Background
-
Which work factors and particularly job stressors are most strongly associated with burnout?
-
Do these work factors or stressors effect burnout more directly or indirectly, and are they mediated by effort-reward and/or work-life imbalance?
-
Can the same contributing factors, predictive effects and direct and indirect paths be observed for the intention to leave the profession as the other outcome under study?
-
Can any differences regarding these relationships and effects between the two major health professions be observed?
Methods
Data and study sample
Measures
Work stressors
-
Temporal workload was measured by the self-reported number of extra hours worked in a standard week, ranging from 0 (no voluntary or required overtime at all) over 1–2, 3–5 and 6–10 extra hours to more than 10 extra hours per week.
-
Physical workload was assessed by asking participants for the amount of time (the whole time, three quarters of the time, half of the time, one quarter of the time or never/almost never) they spend at work a) in painful or tiring positions (poor posture), b) carrying or moving persons, c) carrying or moving heavy loads, d) standing and e) with uniform hand or arm movements.
-
Emotional job stress was measured by the sum score of a 5-item scale taken from the German version of the Copenhagen Psychosocial Questionnaire (COPSOQ) and with questions such as ‘Does your work put you in emotionally disturbing situations?’ (response categories: 4 = always, 3 = often, 2 = sometimes, 1 = seldom, 0 = hardly ever/never) or by asking ‘Do you get emotionally involved in your work?’ and ‘Does your work require that you hide your feelings?’ (response categories: 0 = to a very small extent, 1 = to a small extent, 2 = somewhat, 3 = to a large extent, 4 = to a very large extent).
-
Finally, mental job stress was assessed with the sum score of five 4-point Likert scaled items selected from the 6-item subscale on over-commitment [26], also used in the COPSOQ. Items used were reports or statements of being unable to sleep at night after having left something unfinished at work, being unable to switch off from work when getting home or having a troubled mind due to work problems when waking up etc. (response categories: 0 = strongly disagree, 1 = disagree, 2 = agree, 3 = strongly agree). These items measure “the long arm of the job” and stressful work rather than excessive work engagement or over-commitment to the job as a personal characteristic or personality trait.
Work-related stress
Burnout
Intention to leave the profession
Analyses
Results
Descriptive statistics
Caregivers & nurses (incl. midwives) | Physicians | All health professionals | Total hospital employees | |
---|---|---|---|---|
N = 882 | N = 235 | N = 1441 | N = 1840 | |
Temporal workload | ||||
No regular overtime | 40.0% | 19.1% | 35.3% | 37.5% |
1–2 long hours/week | 42.7% | 22.6% | 39.4% | 37.0% |
3–5 long hours/week | 14.6% | 24.8% | 16.9% | 17.5% |
6+ long hours/week | 2.7% | 33.5% | 8.4% | 8.0% |
Physical workload | ||||
Low (0–1) | 7.9% | 29.7% | 15.3% | 17.8% |
Medium (2–5) | 42.6% | 57.2% | 47.1% | 46.6% |
High (6–10) | 39.1% | 12.2% | 30.3% | 28.6% |
Very high (11–20) | 10.4% | 0.9% | 7.3% | 6.9% |
Emotional job stress | ||||
Low (0–5) | 5.0% | 1.8% | 6.8% | 12.6% |
Medium (6–10) | 35.7% | 27.9% | 34.1% | 35.5% |
High (11–15) | 54.7% | 64.6% | 54.4% | 47.6% |
Very high (16–20) | 4.6% | 5.8% | 4.7% | 4.3% |
Mental job stress | ||||
Low (0–5) | 27.9% | 15.4% | 24.8% | 26.5% |
Medium (6–7) | 42.1% | 38.2% | 41.2% | 40.1% |
High (8–9) | 23.4% | 34.6% | 26.3% | 25.7% |
Very high (10–15) | 6.7% | 11.8% | 7.6% | 7.7% |
Effort-reward imbalance | ||||
(Very) low (≤0.8) | 7.4% | 11.7% | 10.3% | 12.5% |
Moderate (> 0.8–1.0) | 22.2% | 23.4% | 23.0% | 23.9% |
High (> 1.0–1.5) | 56.0% | 53.7% | 53.5% | 51.1% |
Very high (> 1.5) | 14.5% | 11.2% | 13.3% | 12.6% |
Work-life imbalance | ||||
Low (0–5) | 29.4% | 7.9% | 28.7% | 35.2% |
Medium (6–10) | 36.8% | 23.7% | 33.9% | 32.1% |
