Table 3. Multivariable negative binomial regression estimating community-setting pneumonia-associated hospitalization rate ratios—New York State, 2010–2014a.
| Characteristic | Betab | (SE) | RR | (95% CI) |
|---|---|---|---|---|
| Intercept | -7.807 | (0.057) | -- | -- |
| Sexc | ||||
| Male | 0.209 | (0.029) | 1.2 | (1.2, 1.3) |
| Age (years)d | ||||
| 45–64 | 1.434 | (0.044) | 4.2 | (3.8, 4.6) |
| 65–79 | 2.584 | (0.044) | 13.3 | (12.2, 14.4) |
| ≥80 | 3.632 | (0.044) | 37.8 | (34.7, 41.2) |
| Area-based povertye | ||||
| Medium (10%–19% FPL) | 0.379 | (0.039) | 1.5 | (1.4, 1.6) |
| High (20%–29% FPL) | 0.599 | (0.041) | 1.8 | (1.7, 2.0) |
| Very high (≥30% FPL) | 0.747 | (0.045) | 2.1 | (1.9, 2.3) |
| Urbanization groupf,g | ||||
| NYC | 0.252 | (0.047) | 1.3 | (1.2, 1.4) |
| Non-NYC urban areas | 0.356 | (0.047) | 1.4 | (1.3, 1.6) |
| Suburban areas | 0.200 | (0.050) | 1.2 | (1.1, 1.3) |
Abbreviations: CI, confidence interval; FPL, federal poverty level; NYC, New York City; RR, rate ratio; SE, standard error.
aMcFadden’s pseudo R2 was 0.32.
bThe p-values are all <0.01.
cThe reference group is female.
dThe reference group is age 20–44.
eThe reference group is low area-based poverty (<10% FPL).
fThe reference group is rural areas.
gUrbanization groups across New York State were classified using Rural-Urban Continuum Codes (RUCC) based on county of patient residence in NYC (RUCC 1), non-NYC urban areas (RUCC 1–3 other than NYC), suburban areas (RUCC 4–5), or rural areas (RUCC 6–8).