A model predicting health status of patients with heart failure

Objective:
To test the causal relationships among the components of sociodemographic (age, gender, education, and income), illness characteristics (duration of illness, severity of illness, and comorbid diseases), and self-management ability, and health status in the model of health status of patients with heart failure (HSHF).

Design:
Descriptive cross-sectional study.

Materials and Methods:
Participants were 400 heart failure patients hospitalized or attending an out patient clinic at six hospitals in southern Thailand. A survey-interview method was used for data collection. Questionnaires were related to study factors including sociodemographics, duration of illness, the New York Heart Association Functional Classification (NYHA-FC), the Charlson Comorbidity Index, the Self-Care of Heart Failure Index (SCHFI), the Short Form-36 Health Survey (SF-36). The relationship of the study variables was tested and modified under the structural equation modeling (SEM) technique by using LISREL.

Results:
The initial hypothesized model did not fit the data. The modified model adequately fit the data and accounted for 64% of the variance in health status. Age had a direct negative effect on health status (β = −0.20, P < 0.01) and had an indirect negative effect on health status through self-management ability, severity of illness and comorbid disease (β = −0.13, P < 0.01). Education had a direct positive effect on health status (β = 0.12, P < 0.01). Gender and income had indirect negative effects on health status through severity of illness (β = −0.05; −0.05, P < 0.05). Duration of illness had an indirect positive effect on health status through self-management ability (β = 0.09, P < 0.05). Severity of illness and comorbid disease had a direct negative effect on health status (β = −0.31; −0.16, P < 0.01, respectively) and indirect negative effect on health status through self-management ability (β = −0.06; −0.05, P < 0.05, respectively). Self-management ability had a direct positive effect on health status (β = 0.38, P < 0.01).

Conclusions:
This model provides a guideline for explaining and predicting health status of patients with heart failure. Continuity care programs promoting self-management ability should be developed and implemented both in hospital-based and home-based settings in order to improve health status.