Graduate medical education-led continuous assessment of burnout and learning environments to improve residents’ wellbeing

Table 1 Maslach Burnout Inventory scores of all GME residents at 5 points analyzed by Kruskal–Wallis test for trends and MANOVA

	December 2017; n = 130	May 2018; n = 131	December 2018; n = 126	May 2019; n = 133	December 2019; n = 121	Kruskal–Wallis p value	MANOVA p value
Emotional Exhaustion Score	28.12 (0.93)	26.56(0.91)	24.60 (0.98)	25.31(0.95)	24.35 (0.96)	0.04!	0.03*
Depersonalization Score	12.48 (0.58)	11.65(0.51)	10.60 (0.61)	10.81(0.56)	10.65(0.56)	0.074	0.092
Personal Accomplishment Score	38.51 (0.55)	38.14(0.62)	39.84(0.56)	39.41(0.58)	40.05(0.54)	0.113	0.078

Legend:
() = standard error of the mean
N = total number residents who completed the survey for each time period
Results of Kruskal–Wallis test for trends:
! Based on the five tests performed, a Bonferroni-corrected alpha level of p = 0. 0125 (0.05/4) was reported per test as significant for the Kruskal–Wallis test. Thus, the p value of 0.04 was not significant in the Kruskal–Wallis test for trends
Results of MANOVA analysis:
*There was homogeneity of variance-covariances matrices, as assessed by Box’s test of equality of covariance matrices (p = 0.705). Wilks’ Lambda multivariate statistics p value was 0.205 which showed that there were no significant differences between the groups. On tests of between-subjects effects Emotional Exhaustion Score had p value of 0.03 which became non-significant p = 0.055 on post-hoc Bonferroni correction

ISSN: 1472-6920