intervention.