High (11–15) | 27.2% | 45.2% | 28.6% | 25.4% |
Very high (16–20) | 6.6% | 23.2% | 8.8% | 7.3% |
Burnout symptoms | ||||
(Very) few (0–11) | 70.0% | 65.1% | 68.7% | 69.6% |
Some (12–15) | 23.3% | 22.0% | 22.9% | 22.2% |
Many (16–24) | 6.7% | 12.9% | 8.4% | 8.2% |
Thoughts of leaving the profession | ||||
Never (5) | 47.0% | 54.7% | 49.4% | 51.1% |
Several times per year (4) | 34.4% | 31.0% | 33.9% | 32.8% |
Several times per month to daily (1–3) | 18.5% | 14.2% | 16.7% | 16.1% |
Bivariate analyses
Increased burnout symptoms (16–24) | Frequent thoughts of leaving the profession (several times per month to daily) | |||||
---|---|---|---|---|---|---|
% | aORa) | 95% CI | % | aORa) | 95% CI | |
Population at risk / affected | 8.2 | 16.1 | ||||
Temporal workload | ||||||
No regular overtime | 4.0 | 1 | 12.4 | 1 | ||
1–2 long hours/week | 9.1 | 2.47 | 1.53–3.97 | 19.3 | 1.70 | 1.25–2.31 |
3–5 long hours/week | 12.8 | 3.77 | 2.22–6.39 | 17.1 | 1.63 | 1.11–2.39 |
6+ long hours/week | 13.9 | 4.10 | 2.10–8.01 | 17.2 | 1.71 | 1.01–2.88 |
Number of cases in model | 1629 | 1649 | ||||
Physical workload | ||||||
Low (0–1) | 5.0 | 1 | 12.0 | 1 | ||
Medium (2–5) | 5.5 | 1.18 | 0.64–2.16 | 11.9 | 0.96 | 0.64–1.46 |
High (6–10) | 11.5 | 2.78 | 1.49–5.17 | 21.8 | 1.90 | 1.23–2.91 |
Very high (11–20) | 22.6 | 6.57 | 3.20–13.46 | 32.5 | 3.21 | 1.85–5.55 |
Number of cases in model | 1629 | 1649 | ||||
Emotional job stress | ||||||
Low (0–5) | 4.5 | 1 | 8.1 | 1 | ||
Medium (6–10) | 3.2 | 0.79 | 0.36–1.73 | 10.2 | 1.35 | 0.77–2.35 |
High (11–15) | 11.4 | 3.12 | 1.56–6.24 | 20.6 | 3.32 | 1.96–5.63 |
Very high (16–20) | 22.4 | 7.59 | 3.16–18.23 | 36.4 | 8.67 | 4.32–17.41 |
Number of cases in model | 1706 | 1725 | ||||
Mental job stress | ||||||
Low (0–5) | 1.9 | 1 | 6.6 | 1 | ||
Medium (6–7) | 5.0 | 2.71 | 1.28–5.73 | 13.0 | 2.27 | 1.47–3.50 |
High (7–9) | 12.7 | 7.57 | 3.68–15.56 | 23.0 | 4.58 | 2.97–7.07 |
Very high (10–15) | 31.9 | 25.39 | 11.84–54.46 | 39.9 | 10.30 | 6.16–17.21 |
Number of cases in model | 1722 | 1740 | ||||
Effort-reward imbalance | ||||||
(Very) low (≤0.8) | 0.5 | 1 | 3.4 | 1 | ||
Moderate (> 0.8–1.0) | 2.8 | 6.68 | 0.85–52.27 | 5.2 | 1.58 | 0.65–3.85 |
High (> 1.0–1.5) | 7.6 | 20.00 | 2.75–145.6 | 16.7 | 6.54 | 3.00–14.29 |
Very high (> 1.5) | 28.6 | 97.69 | 13.29–717.9 | 46.2 | 30.06 | 13.34–67.73 |
Number of cases in model | 1620 | 1637 | ||||
Work-life imbalance | ||||||
Low (0–5) | 1.9 | 1 | 6.5 | 1 | ||
Medium (6–10) | 3.9 | 2.42 | 1.16–5.07 | 11.7 | 2.00 | 1.32–3.04 |
High (11–15) | 14.7 | 10.43 | 5.35–20.35 | 25.5 | 5.46 | 3.67–8.13 |
Very high (16–20) | 33.6 | 32.64 | 15.74–67.69 | 45.8 | 14.57 | 8.87–23.93 |
Number of cases in model | 1728 | 1747 |
Multivariate analyses
Dependent or outcome variable: Burnout symptoms (CBI score 0–24) | Caregivers & nurses (incl. midwives) | Physicians | All health professionals | Total hospital employees | ||||
---|---|---|---|---|---|---|---|---|
N = 882 | N = 235 | N = 1441 | N = 1840 | |||||
Beta coeff. (β) | Beta coeff. (β) | Beta coeff. (β) | Beta coeff. (β) | |||||
Step 1 | Step 2 | Step 1 | Step 2 | Step 1 | Step 2 | Step 1 | Step 2 | |
Independent or exposure variables: | ||||||||
Temporal workload (number of overtime hours per week 0–10+) | n.s. | n.s. | n.s. | −.13* | n.s. | −.09*** | n.s. | −.07** |
Physical workload (sum score 0–20) | .20*** | .12*** | .16* | n.s. | .15*** | .07** | .17*** | .09*** |
Emotional job stress (sum score 0–20) | .26*** | .13*** | .23*** | n.s. | .22*** | .09** | .17*** | n.s. |
Mental job stress (sum score 0–15) | .25*** | .16*** | .28*** | n.s. | .31*** | .20*** | .34*** | .23*** |
Intervening or mediating variables: | ||||||||
Effort-reward imbalance (ERI ratio) | – | .08* | – | .17* | – | .14*** | – | .12*** |
Work-life imbalance (sum score 0–20) | – | .35*** | – | .48*** | – | .35*** | – | .35*** |
Control variables: | ||||||||
Sex (male) | n.s. | n.s. | n.s. | n.s. | −.05* | −.06* | n.s. | −.05* |
Age (< 25, 25–34, 35–44, 45–54, 55+) | −.13*** | n.s. | n.s. | n.s. | −.13*** | −.09*** | −.13*** | −.10*** |
