Predictors of concordance between digitally recorded and human listener classified lung auscultation and conventional stethoscope classifications, comparing auscultation of any crackles and/or wheeze and neither crackles nor wheeze
| Characteristic | Total N=737 | Agreement n=530 | OR | Adjusted OR* | ||||
| n | n (%) | Estimate | 95% CI | P value | Estimate | 95% CI | P value | |
| Age | ||||||||
| 1–5 months | 315 | 233 (74.0) | 1.37 | 0.84 to 2.23 | 0.208 | 1.46 | 0.87 to 2.44 | 0.152 |
| 6–11 months | 172 | 117 (68.0) | 1.07 | 0.63 to 1.81 | 0.805 | 1.08 | 0.62 to 1.86 | 0.792 |
| 12–23 months | 148 | 112 (75.7) | 1.75 | 0.98 to 3.11 | 0.058 | 1.75 | 0.97 to 3.15 | 0.062 |
| 24–56 months | 102 | 68 (66.7) | Ref | – | – | Ref | – | – |
| Sex | ||||||||
| Male | 422 | 305 (72.3) | Ref | – | – | Ref | – | – |
| Female | 315 | 225 (71.4) | 0.94 | 0.67 to 1.31 | 0.703 | 0.98 | 0.70 to 1.38 | 0.919 |
| Severity | ||||||||
| Very severe pneumonia | 244 | 177 (72.5) | 1.09 | 0.77 to 1.55 | 0.618 | – | – | – |
| Severe pneumonia | 493 | 353 (71.6) | Ref | – | – | – | – | – |
| Site | ||||||||
| Zambia | 234 | 162 (69.2) | Ref | – | – | Ref | – | – |
| Bangladesh | 145 | 112 (77.2) | 1.49 | 0.92 to 2.40 | 0.104 | 1.49 | 0.91 to 2.44 | 0.113 |
| Kenya | 123 | 81 (65.9) | 0.85 | 0.53 to 1.35 | 0.479 | 0.94 | 0.55 to 1.59 | 0.805 |
| South Africa | 92 | 64 (69.6) | 1.16 | 0.65 to 2.09 | 0.616 | 1.25 | 0.66 to 2.39 | 0.495 |
| The Gambia | 79 | 67 (84.8) | 2.37 | 1.21 to 4.67 | 0.012 | 2.40 | 1.20 to 4.80 | 0.014 |
| Thailand | 64 | 44 (68.8) | 0.96 | 0.53 to 1.75 | 0.905 | 1.08 | 0.57 to 2.04 | 0.821 |
| Duration between digital and conventional auscultation | ||||||||
| <2 hours | 383 | 288 (75.2) | 1.36 | 0.98 to 1.89 | 0.060 | 1.14 | 0.75 to 1.75 | 0.541 |
| >2 hours | 332 | 229 (69.0) | Ref | – | – | Ref | – | – |
| Missing time information | 22 | 13 (59.1) | – | – | – | – | – | – |
| Intermittent crying on recording | ||||||||
| Yes | 330 | 244 (73.9) | Ref | – | – | Ref | – | – |
| No | 407 | 286 (70.3) | 1.18 | 0.85 to 1.64 | 0.337 | 1.19 | 0.84 to 1.69 | 0.333 |
Bold values denote statistical significance at the p <0.05 level.
*Adjusted for age, sex, site, duration between digital and conventional auscultation and intermittent crying on recording.