Table 5.
Summary of Effects of Insurance Coverage on Socio-Economic Disparities
| Coefficient on Coverage RD (1) | Disparities at Ages 63−64 |
Percent Change in Disparity Due to Change in Coverage at 65 |
||||||
|---|---|---|---|---|---|---|---|---|
| Outcome | R2 (2) | Low-Ed Minority-Hi-Ed Whites (3) | Black White (4) | Hispanic-White (5) | Low-Ed Minority-Hi-Ed Whites (6) | Black White (7) | Hispanic-White (8) | |
| A. Based on Change in Insurance Coverage at 65 | ||||||||
| 1. Delay in Care Last Year | −0.19 (0.06) |
0.72 | 7.6 | -- | -- | -42.8 | -- | -- |
| 2. No Care Last Year | −0.12 (0.06) |
0.39 | 8.0 | -- | -- | −25.7 | -- | -- |
| 3. Regular Dr. Visit Last Year | 0.32 (0.09) |
0.77 | -7.4 | -- | -- | -73.9 | -- | -- |
| 4. Hospital Stay Last Year | −0.09 (0.08) |
0.26 | 4.7 | -- | -- | −32.7 | -- | -- |
| 5. Total Hospital Admissions | 0.06 (0.18) |
0.74 | -- | 724 | −193 | -- | 1.8 | −0.5 |
| 6. Diagnostic Procedures of the Heart | 0.59 (0.31) |
0.62 | -- | 9 | −25 | -- | 35.6 | 4.5 |
| 7. Bypass Anastomosis of Heart | −0.54 (0.93) |
0.54 | -- | −18 | −14 | -- | −4.9 | −4.8 |
| 8. Joint Replacement Lower of Extremity | −0.19 (0.64) |
0.89 | -- | −7 | −28 | -- | 2.7 | −2.3 |
| B. Based on Change in Incidence of Multiple Coverage at 65 | ||||||||
| 9. Total Hospital Admissions | 0.03 (0.08) |
0.74 | -- | 724 | −193 | -- | −0.1 | 5.8 |
| 10. Diagnostic Procedures of the Heart | −0.21 (0.14) |
0.55 | -- | 9 | −25 | -- | 22.3 | −18.7 |
| 11. Bypass Anastomosis of Heart | 0.46 (0.34) |
0.64 | -- | −18 | −14 | -- | 29.4 | 37.1 |
| 12. Joint Replacement Lower of Extremity | 0.42 (0.21) |
0.94 | -- | −7 | −28 | -- | 59.4 | 25.7 |
Notes: Each entry in Panel A, column 1 is estimated coefficient from regression of RDs in listed health outcome on RDs in insurance coverage over 6 ethnicity/education groups (rows 1−4) or 9 state-ethnicity groups (rows 5−8). All regressions weighted by the inverse sampling variance of the estimated discontinuity in each outcome, and regressions in rows 5−8 include state dummies. Entries in column 2 are corresponding R-squared coefficients from each regression. Entries in columns 3, 4, and 5 are the observed disparities in each health outcome at ages 63−64, and entries in columns 6, 7, and 8 are the percent change in the disparity attributable to the change in insurance coverage based on the coefficient in column 1. Health disparities measured in the NHIS are characterized in terms low-ed minorities versus hi-ed whites, whereas health disparities measured in the hospital discharge data are characterized in terms of black-white or hispanic-white differences. Panel B is similar to Panel A except that the RDs in each health outcome are regressed on the RDs in the incidence of multiple coverage at 65. Panel B regressions are based on data for New York and Florida only (i.e., 6 state-ethnicity groups).