Objective  To explore mediating effect of empathy on the association between stress and doctor-patient relationship among doctors.

  Methods  We conducted a survey among 800 doctors selected from four grade-A tertiary hospitals in Harbin city during March – July, 2019. A self-compiled questionnaire, Perceived Stress Scale (PSS) and Jefferson Scale of Empathy (JSE) were adopted in the study. Descriptive statistics, t-test, chi-square test, Pearson correlation analysis, and logistic regression analysis were performed in data analyses with SPSS 24.0 and Mplus 6.0.

  Results  Among 757 (94.63% of all the doctors surveyed) respondents with valid information, 583 (77.0%) and 174 (23.0%) reported a good and strained doctor-patient relationship and there was a significant gender-difference in the proportions of reporting a good or strained doctor-patient relationship (χ² = 14.644, P < 0.05). The PSS scores were 26.76 ± 4.51 and 26.21 ± 4.53 for male and female doctors, without significant gender difference (t = 1.669, P > 0.05). The JSE scores were 104.85 ± 16.57 and 111.79 ± 11.43 for male and female doctors, with a significant gender difference (t = − 6.411, P < 0.05). Mediation effect analysis revealed that empathy played a full mediation role in the association between stress and doctor-patient relationship among female doctors but a partial mediation role, accounting for 13.10% of total effect, among male doctors.

  Conclusion  For doctors, stress is an important factor leading to strained doctor-patient relationship and empathy mediates the association of stress with doctor-patient relationship, and the degree of the mediating effect differs by gender, with a full mediation in female doctors but a partial mediation in male doctors.