High perceived social support and hospital readmissions in an older multi-ethnic, limited English proficiency, safety-net population.

backgroundEarly readmission amongst older safety-net hospitalized adults is costly. Interventions to prevent early readmission have had mixed success. The role of perceived social support is unclear. We examined the association of perceived social support in 30-day readmission or death in older adults admitted to a safety-net hospital.

methodsThis is an observational cohort study derived from the Support From Hospital to Home for Elders (SHHE) trial. Participants were community-dwelling English, Spanish and Chinese speaking older adults admitted to medicine wards at an urban safety-net hospital in San Francisco. We assessed perceived social support using the Multidimensional Scale of Perceived Social Support (MSPSS). We defined high social support as the highest quartile of MSPSS. We ascertained 30-day readmission and mortality based on a combination of participant self-report, hospital and death records. We used multiple/multivariable logistic regression to adjust for patient demographics, health status, and health behaviors. We tested for whether race/ethnicity modified the effect high social support had on 30-day readmission or death by including a race-social support interaction term.

resultsParticipants (n = 674) had mean age of 66.2 (SD 9.0), with 18.8% White, 24.8% Black, 31.9% Asian, and 19.3% Latino. The 30-day readmission or death rate was 15.0%. Those with high social support had half the odds of readmission or death than those with low social support (OR = 0.47, 95% CI 0.26-0.88). Interaction analyses revealed race modified this association; higher social support was protective against readmission or death among minorities (AOR = 0.35, 95% CI 0.16-0.76) but increased likelihood of readmission or death among Whites (AOR = 3.7, 95% CI 1.07-12.9).

conclusionIn older safety-net patients nearing discharge, high perceived social support may protect against 30-day readmission or death among minorities. Assessing patients' social support may aid targeting of transitional care resources and intervention design. How perceived social support functions across racial/ethnic groups in health outcomes warrants further study.

trial registrationNIH trials registry number ClinicalTrials.gov: NCT01221532 .