Table 3.
Decrease in likelihood of the 14 predictors of the duration of SA spells due to M54
| Predictors | Loss of log-likelihood for the variable | P value | 100×pseudo-R2 |
| Employment status | 1037 | <0.001 | 0.46 |
| Gross SA days in the 365 previous days before start of the SA spell | 542 | <0.001 | 0.24 |
| Age | 375 | <0.001 | 0.17 |
| SA extent at the start of the SA spell | 350 | <0.001 | 0.15 |
| Multimorbidity * | 108 | <0.001 | 0.05 |
| Geographical region | 96 | <0.001 | 0.04 |
| The SA spell was initialised at primary healthcare | 74 | <0.001 | 0.03 |
| Number of specialised outpatient care visits in the 365 previous days | 68 | <0.001 | 0.03 |
| Country of birth | 65 | <0.001 | 0.03 |
| Partial disability pension at start of SA spell | 49 | <0.001 | 0.02 |
| Family status | 47 | 0.004 | 0.02 |
| Educational level | 37 | 0.001 | 0.02 |
| Sex | 32 | <0.001 | 0.01 |
| Number of days spent in inpatient care in the 365 previous days | 13 | 0.961 | 0.01 |
*Defined as at least three different types of medication (different ATC codes) at least once.
ATC, Anatomical Therapeutic Chemical; SA, sickness absence.