The study identifies associations within the sampled workforce; it does not establish clinical diagnoses or causation. The quantitative phase ran from January 2025 to March 2026 at four tertiary public psychiatric hospitals in Xuzhou, Lianyungang, Huai’an and Suqian. Eligible participants were registered, on-duty nurses with at least one year of psychiatric nursing experience who provided direct inpatient care.

Recruitment was by convenience, and 357 of 420 distributed questionnaires passed the investigators’ quality checks, producing a valid response rate of 85.0%. The survey assessed depressive symptoms with the Zung SDS, occupational strain with the Personal Strain Questionnaire of the Occupational Stress Inventory–Revised, resilience with the 25-item Connor–Davidson Resilience Scale and workplace violence with Wang’s Workplace Violence Scale. An SDS index score of 53 or higher was classified as a positive screen, with 53–62, 63–72 and 73 or higher representing mild, moderate and severe symptom strata, respectively.

Violence was counted when a nurse reported at least one physical, verbal or psychological, or sexual incident perpetrated by a psychiatric inpatient or visiting family member during the preceding 12 months. Overall, 143 nurses, or 40.1% of the sample, reported such exposure. Nurses exposed to violence had an unadjusted positive-screen prevalence of 53.6%, compared with 31.5% among nurses who reported no included incident.

In the primary modified Poisson model, which entered violence, night-shift burden, occupational strain and resilience simultaneously and included hospital as a fixed effect, violence exposure was associated with a prevalence ratio of 1.65 for a positive screen, with a 95% confidence interval of 1.31–2.08. The remaining occupational estimates followed distinct patterns. Each standard-deviation increase in average monthly night-shift burden was associated with a prevalence ratio of 1.48, with a 95% confidence interval of 1.33–1.66.

Each standard-deviation increase in occupational strain was associated with a prevalence ratio of 1.45, with a 95% confidence interval of 1.28–1.65. Resilience showed the inverse pattern: each standard-deviation increase was associated with a prevalence ratio of 0.79, with a 95% confidence interval of 0.70–0.88. These adjusted estimates describe prevalence ratios rather than individual risks or causal effects.

A broader sensitivity model reportedly changed the estimates little, while exploratory interaction analyses did not provide a basis for a causal moderation claim. The survey population was predominantly female and mid-career. It included 325 women and 32 men, with a mean age of 34.60 years and mean psychiatric nursing experience of 10.46 years.

Participants averaged 5.86 night shifts per month. Their mean occupational-strain T-score was 68.42, their mean resilience score was 62.15 on the 0–100 Connor–Davidson scale, and their mean depressive-symptom index score was 52.17. Nurses with positive screens reported more night shifts, greater strain and lower resilience than those with negative screens, but the unadjusted group-comparison P values were exploratory and not adjusted for multiple comparisons.

Convenience recruitment from four tertiary hospitals limits transferability, and the cross-sectional survey cannot establish temporal order. All principal measures were self-reported, while sleep, staffing, prior mental health and the clinical complexity of nursing work were not fully measured. The explanatory qualitative phase followed 20 purposively selected nurses for three months.

The group comprised 16 women and four men aged 26–44, with three to 20 years of psychiatric nursing experience, and included five participants from each hospital. Selection sought variation in sex, age, experience, ward, hospital and SDS index score; after severity classifications were recalculated using the study’s prespecified thresholds, the subsample contained seven screen-negative nurses and 13 screen-positive nurses, five with mild, five with moderate and three with severe symptom strata. Each nurse completed a monthly one-to-one semistructured interview, kept a weekly reflective diary and was observed in relevant ward contexts.

The researchers combined these sources into chronological case files and retained divergences among accounts rather than automatically resolving them in favor of one source. Thematic saturation was reached at participant 17; three further planned interviews were completed to retain cross-site representation. Interviews showed that nurses’ distress did not always conform to the survey’s binary violence measure.

Recurrent verbal aggression, near misses, threats and anticipatory vigilance accumulated even when no single major incident occurred. One nurse described automatically expecting something bad before each shift despite the absence of a major incident, while another reported dreams about patients’ cries after repeated exposure to crying, abuse and non-cooperation. The investigators interpreted this as repeated occupational strain: a recurring low-intensity burden rather than a diagnostic construct.

Because the survey counted defined incidents over 12 months, the accounts suggest that exposure categories may not fully represent the persistent vigilance nurses described. Participants also described a mismatch between their professional commitment and external recognition. They drew meaning from psychiatric care while feeling that the specialty’s therapeutic complexity and skills were undervalued.

One nurse said, “We do everything we can to care for patients, yet people still call us ‘nurses who look after mad people’. Sometimes not being understood is harder than the workload itself.” The accounts linked this devaluation to associative stigma and emotional strain. Recognition was not directly measured in the survey, so the interviews expand the interpretation of the occupational-strain association rather than showing that devaluation caused a positive depressive-symptom screen.

Accounts of resilience differentiated several practical capacities: setting emotional boundaries, finding meaning in difficult encounters and recovering between shifts. One participant described absorbing patients’ irritability and feeling helpless every day, whereas another treated small patient improvements as evidence that the work had value. These contrasts accord with the observed inverse association between resilience scores and positive screens, but they also show why a low score cannot be read simply as an individual training deficit.

The findings are compatible with several explanations, including unequal access to recovery time and support, and do not establish that resilience itself reduced depressive symptoms. Similarly, the study’s post hoc strain-by-resilience interaction was hypothesis-generating and did not demonstrate causal moderation. Organizational support was a domain the survey had not measured, yet it emerged strongly in the qualitative material.

Participants knew that support activities existed but considered group lectures or occasional celebrations inadequate after difficult events. They described needing confidential contact, trained counselling, protected debriefing time and follow-up outside performance-management processes. Their objection concerned the timing and specificity of support, not simply its availability.

Night-shift burden, by contrast, was quantitatively supported but comparatively under-examined in interviews because the guide did not separately probe recovery time, commuting or domestic responsibilities associated with night work. Taken together, the results identify safer work organization and responsive post-incident care as practical targets for further testing, rather than as interventions already shown to work in this sample. The authors note that external evidence cited in the paper supports multicomponent prevention combining training, post-incident debriefing, rapid response and environmental modification, as well as regular clinical supervision and confidential counselling.

Voluntary resilience or mindfulness programs may add value, but the participants’ accounts caution against treating individual coping programs as substitutes for safer staffing, working conditions and organizational support. Because the study was observational, based on a convenience sample and restricted to four tertiary hospitals, it identifies relationships and participant experiences rather than establishing temporal order, causation or the effectiveness of a particular workplace program.

Subject of Research: Occupational factors associated with depressive-symptom screening among psychiatric nurses in northern Jiangsu

Article Title: Depressive symptoms and occupational correlates among psychiatric nurses in northern Jiangsu: an explanatory sequential mixed-methods study

Article References: Han, Y., He, J., Wang, Q., & Tang, Y. (2026). Depressive symptoms and occupational correlates among psychiatric nurses in northern Jiangsu: an explanatory sequential mixed-methods study. Frontiers in Psychiatry, 17, Article 1967611. https://doi.org/10.3389/fpsyt.2026.1967611

Image Credits: AI Generated

DOI: 10.3389/fpsyt.2026.1967611

Keywords: psychiatric nursing, depressive symptoms, workplace violence, night-shift work, occupational strain, psychological resilience, mixed-methods research, workforce well-being, Zung SDS, northern Jiangsu, Depressive, symptoms