Table 2.
Effects of user fee removal.
| Policy change in rural districts |
Policy change in peri-urban areas |
|||
|---|---|---|---|---|
| Coefficient (95%CI) | p-value | Coefficient (95%CI) | p-value | |
| Delivered in afacility(institutional delivery) | ||||
| Policy effect | 0.15 (0.11–0.19) | <0.0001 | 0.10 (0.04–0.17) | 0.001 |
| Mean pre-reform in ‘treated’ group | 0.40 | 0.38 | ||
| N | 12927 | 5126 | ||
| R2 | 0.19 | 0.24 | ||
| Deliverywas assisted by qualified health professional(assisted delivery) | ||||
| Policy effect | 0.12 (0.07–0.16) | <0.0001 | 0.08 (0.02–0.14) | 0.012 |
| Mean pre-reform in ‘treated’ group | 0.39 | 0.34 | ||
| N | 12910 | 5119 | ||
| R2 | 0.19 | 0.25 | ||
| Deliveredby Caesarean section | ||||
| Policy effect | 0.01 (−0.02 to 0.04) | 0.457 | 0.01 (−0.02 to 0.04) | 0.570 |
| Mean pre-reform in ‘treated’ group | 0.01 | 0.02 | ||
| N | 12948 | 5132 | ||
| R2 | 0.03 | 0.02 | ||
| Neonataldeaths | ||||
| Policy effect | 0.00 (−0.01 to 0.02) | 0.570 | −0.00 (−0.02 to 0.02) | 0.821 |
| Mean pre-reform in ‘treated’ group | 0.03 | 0.03 | ||
| N | 12,980 | 5144 | ||
| R2 | 0.01 | 0.01 | ||
Notes: Each coefficient comes from an OLS regression that includes year and district fixed effects. Standard errors are clustered at mother level, sampling weights included. The first panel looks at the probability that a woman delivered in a facility. The second panel looks at the probability that the woman delivered assisted by a qualified healthcare professional. The third panel looks at the probability that the delivery was done by C-section. The last panel looks at the probability that the baby died within four weeks of birth (neonatal death).