Scoring system predicts risk for hospital readmission

A prediction model using patient characteristics can help identify those at risk for early readmission to the hospital, according to a new study.

Researchers performed a prospective observational cohort study to determine whether it was possible to predict which patients were more likely to be readmitted to the hospital within 30 days of discharge. The study population involved 10,946 patients, 7,287 in the derivation cohort and 3,659 in the validation cohort, who were discharged from the general medicine service at six academic medical centers. Readmissions were determined from administrative data and from telephone follow-up 30 days after discharge. The authors looked at sociodemographic factors, social support, health condition and health care utilization and used logistic regression to determine which variables were associated with early readmission. The study results appear in the March Journal of General Internal Medicine.

In each cohort, 17.5% of patients were readmitted to the hospital within 30 days. In the derivation cohort, the following factors were associated with early readmission:

• insurance status, marital status, having a regular physician, Charlson comorbidity score, SF12 physical component score, at least one hospital admission within the past year, and a current length of stay longer than two days.

Scores were calculated by multiplying the beta-coefficient by 10 and rounding to the nearest integer. An overall risk score of 25 points or higher identified 5% of patients with a 30% risk for readmission in both the derivation and validation cohorts. Although marital status and having a regular physician seemed to be unusual predictors of early readmission, the former could result in being discharged home rather than an acute care facility and the latter could indicate the presence of more severe illness, the authors wrote.

The authors noted that their model had only fair discrimination and acknowledged that their results should not be generalized to settings other than academic medical centers, among other limitations. However, they concluded that the factors in this model help identify patients who are at risk for being readmitted to the hospital soon after discharge.

“More work is needed to identify additional factors that impact post-discharge health outcomes, optimize the discharge process for all patients, and create interventions tailored to patients' needs in order to prevent potentially avoidable readmissions,” the authors wrote.