**Background:** Work in hospital environments is known to present unhealthy characteristics that affect workers' health, and this intensified during the COVID-19 pandemic. Occupational stress is associated with changes in lifestyle and eating habits, but few longitudinal studies have examined the relationship between occupational stress and dietary patterns using a robust theoretical framework. This study aimed to verify the level of occupational stress before and during the COVID-19 pandemic, and its change and association with dietary patterns among hospital workers.
**Methods:** This was a longitudinal study using baseline and first follow-up data from a cohort of workers at a private hospital in Santo Antônio de Jesus, Bahia, Brazil. Of 371 eligible workers, 218 participated. Data were collected between May and October 2019 (baseline, before pandemic) and between October and November 2020 (first follow-up, during pandemic). Occupational stress was assessed using the Job Content Questionnaire (JCQ) reduced version, applying the Demand–Control Model. Workers were classified as having high occupational stress if they scored above the median on demand and below the median on control. Dietary patterns were identified through principal component factor analysis of 14 food groups derived from the ELSA-Brasil food frequency questionnaire (114 items). Three dietary patterns were identified at both time points. Generalized Estimating Equation (GEE) models were used to assess associations between occupational stress, additional exposures (shift work, weekly workload, COVID-19 infection), and changes in dietary patterns, adjusting for covariates including age, sex, education, income, smoking, alcohol, physical activity, health self-perception, BMI, and waist circumference.
**Key Results:** At baseline, workers' mean age was 32.60 years (SD 8.30), 75.2% were women, 41.7% were health professionals, and 54.6% had high school education. The prevalence of high occupational stress increased 107% from 14.2% before the pandemic to 29.4% during the pandemic (p < 0.001). Shift work increased 26% (32.1% to 39.4%, p = 0.001), and working >44 h/week increased 32% (22.9% to 30.3%, p = 0.02). A total of 67.4% of workers reported testing positive for COVID-19. Three dietary patterns were identified at both time points, explaining 45.51% (pre-pandemic) and 44.47% (during pandemic) of total variance. Pattern A shifted from a traditional Brazilian diet (breads/cereals, legumes, meat, pasta, sweets, fats) before the pandemic to a healthier pattern (tubers, fruits, vegetables, oilseeds, eggs) during the pandemic. Pattern B shifted from a healthy pattern (tubers, fruits, vegetables, oilseeds, eggs) before the pandemic to an unhealthy pattern (breads/cereals, meat, fats, pasta, sweets, sugary drinks) during the pandemic. Pattern C remained relatively stable (breads/cereals, legumes, beverages before; breads/cereals, milk/dairy, fats, beverages during). In GEE models, occupational stress was not significantly associated with changes in any dietary pattern (Pattern A: coefficient 0.26, 95% CI −0.51 to 1.03, p = 0.51; Pattern B: coefficient −0.79, 95% CI −0.78 to 0.20, p = 0.12; Pattern C: coefficient −0.22, 95% CI −0.99 to 0.54, p = 0.57). However, COVID-19 infection was positively associated with changes in Pattern A (coefficient 0.684, 95% CI 0.079 to 1.290, p = 0.027), remaining significant after adjustment for waist circumference (coefficient 0.647, 95% CI 0.044 to 1.241, p = 0.036). Shift work was positively associated with changes in Pattern B (coefficient 0.650, 95% CI 0.053 to 1.248, p = 0.033), remaining significant after adjustment for education (coefficient 0.612, 95% CI 0.016 to 1.207, p = 0.044).
**Clinical Implications:** The study hypothesis that high occupational stress would be associated with changes in dietary patterns was not supported. Instead, COVID-19 infection was associated with a shift toward healthier eating (Pattern A), possibly reflecting increased health awareness and desire to strengthen the immune system. Conversely, shift work was associated with a shift toward unhealthier eating (Pattern B), consistent with literature showing shift workers have difficulty maintaining healthy diets and consume more high-calorie, high-sugar, and high-fat foods. These findings suggest that workplace reorganization during the pandemic—particularly shift work schedules—may be a more important driver of dietary changes than stress itself. The results underscore the need for interventions targeting shift workers' nutrition, such as providing healthier food options, regular meal breaks, and nutritional counseling. The high prevalence of COVID-19 infection (67.4%) among hospital workers also highlights the need for better infection control measures. Limitations include convenience sampling and self-reported COVID-19 status, though the longitudinal design comparing pre- and during-pandemic data is a notable strength.