Objective  To compare the differences of results obtained using different burnout criteria based on the same scale in practical applications and provide a theoretical basis for future burnout research.

Methods  In September 2023, 884 healthcare workers from 11 community health service centers in Jinshan District, Shanghai, were selected using a stratified random sampling method. They were surveyed with the Maslach Burnout Inventory-General Survey(MBI-GS). Three different burnout criteria(i.e., using the scores of three dimensions such as emotional exhaustion, depersonalization, and reduced sense of personal accomplishment and adopting different definition methods to determine the burnout situation of the respondents)used in the literature, were selected to assess the burnout of the survey subjects, and the differences of the three methods of determining the results of the burnout assessment were explored.

Results  A total of 830 valid questionnaires were returned, with a response rate of 93.89%. There were statistically significant differences among the three methods in terms of the dimensions of burnout, overall detection rates, and severity distribution(P < 0.05). The three methods showed very strong agreement in the assessment of the depersonalization dimension(Kappa = 0.907, P < 0.001), but weak or very poor agreement in emotional exhaustion, reduced personal accomplishment, and overall burnout detection rates(Kappa = 0.112, -0.254, 0.057, respectively, all P < 0.001). Multivariate logistic regression analysis with burnout occurrence as the response variable showed that four variables, namely, position, weekly working hours, sleep quality, and self-rated health, entered the equation in Method 1(OR = 0.309 to 0.590, all P < 0.05), while three variables, namely, position, education level, and self-rated health, entered the equation in Method 2(OR = 0.171 to 1.788, all P < 0.05), and three variables of weekly working hours, sleep quality, and self-rated health entered the equation in Method 3(OR = 0.268 to 0.532, all P < 0.05). Different methods yielded different influencing factors and interpretations of the results.

Conclusions  The three different burnout criteria of the MBI-GS resulted in wide differences in burnout assessment. Caution should be exercised when citing, comparing, and extrapolating research results. It is recommended to establish a unified burnout criterion for the same scale.