A school-based programme for tobacco and alcohol prevention in special education: effectiveness of the modified 'healthy school and drugs' intervention and moderation by school subtype.

AIMS: To test the effectiveness of the Healthy School and Drugs (HSD) programme on tobacco and alcohol use in Dutch secondary special education (SE) schools, and whether this depends upon subtypes of SE schools and the level of implementation.

DESIGN: In a quasi-experimental design with baseline and post-treatment follow-up, 363 students were allocated arbitrarily or depending on teacher motivation to either intervention condition (n = 205) or usual curriculum (n = 158).

SETTING: Thirteen secondary SE schools spread throughout the Netherlands.

PARTICIPANTS: Participants were recruited during the autumn of 2013 from three school subtypes: SE for adolescents with intellectual/physical disabilities (SEI; n = 13), behavioural/emotional difficulties (SEB; n = 136) and learning disabilities/developmental disorders (SEL; n = 214).

MEASUREMENTS: Self-reported life-time smoking prevalence and life-time drinking frequency as outcomes, and school subtype (SEL/SEB) and implementation fidelity (high/low) as moderators.

FINDINGS: No significant differences were found at follow-up in life-time smoking [odds ratio (OR) = 1.52; 95% confidence interval (CI) = 0.74-3.12] and drinking frequency (d = 0.01; 95% CI = -0.16 to 0.18). Interaction analyses revealed adverse effects in SEB students for alcohol use (d = 0.43; 95% CI = 0.16-0.69). Effect on tobacco refusal self-efficacy was moderated positively by implementation fidelity (d = 0.35; 95% CI = 0.07-0.63).

CONCLUSION: The Healthy School and Drugs programme adapted for secondary special education in the Netherlands lacked clear evidence for effects on all outcomes. This pilot study suggests further that, within special education, substance use interventions may need to be targeted at school subtypes, as these may have harmful effects among students with behavioural difficulties. Finally, limited evidence was found that programme effectiveness may depend upon implementation fidelity.