Counselors recommend pharmacotherapy use during hospitalization if needed for withdrawal symptoms or and make postdischarge recommendations for those interested in quitting. Participants referred to the TTS during a 24-month period (July 2007 through June 2009) who had smoked cigarettes in the past week and received bedside counseling were eligible for enrollment. Patients were excluded if they received only brief advice about smoking cessation, defined as ��5 min of counseling, had no telephone access, altered mental status, limited English skills or another communication barrier, or were not discharged to home. Procedure Baseline Data Collection After each counseling session, counselors recorded the participant��s admitting service, average number of cigarettes smoked per day (cig/day) during the month before admission, use of NRT for smoking cessation before the current hospitalization (prior use), and counseling session duration (in minutes).

Intention to quit was assessed by asking about the participant��s plan about smoking after hospital discharge. Response options were: ��I will remain quit,�� ��I will try to quit,�� ��I don��t know if I will quit,�� and ��I do not plan to quit.�� Age, gender, length of stay (LOS), and pharmacy orders for NRT were obtained from hospital records. Consent to telephone follow-up was obtained after counseling. Follow-up Data Collection Participants were contacted by telephone at 2 weeks after discharge to assess whether NRT had been used since discharge. Analysis All analyses were conducted using Stata statistical software (StataCorp, 2008).

Baseline differences by NRT use during hospitalization were compared using chi-squared tests, t tests, and Wilcoxon rank sum tests. Cigarettes per day were dichotomized as <10 versus ��10, contrasting light versus heavier smokers. Duration of counseling was dichotomized at the median (��25 vs. >25 min). Intention to quit was classified as ��I will remain quit�� (the strongest intention to quit) versus other responses. Admitting service was categorized as cardiac versus other. LOS was log transformed. We assessed the effects of NRT experience in hospital on NRT use after discharge in a generalized linear model (GLM) using a Poisson distribution, log link function, and robust SEs.

To separate the effects of NRT use before and during the current hospitalization, respondents were divided into four mutually exclusive GSK-3 categories of NRT use before discharge: (a) never (not before or during hospitalization), (b) before the current hospitalization only, (c) during hospitalization only, and (d) both before and during hospitalization. We calculated rate ratios (RR) and adjusted rate ratios (ARRs) with 95% CI for self-reported postdischarge NRT use for each NRT use category using those who had never used NRT before discharge (Group 1) as the reference.