Skip to main content
. 2008 Feb 19;98(6):1046–1052. doi: 10.1038/sj.bjc.6604270

Table 4. Odds ratios and corresponding 95% confidence intervals according to sociodemographic characteristics, GHQ-12 and sources of stress related to job and patients.

  High EE High DP Low PA
Dependent variable ORa (95%CI) ORa (95%CI) ORa (95%CI)
Sociodemographic charateristics
 Age 1.03 (0.99–1.06) 0.98 (0.94–1.01) 0.93* (0.89–0.98)
 Sex 0.93 (0.47–1.84) 0.43* (0.24–0.80) 1.58 (0.69–3.63)
 Profession 1.15 (0.59–2.26) 0.89 (0.48–1.65) 0.89 (0.38–2.09)
       
Sources of stress related to job
 Physical tiredness 2.01* (1.12–3.61) 1.64 (0.94–2.88) 1.00 (0.48–2.07)
 Personal dissatisfaction 1.47 (0.82–2.64) 1.49 (0.86–2.57) 2.13* (1.04–4.36)
 Losing patients 1.29 (0.72–2.29) 0.75 (0.43–1.33) 0.46* (0.22–0.97)
 Low salary 1.35 (0.76–2.41) 1.19 (0.69–2.05) 1.30 (0.64–2.63)
 Lack of free time 1.53 (0.85–2.76) 1.41 (0.81–2.47) 0.43* (0.20–0.93)
       
Sources of stress related to patients
 Work with terminally ill patients 1.67 (0.95–2.95) 1.28 (0.75–2.20) 1.74 (0.85–3.55)
 Work with demanding patients 1.83 (0.87–3.85) 1.54 (0.76–3.11) 2.44* (1.07–5.60)
       
Psichiatric morbidity
 GHQ-12 1.47* (1.32–1.63) 1.17* (1.07–1.28) 1.16* (1.04–1.30)
C-indexb 0.830 0.719 0.798

CI=confidence interval; DP=depersonalisation; EE=emotional exhaustion; GHQ=General Health Questionnaire; OR=odds ratio; PA=personal accomplishment.

*P-value <0.05.

a

Estimated from multiple logistic model including terms for place of work, patient contact time and marital status.

b

C-index provide the predictive power of a logistic regression model.