Predictive Validity of the Emergency Physician and Global Job Satisfaction Instruments

OBJECTIVE: To evaluate the predictive validity of the Emergency Physician Job Satisfaction (EPJS) and Global Job Satisfaction (GJS) instruments. METHODS: Prospective mail survey of 223 Canadian emergency physicians (EPs) using a 42-item questionnaire, including 14 items evaluating their reasons for leaving emergency medicine (EM). Original (1990) EPJS and GJS scores were analyzed using 1-way ANOVA and Scheffe's test comparing the physicians who left EM with those still in their original jobs, and those who had left their original jobs but who stayed in EM. Mean scores on the 14 "reason for leaving" items were compared with scores from an earlier sample of U.S. physicians using a t-test for independent means. Criteria for statistical significance were set at alpha = 0.05 for all analyses. RESULTS: The response rate for the primary study questions was 99.1%. Of the respondents, 29.4% had left their original jobs, and 10.4% had left EM altogether. The GJS scores for the physicians who left EM were significantly different from those for the physicians who stayed (p = 0.004). The EPJS scores for the physicians who left EM were not significantly different from those for the physicians who stayed (p = 0.56). There was no significant difference in scores between the Canadian and U.S. physicians' reasons for leaving EM (all p-values > 0.05). Shiftwork scored the highest as a reason to leave EM. CONCLUSIONS: A low GJS score is associated with physicians' leaving EM, but not with changing jobs. The EPJS instrument was not associated with either outcome. Canadian and U.S. EPs place similar levels of importance on potential reasons for leaving EM.