Table 3. Factors associated with CXR-confirmed pneumonia among patients with IMCI pneumonia.
| CXR-confirmed pneumonia | No CXR-confirmed pneumonia | Unadjusted OR (CI) | p | Adjusted OR (CI) | p | |
|---|---|---|---|---|---|---|
| (n = 42) | (n = 300) | |||||
| Continuous variables | ||||||
| Mean age, months (SD) / OR (CI) | 20.2 (1.97) | 16.4 (0.63) | 1.03 (1.00–1.05) | 0.041 | 1.03 (1.0–1.1) | 0.051 |
| Mean CRP, mg/L (CI) / OR (CI) | 61.5 (9.9) | 26.0 (1.7) | 1.02 (1.01–1.02) | <0.0001 | 1.02 (1.01–1.03) | <0.0001 |
| Mean WBC, 109/L / OR (CI) | 14.5 (1.1) | 12.3 (0.2) | 1.08 (1.0–1.2) | 0.009 | 1.02 (0.9–1.1) | 0.6 |
| Discrete variables | ||||||
| Child perceived as having fast breathing by care taker | 85% (73–96%) | 75% (70–80%) | 1.8 (0.7–4.5) | 0.2 | 1.9 (0.7–5.2) | 0.19 |
| Temperature ≥39.0°C (CI) | 11% (7–15%) | 14% (4–25%) | 1.3 (0.5–3.5) | 0.53 | 1.2 (0.4–4.0) | 0.31 |
| CRPa <20 mg/L (CI) | 36% (21–50%) | 62% (56–67%) | 0.3 (0.2–0.7) | 0.002 | b | b |
| CRPa >80 mg/L (CI) | 24% (11–37%) | 5% (3–7%) | 5.9 (2.4–14.3) | <0.0001 | b | b |
| WBC >20 x 109/L (CI) | 10% (6–18%) | 6% (3–9%) | 1.6 (0.5–5.1) | 0.4 | b | b |
| Pneumococcal urine antigen positivity (CI) | 59% (43–74%) | 58% (53–64%) | 1.0 (0.5–1.9) | 0.99 | 1.3 (0.6–2.9) | 0.48 |
| Pneumococcal NPH PCR positivity (CI) | 88% (78–98%) | 86% (82–90%) | 1.2 (0.4–3.2) | 0.75 | b | b |
| RSV NPH PCR positivity (CI) | 26% (13–40%) | 35% (30–40%) | 0.7 (0.3–1.4) | 0.26 | 0.7 (0.3–1.6) | 0.36 |
| IfA or IfB NPH PCR positivity (CI) | 14% (4–25%) | 18% (14–23%) | 0.7 (0.3–1.8) | 0.52 | 0.6 (0.2–1.8) | 0.38 |
28 of 387 patients with IMCI pneumonia had no CXR performed. Another 17 patients were excluded from analysis because CXR was of too poor quality (n = 16) or missing (n = 1) CXR.
aCXR-confirmed pneumonia was seen in 8% (15/200) of patients with CRP <20 mg/L (p = 0.002), and in 40% (10/25) of those with CRP >80 mg/L (p<0.0001).
bNot included in final multivariate regression analysis