Background
Well-defined measures of therapeutic benefit are essential for evaluating therapies and services. However, there is no single gold standard for defining ‘successful’ outcomes. We therefore examined the potential impact of adopting different success criteria.
Method
We analysed data for 7,064 patients undergoing psychological therapy in a single UK IAPT (Increasing Access to Psychological Therapy) Service, each patient being assessed for depression (PHQ-9) and anxiety (GAD-7) both at the start and end of treatment. Predictors of successful outcomes based on these measures were analysed separately for three different success criteria: based either on assessing clinically significant change, or reliable change, in depression and anxiety.
Results
The choice of criteria had little bearing on which variables predicted successful outcomes. However, the direction of the relationship between initial PHQ-9 or GAD-7 score and outcome success reverses when the criteria used to judge success are changed: successful outcomes are less probable under clinically significant change criteria for patients entering the service with more severe depression and/or anxiety but are more probable for such patients under reliable change criteria.
Conclusion
Relevant for clinicians, researchers, and policymakers, the choice of success criteria adopted can substantially change the incentives for patient selection into a therapy service. Our analysis highlights how the methods used to evaluate treatment outcomes could impact the priorities and organisation of therapeutic services, which could then impact on who is offered treatment. We recommend further investigations of success criteria in other conditions or treatments to determine the reproducibility of the effects we found.