Prevention of Depression in Childhood and Adolescence

 Prevention of Depression in Childhood and Adolescence

Depression is the most common psychiatric disorder and one of the leading causes ofmorbidity and mortality in the United States. Nationally representative epidemiologicstudies indicate that the lifetime prevalence of depression among adolescents aged15 to 18 years is between 11% and 14%, with an estimated 20% of adolescentsexperiencing major depression disorder (MDD) by the time they turn 18 years.1,2Itis also estimated that approximately half of first episodes of depression occur duringadolescence.3In addition, approximately one-quarter of adolescents endorse sub-threshold depressive symptoms that cause impairment in daily functioning.

Several recent systematic reviews and meta-analyses have concluded that there isevidence for beneficial effects of preventive interventions for depression in childrenand adolescents.73–76Effect sizes are generally small but are consistent with, orlarger than, the size of effects obtained for preventive interventions to prevent youthsubstance abuse, human immunodeficiency virus, eating disorders, and obesity.76Mendelson & Tandon10
Merry and colleagues73also noted as “very encouraging” their calculation that thereduction in postintervention depression risk is equivalent to a number needed totreat (NNT; the sample size required in order to affect 1 individual) of 11 individuals(confidence interval, 7–20). Moreover, even small effects can translate into substan-tive benefits over time or across large populations. This section of the article dis-cusses several aspects of the empirical findings to date and suggests directionsfor future research.

Universal Versus Targeted Prevention Approaches

Prior meta-analyses concluded that selective and indicated prevention strategieswere more effective than universal approaches.76,77Researchers suggested that tar-geting youth with subsyndromal symptoms may be more feasible from a resourcestandpoint than intervening with a larger, nonsymptomatic group given that manyyoung participants in a universal prevention program never develop depressive symp-toms or disorder, and extremely large samples may be needed to obtain statisticallysignificant effects.77However, a more recent meta-analysis that included a largernumber of studies concluded that both universal and targeted prevention programsreduced depressive episodes and depressive symptoms at posttest and 12-monthfollow-up.73The finding that universal prevention is effective is an encouraging development,because universal programs have several advantages, including the ability to reachat-risk young people who may not be identified by targeting a given risk factor oradministering a particular screening instrument. Universal prevention strategiesare particularly well suited for schools, where there are often logistical barriers tomental health screening. There is a particular need for universal interventions topromote protective factors in schools serving low-income communities, given thesestudents’ high level of trauma exposure, emotional symptoms, and academic chal-lenges.34For instance, in a study of elementary schools serving impoverished com-munities (n51099 students), 56% of students were identified as having mentalhealth needs.78The public education system is under-resourced and ill equippedto address this level of need using pull-out models, such as indicated pro-grams.34,78Moreover, pull-out approaches may not embed intervention models suf-ficiently within the school ecology to shape school climate.34Universal programsthat can be integrated with school practices are a resource-efficient and logical pre-vention approach for a population of young people at very high risk for pooroutcomes.

Prevention of Depressive Symptoms Versus Major Depressive Disorder

Most studies have assessed reduction in self-reported depressive symptoms, ratherthan prevention of MDD incidence, as an outcome of prevention trials. The additionaltime and expense required for administration of structured clinical interviews makesit difficult to routinely incorporate them into prevention studies, particularly those withlarge sample sizes. As a result, although there is mounting evidence that some pre-vention programs prevent MDD incidence,20,73,76clinicians must be cautious and notassume that findings regarding depressive symptom reductions automatically trans-late into prevention of incidence of depressive disorder. Some prevention re-searchers77have additionally argued that reductions in depressive symptomsrelative to a lack of change in the control group should be viewed as a treatmentrather than a prevention effect; by contrast, prevention should signify symptomreduction or lack of change in the intervention group relative to an increase in thecontrol group. A further complication is that some young people enrolled inDepression Prevention in Childhood and Adolescence11
prevention trials have had prior episodes of depression; for these participants, inter-vention effects are perhaps most accurately conceptualized as relapse prevention.73Despite these challenges in categorizing prevention, it is clear that young people canbenefit substantively from programs that reduce current depressive symptoms andprovide mood management strategies that increase long-term resilience againststress. 

Moderators of Intervention Effects

Understanding moderators of intervention effects is key to effectively disseminatingand implementing prevention programs. Potential moderators (ie, prevention type,depression outcome) were discussed earlier. There is additional heterogeneity perthe current literature, including variability in intervention content and length, settings,delivery modalities, and measurement. This diversity makes it more difficult to drawgeneral conclusions about intervention efficacy and suggests that more nuanced eval-uation of moderators may be beneficial in identifying the most promising interventions,study designs, and target populations.Some individual studies have evaluated hypothesized moderators. For instance,some universal prevention programs have examined intervention outcomes byhigh-risk versus low-risk participant status based on participants’ initial symptomlevels.45Meta-analysis is also a valuable tool for understanding potential moderators,including gender, intervention content, and intervention delivery. Although someearlier reviews found evidence that depression prevention programs were moreeffective for girls than for boys,76,77the most recent Cochrane Review concludedthat depression prevention effects did not differ by gender.73Program content mayalso modify outcomes, with some evidence for the efficacy of cognitive behavioral ap-proaches.73Stice and colleagues76reported that shorter prevention programs weremore efficacious than longer programs; by contrast, Horowitz and Garber77andMerry and colleagues73did not find a significant effect for program length acrossall programs, although Merry and colleagues73reported that, for universal programsonly, offering 8 or more sessions was superior to fewer sessions. Intervention deliverymodality has also been explored as a possible moderator of prevention effects, withsome evidence for stronger effect sizes when interventions were implemented bymental health professionals rather than teachers.75,76Fewer Internet-based pro-grams have been tested, but future research should compare Internet-based deliverywith more traditional delivery models