No abstract

KK, MW, and SC oversaw the study design. KK and SC led data acquisition through the literature reviews, Delphi surveys, and Open Review surveys, as well as data analysis and interpretation. KK takes responsibility for the integrity of the data and the accuracy of data analysis. All authors attended teleconferences, and all but the core project team (KK, MW, SC) completed anonymous votes and feedback surveys as part of the Delphi process and formed consensus on the final recommendation. KK drafted the manuscript with support from SC, and all co-authors critically reviewed the working draft and agreed to the revisions and final submission. Administrative, technical, and organisational support was provided by SC. Declaration of interests AMA receives royalties fromOxford University Press for the Anxiety Disorders Interview Schedule (ADIS), Child and Parent Versions. PB is involved with the development across Australia of routine outcome measurement in public mental health. He chairs the National Mental Health Child and Adolescent Information Development Expert AdvisoryGroup. There is an interest in supporting routine outcome measurement and benchmarking between organisations.

Objective: Depression and anxiety are the most prevalent mental health problems in youth, yet almost nothing is known about what outcomes are to be expected at the individual level following routine treatment. This paper sets out to address this gap by undertaking a systematic review of outcomes following treatment as usual (TAU) with a particular focus on individual-level outcomes. Method: MEDLINE, Embase and PsycInfo were searched for articles published between 1980 and January 2019 that assessed TAU outcomes for youth depression and anxiety accessing specialist mental health care. Meta-analysis considered change at both group-level preÀpost effect size (ES) and individual-level recovery, reliable change, and reliable recovery. Temporal analysis considered stability of primary and secondary outcomes over time. Subgroup analysis considered the moderating effect of informant; presenting problem; study design; study year; mean age of youth; use of medication; intervention dosage and type of treatment offered on outcomes. A protocol was preregistered on PROSPERO (CRD42017063914). Results: Initial screening of 6,350 publications resulted in 38 that met the inclusion criteria, and that were subsequently included in meta-analyses. This resulted in a final full pooled sample of 11,739 young people (61% of whom were female, mean age 13.8 years). The preÀpost ES (Hedges' g) at first/final outcome (13/26 weeks) was À0.74/À0.87. The individual-level change on measures of self-report was 38% reliable improvement, 44% no reliable change, and 6% reliable deterioration. Outcomes varied according to moderators, informant, problem type and dosage. Conclusion: Poor data quantity and quality are limitations, but this is the first study that indicates likely rates of reliable improvement for those accessing TAU. We propose the need for improved reporting of both individual-level metrics and details of TAU to enable greater understanding of likely current outcomes from routine care for youths with depression and anxiety in order to allow the potential for further improvement of impact.