Table 2.
Full Instrumental Variables Regression Results
| Hip Fracture Patients | Stroke Patients | |||||
|---|---|---|---|---|---|---|
| Costs | Length of Stay | Dead or Institutionalized (180 days) | Costs | Length of Stay | Dead or Institutionalized (180 days) | |
| Coefficient (SE) | Coefficient (SE) | Marginal Effect (SE) | Coefficient (SE) | Coefficient (SE) | Marginal Effect (SE) | |
| Competition | 0.283** | 0.509** | 0.921** | −0.439** | −0.253** | 0.599** |
| (0.099) | (0.084) | (0.221) | (0.069) | (0.059) | (0.117) | |
| Patient characteristics | ||||||
| Age | 0.007** | 0.008** | 0.022** | 0.002** | 0.002** | 0.021** |
| (0.000) | (0.000) | (0.001) | (0.000) | (0.000) | (0.001) | |
| Female | −0.006 | 0.005 | −0.163** | 0.009 | 0.018** | 0.023* |
| (0.007) | (0.006) | (0.018) | (0.006) | (0.006) | (0.013) | |
| Hispanic | 0.041 | 0.014 | 0.045 | −0.015 | −0.037 | 0.095 |
| (0.030) | (0.027) | (0.079) | (0.029) | (0.027) | (0.060) | |
| Black | 0.092** | 0.077** | 0.408** | 0.005 | 0.013 | 0.234** |
| (0.025) | (0.023) | (0.068) | (0.018) | (0.018) | (0.042) | |
| White | 0.048** | 0.040** | 0.286** | −0.021 | −0.015 | 0.224** |
| (0.020) | (0.018) | (0.057) | (0.016) | (0.016) | (0.038) | |
| Medicaid beneficiary | 0.028** | 0.029** | 0.411** | −0.008 | −0.003 | 0.375** |
| (0.009) | (0.008) | (0.023) | (0.009) | (0.008) | (0.018) | |
| Condition-specific factors | ||||||
| Pertrof fracture | 0.135** | 0.128** | 0.050** | – | – | – |
| (0.008) | (0.007) | (0.021) | ||||
| Total hip replacement | 0.041** | 0.035** | −0.027 | – | – | – |
| (0.014) | (0.013) | (0.042) | ||||
| Partial hip replacement | 0.054** | 0.034** | −0.021 | – | – | – |
| (0.008) | (0.007) | (0.022) | ||||
| Revision hip replacement | −0.112** | −0.105* | −0.270 | – | – | – |
| (0.055) | (0.058) | (0.208) | ||||
| Basilar artery infarct | – | – | – | 0.115** | 0.125** | −0.165* |
| (0.039) | (0.038) | (0.095) | ||||
| Hemorrhagic stroke | – | – | – | 0.078** | 0.068** | 0.003 |
| (0.011) | (0.010) | (0.022) | ||||
| Carotid, vertebral, or multiple artery | – | – | – | 0.059** | 0.044** | 0.101** |
| (0.013) | (0.013) | (0.027) | ||||
| Facility characteristics | ||||||
| Freestanding IRF | −0.032** | 0.117** | 0.020 | 0.020** | 0.140** | 0.028* |
| (0.010) | (0.009) | (0.020) | (0.009) | (0.008) | (0.017) | |
| Rural IRF | 0.131** | 0.027 | −0.057 | 0.130** | 0.013 | 0.025 |
| (0.049) | (0.047) | (0.104) | (0.035) | (0.032) | (0.066) | |
| Percentage low-income patients | 0.481** | 0.379** | 0.533** | 0.091** | 0.062* | 0.311** |
| (0.041) | (0.038) | (0.104) | (0.039) | (0.035) | (0.077) | |
| Volume of Medicare patients | 0.013 | 0.361** | 0.440** | −0.413** | −0.079** | 0.211** |
| (0.037) | (0.028) | (0.071) | (0.030) | (0.026) | (0.053) | |
| Teaching facility | 0.008 | −0.016* | −0.004 | 0.019 | −0.018* | −0.005 |
| (0.014) | (0.010) | (0.025) | (0.012) | (0.010) | (0.020) | |
| Wage index | 0.500** | −0.024 | −0.208** | 0.812** | 0.173** | −0.071 |
| (0.042) | (0.032) | (0.080) | (0.036) | (0.030) | (0.062) | |
| Nonprofit IRF | −0.040** | −0.118** | −0.070** | 0.020** | −0.094** | −0.060** |
| (0.011) | (0.008) | (0.020) | (0.009) | (0.008) | (0.018) | |
| Government IRF | 0.047** | −0.035** | 0.015 | 0.041** | −0.063** | −0.017 |
| (0.018) | (0.015) | (0.038) | (0.017) | (0.015) | (0.032) | |
| Medium IRF | −0.091** | 0.114** | −0.035 | −0.078** | 0.109** | −0.017 |
| (0.015) | (0.012) | (0.027) | (0.014) | (0.012) | (0.023) | |
| Large IRF | −0.105** | 0.185** | −0.055* | −0.071** | 0.185** | 0.004 |
| (0.016) | (0.012) | (0.029) | (0.015) | (0.013) | (0.025) | |
Notes. Complications, comorbidities, zip code characteristics, census regions, and a quarterly time trend were included in the model but are not reported in the table.
IRF, inpatient rehabilitation facility.