Does stage tailoring matter in brief alcohol interventions for job-seekers? A randomized controlled trial.

Abstract

To investigate whether or not a stage tailored intervention is more effective than a non-stage tailored intervention of the same intensity in reducing alcohol use among job-seekers with unhealthy alcohol use, and whether initial motivation to change is a moderator of efficacy. A three-group randomized controlled trial with 3-, 6- and 15-month follow-ups. Three job agencies in Germany. A total of 1243 job-seekers with unhealthy alcohol use were randomized to (i) stage tailored intervention based on the transtheoretical model of change (ST), (ii) non-stage tailored intervention based on the theory of planned behaviour (NST) and (iii) assessment only (controls). Participants received feedback letters and manuals at baseline and 3 months later. Piecewise latent growth models were calculated measuring change in 'alcohol use' from baseline to month 3 (active intervention phase) and from months 3 to 15 (post-intervention phase, primary outcome). Motivation to change was included as a 4-point continuous measure. All groups reduced alcohol use from months 0 to 3 (controls: mean = -0.866, NST: mean = -0.883, ST: mean = -0.718, Ps ≤ 0.001). Post-intervention (months 3-15), low-motivated individuals in the ST group showed a greater reduction than those in the control group (β = 0.135, P = 0.039, Cohen's d = 0.42) and in the NST group (β = 0.180, P = 0.009, Cohen's d = 0.55). In contrast, compared to the ST group (β = 0.030, P = 0.361), alcohol use decreased more strongly with higher initial motivation in the NST group (β = -0.118, P = 0.010). Among job-seekers with high levels of alcohol consumption, an intervention tailored to motivational 'stage of change' was more effective than a non-stage tailored intervention for reducing alcohol use 15 months after baseline assessment in participants with low initial motivation to change.