Table 3.
Risk of pneumonia by type of treatment (FP, other ICS, no ICS use) at baseline
| Treatment at baseline | Comparison | |||||
|---|---|---|---|---|---|---|
| FP | Other ICS | No ICS | FP versus other ICS HR* or RR† (95% CI); p-value |
FP versus no ICS HR* or RR† (95% CI); p-value |
Other ICS versus no ICS HR* or RR† (95% CI); p-value |
|
| Use at baseline, n | 1334 | 1334 | 1334 | – | – | – |
| Patients with pneumonia event, n (%)# | 207 (15.5) | 184 (13.8) | 169 (12.7) | HR 1.20 (0.99, 1.47); p = 0.067 |
HR 1.30 (1.06, 1.59); p = 0.012 |
HR 1.08 (0.88, 1.33); p = 0.467 |
| Pneumonia events, n | 286 | 220 | 214 | – | – | – |
| Adjusted rate of pneumonia events (per patient-year), mean (95% CI) | 0.10 (0.08, 0.12) |
0.07 (0.06, 0.09) |
0.07 (0.05, 0.09) |
RR 1.35 (1.05, 1.74); p = 0.020 |
RR 1.37 (1.06, 1.78); p = 0.015 |
RR 1.02 (0.77, 1.34); p = 0.907 |
| Patients with hospitalized pneumonia events, n (%) | 140 (10.5) | 114 (8.5) | 100 (7.5) | HR 1.31 (1.02, 1.68); p = 0.032 |
HR 1.48 (1.14, 1.91); p = 0.003 |
HR 1.13 (0.86, 1.47); p = 0.382 |
| Hospitalized pneumonia events, n | 172 | 133 | 121 | – | – | – |
| Adjusted rate of hospitalized pneumonia events (per patient-year), mean (95% CI) | 0.06 (0.04, 0.08) |
0.04 (0.03, 0.06) |
0.04 (0.03, 0.05) |
RR 1.34 (0.99, 1.81); p = 0.055 |
RR 1.46 (1.07, 1.98); p = 0.016 |
RR 1.09 (0.79, 1.51); p = 0.607 |
| Patients with pneumonia resulting in death, n (%) | 8 (0.60) | 18 (1.35) | 15 (1.12) | HR 0.46 (0.20, 1.05); p = 0.066 |
HR 0.55 (0.23, 1.30); p = 0.174 |
HR 1.20 (0.61, 2.39); p = 0.595 |
Abbreviations: CI confidence interval, FEV1 forced expiratory volume in 1 s, FP fluticasone propionate, GOLD Global Initiative for Chronic Obstructive Lung Disease, HR hazard ratio, ICS inhaled corticosteroid, RR rate ratio. Matched-subgroup population. *Cox regression analysis with ICS use at baseline and matching factors as covariates. †Poisson regression with ICS use at baseline and matching factors (age, FEV1% predicted, GOLD stage, emphysema diagnosis, and courses of antibiotics) as covariates. #32 patients treated with FP at baseline received this treatment for non-pulmonary (mostly nasal) indication only; four of these patients had a pneumonia event during the study. In the “other ICS” group, these numbers were 43 and 11, respectively