Background: Reducing readmission rates may only be successfully realized if all factors that influence readmissions are thoughtfully considered. Beyond medical elements, nontraditional risk factors such as social and environmental variables may be associated with a higher likelihood of being readmitted. In this study we evaluated the effect of distance between home and ‘key resources’ on 30 day all cause readmission rates in an underserved urban population.
Methods: We conducted a retrospective study on all consecutive patients admitted to the medicine service between September 1, 2012 and August 31, 2013 at a single academic medical center in an area designated by the U.S. Health Resources and Services Administration as a Medically Underserved Population (“MUP”). Patients were stratified based on the presence or absence of at least one 30 day all cause readmission during the study period. Using a randomization software, a random sample of 400 individuals from both the readmitted and non-readmitted group were selected. Key resources studied that were believed to be supportive of patient recovery and wellness after hospitalization included pharmacy, grocery store and PMD location. To date, seventy five charts from each group were reviewed. Patient characteristics were compared between the two groups using chi-square or student t-test as appropriate. Multivariate logistic regression models were constructed to evaluate the association of distance between home and key resources with 30 day all cause readmission while adjusting for age, gender, marital status, insurance status, having a primary-care provider, case-mix index, yearly income and length of stay.
Results: Compared to non-readmitted patients, readmitted patients had comparable age (60 vs 61 years, p=0.88), yearly income ($48,754 vs $48,529, p=0.88), and rates of insurance (50% vs 51%, p=0.77,). Patients who were readmitted were more likely to have Medicare or Medicaid as their primary insurance and less likely to be females (40% vs 60%, p=0.014). Length of stay was significantly higher in the readmitted patient group (5.5 vs 3.0 days, p<0.01). Readmitted patients had a greater distance between home and pharmacy than non-readmitted patients (0.55 miles vs 0.39 miles; p= 0.001). The distances between home and PMD was greater for the non-readmitted patients (7.17 miles vs 4.33 miles; p= 0.004), as was the case for distance between home and grocery store (0.86 miles vs 0.84 miles; p=0.041). These associations remained significant in the multivariate analyses.
Conclusions: Preliminary data from our study show an association between access to key resources after discharge and readmission. If these findings hold up in more robust analyses, it may suggest that patients who are more remote or isolated in our cities need more attention when planning their discharge.