Skip to main content

Table 2 Comparing strength of predictors of 30-day vs. 7-day readmissionsa

From: Can we predict early 7-day readmissions using a standard 30-day hospital readmission risk prediction model?

  Adjusted Odds Ratio (95% CI)
Original 30-day Readmission Model New 7-day Readmission Model
Baseline factors
 Age in years, per 10 yearsb 1.07 (1.04–1.10) 1.08 (1.03–1.14)
 Widow 1.27 (1.11–1.45) 1.13 (0.92–1.40)
 Medicaidb 1.55 (1.31–1.83) 1.37 (1.06–1.78)
 Prior ED visit, per visitb 1.04 (1.02–1.06) 1.03 (1.01–1.04)
 Prior hospitalization, per hospitalizationb 1.16 (1.12–1.20) 1.13 (1.08–1.18)
Factors from admission day
 Nonelective admissionb 1.42 (1.22–1.65) 1.40 (1.09–1.80)
 Charlson Comorbidity Index, per pointb 1.06 (1.04–1.09) 1.04 (1.01–1.08)
Laboratory abnormalities within 24 h of admission
  Albumin < 2 g/dLb 1.52 (1.05–2.21) 1.75 (1.06–2.87)
  Albumin 2–3 g/dL 1.20 (1.06–1.36) 1.04 (0.86–1.27)
  Aspartate aminotransferase > 40 U/Lb 1.21 (1.06–1.38) 1.34 (1.09–1.63)
  Creatine phosphokinase < 60 mcg/Lb 1.28 (1.11–1.46) 1.40 (1.14–1.72)
  Mean corpuscular volume > 100 fL/red cell 1.32 (1.07–1.62) 0.86 (0.60–1.23)
  Platelets < 90 × 103/μL 1.56 (1.23–1.97) 1.36 (0.94–1.96)
  Platelets > 350 × 103/μL 1.24 (1.08–1.44) 1.18 (0.94–1.49)
  Prothrombin time > 35 s 1.92 (1.73–2.90) 1.57 (0.84–2.94)
Factors from hospital stay
 Discharge to hospice 0.23 (0.13–0.40) 0.41 (0.20–0.86)
Hospital complications
  Clostridium difficile infectionb 2.03 (1.18–3.48) 1.96 (0.96–4.00)
  Pressure ulcerb 1.64 (1.15–2.34) 1.68 (1.01–2.79)
  Venous thromboembolism 1.55 (1.03–2.32) 1.40 (0.76–2.58)
Factors from discharge day
Laboratory abnormalities at discharge
  Blood urea nitrogen > 20 mg/dLb 1.37 (1.24–1.52) 1.38 (1.17–1.62)
  Sodium < 135 mEq/Lb 1.34 (1.18–1.51) 1.49 (1.24–1.79)
  Hematocrit <= 27% 1.22 (1.05–1.41) 1.16 (0.92–1.46)
 Vital sign instability at discharge, per instabilityb 1.25 (1.15–1.36) 1.32 (1.17–1.50)
 Length of stay, per dayb 1.06 (1.04–1.07) 1.06 (1.04–1.08)
  1. Abbreviation: ED Emergency department
  2. a Values reflect adjusted odds ratios and 95% CI for each variable after adjustment for all other variables listed in the table separately for our re-derived early 7-day model and our original validated 30-day readmission model
  3. b Indicates variables that are still statistically significant in the 7-day model. All variables in the 30-day model were statistically significant