Evaluation of the Associations Between Unplanned Readmissions and the LACE Index and Other Variables

Background: Unplanned readmissions, within 30 days following an inpatient hospital admission, are common and costly. Because implementing effective interventions, such as advanced nurse practitioner discharge planning and nurse transition coaches, that are known to reduce readmissions, can also be costly, hospitals often aim to target interventions to the most high-risk patients. Research has identified factors that predict readmissions, and predictive algorithms, such as the LACE index, have been studied and widely adopted by hospitals despite demonstrated variability in predictive ability.

Objectives: To examine the associations between unplanned readmissions and the LACE index, and other variables that reflect patient- and encounter-level factors not currently incorporated in the LACE index, including race, marital status, age, gender, preferred language, payer source, index disposition, and index Diagnosis-Related Group (DRG).

Methods: A quantitative, correlational, and retrospective analysis was conducted utilizing data from electronic health records of 17, 082 inpatients discharged from a large quaternary hospital located in the southeastern United States between January 1 and June 30, 2017. The associations between readmissions and each variable were separately examined utilizing chi-square test.

Results: Of the 17,082 inpatients, 1,695 (9.9%) patients were readmitted with an unplanned readmission within 30 days. Positive, statistically significant associations (p<0.01), were found between unplanned readmission and each of the following: LACE index, race, marital status, payer source, index disposition, and index Diagnosis-Related Group (DRG). No association was found with age, gender, or preferred language.

Conclusions: The LACE index, race, marital status, payer source, index disposition, and DRG were associated with unplanned readmission. Age, gender, and preferred language were not associated with unplanned readmission. Utilizing other factors, in addition to the LACE index, may provide clinically useful information to better predict readmissions, to target resources to prevent the readmission from occurring, and to improve the quality of hospital care and transition at discharge.