Purpose:To retrospectively examine depression and social anxiety improvement in patients on sick leave due to depression who participated in a return-to-work intervention (RTW-I) program.Methods:Patients visited a psychiatric outpatient clinic simulating workplaces to learn recurrence prevention skills through RTW-Is, including group cognitive behavioral therapy, from April 1, 2013, to September 30, 2017. The Beck Depression Inventory-Second Edition (BDI-II), Social Adaptation Self-Evaluation Scale (SASS), and Liebowitz Social Anxiety Scale (LSAS) scores of 112 patients were analyzed before and after the intervention program. Missing postprogram data were substituted using the last observation carried forward scores. Next, 45 patients who responded to the work continuity survey 1 year after RTW-I were categorized into Group A (patients who continued working: 37) and Group B (those who did not continue: 8).Results:The mean BDI-II scores significantly decreased from preintervention 19.4 to postintervention 7.9 (t = 13.303, P < .001). The mean SASS scores significantly increased from preintervention 31.9 to postintervention 36.0 (t = −5.953, P < .001). The mean LSAS scores significantly decreased from preintervention 54.7 to postintervention 37.0 (t = 8.682, P < .001), and all scores demonstrated an improvement. Patients who continued working showed improved depressive and social anxiety symptoms. The BDI-II and SASS scores showed no significant differences between the groups, but the postintervention LSAS scores were significantly different (P = .041). LSAS score changes: Group A = −26.2; Group B = −9.8; estimated difference: −17.920, 95% CI: −32.181 to −3.659, P = .015.Conclusions:The RTW-I program improved depressive and social anxiety symptoms. Patients with improved scores continued working for 1 year after the intervention.Trial registration: This trial was retrospectively registered with the UMIN Clinical Trial Registry (UMIN-CTR) (ID: UMIN000037662) on August 10, 2019.