Computer-administered cognitive behavioural self-help intervention for adolescents with mild to moderate depressive symptoms: programme development and examination of feasibility, efficacy and acceptability

Abstract

The aim of this project was to design and evaluate the feasibility, efficacy and acceptability of a developmentally appropriate computerised self-help programme for adolescents with mild to moderate depressive symptoms. The project consisted of three phases. The first phase involved the development of a conceptual framework based on learning theories and relevant instructional and media design concepts. Subsequently a prototype of a computer-administered cognitive behavioural therapy (CCBT) was designed. A matching computerised psychoeducational (CPE) programme was also developed to control for non-specific factors. The second phase involved a small randomised controlled trial to pilot the efficacy of CCBT versus CPE. Thirty-four adolescents who scored at least 30 on Child Depression Rating Scale-Revised (CDRS-R) or 76 on Reynolds Adolescent Depression Scale-2 (RADS-2) were randomly assigned to either CCBT or CPE condition. 94% of CCBT and 82% of CPE participants completed the intervention. Controlling for age and gender, repeated measures ANOVA showed a significant decrease on CDRS-R for the older (16-18 year old) adolescents compared with CPE condition (F(2,24)=13.890, p<0.000). The RADS-2 scores were also significantly improved for older CCBT adolescents (F(2,24)=4.833, p<0.017) compared with their CPE counterparts. Significant improvements in CDRS-R (F(2,36)=23.911, p<0.000) and RADS-2 (F(2,36)=6,223, p=0.001) were observed in the younger (13-15 year old) adolescents irrespective of group allocation. CDRS-R scores improved more for males than for females regardless of group allocation (F(2,56)=6.30, p=0.003). Problem solving measured by Adolescent Coping Scale (ACS) improved significantly for older adolescents in the CCBT group (F(2,24)=9.410, p<0.005). ACS Non-productive Coping was significantly reduced by CCBT compared with CPE in younger and older adolescents (F(2,56)=5.19, p=0.009). Pediatric Quality of Life measure showed a significant improvement for all participants but showed no group differences (F(2,56)=4.56, p=0.014). The third phase of the project aimed to generate user feedback through a questionnaire (n=21) and in-depth interviews (n=14) with adolescents who had completed either of the programme. Adolescents found the programmes equally useful, easy to use and engaging. According to adolescents the programmes communicated three key messages: education about depression, validation of feelings and hope for recovery. Content specific feedback suggested that CCBT taught some adolescents cognitive restructuring and problem solving. Among the most useful CPE topics were anger management and stress reduction. Suggestions for improvement included more use of interactive/game-like technology, programme customisation and reduction of text in favour of multimedia presentation. New topics such as dealing with drugs/alcohol or violence and recovery stories from credible role models were also suggested. The results and implications of the study are discussed in the light of a growing interest in the use of computer-administered mental health interventions. Suggestions for future research and programme development are outlined