Potential explanations for conflicting findings on abrupt versus gradual smoking cessation: a population study in England.

BACKGROUND AND AIM Observational and trial evidence conflict on the efficacy of two contrasting behavioural approaches to quitting smoking - gradual and abrupt. Observational data suggests an abrupt approach to quitting is superior to a gradual approach, whilst trials show no difference. One potential explanation is self-selection in observational data, whereby people can choose their quit approach and those who find it harder to quit may be more likely to choose a gradual quit approach. This study aims to investigate potential explanations for these conflicting findings. METHODS We used observational data from a nationally representative sample of adults in England from November 2006 to February 2020 who reported smoking and had made at least one quit attempt in the past year (n=21,542). We used logistic regression models to assess the association between abrupt versus gradual quit attempts and quit success, adjusting for sociodemographic, smoking and quit attempt characteristics. FINDINGS Abrupt, versus gradual, attempts were associated with improved quit success in an unadjusted model (OR=2.02, 95% CI=1.86-2.19). This association remained after adjusting for a broad range of relevant confounders (OR=1.75, 95% CI=1.59-1.93). CONCLUSIONS Among a representative sample of adults who had smoked and made a quit attempt in the past year, there was evidence of an association between abrupt attempts and quit success before and after adjusting for relevant confounders. This suggests that the differences in quit success seen between abrupt and gradual quit attempt types are not completely driven by self-selection in observational data. IMPLICATIONS We investigated explanations for conflicting findings on the efficacy of gradual versus abrupt approaches to quitting smoking between trial and observational data. Despite adjusting observational data for sociodemographic, smoking and quit attempt characteristics, an association between abrupt quitting and quit success remained. Therefore, differences in quit success were not completely driven by self-selection of a gradual approach by people who found it especially difficult to quit or differences in the use of quitting aids. However, characteristics adjusted for were limited by the data available, and future research should continue to investigate the difference in findings across study types to inform cessation support.