Population-based approaches to reducing depression in adolescents in New Zealand

Abstract

Studies in this thesis address two questions:

What could be done to reduce depression in adolescents across the population in New Zealand? How could effectiveness of interventions be measured effectively? Depressive disorder is common and a major cause of disability. Prevalence rises steeply in adolescence and has a negative impact on development; therefore adolescence is a good time to intervene. A multi-media computer-assisted survey of 9567 randomly selected secondary school students in New Zealand was carried out and proved to be an effective method of measuring depression nationally. On the Reynolds Adolescent Depression Scale (RADS) 13.9% of students were shown to have significant depression symptoms. Less than a fifth had sought help from a doctor. The validity and reliability of RADS in different ethnic groups in New Zealand was supported, with Cronbach’s alpha of 0.94, median item/total score correlations of 0.62, and a factor structure similar to the original studies. A meta-analysis of specific serotonin re-uptake inhibitors (SSRIs) in adolescents showed a favourable risk benefit ratio only for fluoxetine. A national campaign to increase treatment of depression with SSRIs was not supported. A meta-analysis of depression prevention programmes showed that targeted psychological interventions were effective in the short-term compared with non-intervention (standardized mean difference (SMD) -0.26, 95% confidence interval (CI) -0.40 to -0.13). Effect sizes were small but the "numbers needed to treat" (NNT) to prevent one person developing clinical depression was 10. The lack of a placebo was a flaw in most studies in the analysis. A bi-cultural randomized placebo controlled trial of a universal school based depression prevention programme (RAP-Kiwi)for 394 participants was carried out. Follow-up was to 18 months. Immediately after intervention, depression scores were reduced significantly more by RAP-Kiwi than by placebo, (mean difference in change from baseline between groups was 1.5 on BDI-II (CI> 0.38, p=.01)and 2.24 on RADS (CI> 0.22,p=.04)).Categorical analysis confirmed significant benefit (absolute risk reduction 3% (95% CI, 1-11%, McNemar X2, p = .03), NNT for short-term benefit was 33). Significant differences persisted for 18 months on RADS but not on BDI-II. There were no differences in effectiveness by ethnicity. Retention rates were 91% at 6 months and 72% at 18 months. In conclusion these studies showed rigorous critical appraisal of interventions is essential before implementation. Prevention of depression is potentially an effective public health measure.