Skip to main content

Table 7 Multivariable Cox proportional hazards regression on hospitalization in the subsequent period

From: Change in forced vital capacity and associated subsequent outcomes in patients with newly diagnosed idiopathic pulmonary fibrosis

  Hospitalizationa
  Hazard Ratio (95 % CI) P-value*
Race
 White vs. non-white 0.50 (0.32–0.79) 0.003*
Lung-function decline group
 Marginal decline vs. stable 2.50 (1.06–5.91) 0.033*
 Significant decline vs. stable 3.37 (1.62–7.00) <0.001*
BMI
 25–30 vs. <25 0.59 (0.34–1.03) 0.061
 ≥30 vs. <25 0.95 (0.51–1.74) 0.861
Comorbidities
 Cardiac disorder vs. no cardiac disorder 1.87 (1.06–3.32) 0.032*
 Pulmonary hypertension vs. no pulmonary hypertension 2.09 (1.15–3.83) 0.017*
 Emphysema vs. no emphysema 1.44 (0.62–3.36) 0.399
 Gastroesophageal reflux disease vs. no gastroesophageal reflux disease 1.08 (0.67–1.73) 0.746
Smoking status
 History of smoking vs. no history of smoking 1.02 (0.61–1.70) 0.935
Suspected AEx in the concurrent period
 Yes vs. no 1.86 (1.06–3.26) 0.030*
Use of prednisone and azathioprine in the concurrent period
 Both vs. neither 1.24 (0.59–2.62) 0.575
 Prednisone only vs. neither 1.25 (0.72–2.16) 0.428
 Azathioprine only vs. neither 0.74 (0.09–6.38) 0.780
Symptoms at initial IPF diagnosis
 Dyspnea vs. no dyspnea 1.43 (0.43–4.80) 0.562
 Weight loss vs. no weight loss 1.45 (0.77–2.76) 0.250
Physician's main practice setting
 Academic vs. non-academic 0.89 (0.51–1.55) 0.680
GAP index (per unit increase)b 1.23 (1.04–1.46) 0.018*
  1. aMultivariable Cox proportional hazard regression was used to estimate the hazard ratio, 95 % confidence interval and p-value, accounting for physician clustering using generalized estimating equations
  2. bThe hazard ratio for the GAP index was estimated for every one point increase in GAP index
  3. *P-values less than 0.05 are indicated with an asterisk (*)