Table 5.
Estimated effect of CBHI membership on outpatient and inpatient care utilization among CBHI members
| Variable | Modela | Sample | CBHI members | Non-members | Difference | ||
|---|---|---|---|---|---|---|---|
| Estimate | 95% CI | ||||||
| LL | UL | ||||||
| Number of OPD visit per capita per year | Model 1 | Unmatched | 1.93 | 1.6 | 0.33 | ||
| ATT | 1.95 | 1.59 | 0.36 | 0.25 | 0.44 | ||
| Model 2 | Unmatched | 1.93 | 1.81 | 0.12 | |||
| ATT | 2.03 | 1.86 | 0.17 | -0.04 | 0.19 | ||
| Probability of treatment seeking for an illness in the last one month | Model 1 | Unmatched | 74.3 | 67.0 | 7.3% | ||
| ATT | 73.3 | 65.2 | 8.1 | 7.9 | 12.0 | ||
| Model 2 | Unmatched | 74.3 | 61.0 | 13.3 | |||
| ATT | 73.3 | 65.9 | 7.4 | 6.1 | 10.6 | ||
| % of households with at least one IPD admission in the past 1 year | Model 1 | Unmatched | 0.061 | 0.08 | -0.019 | ||
| ATT | 0.06 | 0.079 | -0.019 | -4.8 | 0.9 | ||
| Model 2 | Unmatched | 0.061 | 0.057 | 0.004 | |||
| ATT | 0.065 | 0.058 | 0.007 | -2.1 | 4.1 | ||
| OOP health payment (direct medical spending) | Model 1 | Unmatched | 1610.2 | 2178.3 | -568.1 | ||
| ATT | 1567.8 | 2179.9 | -612.2 | -1141.5 | -59.4 | ||
| Model 2 | Unmatched | 1610.2 | 2423.9 | -813.7 | |||
| ATT | 1596.9 | 2834.7 | -1237.8 | -1675.0 | -183.5 | ||
| OOP non-medical spending (in birr) | Model 1 | Unmatched | 279.6 | 196.9 | 82.7 | ||
| ATT | 272.6 | 205 | 67.7 | -27.0 | 173.1 | ||
| Model 2 | Unmatched | 279.6 | 140.6 | 139 | |||
| ATT | 261 | 142 | 119 | 45.9 | 230.0 | ||
| Total OOP spending (in birr) | Model 1 | Unmatched | 1889.8 | 2375.2 | -485.4 | ||
| ATT | 1840.4 | 2384.9 | -544.5 | -1111.5 | 56.8 | ||
| Model 2 | Unmatched | 1889.8 | 2564.5 | -674.7 | |||
| ATT | 1857.9 | 2976.7 | -1118.8 | -1067.2 | -340.2 | ||
| % of households with catastrophic health expenditureb | Model 1 | Unmatched | 6.4 | 8.7 | -2.3 | ||
| ATT | 6.4 | 9.5 | -3.1 | -5.98 | -0.38 | ||
| Model 2 | Unmatched | 6.4 | 9.9 | -3.5 | |||
| ATT | 6.3 | 11.8 | -5.5 | -7.48 | -0.85 | ||
| % of households falling below poverty line after OOP health payment | Model 1 | Unmatched | 1.39 | 1.02 | 0.37 | ||
| ATT | 1.4 | 1.2 | 0.2 | -0.74 | 1.18 | ||
| Model 2 | Unmatched | 1.4 | 2.2 | -0.8 | |||
| ATT | 1.3 | 3.1 | -1.8 | -3.42 | 0.14 | ||
| Increase in poverty gap after OOP health payment | Model 1 | Unmatched | 0.51 | 0.48 | 0.03 | ||
| ATT | 0.51 | 0.62 | -0.11 | -0.45 | 0.27 | ||
| Model 2 | Unmatched | 0.51 | 0.80 | -0.29 | |||
| ATT | 0.54 | 0.85 | -0.31 | -0.73 | 0.03 | ||
aComparison groups for model 1 and model 2 were non-members in CBHI and non-CBHI woredas, respectively
bCatastrophic health expenditure: household’s annual direct medical cost > 10% of total expenditure