Table 3.
Effects of occupational group on number and patterns of sickness absence spells.
All spells1) |
Short spells (1-3 days)2) |
Medium spells (4-14 days)3) |
Long spells (>14 days)4) |
"Normal" absence pattern versus no absence5) |
"Abnormal" absence pattern versus no absence6) |
|||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Start model |
Final model |
Start model |
Final model |
Start model |
Final model |
Start model |
Final model |
Start model |
Final model |
Start model |
Final model |
|
Occupational group7) |
RR | RR (95% CI) |
RR | RR (95% CI) |
RR | RR (95% CI) |
RR | RR (95% CI) |
OR | OR (95% CI) |
OR | OR (95% CI) |
Doctors | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
Physio- therapists |
1.45 |
1.43 (1.13-1.83) |
1.39 |
1.36 (1.06-1.74) |
1.56 |
1.52 (1.00-2.32) |
1.19 |
1.03 (0.33-3.21) |
1.19 |
1.29 (0.70-2.37) |
2.13 |
2.27 (1.07-4.83) |
Nurses | 1.58 |
1.68 (1.36-2.08) |
1.35 |
1.46 (1.18-1.82) |
2.20 |
2.29 (1.59-3.30) |
1.49 |
1.41 (0.50-3.95) |
1.98 |
2.35 (1.39-3.97) |
4.42 |
5.54 (2.87-10.7) |
Medical secretaries |
1.64 |
1.72 (1.37-2.16) |
1.34 |
1.43 (1.14-1.81) |
2.65 |
2.81 (1.92-4.13) |
1.10 |
1.06 (0.35-3.24) |
1.71 |
2.03 (1.12-3.67) |
4.83 |
6.31 (3.06-13.0) |
Nursing assistants |
1.99 |
1.95 (1.58-2.41) |
1.43 |
1.47 (1.18-1.82) |
3.41 |
3.34 (2.33-4.80) |
2.33 |
1.89 (0.68-5.28) |
2.54 |
2.90 (1.65-5.09) |
9.52 |
10.5 (5.30-20.8) |
Cleaners/ porters |
1.52 |
1.41 (1.10-1.80) |
0.83 |
0.79 (0.60-1.03) |
4.30 |
4.19 (2.84-6.19) |
1.63 |
1.37 (0.45-4.13) |
2.09 |
2.43 (1.25-4.70) |
6.60 |
7.39 (3.33-16.4) |
p8) | <0.0001 | <0.0001 | <0.0001 | <0.0001 | <0.0001 | <0.0001 | 0.10 | 0.13 | 0.006 | 0.001 | <0.0001 | <0.0001 |
Mean change of estimates compared to start model |
0% | 2.2% | 0.6% | -11% | 15% | 17% |
Results from multiple Poisson regression analyses (rate ratios (RR) and their 95% confidence intervals (CI)) of all spells, short, medium and long spells, and from multiple logistic regression analysis (odds ratios (OR) and their 95% CI)) of "normal" and "abnormal" absence patterns versus no absence spells.
1) Start model adjusted for age, gender, work-place unit. Final model as start model and for violence, job satisfaction, weekly work hours, being single and general health.
2) Start model adjusted for age, gender, work-place unit, any medium absence, any long absence. Final model as start model and for quality of leadership, social support at work, job satisfaction, weekly work hours, being single and general health.
3) Start model adjusted for age, gender, work-place unit, any short absence, any long absence. Final model as start model and for overall demands, control, job strain, job satisfaction, being single and general health.
4) Start model adjusted for age, gender, work-place unit, any short absence, any medium absence. Final model as start model and for violence and general health.
5) Start model adjusted for age, gender, work-place unit, days at risk. Final model as start model and for overall demands, control, job strain, weekly work hours, overtime work and general health.
6) Start model adjusted for age, gender, work-place unit, days at risk. Final model as start model and for overall demands, control, job strain, weekly work hours and general health.
7) See table 1.
8) Probability of no difference between occupational groups, χ2-test.