Robust linear regression analyses. Estimation of differences in sickness absence days over 6 months between UC and UC+MI or UC+SVAI
Variable | Unadjusted ITT analysis (n=509) | Adjusted complete case analysis* (n=479) | Adjusted primary ITT analysis with imputations*† (n=509) | ||||||
Coef. B | 95% CI | Coef. B | 95% CI | Coef. B | 95% CI | ||||
UC+MI | −7.3 | −16.6 | 1.9 | −9.2‡ | −17.9 | −0.4 | −6.6 | −15.0 | 1.8 |
UC+SVAI | −9.3‡ | −18.5 | −0.1 | −9.4‡ | −18.0 | −0.7 | −7.0 | −15.4 | 1.4 |
Sex, male | 11.2‡ | 3.8 | 18.7 | 11.8‡ | 4.6 | 19.1 | |||
Age | −0.1 | −0.4 | 0.3 | −0.0 | −0.4 | 0.3 | |||
Secondary school§ | 2.6 | −9.6 | 14.8 | 1.9 | −9.7 | 13.5 | |||
Higher education <4 years§ | 2.8 | −10.3 | 16.0 | 2.9 | −9.7 | 15.5 | |||
Higher education ≥4 years§ | −11.0 | −26.9 | 4.9 | −10.3 | −25.4 | 4.8 | |||
Meeting or follow-up plan¶ | −7.2 | −15.3 | 0.9 | −5.6 | −13.5 | 2.2 | |||
Meeting and follow-up plan¶ | −5.0 | −14.4 | 4.4 | −3.5 | −12.6 | 5.7 | |||
Physical activity 1–2 days** | 0.7 | −8.8 | 10.1 | 1.1 | −8.1 | 10.3 | |||
Physical activity 3–4 days** | 6.5 | −3.2 | 16.3 | 3.5 | −4.1 | 18.0 | |||
Physical activity 5–7 days** | 8.1 | −3.3 | 19.5 | 7.0 | −4.1 | 18.0 | |||
Work ability†† | −3.5‡ | −5.0 | −2.1 | −3.8‡ | −5.2 | −2.4 | |||
Musculoskeletal health‡‡ | −0.8‡ | −1.3 | −0.3 | −0.7‡ | −1.2 | −0.02 | |||
Sickness absence days previous year§§ | 19.5‡ | 13.1 | 25.8 | 19.1‡ | 12.9 | 25.3 |
n, number of participants in analysis (ITT analysis: UC n=171, UC+MI n=169, UC+SVAI n=169, complete case analysis: UC n=158, UC+MI n=157, UC+SVAI n=159)
*Multiple robust linear regression analyses adjusted for predefined possible confounding factors.
†Values for missing on the independent variables were imputed with multiple imputations by chained equations with 10 imputations. Imputations were not conducted for the five missing outcome values.
‡p<0.05.
§Education: dummy variables compared with compulsory education.
¶Follow-up from employer, dummy variables compared with no follow-up.
**Physical activity 1 week prior to baseline, dummy variables compared with no physical activity.
††Measured with single question from the Work Ability Index (0–10).
‡‡Measured with the Musculoskeletal Health Questionnaire (0–56).
§§Number of days away from work due to sickness absence 12 months prior to baseline, logarithmic transformed variable.
Coef., Coefficient.; ITT, intention-to-treat; MI, motivational interviewing; SVAI, stratified vocational advice intervention; UC, usual case management.