Factors affecting frequencies of ED visits and hospitalizations during the followup period (n = 204 patients who had followup 2013–2015). Other nonsignificant covariates not showing in the table include age, sex, ethnicity, education levels, and followup doctors.
| Multivariable Poisson Regression | Relative Risk | 95% CI of Relative Risk | p |
|---|---|---|---|
| No. ED visits 2013–2015 | |||
| Employed vs unemployed, 2011–2012 | 0.586 | 0.382–0.899 | 0.014 |
| SDI ≥ 1, 2011–2012 | 1.576 | 1.073–2.315 | 0.020 |
| No. ED visits, 2011–2012 | |||
| No. hospitalizations, 2013–2015 | |||
| Employed vs unemployed, 2011–2012 | 0.449 | 0.263–0.764 | 0.003 |
| SDI ≥ 1, 2011–2012 | 3.091 | 1.843–5.182 | < 0.001 |
| Prednisone dose, mg/day, 2011–2012 | 1.034 | 1.018–1.049 | < 0.001 |
| No. hospitalizations, 2011–2012 | 1.205 | 1.059–1.371 | 0.005 |
ED: emergency department; SDI: Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index.