Table 6.
Multivariable analysis for the Press Ganey Total Score Forward Selection
| Factor* | Odds ratio (OR) | Coefficient | Coefficient standard error | p value | |
|---|---|---|---|---|---|
| OR | 95% confidence interval | ||||
| Age† | 1.013 | (1.011 · 1.014) | 0.012 | 0.001 | < 0.001 |
| Area deprivation Indexc | 0.999 | (0.998 · 1.000) | − 0.001 | < 0.001 | 0.008 |
| Insurance | – | – | – | – | – |
| Commercial | Reference category | – | – | – | – |
| Medicaid | 1.324 | (1.142 · 1.536) | 0.281 | 0.075 | < 0.001 |
| Medicare | 1.078 | (1.022 · 1.137) | 0.075 | 0.027 | 0.006 |
| Other Government Insurance | 1.137 | (1.077 · 1.201) | 0.129 | 0.028 | < 0.001 |
| Workers' Compensation | 0.786 | (0.514 · 1.201) | − 0.241 | 0.216 | 0.265 |
| Race | – | – | – | – | 0.001‡ |
| White | Reference category | – | – | – | – |
| Other | 1.106 | (1.025 · 1.193) | 0.100 | 0.039 | 0.009 |
| Asian | 0.765 | (0.672 · 0.871) | − 0.267 | 0.066 | < 0.001 |
| Black or African American | 1.012 | (0.813 · 1.260) | 0.012 | 0.112 | 0.915 |
| American Indian and Alaska Native | 0.803 | (0.603 · 1.096) | − 0.207 | 0.153 | 0.174 |
| Native Hawaiian and Other Pacific Islander | 1.612 | (1.249 · 2.081) | 0.477 | 0.130 | < 0.001 |
| Provider type | – | – | – | – | < 0.001§ |
| Physician | Reference category | – | – | – | – |
| Nurse | 1.048 | (0.964 · 1.140) | 0.047 | 0.043 | 0.271 |
| Other | 1.069 | (0.988 · 1.156) | 0.067 | 0.040 | 0.095 |
| Physician Assistant | 1.169 | (1.098 · 1.244) | 0.156 | 0.031 | < 0.001 |
| Wait time|| | 0.937 | (0.935 · 0.939) | − 0.066 | 0.001 | < 0.001 |
| Specialty type | – | – | – | – | – |
| Internal medicine | Reference category | – | – | – | – |
| Medicine other | 0.932 | (0.885 · 0.981) | − 0.070 | 0.026 | 0.008 |
| Surgical | 1.002 | (0.950 · 1.057) | 0.002 | 0.027 | 0.934 |
Bolded p-values denote statistial significance
*Sex was included as a predictor in the model, but was insignificant based on forward step-wise term selection threshold of α = 0.10
†Per 5 years of additional age
‡Per additional 10 percentile points
§p values for the overall univariate binary logistic regression model. Subsequent p values listed are for pairwise comparisons
||Per additional 5 min
cp-values for the variable category in the multivariable binary logistic regression model. Subsequent p-values listed are for individual comparisons