The impact of a social network intervention on retention in Belgian therapeutic communities: a quasi-experimental study


Veerle Soyez, Ghent University, Department of Orthopedagogics, H. Dunantlaan 1, 9000 Gent, Belgium. E-mail:


Background  Although numerous studies recognize the importance of social network support in engaging substance abusers into treatment, there is only limited knowledge of the impact of network involvement and support during treatment. The primary objective of this research was to enhance retention in Therapeutic Community treatment utilizing a social network intervention.

Aims  The specific goals of this study were (1) to determine whether different pre-treatment factors predicted treatment retention in a Therapeutic Community; and (2) to determine whether participation of significant others in a social network intervention predicted treatment retention.

Design, setting and participants  Consecutive admissions to four long-term residential Therapeutic Communities were assessed at intake (n = 207); the study comprised a mainly male (84.9%) sample of polydrug (41.1%) and opiate (20.8%) abusers, of whom 64.4% had ever injected drugs. Assessment involved the European version of the Addiction Severity Index (EuropASI), the Circumstances, Motivation, Readiness scales (CMR), the Dutch version of the family environment scale (GKS/FES) and an in-depth interview on social network structure and perceived social support. Network members of different cohorts were assigned to a social network intervention, which consisted of three elements (a video, participation at an induction day and participation in a discussion session).

Findings  Hierarchical regression analyses showed that client-perceived social support (F1,198 = 10.9, P = 0.001) and treatment motivation and readiness (F1,198 = 8.8; P = 0.003) explained a significant proportion of the variance in treatment retention (model fit: F7,197 = 4.4; P = 0.000). By including the variable ‘significant others’ participation in network intervention’ (network involvement) in the model, the fit clearly improved (F1,197 = 6.2; P = 0.013). At the same time, the impact of perceived social support decreased (F1,197 = 2.9; P = 0.091).

Conclusions  Participation in the social network intervention was associated with improved treatment retention controlling for other client characteristics. This suggests that the intervention may be of benefit in the treatment of addicted individuals.