Martin Olsson

This study deals with the problem of breakdown in out‐of‐home placements for Danish teenagers. How often are these placements prematurely terminated against the requests and intentions of the child welfare authorities? Which factors seem to increase and decrease placement breakdown? The sample consists of 225 youths and 367 placements during the years 2004–2008. Among the 225 youths, 44% experienced a breakdown, and 33% of the 367 placements ended with a breakdown. A large proportion of the placement breakdowns (20%) occurred before 4 months, and within the first year, 62% of the placement breakdowns had taken place. Replacements took place in 61% of all placement breakdowns. The analysis showed that type of placement and number of youth at the same placement had significant association to placement breakdown, but variables based on the social worker's clinical assessment showed no association to breakdown. However, while the logistic regression model was significant, it explained only 6% of the variance. Earlier studies have indicated that breakdown is a substantial problem of child welfare in several other western countries, Denmark is no exception. These corresponding breakdown rates point out that no child welfare system yet have found a method to reduce breakdown substantially.

This study examines Swedish young adults (mean age 21) with a history of conduct disorder (CD) as adolescents. Using medical records, this study explores the relationship between adolescent inpatients and their outcomes in adulthood. Two outcome variables were used: an indication of non-successful outcome variable (seven undesirable outcomes) and sense of coherence. Using multiple regression analyses, this study showed that extracted data from the medical case record could significantly explain small variance depending on output variable. The small variance could be related to the homogeneous clinical sample, the follow-up time, the outcome variables and the absence of a biological perspective. This study suggest, clinicians should be very careful when predicting outcome in young adulthood, if they should predict outcome at all. The positive conclusion in this matter is that as far as we know any teenager with CD could have a positive outcome in young adulthood.

The present study is one of a few that have used ‘sense of coherence’ (SOC) as a dependent variable in an explanatory model. After studying three different samples – 680 students, 180 parents and 315 couples – we conclude that family relational and psychopathological variables contribute significantly to the explanation of SOC (explained variance between 10–27 and 26–50 per cent). In total, we obtained an explained variance of between 42 and 64 per cent. This leads us to the conclusion that in all three samples, SOC is multifaceted and thereby is more than simply an opposite state to depression. Context may play an important part in the explanation of SOC.

This study focuses on young male and female adults (n=290) who were diagnosed with conduct disorder (CD) during adolescence at the inpatient child and adolescent psychiatric unit in Lund, Sweden. Their adulthood is described using seven outcome variables: in custody, crime, illicit drug use, mental health, teenage parenthood, transference income and educational attainment. Together, the seven outcome variables represent a wide perspective of individual life and known risks for both genders. The variables are compared, scaled and described in an outcome that identifies successful outcome. Although this was a severe clinical inpatient group of adolescents, a relatively large number-about a third of the males and close to half of the females-exhibit a successful outcome by their early twenties. This study highlighted the merits use gender-sensitive outcome variables. The take-home message from a societal perspective is that many improvements are needed, a message that has been voiced in several Swedish commissions.