Table A1.
Results (odds ratios) from multinomial logit (Model 1 & 2) and mixed multinomial logit (Model 3 & 4) regression for provider choice.
| International (1978) | Berman (2000) | Perry et al. (2014) | Governmet of India (2018) | |
|---|---|---|---|---|
| Provider attributes | ||||
| Clinical competence (0–100) | 1.00 [0.99,1.00] | 1.00 [0.99,1.02] | 1.00 [0.99,1.00] | 1.01 [0.98,1.02] |
| Distance (Km) | 0.99** [0.98,1.00] | 0.99* [0.98,1.00] | 0.97* [0.93,1.00] | 0.93** [0.91,0.96] |
| Log cost (Rs.) | 0.97 [0.90,1.05] | 0.39** [0.37,0.42] | 1.12** [1.01,1.24] | 0.36** [0.33,0.39] |
| SD (Log cost) | 1.47** [1.40,1.55] | 1.03** [1.00,1.05] | ||
| SD (Distance) | 1.05* [1.00,1.10] | 1.10** [1.06,1.14] | ||
| Residual control function | 2.39** [2.34,2.44] | 2.56** [2.44,2.69] | ||
| Patient attributes | ||||
| Private clinic (Ref: Primary health center) | ||||
| Male (Ref: female) | 0.77** [0.64,0.94] | 0.87** [0.78,0.98] | 0.75** [0.60,0.93] | 0.87** [0.77,0.98] |
| Age (years) | 0.99** [0.99,0.99] | 0.99** [0.99, 1.00] | 0.99** [0.99,0.99] | 0.99** [0.99,1.00] |
| Illness severe (Ref: not severe) | 0.80* [0.63,1.02] | 1.31** [1.13,1.52] | 0.83 [0.63,1.10] | 1.38** [1.17,1.62] |
| Wealth quintile (Ref: Poorest 20%) | ||||
| Wealth quintile 2 | 1.10 [0.84,1.45] | 0.58** [0.49,0.68] | 1.07 [0.78,1.47] | 0.54** [0.45,0.65] |
| Wealth quintile 3 | 1.35 [0.92,1.97] | 0.74** [0.59,0.94] | 1.36 [0.89,2.09] | 0.75** [0.58,0.97] |
| Wealth quintile 4 | 0.86 [0.65,1.16] | 0.36** [0.31,0.43] | 0.80 [0.57,1.12] | 0.37** [0.30,0.44] |
| Richest 20% | 0.92 [0.68,1.24] | 0.32** [0.27,0.39] | 0.84 [0.59,1.19] | 0.34** [0.27,0.42] |
| Private hospital (Ref: Primary Health Center) | ||||
| Male (Ref: female) | 0.73** [0.56,0.95] | 0.73** [0.59,0.90] | 0.73** [0.53,0.99] | 0.74** [0.58,0.95] |
| Age (years) | 0.99** [0.99,1.00] | 0.99** [0.99,0.99] | 0.99* [0.99,1.00] | 0.99** [0.98,0.99] |
| Illness severe (Ref: not severe) | 1.81** [1.34,2.44] | 2.20** [1.75,2.77] | 2.28** [1.55,3.34] | 2.62** [1.98,3.46] |
| Wealth quintile (Ref: Poorest 20%) | ||||
| Wealth quintile 2 | 1.73** [1.14,2.63] | 1.04 [0.73,1.47] | 1.83** [1.10,3.03] | 1.16 [0.76,1.77] |
| Wealth quintile 3 | 2.29** [1.36,3.86] | 1.25 [0.82,1.91] | 2.58** [1.38,4.84] | 1.46 [0.88,2.42] |
| Wealth quintile 4 | 1.63** [1.06,2.51] | 0.66** [0.46,0.94] | 1.69* [0.99,2.91] | 0.81 [0.52,1.25] |
| Richest 20% |
2.32** [1.51,3.57] |
0.82 [0.57,1.18] |
2.54** [1.47,4.38] |
1.06 [0.68,1.64] |
| N (observations) | 19398 | 19398 | 19398 | 19398 |
| N (patients) | 7279 | 7279 | 7279 | 7279 |
| Log likelihood | −3128 | −1696 | −3083 | −1689 |
Note (International, 1978): Figures in parenthesis are 95% CI (Berman, 2000); *p < 0.10, **p < 0.05 (Perry et al., 2014); Model 1 and 3 have endogenous cost and models 2 and 4 use the control function for addressing endogeneity in costs (Governmet of India, 2018); Residual control function is the residual of the cost control function (GOI. Ayushman Bharat Health, 2022); Observations are clustered at the individual level.