Implementing guidelines for interventions in Swedish substance abuse treatment: A national evaluation of outcomes from 2010 to 2014

Abstract In 2009, the Swedish government launched a nationwide dissemination and implementation project regarding substance abuse treatment, Knowledge to Practice (KTP), which was commissioned by the Swedish National Board of Health and Welfare (SoS). Aim The study aimed to evaluate the effectiveness of a national project to disseminate and implement evidence-based assessment instruments and treatment methods, and to promote cross-sectional collaboration in substance abuse treatment. Design and recruitment This study was a naturalistic prospective follow-up study of a sample of 3,063 participants in 203 municipalities via five consecutive online surveys. Of the 21 Swedish regions, we included 20. All participants, representing 79.5% of the population, were employed in substance abuse treatment/care within social work and psychiatry. We used individual background variables and factors in the organizational readiness for change (ORC) questionnaire for prediction. Methods We calculated descriptive statistics using SPSS 24.0. We applied latent profile analysis (LPA) to identify respondents with distinct growth profiles across five assessment points. Next, we calculated multilevel growth curve analyses using full information maximum likelihood analysis (FIML) in Mplus 8.3. In the analyses, we nested “within-person change” (level 1) in “between-persons analysis” (level 2). We, in turn, nested these in municipalities/regions (level 3). We included both respondents and organizational predictors in levels 2 and 3. We tested four statistical models in the multilevel analyses to predict the outcome. Results The annual response rate was 59% on average across the five years. Respondents were a homogenous group with regard to education and years of employment, and two thirds were women. We identified four different trajectories nationwide, which showed different growth profiles for the use of methods and assessment instruments. Our results showed a larger between-region variation than within-region variation in outcome indicators. Some regions were more successful than others over the five years. At the respondent level, the years of employment in treatment work was associated with increased use of assessment instruments, while the ORC factor, institutional resources, predicted decreased use of assessment instruments. Specialist competence and the ORC factors staff attributes (job satisfaction, efficacy, and influence) and institutional resources predicted increased use of psychosocial treatment methods. The ORC factor organizational climate was positively associated with increased cross-sectional collaboration activities.