Self-Reported Clinical Behavior Shows Weak Correlation With Objective Guideline Adherence

In an assessment of the Effectiveness of Guidelines for Dissemination and Education in Psychiatric Treatment (EGUIDE) project—a nationwide Japanese educational initiative promoting guideline-concordant pharmacotherapy—investigators examined whether psychiatrists' self-reported clinical behavior (CB) reflects comprehensive objective measures of guideline adherence. The multicenter observational study analyzed 1897 EGUIDE-trained psychiatrists whose post-training online CB scores were linked to discharge records. Guideline Adherence Rates (GAR) were evaluated across 5328 patients with schizophrenia and 3538 patients with major depressive disorder (MDD).

Unadjusted patient-level Spearman correlations showed positive but very weak associations between post-training CB scores and GAR for both schizophrenia (ρ = 0.09–0.14, all p < 0.001) and MDD (ρ = 0.08–0.11, all p < 0.001). Squared coefficients (approximately 0.006–0.020) indicated that self-reported CB accounted for less than 2% of the rank-order variation in GAR. These associations remained statistically significant in sensitivity analyses that aggregated data to the psychiatrist-year level, adjusted for institution type, psychiatrist clinical experience, and patient age, and used generalized estimating equations (GEE) to correct for patient clustering. The authors concluded that, given the cross-sectional design, unadjusted primary analysis, and non-independence of patients treated by the same clinician, post-training self-reported CB scores do not establish a reliable proxy for actual clinical practice.