Table 2.
Association between ownership status and home health care discharge: All patients
| Dependent variable | Absolute percentage point change for patients in for-profit agencies vs. nonprofit agencies |
Absolute percentage point change after the IPS for patients in for- profit agencies vs. nonprofit agencies |
N |
|---|---|---|---|
| Length of use < 30 days (vs. ≥30 days) | −6.5** (0.006) |
−8.1* (0.037) |
14,088 |
| Length of use > 99 days (vs. ≤99 days) | 3.4 (0.064) |
−4.3 (0.234) |
14,088 |
| Discharged: death (vs. goals met) | −0.2 (0.677) |
−1.5 (0.581) |
9,487 |
| Discharged: institution (vs. goals met) | 3.4 (0.057) |
−5.0 (0.098) |
11,137 |
| Discharged: other (vs. goals met) | 2.3 (0.142) |
−3.6 (0.254) |
10,496 |
Notes: Estimates in Column 1 are the main effects from a multivariate probit model and estimates from column 2 are the interaction terms (for-profit * Interim Payment System) from a linear probability model. The regressions adjust for all the variables in Table 1. Marginal probability effects are presented and can be interpreted as the percentage point change in the dependent variable. P-values are presented in parentheses. The discharge status models include dummy variables controlling for length of use.
**
p<.01
*
p< 0.05