Rural Patients With Heart Failure And Multiple Chronic Conditions: A Cross-Sectional, Descriptive Study

Introduction Living in a rural area increases the risk of adverse heart failure (HF)-related outcomes including decreased quality of life (QOL). Multiple chronic conditions (MCC) are also known to affect QOL. HF is frequently accompanied by MCC including hypertension (HTN), diabetes (DM), pulmonary disease, and/or renal dysfunction. However, little is known about the occurrence of these HF+MCC clusters and their impact on QOL in rural patients. Therefore, we aimed to identify and describe HF+MCC clusters and examine characteristics including demographics, ejection fraction (EF), and QOL across clusters in a sample of rural adults with HF. Methods We used data previously collected by the RICH Heart Program at the University of Kentucky. Subjects in the database with rural-urban commuting area codes of 7-10 were considered rural. We extracted demographics, EF, MCC (coronary heart disease (CHD), HTN, peripheral vascular disease, stroke, pulmonary disease, DM, peptic ulcer disease (PUD), dementia, renal disease, connective tissue disease, liver disease, dementia, AIDS, and cancer), and responses to the Minnesota Living with Heart Failure Questionnaire (MLHFQ) for analysis. Latent class analysis, one-way ANOVA, and Chi-square tests were applied as appropriate. Results Subjects (n=839) were mostly male (513/839, 61.1%), non-Hispanic (601/839, 71.9%), White/Caucasians (715/839, 85.2%) with a mean age of 64.8 ± 12.8 and EF of 38.2 ±15. HTN was the most reported MCC followed by CHD. Mean MLHFQ for all subjects was 44.6 ± 26.9. A three-class (HF+MCC cluster) solution was the best fitting latent model. Discriminatory MCC were DM and PUD. HF+MCC Cluster 1 (n=367, 43.7%) was defined by DM and the largest proportion of members with HTN. HF+MCC Cluster 2 (n=70, 8.3%) had the greatest proportion of comorbidities overall, the highest proportion of PUD, and zero members with DM. HF+MCC Cluster 3 (n=402, 47.9%) had the least proportion of comorbidities and was differentiated by no class members with PUD. We compared the MLHFQ total QOL score (lower scores reflect better QOL) across clusters. Mean scores (Cluster 1 - 51.4 ± 26.4, Cluster 2- 48.8 ± 26.5, Cluster 3- 37.7 ± 25.7) were significantly different (F(2,834)=27.25, p= Conclusions Most subjects in this sample reported MLHFQ scores that indicate poor QOL (scores >45). Cluster 3 members reported better QOL (lower scores) which is likely related to the low HF+MCC burden in this class. Interestingly, Cluster 1 members reported the lowest QOL (highest MLHFQ scores) despite Cluster 2 including a greater HF+MCC burden. Cluster 1 included the highest proportion of members with DM which may have driven this finding. QOL interventions in rural patients with HF should be designed to account for the number of MCC and the influence of HF+MCC clusters.