Skip to main content
. 2013 Jun 18;8(6):e66127. doi: 10.1371/journal.pone.0066127

Table 2. Variables associated with active role preference on the eCPS among 87 people with MS in univariate and multivariate models. Age was categorized into two classes, with median as cutoff.

Characteristic Sub-characteristic No at risk, Event (%) Crude OR (95% CI) p value Adjusted OR (95% CI)* p value
Age (years) <38 44, 16 (36.4) 1 0.54 1 0.87
≥38 43, 13 (30.2) 0.76 (0.31–1.86) 0.91 (0.31–2.72)
Sex Men 30, 11 (36.7) 1 0.63 1 0.35
Women 57, 18 (31.6) 0.80 (0.31–2.02) 0.59 (0.20–1.78)
Education Primary 22, 6 (27.3) 1 0.48 1 0.05
Secondary or higher 65, 23 (35.4) 1.46 (0.50–4.24) 3.75 (1.00–14.05)
Followed at the MS center ≤5 years 49, 21 (42.9) 1 0.03 1 0.39
>5 years 38, 8 (21.1) 0.35 (0.13–0.93) 0.61 (0.19–1.88)
Country Italy 50, 8 (16.0) 1 <0.001 1 <0.001
Germany 37, 21 (56.8) 6.89 (2.54–18.68) 10.30 (3.10–34.15)

OR is odds ratio, and 95% CI the OR confidence interval, estimated by unconditional logistic-regression.

Multivariate model including all explanatory variables, with Hosmer-Lemshow goodness-of-fit test χ2 = 7.7 (degrees of freedom = 22), p = 0.47.