Table 3.
Independent Effect of Patient, Hospitalization, and Follow-up Factors on 30-Day Death or Nonelective Readmission
| Adjusted Hazard Ratio (95% CI) | |
|---|---|
| Patient Factors | |
| Age increased by decade | 1.16 (1.16 to 1.17) |
| Patient is male | 1.07 (1.06 to 1.09) |
| Charlson score increased by 1 | 1.21 (1.21 to 1.22) |
| Study year increased by 1 | 0.97 (0.96 to 0.98) |
| Hospitalization in last 6 months | 1.73 (1.69 to 1.76) |
| Pre-admission MD visits increased by 1 | 1.01 (1.01 to 1.01) |
| Pre-admission MDs increased by 1 | 1.04 (1.03 to 1.04) |
| Diagnosis risk score increased to next quartile | 1.45 (1.44 to 1.46) |
| Hospital Factors | |
| Length of stay increased 1 day | 1.01 (1.01 to 1.01) |
| Complication during admission | 1.11 (1.08 to 1.14) |
| Procedure during admission | 0.96 (0.94 to 0.98) |
| Medical admission | 1.75 (1.70 to 1.80) |
| Resource Intensity Weighting increased 1 unit | 1.07 (1.06 to 1.08) |
| Number of hospital MDs increased by 1 | 1.01 (1.00 to 1.01) |
| Friday discharge | 1.06 (1.03 to 1.08) |
| Follow-up Factors | |
| One more visit with hospital MD vs community MD | 0.95 (0.95 to 0.96) |
| One more visit with hospital MD vs specialist | 0.97 (0.97 to 0.98) |
The hazard ratio indicates the association of each factor with time to death or readmission after adjusting for all other factors in the table. The hazard ratio is stratified by hospital. Resource Intensity Weighting is a measure of hospitalization costliness. A hospital physician is one who saw the patient during their hospitalization. A community physician is one who saw the patient in the 3 months prior to their admission. A specialist is a physician whose specialty was anything other than general or family practitioner. This model controls for the total number of all physician visits.
CI, confidence interval.