1037. The Impact of Health Literacy and Social Support on Healthcare Utilization
Ahsan M Arozullah, MD, MPH, VA Chicago Healthcare System, AM Khan, VA Chicago Healthcare System, T Khan,
VA Chicago Healthcare System, S Kurup,
VA Chicago Healthcare System, S Lee,
University of North Carolina
Objectives: To determine the impact of health literacy and social support on hospital and outpatient care utilization.
Methods: We enrolled 400 medical service patients admitted to VA Chicago (8/1/01-4/1/03). Patients with dementia, blindness, deafness, or VA care for <6 months were excluded. The Rapid Estimate of Adult Literacy in Medicine was used to classify subject literacy as inadequate (<4th grade), marginal (4th-8th grade), or adequate (>9th grade). The MOS Social Support Scale assessed social support. Outcomes obtained from electronic records included: hospital admissions (0 vs. >1), emergency room (ER) visits, outpatient visits and no-show rates. We used hierarchical regression models, adjusting for health status, to determine the association of health literacy and social support with each outcome.
Results: 42% of subjects had adequate literacy, 58% had previous hospitalization or ER visit. Inadequate literacy was associated with hospitalization (OR=4.5, 95% CI, 1.4-14.1) and was marginally associated with ER visit (OR=2.5, 95% CI, 0.9-6.9) compared with adequate literacy. Higher self-reported health responsibility was significantly associated with higher outpatient visits (P<0.001). Higher affectionate social support was significantly associated with higher visits (p=0.02) and lower no-show rates (p=0.02). Health literacy was not associated with outpatient visits or no-show rates.
Conclusions: Inadequate literacy is associated with higher hospital and ER use. Affectionate social support is associated with higher outpatient visits and lower no-show rates. Higher health responsibility is associated with increased outpatient visits, but not with lower no-show rates.
Impact: Assessing health literacy and social support may improve our ability to predict healthcare utilization including no-show rates.