Effect sizes and confidence intervals for resilience, control, and treatment-as-usual interventions for high risk populations
Description
We hypothesised that resilience interventions would lead to greater improvements in mental and physical health outcomes than control or treatment-as-usual comparison conditions. Eight online databases were searched, identifying twelve primary studies. Within and between group effect sizes were calculated for all relevant outcomes. Eight studies demonstrated intervention-related improvements and less often, control intervention or treatment-as-usual improvements. Few studies included follow-up and no studies assessed outcome following exposure to stress or trauma after the intervention. Thus, no conclusions could be made about the protective effects of the interventions. Methodological quality was low, effect sizes were modest, and risk of bias was unclear or high. The current empirical evidence lacks theoretical grounding, presenting poor translatable evidence. Future work would benefit from adopting a systematic approach to intervention development, identifying then targeting known predictors of mental and physical ill health, and evaluating interventions in comparison to alternative training as well as to wait-list with sufficient follow-up so that intervention specific and protective effects can be determined.