15581-m-nanninga

CHAPTER 3 56 Compared to the original BTPS, the BTPS-exp showed higher Cronbach’s alphas and better fit of the data to the scale structure [20]. Scores of parents and adolescents on the BTPS-exp were correlated, though not strongly. Adolescents of both the community and the care samples scored higher on all scales of the BTPS-exp than their parents, i.e. expected more barriers. This might be explained by adolescents’ tendency to avoid interference of others, among them their parents but probably also psychosocial care professionals. Their developmental stage is characterized by a growth of autonomy and of self-directedness, and by a desire to solve problems on their own [25, 43, 44]. Ultimately, the adolescents are requested to talk about their personal problems during the treatment, and not the parents. These findings also showed that each informant had his/her own unique information, which is in line with the level of parent-child agreement, when symptoms of psychopathology are evaluated [45, 46, 47]. Regarding criterion validity, the BTPS-exp parent version was associated, as assumed, with child and family background characteristics. The presence of child psychosocial problems was the only characteristic significantly associated with expecting multiple barriers in the BTPS-exp adolescent version. This might imply that the BTPS-exp does not cover all barriers of importance for some groups of adolescents, such as stigma attached to having psychosocial problems and care. Item 16 (see Appendix) of the BTPS- exp measures stigma to some extent, but this could be measured more extensively [48]. Also, lack of parental support in seeking and using psychosocial care may become a barrier to care, since adolescents are still mainly dependent on their parents in attending care [49, 50]. Contrary to our assumption, the BTPS-exp did not affect the association between children’s psychosocial problems and care enrolment. This most likely implies that the enrolment of children and adolescents in psychosocial care is more complex than we had assumed. This is supported by Andersen’s Health Behaviour Model and Brannan’s model for children’s mental health service utilization, which presents care utilization as the result of a complex interaction of factors, such as child emotional and behavioural challenges, stressful life events, families’ resources and perceptions, caregiver strain, and service and system factors [18, 51]. Strengths and limitations A major strength of our study is the inclusion of children enroled in psychosocial care as well as children from the community, its large sample size, extensive recruitment procedure, and successful actions to reduce missing data. Potential selection bias may have affected the results of our study. However, we found only small differences between respondents and non-respondents, which