Abstract Background Nurses, the largest frontline healthcare group in the world, experience a high incidence of moral distress. Enhancing moral sensitivity (MS) can effectively alleviate this distress. However, MS levels among nurses have not been clearly defined. Therefore, this study aimed to assess the level of MS among nurses and provide evidence-based insights to improve their moral practices. Methods This review searched multiple databases, including PubMed, Cochrane Library, Embase, Web of Science, CINAHL, Scopus, Medline, China Knowledge Resource Integrated Database, Wanfang Database, VIP Database, Chinese Biomedical Database, Chinese Medical Journal Full Text Database, Google Scholar, and OpenGrey, from inception to December 31, 2024. Two reviewers (Ting Zhao and Shi Chen) independently screened the literature and extracted data. Their quality was assessed using the Joanna Briggs Institute’s Critical Appraisal Tool. Data were analyzed using Stata software (version 17.0) to synthesize the mean scores of the moral sensitivity questionnaire (MSQ). Subgroup and meta-regression analysis were performed to identify the sources of heterogeneity, and a sensitivity analysis was conducted to assess the robustness of the results. Results A total of 99 studies involving 29,387 nurses were included. The pooled mean score for MS was 4.49 [95% CI (4.29, 4.70)], indicating a moderate level. Meta-regression revealed the assessment instrument, country, and department as potential sources of heterogeneity. The mean scores by country were as follows: China (5.22), Korea (4.82), Iran (4.44), and Turkey (3.28). The scores for the different assessment instruments varied, with the MSQ-revised version (5.46) having the highest scores. Conclusion Nurses demonstrated moderate MS levels, indicating opportunities for further improvement. This review offers useful insights for nurse managers and educators in shaping strategies to improve moral training.