Table 2

Multivariable logistic regression of factors associated with self-perceived medical errors

VariablesOdds ratio (95% CI)P value
Age*0.79 (0.73 to 0.86)<0.0001
Sex (ref=male)
 Female0.70 (0.64 to 0.78)<0.0001
Education level (ref=associate degree or vocational diploma)
 Bachelor degree1.17 (1.04 to 1.34)0.0125
 Master degree or higher1.56 (1.29 to 1.88)<0.0001
Self-perceived physical health (ref=good)
 Fair1.40 (1.21 to 1.61)<0.0001
 Bad1.61 (1.37 to 1.89)<0.0001
The level of hospital (ref=primary hospital and others)
 Secondary hospital1.17 (1.05 to 1.31)0.0051
 Tertiary hospital1.22 (1.09 to 1.38)0.0009
Job title (ref=resident)
 Physician-in-charge1.12 (1.00 to 1.24)0.0446
 Chief physician1.31 (1.10 to 1.56)0.0023
Shortage of physicians (ref=no)
 Yes1.12 (1.01 to 1.24)0.0337
Workplace verbal aggression (times) (ref=no)
 1~31.66 (1.45 to 1.90)<0.0001
 4~62.08 (1.78 to 2.44)<0.0001
 7~2.33 (2.00 to 2.70)<0.0001
Effort-reward imbalance (ref=no)
 Yes1.04 (0.92 to 1.17)0.5595
 Overcommitment†1.12 (1.06 to 1.18)<0.0001
Depressive symptoms (ref=no)
 Yes1.05 (0.95 to 1.17)0.3532
Negative affect†1.41 (1.34 to 1.49)<0.0001
Satisfaction with job†0.96 (0.91 to 1.02)0.1676
Self-efficacy†0.81 (0.77 to 0.85)<0.0001
  • *This result meant each 10-year increase in the physicians’ age was associated with a 21% reduction in the odds of a self-reported error in the previous 3 months.

  • †These results showed the association between each 1-point change of continuous variables and the odds of self-reported medical errors in the previous 3 months.