Table 3.
Predictors of work outcomes from univariable models
| Left work, OR (95% CI) |
Sick-leave, OR (95% CI) |
No presenteeism, OR (95% CI) |
Presenteeism Score, IRR (95% CI) |
|||||
|---|---|---|---|---|---|---|---|---|
| Baseline predictor | MTX-starters (N = 463) | Biologic-starters (N = 260) | MTX-starters (N = 404) | Biologic-starters (N = 217) | MTX-starters (N = 325) | Biologic-starters (N = 207) | MTX-starters (N = 325) | Biologic-starters (N = 207) |
| Age | 1.01 (0.97, 1.04) | 1.05 (1.00, 1.11) | 0.98 (0.96, 1.00) | 1.00 (0.97, 1.04) | 1.01 (0.98, 1.05) | 1.01 (0.99, 1.04) | 1.00 (0.99, 1.01) | 1.00 (0.99, 1.01) |
| Women vs men | 1.18 (0.63, 2.19) | 0.76 (0.32, 1.77) | 1.57 (0.95, 2.61) | 0.94 (0.43, 2.05) | 1.00 (0.56, 1.79) | 0.97 (0.52, 1.80) | 1.43 (1.16, 1.77) | 1.10 (0.85, 1.41) |
| Smoking vs never | ||||||||
| Former | 1.10 (0.51, 2.35) | 0.94 (0.41, 2.16) | 0.84 (0.51, 1.37) | 1.32 (0.64, 2.69) | 0.76 (0.42, 1.35) | 1.45 (0.80, 2.64) | 0.93 (0.74, 1.21) | 1.15 (0.92, 1.42) |
| Current | 3.78 (1.88, 7.60) | 1.25 (0.45, 3.49) | 1.09 (0.59, 1.99) | 1.58 (0.64, 3.91) | 0.48 (0.20, 1.15) | 1.19 (0.53, 2.68) | 0.94 (0.74, 1.21) | 1.21 (0.91, 1.63) |
| SJC28 | 1.02 (0.97, 1.07) | 1.02 (0.94, 1.10) | 1.01 (0.97, 1.05) | 1.00 (0.94, 1.07) | 1.00 (0.95, 1.06) | 1.09 (1.03, 1.16) | 1.01 (0.99, 1.03) | 1.01 (0.99, 1.03) |
| HAQ | 1.63 (1.08, 2.46) | 1.83 (0.96, 3.49) | 2.08 (1.50, 2.89) | 2.86 (1.63, 5.05) | 0.43 (0.28, 0.68) | 0.47 (0.29, 0.76) | 1.37 (1.22, 1.54) | 1.55 (1.30, 1.86) |
| Pain-VAS | ||||||||
| Natural scale | 1.01 (0.995, 1.02) | 0.99 (0.98, 1.01) | 1.02 (1.01, 1.03) | 1.01 (0.998, 1.03) | 0.98 (0.97, 1.00) | 0.99 (0.98, 1.01) | 1.01 (1.00, 1.01) | 1.01 (1.00, 1.01) |
| Standardized scale | 1.18 (0.88, 1.57) | 0.83 (0.58, 1.18) | 1.53 (1.22, 1.93) | 1.38 (0.97, 1.95) | 0.65 (0.47, 0.90) | 0.88 (0.66, 1.17) | 1.23 (1.11, 1.36) | 1.16 (1.03, 1.30) |
| Fatigue-VAS | ||||||||
| Natural scale | 1.01 (0.997, 1.02) | 1.00 (0.98, 1.01) | 1.02 (1.01, 1.03) | 1.02 (1.00, 1.03) | 0.97 (0.96, 0.98) | 0.99 (0.98, 1.00) | 1.01 (1.01, 1.01) | 1.01 (1.00, 1.02) |
| Standardized scale | 1.22 (0.91, 1.64) | 0.90 (0.63, 1.30) | 1.80 (1.41, 2.31) | 1.50 (1.03, 2.19) | 0.46 (0.34, 0.63) | 0.83 (0.63, 1.10) | 1.31 (1.19, 1.45) | 1.26 (1.10, 1.45) |
| HADS Depression | 1.11 (1.03, 1.19) | 1.05 (0.95, 1.15) | 1.13 (1.07, 1.19) | 1.14 (1.05, 1.23) | 0.83 (0.75, 0.92) | 0.89 (0.82, 0.96) | 1.06 (1.03, 1.08) | 1.04 (1.02, 1.06) |
| HADS Anxiety | 1.12 (1.05, 1.19) | 1.04 (0.95, 1.14) | 1.12 (1.06, 1.18) | 1.09 (1.01, 1.18) | 0.85 (0.78, 0.93) | 0.88 (0.82, 0.95) | 1.04 (1.02, 1.06) | 1.03 (1.01, 1.05) |
