Table 5.
Results for a 10% increase in the following indicators: |
|||||
---|---|---|---|---|---|
Government health spending per capita | VHI health spending per capita | OOP health spending per capita | OOP health spending (share of total) | Immunization coverage rate | |
Under-5 mortality rate | (−) 7.9 per 1000 | No effect | No effect | No effect | Negative significant effect not robust |
Female mortality rate (adult) | (−) 1.6 per 1000 | No effect | (−) 4.4 per 1000 | (+) 11.6 per 1000 | (−) 8.5 per 1000 |
Male mortality rate (adult) | (−) 1.3 per 1000 | No effect | (−) 2.9 per 1000 | (+) 13.6 per 1000 | (−) 6.8 per 1000 |
The table presents the baseline estimated incremental effect, on each health outcome, for a 10% increase in the corresponding coverage indicator (relative to the observed average in the data). Models were estimated through two-stage least squares (method IV-2SLS). Incremental effects are expressed in deaths per 1000. ‘(+)’ denotes an increase (a positive regression coefficient) and ‘(−)’ denotes a decrease (a negative regression coefficient). ‘No effect’ denotes that no statistically significant effect was found in the baseline model. ‘Significant effect not robust’ denotes that a statistically significant effect was found in the baseline model but not across robustness tests.