Table 3.
Fixed effects regression results—determinants of UHC service coverage Index.
| Variable | Model 1 | Model 2 | Model 3 | Model 4 |
|---|---|---|---|---|
| Health Financing | ||||
| Government Health Expenditure per capita (log) | 3.42*** (0.85) | 3.15*** (0.82) | 2.87*** (0.79) | 2.64*** (0.76) |
| Out-of-pocket Health Expenditure (% of THE) | −0.24*** (0.06) | −0.22*** (0.06) | −0.20*** (0.05) | −0.18*** (0.05) |
| Social Health Insurance Coverage (% population) | 0.18*** (0.04) | 0.16*** (0.04) | 0.15*** (0.04) | 0.14*** (0.04) |
| Health Workforce | ||||
| Physician Density (per 10,000, log) | 2.85** (1.12) | 2.67** (1.09) | 2.48** (1.07) | 2.31** (1.04) |
| Nurse/Midwife Density (per 10,000, log) | 1.92* (0.98) | 1.84* (0.95) | 1.76* (0.93) | 1.68* (0.91) |
| Governance | ||||
| Government Effectiveness Index | 4.28*** (1.24) | 4.05*** (1.21) | 3.82*** (1.18) | 3.61*** (1.15) |
| Control of Corruption Index | 3.76*** (1.15) | 3.54*** (1.12) | 3.33*** (1.09) | 3.14*** (1.06) |
| Digital Health | ||||
| Digital Health Adoption Index | 0.31*** (0.08) | 0.29*** (0.08) | 0.27*** (0.07) | 0.25*** (0.07) |
| Service Delivery | ||||
| PHC Facility Density (per 100,000, log) | 1.46** (0.58) | 1.38** (0.56) | 1.30** (0.54) | 1.23** (0.52) |
| Essential Medicines Availability (%) | 0.21*** (0.05) | 0.20*** (0.05) | 0.19*** (0.05) | 0.18*** (0.05) |
| Control Variables | ||||
| GDP per capita (log) | - | 2.45*** (0.72) | 2.18*** (0.69) | 1.94*** (0.66) |
| Female Education (mean years) | - | - | 0.87*** (0.26) | 0.79*** (0.25) |
| Urbanization Rate (%) | - | - | - | 0.12** (0.05) |
| Constant | 24.38*** (3.42) | 18.72*** (3.18) | 15.63*** (2.94) | 13.25*** (2.71) |
| Model Statistics | ||||
| R-squared (within) | 0.64 | 0.67 | 0.69 | 0.71 |
| Countries | 135 | 135 | 135 | 135 |
| Observations | 3,240 | 3,240 | 3,240 | 3,240 |
| Country FE | Yes | Yes | Yes | Yes |
| Year FE | Yes | Yes | Yes | Yes |
Source: Author’s calculations using Stata 18.0. Dependent variable, UHC Service Coverage Index (0–100). Robust standard errors clustered at country level in parentheses.
THE, total health expenditure; PHC, primary health care. All continuous variables except percentages are in natural logarithm form. Model 1 includes only health system variables, with controls added sequentially in Models 2–4.
*
p < 0.1.
**
p < 0.05.
***
p < 0.01.