The Role of Socioeconomic Deprivation in Explaining Neighborhood and Clinic Effects in the Effectiveness of Psychological Interventions

Objective: Treatment outcomes are known to vary according to therapist and clinic/organization (therapist effect, clinic effect). Outcomes may also vary according to the neighborhood where a person lives (neighborhood effect), but this has not previously been formally quantified. Evidence suggests that deprivation may contribute to explaining such cluster effects. This study aimed to (a) simultaneously quantify neighborhood, clinic, and therapist effects on intervention effectiveness and (b) determine the extent to which deprivation variables explain neighborhood and clinic effects. Method: The study used a retrospective, observational cohort design with a high intensity psychological intervention sample (N = 617,375), and a low intensity (LI) psychological intervention sample (N = 773,675). Samples each included 55 clinics, 9,000-10,000 therapists/practitioners, and over 18,000 neighborhoods in England. Outcomes were postintervention depression and anxiety scores and clinical recovery. Deprivation variables included individual employment status, domains of neighborhood deprivation, and clinic-level mean deprivation. Data were analyzed using cross-classified multilevel models.Results: Unadjusted neighborhood effects of 1%-2% and unadjusted clinic effects of 2%-5% were detected, with proportionally larger effects for LI interventions. After controlling for predictors, adjusted neighborhood effects of 0.0%-0.1% and clinic effects of 1%-2% remained. Deprivation variables were able to explain a significant proportion of the neighborhood effect (80%-90% of neighborhood variance) but not clinic effect. The majority of neighborhood variance could only be explained by a shared effect of baseline severity and socioeconomic deprivation variables. Conclusions: People in different neighborhoods respond differently to psychological intervention, and this clustering effect was mainly explained by socioeconomic factors. People also respond differently according to the clinic they access, but this could not be completely explained by deprivation in the present study.