Readmissions to medical wards: Analysis of demographic and socio-medical factors

BackgroundUnplanned readmissions affect occupancy rates in medical wards and these have been constantly increasing in Israel. We examined the frequency and risk factors affecting the likelihood of 30-day emergency medical readmissions.MethodsWe compared the clinical, epidemiological, and socioeconomic characteristics of readmitted patients during 2009 to a control group matched by age, sex, and primary medical diagnosis.ResultsRate of unplanned readmissions within 30 days was 12.2%. The mean time to readmission was 12.8 days. The mean length of hospital stay at index admission was 4.4 and 3.8 days for the study and control groups, respectively, and 4.99 days in the second admission (study group only). By simple univariate logistic regression, living in a nursing home, chronic kidney disease, ischemic heart disease, previous cerebrovascular accident, number of chronic medications, length of hospital stay at index admission, and hospitalization in the previous year prior to index admission were significantly associated with risk of readmission. In multivariate logistic regression model, only living in a nursing home (OR = 2.94, 95%CI = 1.15–7.48), presence of chronic kidney disease (OR = 1.62, 95%CI = 1.06–2.46), length of index admission ≥ 3 days (OR = 1.53, 95%CI = 1.07–2.2), and hospitalization in the previous year (OR = 3.33, 95%CI = 2.34–4.74) were found to be significantly associated with likelihood of readmission.ConclusionRisk factors affecting 30-days readmission at our medical centre are similar to previous observations, and yet, some are perhaps unique to our region.