**Background:** Healthcare workers experience high levels of occupational stress, which increased dramatically during the COVID-19 pandemic. Existing pandemic-related stress scales were developed for the general population and do not capture the specific individual and organizational stressors faced by healthcare workers. The authors aimed to develop and validate a brief, two-dimension instrument—the Individual- and Organization-Related Stressors in Pandemic Scale for Healthcare Workers (IOSPS-HW)—to separately assess individual health-related stressors and organization-related stressors in this population.
**Methods:** The study was conducted in two waves at Careggi University Hospital in Florence, Italy. Study 1 (T0, June–September 2020) included 950 healthcare workers (77% female, mean age 47.76 years, SD=10.20). An initial pool of 43 items was generated from literature review and expert consultation. After item reduction using exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), a final 20-item version was obtained. Reliability was assessed using omega coefficients, and validity was examined via correlations with the Impact of Event Scale-Revised (IES-R) and by comparing scores between workers in COVID-19 units (n=346) vs. non-COVID-19 units (n=604). Study 2 (T1, February–May 2021) included 843 healthcare workers (76% female, mean age 47.91 years, SD=10.66), of whom 321 participated in both waves. Measurement invariance across time was tested using multigroup CFA. Criterion validity was assessed using correlations with the GAD-7 (anxiety), PHQ-9 (depression), WHO-5 (well-being), and IES-R (post-traumatic stress). Predictive validity was tested via hierarchical linear regression with IOSPS-HW subscale scores at T0 predicting PTSD symptoms at T1. Test-retest reliability was calculated for the 297 participants with complete data at both time points.
**Key Results:** The final IOSPS-HW consists of 20 items: 8 items for Individual- and Health-related Stressors (IH-S) and 12 items for Organization-related Stressors (O-S). The two-factor model showed good fit at T0 (TLI=0.902, CFI=0.913, RMSEA=0.069, 90% CI [0.064, 0.073]) and T1 (TLI=0.914, CFI=0.924, RMSEA=0.066, 90% CI [0.060, 0.072]). Measurement invariance across time was supported (ΔCFI ≤ 0.004). Omega reliability was 0.93 (95% CI [0.92, 0.93]) for the total scale, 0.91 (95% CI [0.90, 0.92]) for O-S, and 0.84 (95% CI [0.82, 0.85]) for IH-S. Test-retest correlations were 0.70 for the total score, 0.67 for IH-S, and 0.65 for O-S (all p<0.001). The IOSPS-HW total score correlated positively with PTSD symptoms (r=0.61 at T0; r=0.63 at T1), anxiety (r=0.54), depression (r=0.50), and negatively with well-being (r=-0.38) (all p<0.001). Healthcare workers in COVID-19 units scored significantly higher on both subscales and total score compared to those in non-COVID-19 units (total score: 65.39 vs. 60.19, t=4.45, p<0.001, Cohen's d=0.30). In predictive analysis, IH-S at T0 significantly predicted PTSD symptoms at T1 (β=0.56, p<0.001), and adding O-S significantly improved the model (ΔR²=0.03, p<0.001), with both subscales remaining significant predictors (IH-S: β=0.38, p<0.001; O-S: β=0.25, p<0.001).
**Clinical Implications:** The IOSPS-HW is a valid and reliable tool that separately assesses individual and organizational pandemic-related stressors in healthcare workers. Its two-factor structure allows for targeted interventions: individual stressors (e.g., fear of illness, ethical dilemmas, family concerns) can be addressed through psychological support and psychoeducation, while organizational stressors (e.g., workload, communication, role clarity, resource availability) can inform systemic changes in hospital management and emergency preparedness. The scale's brevity (20 items) makes it suitable for large-scale monitoring in multivariate studies. Its items are not COVID-19-specific, enabling use in future pandemics or epidemics. The authors note limitations including the single-country (Italy) sample and lack of discrimination between professional categories or critical care settings. Future research should test cross-language invariance and applicability in other healthcare contexts.