### INTRODUCTION

### MATERIALS AND METHODS

### Measures and variables

### Operational definitions

#### Work intensity

#### Work-life balance

#### Burnout

### Online survey using the questionnaires

### Statistical analysis

### RESULTS

### Socio-demographic characteristics and objective work environment (Table 1)

### Subjective work environment (Table 1)

### Special questions related to the Special Act on Korean Medical Residents

### Validity analysis of the variables

*p*<0.001). The factor loadings derived from the factor analysis on work intensity ranged in value from 0.617 to -0.888, which were greater than 0.50; and one factor was extracted which explained 65.45% of the total variance.

*p*<0.001). Therefore, these data were appropriate for conducting the factor analysis. The factor analysis resulted in a two-factor solution; the factors were named as ‘work-life balance-family and leisure,’ and ‘work-life balance-growth,’ and the factor loadings ranged in value from 0.469 to -0.801, which were greater than 0.50 with the exception of one question. The two factors accounted for 61.28% of the total variance.

*p*<0.001). The results of the factor analysis on burnout confirmed the three-factor solution. The factor names were consistent with previous research : ‘emotional exhaustion,’ ‘depersonalization,’ and ‘reduced personal achievement.’ The factor loadings ranged in value from 0.522 to -0.876, which were greater than 0.50. The three factors accounted for 59.15% of the total variance.

### Reliability analysis & normality test

### Analysis of the relationships between the socio-demographic characteristics and work intensity, work-life balance, and burnout

#### Work intensity

#### Work-life balance

#### Burnout

### Linear correlation analysis

*p*<0.001), work intensity and work-life balance-family and leisure was r=0.586 (

*p*<0.001), and work intensity and work-life balance-growth were r=0.409 (

*p*<0.001). In the relationship between the independent variables and dependent variables, statistically significant, positive correlations were found : work intensity and burnout r=0.620 (

*p*<0.001), and work-life balance and burnout r=0.693 (

*p*<0.001); work-life balance-family and leisure and burnout r=0.674 (

*p*<0.001), and work-life balance-growth and burnout r=0.597 (

*p*<0.001).

### Regression analysis

*p*=0.046), average working hours per day on weekdays (

*p*<0.001), vacation days per year (

*p*=0.035), and working-as-usual after overnight surgery (

*p*=0.002) had significant effects of burnout. Furthermore, female respondents (β=0.285), respondents with more average working hours per day on weekdays (β=0.243), fewer vacation days per year (β=-0.017), and frequent days of working-as-usual after overnight surgery (β=0.088), showed higher levels of burnout; 24.5% (

*p*<0.05) of the variance was explained. As a result of the goodness-of-fit test of the model containing only primary control variables, Model 1 was found suitable for the normality (

*p*=0.082) and equal variance (

*p*=0.373) tests of the standardized residuals.

*p*<0.001) and satisfaction with human relationships (

*p*=0.001) had significant effects on burnout. In the case of high level of job stress (β=0.362) and low level of satisfaction with human relationships (β=-0.098), indicating a higher level of burnout; 19.2% (

*p*<0.05) of the variance was explained. As a result of the goodness-of-fit test of this model, which contained secondary control variables, Model 2 was found suitable for the normality (

*p*=0.200) and equal variance (

*p*=0.467) tests of the standardized residuals.

*p*<0.001), work-life balance-family and leisure (

*p*<0.001), and work-life balance-growth (

*p*<0.001) had significant effects on burnout. In the case of high work intensity (β=0.314), negative work-life balance-family and leisure (β=0.216), and negative work-life balance-growth (β=0.147), the results were a high level of burnout; 14.5% (

*p*<0.05) of the variance was explained. In Model 3, the explanatory power increased significantly by 14.5% (

*p*<0.05); the independent variables including work intensity, work-life balance-family and leisure, and work-life balance-growth had significant effects on burnout. As a result of the goodness-of-fit test of the model containing independent variables, Model 3 was found suitable for the normality (

*p*=0.200) and equal variance (

*p*=0.409) tests of the standardized residuals.

### DISCUSSION

*p*=0.083, not described separately), so work intensity and stress level cannot explain the lower level of work-life balance experienced by neurosurgeon with specializations in brain and spine. There was no significant difference in stress level by gender (

*p*=0.463, not described separately), so work intensity and stress level cannot explain the lower work-life balance and high level of burnout of females. These two factors are beyond the scope of this study, so additional research on these topics should be undertaken in the future.

*p*<0.05), and the explanatory power of model 3 with the addition of independent variables such as work intensity and work-life balance was 14.5% (

*p*<0.05). In other words, it was found that work intensity, work-life balance-family and leisure, work-life balance-growth, job stress, and satisfaction with human relationships in the workplace had significant effects on burnout in this order.