Statistical Relations Among the CORE ‐10, PHQ ‐9 and GAD ‐7 in an English Urban Community Counselling and Psychotherapy Clinic

ABSTRACT Background Three widely used brief self‐report symptom intensity measures used to assess clients seeking psychological therapy are the Patient Health Questionnaire‐9 (PHQ‐9), which focuses on depression, the Generalised Anxiety Disorder‐7 (GAD‐7), which focuses on anxiety, and the Clinical Outcomes in Routine Evaluation‐10 (CORE‐10), which focuses on general distress. Aim This study aimed to examine the statistical relations among the PHQ‐9, GAD‐7 and CORE‐10, their level of inter‐agreement and ability to predict clinical caseness. Method All three measures were administered contemporaneously to 1103 clients seeking psychological therapy at a community clinic in England. Data were anonymised and analysed following ethical guidelines. The data analysis plan was preregistered via the Open Science Framework. Results Correlations among the three measures ranged from r = 0.71 to 0.78, p < 0.001. The percent of clients meeting caseness criteria was highest for the CORE‐10 (87.3%), followed by the GAD‐7 (74.8%) and the PHQ‐9 (71.0%). Agreement between pairs of measures ranged from fair to moderate ( κ = 0.45 to 0.59), with the 3‐way agreement also moderate ( κ = 0.58). Structural Equation Modelling indicated these scales measured a common latent construct of psychological distress. The GAD‐7 showed a modest ceiling effect, whereas the CORE‐10 and PHQ‐9 did not. Conclusions The high level of overlap among the three measures suggests they may not distinguish sharply among types of distress. Pending future repeated‐measures research on the instruments' sensitivity to change, these findings suggest a single measure may suffice as a baseline measure of distress.