Educational level (low, medium, high, very high) | – | – | – | – | n.s. | n.s. | n.s. | −.05* |
Adjusted R square
| .300 | .400 | .228 | .433 | .276 | .383 | .278 | .376 |
No. cases in model | 763 | 708 | 204 | 187 | 1242 | 1157 | 1551 | 1438 |
Dependent or outcome variable: Thinking of leaving the profession (ordinal scale from 1 ‘never’ to 5 ‘daily’) | Caregivers & nurses (incl. midwives) | Physicians | All health professionals | Total hospital employees | ||||
---|---|---|---|---|---|---|---|---|
N = 882 | N = 235 | N = 1441 | N = 1840 | |||||
Beta coeff. (β) | Beta coeff. (β) | Beta coeff. (β) | Beta coeff. (β) | |||||
Step 1 | Step 2 | Step 1 | Step 2 | Step 1 | Step 2 | Step 1 | Step 2 | |
Independent or exposure variables: | ||||||||
Temporal workload (number of overtime hours per week 0–10+) | n.s. | −.09* | n.s. | n.s. | n.s. | −.10*** | n.s. | −.11*** |
Physical workload (sum score 0–20) | .12*** | n.s. | .15* | n.s. | .12*** | n.s. | .11*** | n.s. |
Emotional stress (sum score 0–20) | .21*** | n.s. | n.s. | n.s. | .17*** | n.s. | .18*** | n.s. |
Mental stress (sum score 0–15) | .18*** | .08* | .29*** | .19* | .21*** | .10*** | .21*** | .09*** |
Intervening or mediating variables: | ||||||||
Effort-reward imbalance (ERI ratio) | – | .31*** | – | .36*** | – | .34*** | – | .34*** |
Work-life imbalance (sum score 0–20) | – | .25*** | – | n.s. | – | .20*** | – | .19*** |
Control variables: | ||||||||
Sex (male) | n.s. | n.s. | n.s. | n.s. | n.s. | n.s. | n.s. | n.s. |
Age (< 25, 25–34, 35–44, 45–54, 55+) | −.13*** | −.09** | n.s. | n.s. | −.10*** | −.11*** | −.11*** | −.11*** |
Educational level (low, medium, high, very high) | – | – | – | – | n.s. | n.s. | −.06* | −.06* |
Adjusted R square
| .166 | .287 | .110 | .221 | .143 | .262 | .140 | .251 |
No. cases in model | 776 | 715 | 204 | 187 | 1260 | 1168 | 1573 | 1450 |
Discussion
Main findings
Strengths and limitations
-
The study was not focused on a single health profession or an individual (health) outcome, as is usually the case, but was integrative as regards the study population, the outcomes under study and the work factors and stress models that were included. Nurses and physicians were considered together. Emotional and mental job stresses were supplemented by temporal and physical workloads and jointly taken into account as possible predictors. Both effort-reward imbalance and work-life imbalance were included simultaneously. And burnout symptoms were supplemented by thoughts of leaving the profession as a second major outcome.
-
Sample size and numbers of cases of the two main health professions (nurses, physicians) were sufficiently large to allow for stratified and multivariate association analyses simultaneously, i.e. for comparisons between different health professions or occupational groups and for adjustments for different control variables and possible confounders and/or mediators.
-
The use of mostly well established and validated measures and (sub)scales in the written questionnaire broadly insured the validity and reliability of the study findings.
-
Due to its cross-sectional design, causal conclusions cannot be drawn, even though there was repeatedly talk of effects, predictors or dose-response relationships.
-
Since the participating hospitals and rehab clinics were self-selected and hence not randomly selected, and the return rate of the questionnaire-based survey was rather low, the study sample is not representative for hospital staff or health professionals in German-speaking Switzerland. A selection bias due to a systematic self-exclusion of the stressed, heavily loaded and dissatisfied employees cannot be ruled out. This may possibly lead to an underestimation of the true burden of stress and disease and particularly the prevalence of burnout or the intention to change the profession among hospital staffs or certain health professions. Therefore the findings can only be generalized to a limited extent, and the prevalence rates must be treated with caution.