Table 2.
Characteristics of Newborns Observed in Comparison (Paper form: pCNCP) and Intervention Group (Mobile App: mCNCP)
| Expert classification of pCNCP cohort | Expert classification of mCNCP cohort | P-Value† | CHW classification of pCNCP cohort | CHW classification of mCNCP cohort | P-Value† | |
|---|---|---|---|---|---|---|
| Female | 26/84 (31.0%) | 29/86 (33.7%) | 0.70 | 26/82‡ (31.7%) | 32/86 (37.2%) | 0.45 |
| Infant post-natal age in days (mean) | 6.5 (6.7 SD) | 6.5 (6.7 SD) | 0.98 | 6.6 (7.2 SD) | 5.6 (6.2 SD) | 0.90 |
| Gestational age at birth in weeks (mean) | 40.1 (3.2 SD) | 40.0 (2.6 SD) | 0.89 | 39.6 (2.2 SD) | 39.4 (2.2 SD) | 0.86 |
| Low birth-weighta* | 41/84 (48.8%) | 45/86 (52.3%) | 0.65 | 0/84 (0%) | 45/86 (52.3%) | < 0.001 |
| Smallb* | 41/84 (48.8%) | 45/86 (52.3%) | 0.65 | 9/84 (10.7%) | 46/86 (53.5%) | < 0.001 |
| Birth-weight or Current Weight < 1500 g* | 6/84 (7.1%) | 8/86 (9.3%) | 0.61 | 18/84 (21.4%) | 7/86 (8.1%) | 0.014 |
| Severe Weight Lossc* | 21/84 (25.0%) | 21/86 (24.4%) | 0.93 | 14/84 (16.7%) | 20/86 (23.3%) | 0.28 |
| Moderate Weight Lossd* | 18/84 (21.4%) | 18/86 (20.9%) | 0.94 | 22/84 (26.2%) | 17/86 (19.8%) | 0.32 |
| Poor Movement: Unconscious or moves ONLY when Stimulated | 3/84 (3.6%) | 3/86 (3.5%) | 0.98 | 34/84 (40.5%) | 4/86 (4.7%) | < 0.001 |
| History of Convulsions | 19/84 (22.6%) | 19/86 (22.1%) | 0.93 | 23/84 (27.4%) | 25/86 (29.1%) | 0.81 |
| Poor Feeding: Unable to feed OR Stopped feeding well | 52/84 (61.9%) | 52/86 (60.5%) | 0.85 | 35/84 (41.7%) | 47/86 (54.7%) | 0.09 |
| Severe Chest In-drawing | 3/84 (3.6%) | 5/86 (5.8%) | 0.49 | 9/84 (10.7%) | 17/86 (19.8%) | 0.10 |
| Fast Breathinge* | 18/84 (21.4%) | 15/86 (17.4%) | 0.51 | 9/84 (10.7%) | 2/86 (2.3%) | 0.026 |
| Umbilicus: red or pus | 13/84 (15.5%) | 13/86 (15.1%) | 0.95 | 4/84 (4.8%) | 18/86 (20.9%) | 0.002 |
| Skin Pustules | 0/84 (0%) | 0/86 (0%) | – | 2/84 (2.4%) | 6/86 (7.0%) | 0.16 |
| Jaundice on Soles or Body | 9/84 (10.7%) | 10/86 (11.6%) | 0.85 | 18/84 (21.4%) | 23/86 (26.7%) | 0.42 |
| Eye infection | 0/84 (0%) | 0/86 (0%) | – | 2/84 (2.4%) | 5/86 (5.8%) | 0.26 |
| Feverf* | 13/84 (15.5%) | 13/86 (15.1%) | 0.95 | 19/84 (22.6%) | 10/86 (11.6%) | 0.06 |
| Hypothermia g* | 5/84 (6.0%) | 5/86 (5.8%) | 0.97 | 23/84 (27.4%) | 5/86 (5.8%) | < 0.001 |
| Feeding Intoleranceh | 7/84 (8.3%) | 6/86 (7.0%) | 0.74 | 22/84 (26.2%) | 12/86 (14.0%) | 0.046 |
| Problems with breastfeedingi | 32/84 (38.1%) | 32/86 (37.2%) | 0.91 | 62/84 (73.8%) | 32/86 (37.2%) | < 0.001 |
aBirth-weight < 2500 g; bBirth-weight < 2500 g or foot- length < 74 mm; cWeight loss > 10% for small infant or > 15% for non-small infant; dWeight loss is 8–10% for small infant or 10–15% for non-small infant; eRespiratory rate > 60 breaths per min; fTemperature > 38 °C or 100.4 °F; gTemperature < 35.5 °C or 95.9 °F; hInfant has at least one condition: chokes, turns blue or pale when feeding, vomits frequently, has distended or tender abdomen, or bloody stools.; iInfant has problems with at least one: waking easily for feeds, breastfeeding for 10+ minutes for per side, sleeping comfortably between feeds, having 5+ wet diapers per day, mother’s breasts haven’t softened.; N: infant participants.; *For mCNCP, these danger signs were identified by the built-in algorithms based on guideline thresholds and CHW assessment responses.; †P-values are from two-sided t-tests for differences in means for infant post-natal age and gestational age at birth, and from chi-squared tests of proportions for all other variables. ‡2 CHWs using pCNCP did not record the infants’ sex.; Of the 170 assessments performed on 82 unique infants, 84 assessments were conducted by CHWs using pCNCP and 86 assessments were conducted by CHWs using mCNCP. Expert classifications were gold standard assessors’ determinations of danger signs (present or not present) using mCNCP. This table shows that the baseline characteristics of infants in the study are similar when comparing proportions of newborn danger signs in the intervention (mCNCP) and comparison (pCNCP) cohorts as determined by the gold standard experts (physicians or paramedics). When comparing the classification of infants in the assessment method cohorts as determined by the CHWs, however, the CHWs tended to over- or under-estimate conditions, and the baseline characteristics thus appear different across the intervention and comparison groups