Prediction of Patient Readmission by LACE Index components at Cardiac Care Unit of an Iranian Hospital: A Cohort Study

Background: One approach to improve efficiency in health care is to identifypatients with high risks of readmission so that resources should be distributed in away they would benefit targeted care. A model named LACE (length of stay,acuity of admission, Charlson comorbidity index (CCI) , and number of emergencydepartment visits in preceding 6 months) has been proposed to predict patientreadmission which is widely used due to its simplicity to rank factors’ risks. Theaim of this study is to determine if LACE Index could be used to predict Iranianhospital readmission.Methods: This was a prospective cohort study in which the prediction ofreadmission for patients admitted to the cardiac intensive care of Shahid BeheshtiHospital of Qom during April to June 2012 within one month after the dischargewas evaluated based on 4 items of LACE index. Following-up readmission statesby making calls within a month after discharge. Purposive sampling was used toselect the sample, patients having four most prevalent chronic heart diseases in theCCU of the hospital were selected and at last sample size was 109 patients. Weused logistic regression, the phi and Spearman correlation coefficient to analyzedata using SPSS18. the significance level was considered as 5% in all tests.Results: Among the items of LACE model, 48.6% of patients stayed at the hospitalfor 4 to 6 days. Only 11 patients (10.09%) referred to the hospital after a month.None of the components of the LACE index could enter the stepwise logisticregression model.Conclusions: Considering that LACE model with its four items is a weak inpredicting readmission, in order to improve the model in predicting thereadmission of cardiac patients, it is recommended that individual variables andfactors associated with the service providers be added to it.

Keywords:

Readmission , LACE Index , CCU , Hospital , Iran