Gender Differences for the Associations among Housework, Self-Stigma, and Psychiatric Symptoms among Community Severe...

Previous studies have suggested housework might reduce self-stigma and psychiatric symptoms, but it had not been validated in severe mental illness (SMI) patients. Considering gender differences, we examined the relationship between housework, self-stigma, and psychiatric symptoms to link family, social, and disease factors, aiding rehabilitation from a "life-oriented" perspective, promoting patients from "passive treatment" to "active life reconstruction".

We collected 486 questionnaires from SMI patients in Shandong, China (210 men, 276 women, 124 in 0-40 age group, 244 in 41-64 age group, and 118 in 65 or older age group). Brief Psychiatric Rating Scale, Self-Stigma Scale for Chronic Illness 8-item version were used to measure psychiatric symptoms and self-stigma. Housework and demographic-variables were collected through a questionnaire complied by our research group. Linear regression explicit the relationship between the study variables, based on which a mediation analysis is established to verify the internal mechanism which controls confounding variables (age, religion, marriage, education, occupation, labor capacity, living alone, liability, chronic disease, disease status).

Most participants did housework for less than 1 hour, with men more likely than women (76.2% vs 56.2%). Housework reduced psychiatric symptoms in both genders (men: β = -5.563 (95% CI = -9.513, -1.613), P <0.01, and women: β = -4.088 (95% CI = -7.706, -0.469), P <0.01). However, housework only lowered self-stigma in women (β = -2.322 (95% CI = -3.922, -0.723), P <0.01). Self-stigma fully mediated the housework and psychiatric symptoms (indirect effect = -2.228, 95% CI = -4.046, -0.716).

Housework alleviated psychiatric symptoms in both genders, but only reduced the self-stigma of disease in women, suggesting its potential as a modifiable rehabilitation intervention.

The cross-sectional design precludes causal inferences, self-reported data may introduce recall bias, the Shandong-based sample limits generalizability, unmeasured confounders (eg, medications, comorbidities) warrant further study, and the small sample size may reduce statistical power.