S. Baker1,2, L. Graham1,2, E. Dasinger1,2, T. Wahl1,2, J. Richman1,2, L. Copeland3, E. Burns4, J. Whittle4, M. Hawn5, M. Morris1,2 1University Of Alabama at Birmingham,Birmingham, AL, USA 2VA Birmingham Healthcare System,Birmingham, AL, USA 3VA Central Western Massachusetts Health Care System,Leeds, MA, USA 4Milwaukee VA Medical Center,Milwaukee, WI, USA 5VA Palo Alto Healthcare Systems,Palo Alto, CA, USA
Introduction: Hospital readmissions following surgery can be expensive and taxing on patients. Identifying mutable factors in predicting readmissions would be advantageous to both patients and healthcare systems. We hypothesized that patients with lower health literacy (HL) were more likely to be readmitted to the hospital following surgery.
Methods: We enrolled 734 patients undergoing general, vascular, or thoracic surgery at 4 Veterans Affairs (VA) Medical Centers, August 2015-June 2017. Patients were eligible if their post-operative hospital stay was more than 48 hours and they were discharged alive. Trained interviewers assessed patients’ overall health on the day of discharge using the Veterans Health Survey (VR12) Physical and Mental Component Scores (PCS; MCS). Health literacy was assessed by the 3-question Chew Health Literacy Questionnaire (HLQ), and the quality of the discharge transition by the Care Transition Measure (CTM-15). Patients were followed for 30 days post-discharge for readmission or emergency department (ED) use. A follow-up telephone interview at day 30 identified readmissions to non-VA hospitals. The HLQ summed three 5-point items (range 0-12); scores of 0-3 indicated adequate health literacy while scores of 4-12 indicated marginal or possibily inadequate health literacy. Bivariate and multivariable analyses examined correlations between HL and each outcome, 30-day readmission or ED use. Logistic regression models adjusted for clinical and demographic covariates.
Results: At the time of discharge, 33% of patient responses were consistent with inadequate HL (HL-low, n=245). Patients with adequate HL (HL-high) had better overall physical and mental health compared to patients with HL-low (PCS 32.0 vs. 29.5, p=0.01; MCS 49.7 vs 45.7, p<0.01) and reported higher-quality discharges (CTM-15 Mean: 3.3 vs 3.2, p<0.01). The overall 30-day readmission rate was 16% (n=124), however, it was 14% for patients with HL-high compared to 21% with HL-low (p<0.01). After adjusting for overall health (VR12), patients with HL-low were 1.5 times more likely to experience a readmission versus HL-high (OR=1.5, 95% CI=1.0-2.2); patterns of ED use were similar (OR for HL-low =1.38; 95% CI=0.95-2.01). Among the HL factors, patients who reported: (1) always having difficulty understanding written information were 2.8 times more likely to be readmitted (95% CI= 1.0-2.3), (2) not always confident filling out medical forms were 1.6 times more likely to be readmitted (95% CI= 1.1-2.4), and (3) ever requiring help to read hospital materials were 1.5 times more likely to be readmitted (95% CI= 1.2-6.5).
Conclusion: Low health literacy is common among VA surgery patients and an important contributor to readmission. Future work should focus on early identification of inadequate HL and the development of interventions to educate and empower this vulnerable population prior to discharge.