Temporal and Contextual Variations in Job Satisfaction Between Physicians and Nurses: A Systematic Review and Meta-Analysis.

Objectives: This systematic review and meta-analysis evaluated differences in job satisfaction scores between nurses and physicians, examining variation by (a) care setting (hospital, emergency department, outpatient, mixed), and (b) time period (pre-COVID, during COVID, post-COVID). Methods: We systematically searched PubMed, Scopus, ScienceDirect, Web of Science, and CINAHL for studies published between January 2020 and July 2025. Eligible studies reported mean and standard deviation values for job satisfaction among physicians and nurses in healthcare settings across the specified timeframes. Studies were excluded if they assessed other types of satisfaction or combined data across COVID periods. Pooled standardized mean difference (SMD) was calculated using random-effects models in R. Results: Before COVID-19, the SMD was -2.40 (95% CI -8.05 to 3.26; I2 = 98%). During the pandemic, the estimate was 1.39 (95% CI -0.57 to 3.35; I2 = 91.5%), and post-pandemic, it remained small (SMD = 0.29; 95% CI -1.63 to 2.22; I2 = 95.8%). Emergency care during COVID showed a significant advantage for physicians (SMD = 0.29; 95% CI 0.05 to 0.52; I2 = 0%). Post-COVID, mixed settings slightly favored physicians (SMD = 0.06), while primary care favored nurses (SMD = -0.30); subgroup differences were significant. Conclusions: The findings reveal that job satisfaction is not solely determined by professional role but is significantly influenced by temporal and contextual factors. Job satisfaction is shaped more by temporal and contextual factors than by professional role. While no consistent differences were observed pre-pandemic, emergency care favored physicians during COVID, and post-pandemic trends showed modest advantages for nurses in primary care and physicians in mixed settings. Due to the methodological limitations of this meta-analysis, including high heterogeneity, reliance on cross-sectional data, and very low/low certainty of evidence, these results should be interpreted with caution.