Table 8.
Sensitivity analysis including missing data
| Number (%) of mothers in intervention villages | Number (%) of mothers in control villages | Crude mean difference (95% CI) | P value | Adjusted mean differencea (95% CI) | P value | |
|---|---|---|---|---|---|---|
| Facility delivery quintile 3 | ||||||
| Excluding missing data | n = 878 | n = 801 | ||||
| Baseline (n = 455) | 190/290 (65.5%) | 99/165 (60.0%) | ||||
| Endline (n = 423) | 149/200 (74.5%) | 153/223 (68.6%) | ||||
| Difference | 9.0% | 8.6% | 0.4% (−0.12–0.13) | 0.954 | −3.0% (− 0.27–0.20) | 0.781 |
| Including missing data | n = 739 | n = 660 | n = 1399 | n = 1256 | ||
| Baseline (n = 621) | 245/377 (65.0%) | 132/224 (58.9%) | ||||
| Endline (n = 798) | 259/362 (71.6%) | 283/436 (64.9%) | ||||
| Difference | 6.6% | 6.0% | 0.6% (−0.98–0.11) | 0.912 | −0.5% (− 0.18–0.17) | 0.951 |
| Excluding missing data | n = 353 | n = 384 | n = 872 | n = 800 | ||
| Baseline (n = 448) | 179/287 (62.4%) | 151/161 (93.8%) | ||||
| Endline (n = 424) | 174/201 (86.6%) | 143/223 (64.1%) | ||||
| Difference | 24.2% | −29.7% | 53.9% (0.43–0.64) | < 0.001 | 58.0% (0.23–0.94) | 0.004 |
| Including missing data | n = 739 | n = 655 | n = 1394 | n = 1256 | ||
| Baseline (n = 595) | 233/376 (62.0%) | 205/219 (93.6%) | ||||
| Endline (n = 799) | 282/363 (77.7%) | 285/436 (65.4%) | ||||
| Difference | 15.7% | −28.2% | 43.9% (0.35–0.53) | < 0.001 | 49.4% (0.21–0.77) | 0.003 |
| PNC quintile 3 | ||||||
| Excluding missing data | n = 482 | n = 383 | n = 865 | n = 791 | ||
| Baseline (n = 444) | 142/282 (50.4%) | 115/162 (71.0%) | ||||
| Endline (n = 421) | 163/200 (81.5%) | 133/221 (60.2%) | ||||
| Difference | 31.1% | −10.8% | 41.9% (0.30–0.54) | < 0.001 | 45.8% (−0.47–096) | 0.071 |
| Including missing data | n = 732 | n = 652 | n = 1384 | n = 1245 | ||
| Baseline (n = 591) | 177/372 (47.6%) | 166/219 (75.8%) | ||||
| Endline (n = 793) | 250/360 (69.4%) | 252/433 (58.2%) | ||||
| Difference | 21.8% | −17.6% | 39.4% (0.29–0.50) | < 0.001 | 49.8% (−0.01–1.00) | 0.054 |
| CHW home visiting quintile 3 | ||||||
| Excluding missing data | n = 489 | n = 377 | n = 866 | n = 790 | ||
| Baseline (n = 456) | 142/291 (48.8%) | 66/165 (40.0%) | ||||
| Endline (n = 410) | 73/198 (36.9%) | 61/212 (28.8%) | ||||
| Difference | −11.9% | −11.2% | −0.7% (−0.14–0.12) | 0.916 | −2.8% (− 0.60–0.55) | 0.917 |
| Including missing data | n = 740 | n = 642 | n = 1382 | n = 1239 | ||
| Baseline (n = 599) | 184/380 (48.4%) | 111/219 (50.7%) | ||||
| Endline (n = 783) | 130/360 (36.1%) | 99/423 (23.4%) | ||||
| Difference | −12.3% | −27.3% | 15.0% (0.04–0.25) | 0.005 | 16.0% (−0.49–0.81) | 0.596 |
aAdjusted for clustering by district, maternal age, parity, maternal education, quintile and access to health facilities