| EQ5D (standardized) | 0.74 (0.58, 0.96) | 0.82 (0.57, 1.17) | 0.61 (0.50, 0.75) | 0.61 (0.45, 0.83) | 1.94 (1.24, 3.02) | 1.46 (1.08, 1.98) | 0.80 (0.74, 0.87) | 0.88 (0.80, 0.96) |
| NS-SEC | ||||||||
| Class 2 vs class 1 | 1.22 (0.55, 2.71) | 0.79 (0.30, 2.06) | 0.68 (0.38, 1.22) | 1.89 (0.84, 4.29) | 1.32 (0.70, 2.51) | 0.94 (0.44, 2.02) | 0.79 (0.62, 0.99) | 1.31 (1.03, 1.67) |
| Class 3 vs class 1 | 2.22 (1.05, 4.70) | 0.62 (0.24, 1.64) | 1.32 (0.78, 2.23) | 1.89 (0.84, 4.28) | 0.93 (0.48, 1.80) | 1.45 (0.72, 2.90) | 0.99 (0.80, 1.21) | 1.31 (1.02, 1.69) |
| Comorbidity | ||||||||
| 1 comorbidity vs 0 | 0.81 (0.41, 1.60) | 1.45 (0.61, 3.42) | 1.19 (0.73, 1.94) | 1.25 (0.60, 2.59) | 0.79 (0.44, 1.42) | 0.72 (0.38, 1.35) | 1.11 (0.91, 1.36) | 1.03 (0.81, 1.31) |
| ≥2 comorbidities vs 0 | 1.62 (0.79, 3.35) | 1.81 (0.64, 5.16) | 1.33 (0.71, 2.49) | 1.51 (0.58, 3.94) | 0.66 (0.27, 1.62) | 0.37 (0.16, 0.86) | 1.17 (0.91, 1.50) | 1.18 (0.88, 1.56) |
The ORs and IRRs are interpreted as the change in odds or the relative change in the outcome per unit increase in the predictor. For the natural scale of the Pain- and Fatigue-VAS, a 1 unit change is very small. Hence, the results are also given on a standardized scale. Here, a 1 unit increase in the scale corresponds to 1 s.d. change in the visual analogue scale score (approx. 20 unit change in this case). Interpretation of ZINB regression output used to model presenteeism scores: ZINB models data with a high number of zeros (i.e. presenteeism) by splitting the data into two portions, the excess zero scores and the count scores. Then the model predicts whether or not a patient had an excess zero for presenteeism or not using a logistic regression model, and gives OR for no presenteeism (with a score <1 indicating that, as the variable increases, patients are less likely to have no presenteeism). After this, the model predicts the count scores, using a negative binomial regression model and gives IRRs (with a score >1 indicating that as the predictor increases, presenteeism score increases; for example, an IRR of 1.55 for HAQ scores of biologic-starters indicates that, for each unit increase in HAQ score, the average presenteeism score increases by 55%). HADS: Hospital Anxiety and Depression Scale; IRR: incidence rate ratio; NS-SEC: The National Statistics Socio-Economic Classification (see Table 2 for definition of classes); OR: odds ratio; SJC: swollen joint count; VAS: visual analogue scale; Natural scale: the unadjusted scores from the visual analogue scales ranging from 0–100; ZINB: zero-inflated negative binomial.