Predicting treatment drop-out

Dropout represents one of the largest problems in substance abuse treatment. International and Nordic research show that only 20 – 40 % of substance abusers complete treatment as intended. At the same time, one of the most consistent factors of favourable post-treatment outcome is treatment completion (Ravndal, Vaglum & Lauritzen, 2005). In spite of the serious and continuous challenge dropout represents the phenomena is not well understood and there is a need to explore more of the factors that influence dropout and how it can be counteracted. As also stated: “…effective methods for reducing the problem of dropouts from treatment is one more area in need of further research” (NOU 2003:4, s 77).

Even relatively small reductions in dropout rate would be greatly beneficial for society and for the patients. In the case of patients, reduced dropout means reduced chance of relapse, which in turn means less suffering and reduced chance of early death due to overdose or drug related diseases. Family members and friends of the patients would also benefit directly from reduced dropout rate and higher rates of successful treatment.

The main goal of the study is to examine the usefulness of feedback with regard to predicting and preventing dropout from substance abuse treatment. Subsequent goals are to examine whether feedback will optimize the length of inpatient substance abuse treatment, strengthen the therapeutic alliance and serve as a positive intervention on its own, to examine reasons for dropping out of treatment and to explore the predictive strength of OQ45.2.

To test the above hypotheses and research questions, a research design that combines studying interventions in a naturalistic design, series of controlled trials and qualitative methods. This mixed design will help ensure information about the ecological validity of the instrument as well as measurement of effect (Rønnestad, 2008). Because of findings showing age and gender specific differences in response to the OQ45.2 (Whipple et al, 2003), variation across age and gender will also be assessed. In addition we will study possible differences in effect due to type of substance abuse.

Collaborators: Espen Ajo Arnevik (OUS), Hanne Brorson (UiO), Kim Rand Henriksen (A-hus), Espen Walderhaug (OUS), Fanny Duckert (UiO)


  • Arnevik, Espen Ajo. (2016). Selection of a Progress Monitoring Instrument for Substance Use Disorder Treatment. International Archives of Addiction Research and Medicine. Vol 2, Iss 2
  • Nordheim, Kristoffer; Walderhaug, Espen; Alstadius, Ståle; Kern-Godal, Ann; Arnevik, Espen Kristian; Duckert, Fanny. (2016). Young adults' reasons for dropout from residential substance use disorder treatment. Qualitative Social Work.
  • Amble, I.,  Gude, T., Stubdal, S., Oktedalen, T., Skjorten, A.M., Andersen, B.J., Solbakken, O.A.,  Brorson, H.H., Arnevik, E.A., Lambert, M.J. & Wampold B E. (2013) , Psychometric properties of the Outcome Questionnaire-45.2: The Norwegian version in an international context. Psychotherapy Research, doi: 10.1080/10503307.2013.849016.


Page visits: 495