Denoting treatment outcome in child and adolescent psychiatry: a comparison of continuous and categorical outcomes

Various approaches have been proposed to denote treatment outcome, such as the effect size of the pre-to-posttest change, percentage improvement, statistically reliable change, and clinical significant change. The aim of the study is to compare these approaches and evaluate their aptitude to differentiate among child and adolescent mental healthcare providers regarding their treatment outcome. Comparing outcomes according to continuous and categorical outcome indicators using real-life data of seven mental healthcare providers, three using the Child Behavior Checklist and four using the Strengths and Difficulties Questionnaire as primary outcome measure. Within each dataset consistent differences were found between providers and the various methods led to comparable rankings of providers. Statistical considerations designate continuous outcomes as the optimal choice. Change scores have more statistical power and allow for a ranking of providers at first glance. Expressing providers' performance in proportions of recovered, changed, unchanged, or deteriorated patients has supplementary value, as it denotes outcome in a manner more easily interpreted and appreciated by clinicians, managerial staff, and, last but not least, by patients or their parents.