Table 5

Intervention scenarios and corresponding predicted marginal probabilities of selected outcomes during the TSU phase: Bihar, 2014–2017

Intervention scenariosInstitutional deliveryANC3+EIBF
Predicted marginal probabilities at mean values0.735
(0.726 to 0.744)
0.417
(0.408 to 0.426)
0.675
(0.667 to 0.682)
No woman gets FHW visits during pregnancy0.696
(0.685 to 0.706)
0.396
(0.386 to 0.405)
0.630
(0.621 to 0.638)
All women get FHW visits during pregnancy0.772
(0.761 to 0.784)
0.425
(0.413 to 0.438)
0.693
(0.683 to 0.702)
All women get 2+ FHW visits during third pregnancy trimester with low-to-moderate quality of interactions with FHWs, but no outcome-specific advice0.783
(0.768 to 0.797)
0.408
(0.392 to 0.424)
0.693
(0.682 to 0.705)
All women get 2+ FHW visits during last trimester with high quality of interactions with FHWs, but no outcome-specific advice0.798
(0.780 to 0.817)
0.450
(0.429 to 0.472)
0.735
(0.716 to 0.755)
All women get 2+ FHW visits during last trimester with high quality of interactions with FHWs and outcome-specific advice0.822
(0.809 to 0.835)
0.485
(0.469 to 0.501)
0.819
(0.807 to 0.832)
All women get 2+ FHW visits during last trimester with high quality of interactions with FHWs and outcome-specific advice and all deliveries occur in health facilityn/an/a0.849
(0.838 to 0.860)
100% deliveries in a health facilityn/an/a0.733
(0.726 to 0.741)
100% mothers are members of self-help groups0.751
(0.739 to 0.762)
0.423
(0.410 to 0.435)
0.695
(0.684 to 0.707)
100% mothers are literate0.784
(0.774 to 0.794)
0.474
(0.463 to 0.484)
0.673
(0.664 to 0.682)
  • Predicted marginal probabilities are from multivariable regression models fitted using the scale-up phase data from all 38 districts and shown in table 4.

  • ANC, antenatal care; EIBF, early initiation of breast feeding; FHW, frontline health worker; n/a, not available; TSU, Technical Support Unit.