Relapse to smoking during unaided cessation: clinical, cognitive, and motivational predictors

Rationale: Neurobiological models of addiction suggest that abnormalities of brain reward circuitry distort salience attribution and inhibitory control processes, which in turn contribute to high relapse rates. Objectives: To determine whether impairments of salience attribution and inhibitory control predict relapse in a pharmacologically unaided attempt at smoking cessation. Methods: 141 smokers were assessed on indices of nicotine consumption / dependence (e.g. the FTND, cigarettes per day, salivary cotinine), and three trait impulsivity measures. After overnight abstinence they completed experimental tests of cue reactivity, attentional bias to smoking cues, response to financial reward, motor impulsiveness, and response inhibition (antisaccades). They then started a quit attempt with follow-up after 7 days, 1 month, and 3 months; abstinence was verified via salivary cotinine levels $\leq$ 20ng/ml. Results: Relapse rates at each point were 52.5%, 64% and 76.3%. The strongest predictor was pre-cessation salivary cotinine; other smoking / dependence indices did not explain additional outcome variance and neither did trait impulsivity. All experimental indices except responsivity to financial reward significantly predicted one week outcome. Salivary cotinine, attentional bias to smoking cues and antisaccade errors explained unique as well as shared variance. At one and three months, salivary cotinine, motor impulsiveness and cue reactivity were all individually predictive; the effects of salivary cotinine and motor impulsiveness were additive. Conclusions: These data provide some support for the involvement of abnormal cognitive and motivational processes in sustaining smoking dependence and suggest that they might be a focus of interventions, especially in the early stages of cessation.