Skip to main content

Advertisement

Table 2 Risk Scores’ calibration performance

From: Predicting 7-day, 30-day and 60-day all-cause unplanned readmission: a case study of a Sydney hospital

  7-day Readmission (β0 =  − 4.72,  β1 = 0.14) 30-day Readmission (β0 =  − 4.02,  β1 = 0.13) 60-day Readmission (β0 =  − 3.79,  β1 = 0.13)
Number of admissions Observed readmission rate (%) Expected readmission rate (%) Number of admissions Observed readmission rate (%) Expected readmission rate (%) Number of admissions Observed readmission rate (%) Expected readmission rate (%)
score = 0 641 0.6 0.9 185 1.1 1.8 135 2.2 2.2
score = 10 1003 3.4 3.4 893 5.8 6.3 707 8.5 7.6
score = 20 98 12.2 12.0 101 21.8 20.4 156 22.4 23.2
score = 30 7 28.6 34.9 10 50.0 49.1 18 55.6 52.5
  1. Readmission \( Risk\ (S)=\frac{e^{\left({\beta}_0+S\ast {\beta}_1\right)}}{1+{e}^{\left({\beta}_0+S\ast {\beta}_1\right)}} \), where S = score, β0=intercept, β1=normalization parameter (full details in Additional file 1: Table S7)