Association between Inpatient Heart Failure (HF) Group Self-Care Class Attendance and Clinical Outcomes

Abstract

Introduction In HF, ongoing self-care behaviors may be challenging for patients to master, and non-adherence may contribute to rehospitalization. An inpatient group class may provide new knowledge that reduces early rehospitalization. Research Questions Is there a difference 30- and 90-day all-cause rehospitalization rates based on group class attendance? Of class attendees, does family member attendance or attendance within 48 hours of discharge affect post-discharge rehospitalization rates? After controlling for patient characteristics that differed between groups, does rehospitalization differ by HF class attendance? Methods In this retrospective, medical records research project, a single cohort of hospitalized patients with HF who were treated between January 2012-December 2016 and discharged alive were separated into 2 groups - usual care + attended the group education class [cases] or usual care [controls]. Controls were matched to cases based on date of HF group education class, type of HF, condition severity, and Quan-Charlson Comorbidity Index (CCI). Group comparisons used Pearson Chi-square tests (categorical variables), ANOVA tests (normally distributed continuous variables) or Kruskal-Wallis tests (non-normally distributed continuous variables). Logistic regression models were performed for multivariable analyses, and included all variables that significantly differed between groups. Analyses were performed using SAS software (version 9.4; Cary, NC). Results Of 3,758 patients, 1,879 cases and 1,879 control subjects, attendance was not associated with all-cause 30-day rehospitalization, including attending within 48 hours of discharge. Cases were more likely to be readmitted within 90 days of discharge compared to control subjects: after adjustment for patient characteristics that differed between groups (race, marital status and length of stay), the odds of 30-day and 90-day rehospitalization were 0.93 (95% CI: 0.74, 1.16), p=0.52 and 1.32 (1.11, 1.58), p=0.002, respectively. When family members attended class, class attendees were less likely to be readmitted within 90 days of discharge, compared to control subjects (27.5% versus 34.1%, p= 0.008). Conclusion A group education class was not a powerful enough strategy in promoting patients’ self-care post discharge to be associated with all-cause early rehospitalization. More research is needed to learn if a revision in education content or teaching strategies would improve outcomes. Since family members’ presence was associated with reduced rehospitalization, they should be encouraged to attend group education sessions.