The most well-known readmission risk score is the LACE score, which was developed in 2010 by Canadian researchers (van Walraven et al., 2010). "LACE" stands for the four predictors used to calculate the score, and the full score calculation ranges from 0-19 and is shown in the following table. The Charlson comorbidity index is a score that assigns scores to patients based on the presence of certain illnesses in their past medical history, including myocardial infarction, cancer, CHF, and HIV infection:

To derive this index, the study authors entered over a dozen variables thought to be related to readmission risk for 2,393 patients into a multivariable logistic regression model, and were left with these four variables as the significant predictors in the model. They then developed the scoring system and externally validated it using 1,000,000 records from a Canadian patient database. Their own reported C-statistic for predicting early death or urgent readmission within 30 days was 0.684.

Another example of a hospital readmission risk score is the more recently developed HOSPITAL score (Donze et al., 2013; Donze et al., 2016). Again "HOSPITAL" stands for the seven predictors used in the score as listed in the following table:

The HOSPITAL score ranges from 0-13 points. The score was derived using the seven factors independently related to hospital readmission, also using a multivariable logistic regression model (Donze et al., 2013). The score was externally validated using 117,065 discharges from four countries and the authors reported a C-statistic of 0.72 (Donze et al., 2016).