Table 3

Adjusted incidence rate ratios (IRR) of developing a serious infection in the first or second year

Model AModel BFully adjusted model C
IRR95% CIIRR95% CIIRR95% CIp Value
Age >60 years (yes vs no)1.71.3 to to to 2.40.012
Chronic lung disease (yes vs no)2.61.8 to to to 2.60.014
Chronic renal disease (yes vs no)2.31.5 to to to 2.80.14
High number of treatment failures (>5) at baseline (yes vs no)1.20.8 to to to 2.30.027
History of serious infection at follow-up (yes vs no)2.11.0 to 4.30.038
FFbH (0–100%) per 10% at baseline0.920.87 to 0.98
FFbH (0–100%) per 10% at follow-up0.900.86 to 0.950.900.85 to 0.960.0023
Glucocorticoids 7.5–14 mg/day at baseline (yes vs no)1.00.7 to 1.4
Glucocorticoids 7.5–14 mg/day at follow-up (yes vs no)1.91.4 to to 3.20.0002
Glucocorticoids ≥ 15 mg/day at baseline (yes vs no)1.51.0 to 2.1
Glucocorticoids ≥ 15 mg/day at follow-up (yes vs no)3.62.2 to to 9.4<0.0001
Treatment with TNF inhibitors (yes vs no)1.61.2 to to to 2.70.0027
Trend (IRR TNF year 2/IRR TNF year1)0.690.50 to 0.960.790.57 to 1.10(1.0)(0.5 to 2.0)(0.93)
  • Model A: only baseline characteristics were used for calculation and adjustment.

  • Model B: time-dependent use of glucocorticoids and FFbH were considered in the model. No adjustment for DMARD/anti-TNF treatment adaptations and dropout processes at follow-up.

  • Model C: fully adjusted model. Adjustment for time-varying risk factors as in model B and for treatment adaptations and dropout processes (see Methods section for further details).

  • DMARD, disease-modifying antirheumatic drug; FFbH, Hannover Functional Status Questionnaire measuring functional capacity as percentage of full function; TNF, tumour necrosis factor.