Skip to main content

Table 1 The determinants of institutional delivery in rural India: adjusted odds ratios and predicted probabilities from random effects logistic regression model of NFHS 2 data.

From: Institutional delivery in rural India: the relative importance of accessibility and economic status

  n % Institutional delivery Unadjusted OR & 95% CIs Adjusted OR & 95% CIs Adjusted Predicted Probability (%)
Geographical Access      
Distance to hospital Change in AIC due to removal of distance to hospital from model: +66.3
31 + km 3027 13.4 1 1 9.7
16-30 km 4544 18.5 1.98 (1.47-2.69) 1.40 (1.09-1.77) 13.0
6-15 km 8180 23.8 2.81 (2.13-3.72) 1.79 (1.43-2.24) 16.1
Up to5 km 6160 32.0 5.08 (3.81-6.78) 2.43 (1.93-3.06) 20.7
Economic Status      
Wealth Quartiles Change in AIC due to removal of wealth from model: +361.9
1 (poorest) 5490 9.5 1 1 9.2
2 5429 15.7 1.76 (1.56-1.99) 1.40 (1.24-1.59) 12.4
3 5152 24.9 3.19 (2.81-3.61) 1.95 (1.71-2.22) 16.4
4 (richest) 5840 44.0 8.44 (7.42-9.62) 3.76 (3.26-4.34) 27.6
Indicators of preference for place of delivery      
Maternal Education Change in AIC due to removal of maternal education from model: +376.1
None 13009 13.0 1 1 11.6
Primary 3540 26.8 2.32 (2.08-2.59) 1.60 (1.43-1.79) 17.3
Secondary 4435 44.2 5.60 (5.06-6.20) 2.60 (2.33-2.91) 25.4
Higher 927 67.4 14.21(11.76-17.17) 4.79 (3.90-5.87) 38.6
Region Change in AIC due to removal of region from model: +685.5
North 12983 15.6 1 1 9.3
East 4633 22.2 2.99 (2.47-3.62) 2.80 (2.35-3.34) 22.3
West 1440 34.7 4.89 (3.95-6.04) 3.35 (2.76-4.06) 25.6
South 2855 58.7 14.51 (12.15-17.33) 8.77 (7.45-10.32) 47.3
Birth Order Change in AIC due to removal of birth order from model: +475.8
1 5916 39.4 2.68 (2.46-2.93) 2.41 (2.20-2.64) 26.9
2-3 9267 23.0 1 1 13.2
4-6 5325 11.9 0.53 (0.47-0.59) 0.80 (0.71-0.90) 10.9
7 + 1403 9.3 0.54 (0.44-0.67) 1.00 (0.80-1.24) 13.2
rho (P-value)     0.27 (P < 0.001)  
  1. Note. Significant associations (P < 0.05) are shown in bold
  2. Adjusted odds ratios are calculated by adjustment for all other factors in the model, and including random effects
  3. Akaike's Information Criterion (AIC)