Importance: The optimal configuration of a smoking cessation intervention in a lung cancer screening (LCS) setting has not yet been established.
Objective: To evaluate the efficacy of 3 tobacco treatment strategies of increasing integration and intensity in the LCS setting.
Design, setting, and participants: In this randomized clinical trial, LCS-eligible current smokers were randomized into 3 treatments: quitline (QL), QL plus (QL+), or integrated care (IC). The study was conducted from July 2017 to June 2022 at a hospital-based tobacco treatment clinic in Houston, Texas.
Interventions: The QL intervention group had quitline referral and 12-week nicotine replacement therapy (NRT). The QL+ group had quitline referral plus 12-week NRT or pharmacotherapy prescribed by the LCS clinician. The IC group had 12-week NRT or prescription pharmacotherapy and counseling provided by tobacco treatment specialists within the LCS health care environment.
Main outcomes and measures: The original primary outcome was biochemically verified 7-day point prevalence abstinence at 6 months; however, this was changed to self-reported abstinence during the conduct of the study due to COVID-19 pandemic restrictions.
Results: Of 630 participants, 320 (50.8%) were male, and the median (IQR) age was 59 (55-64) years. Participants smoked a median (IQR) of 20 (15-25) cigarettes per day. Each cohort (QL, QL+, and IC) was composed of 210 participants. The median (IQR) number of counseling sessions was 4 (2-5) sessions for both QL and QL+ and 8 (7-9) sessions for IC. At 3 months, 53 participants (25.2%) in QL, 57 (27.1%) in QL+, and 78 (37.1%) in IC reported abstinence. IC outperformed both QL (odds ratio [OR], 1.75 [95% CI, 1.15-2.66]; P = .01) and QL+ (OR, 1.58 [95% CI, 1.05-2.40]; P = .03). At 6 months, IC maintained the highest rate of abstinence with 68 individuals (32.4%), followed by QL+ at 58 (27.6%) and QL at 43 (20.5%). IC outperformed QL at this time point (OR, 1.86 [95% CI, 1.19-2.89]; P = .01). In the bayesian analysis, IC demonstrated a higher probability of positive absolute risk differences (ARDs) in abstinence at 3 months vs QL (ARD, 0.12) with 99% probability of positive ARD, and QL+ (ARD, 0.10) with 98% probability of positive ARD. This advantage was maintained at 6 months with ARDs of 0.12 for QL (probability of positive ARD, 99%) and 0.05 for QL+ (probability of positive ARD, 86%).
Conclusions and relevance: In this randomized clinical trial, IC involving medication and intensive counseling provides the best opportunity for smoking cessation relative to QL counseling, with or without LCS clinician-managed medication. Although IC consistently outperformed QL and QL+, differences with QL+ were reduced at 6 months, suggesting QL+ could be considered in low-resource settings.
Trial registration: ClinicalTrials.gov Identifier: NCT03059940.