The purposes of this dissertation were to: (1) examine the psychometric properties of the Minnesota Living with Heart Failure Questionnaire (LHFQ), an instrument with a total scale score, and emotional and physical scale scores; (2) identify variables associated with health-related quality of life (HRQOL) in patients with heart failure; and (3) determine whether there were gender differences in HRQOL. Data consisting of 638 patients and a subgroup of 196 patients were used to analyze the psychometric properties of the LHFQ using Cronbach's alpha, item analyses, and construct validity using factor analysis and hypothesis testing related to proposed relationships between the LHFQ and functional status, symptom status, health perception, and social support. A data set consisting of 293 patients was used to identify variables associated with HRQOL in patients with heart failure using all possible regression model analysis. Gender differences in HRQOL were tested using the independent t-test.
The results of the psychometric tests provided additional support for the reliability and some evidence for the validity of the LHFQ. However, the results of the factor and item analyses suggested that rewording or elimination of several items may improve the reliability and validity of the LHFQ and provide researchers and clinicians with a more useful measure of HRQOL in patients with heart failure.
In the investigation of variables related to HRQOL, the multivariate model with the LHFQ total scale score as the dependent variable included the following variables: health perception, symptom status, New York Heart Association functional classification (NYHA), and age. The model for the emotional scale score included the same variables with the exception of NYHA. The model for the physical scale score included symptom status, health perception, NYHA, and etiology. Women reported worse HRQOL than men, and slightly different variables were included in the final models with HRQOL in men and women. Subjective variables including health perception, symptom status, and functional status were more consistently related to HRQOL than objective variables including etiology. The most influential variables associated with all scales of the LHFQ were health perception and symptom status.