Objective: Quality of life (QOL) is a multidimensional construct and it is an appropriate outcome for measuring the impact of patient care. Few in-depth analyses have demonstrated its conceptual unity. The dimensions identified in QOL frameworks, for the most part, have little theoretical basis. An atheoretical approach to conceptualizing QOL is problematic and has resulted in the inability to adequately define, measure and ultimately address QOL in a systematic fashion in clinical practice. The purpose of this paper is to discuss conceptual issues surrounding QOL and to provide an example of how structural equation modeling can address some of these issues.
Design: This study was descriptive and correlational using secondary data analysis.
Sample: The sample for this analysis was data collected as part of the AIDS Time-Oriented Health Outcome Study (ATHOS) between 1992 and 1994. ATHOS was a longitudinal observational database of persons with HIV-associated illness cared for by community-based providers in the San Francisco Bay Area, two private practices in Los Angeles, and five community clinics in San Diego. The sample size was 1410, 84% were Caucasian, and 95% were male.
Concepts: QOL was conceptualized as a second-order structure, consisting of one higher order general factor (QOL) and four first-order factors (cognition, vitality, mental heath, and health worry). This representation is supported by the work done during the Medical Outcomes Study (Stewart & Ware, 1992). Cognition was defined as day-to-day problems in cognitive functioning of which the patient was aware. Cognition was measured by using responses to: Do you react slowly to things that were said or done?; Become confused and start several actions at a time?; Forget where you put things or appointments?; Have difficulty concentrating?; and Have difficulty reasoning and solving problems? Vitality was defined as the positive and negative ends of the energy/fatigue continuum. It was measured by responses to: Do you Feel full of pep?; Feel worn out?; Have enough energy to do the things you want?; and Feel tired? Mental health was assessed as psychological distress and well-being. It was measured by responses to: Do you feel calm and peaceful?; Feel downhearted and blue?; Feel very happy?; Feel very nervous?; and Feel so down in the dumps nothing could cheer you up? Health worry was defined as the extent to which problems cause people to worry or be greatly concerned about their health. Health worry was measured by responses to: Were you frustrated about your health?; Were you afraid because of your health?; and Was your health a worry in your life?
Methods: Structural equation modeling was used to confirm the hypothesized second-order factor structure. Analysis was conducted using the LISREL program and maximum likelihood estimation procedures. Chi-square likelihood ratio statistic was used in the assessment of model fit as well as several goodness of fit indices.
Findings: The large sample size resulted in a significant chi-square statistic for the hypothesized model (p<.0001), indicating a lack of fit. However, the other indicators of fit, RMSEA (.07), SRMR (.05), and CFI (.95), suggest that the hypothesized model provides an adequate description of the components that make up the QOL construct in the sample.
Conclusions: Based on the fit indexes (RMSEA, SRMR, and CFI), it is suggestive that the data fits this model, such that QOL can be defined by indicators of vitality, cognition, mental health, and health worry.
Implications: The purpose of this paper was to discuss conceptual issues surrounding the measurement of QOL and to provide methodological support for the precision of structural equation modeling as a solution to some of these issues. Measuring QOL without the development of conceptual models has hindered the advancement of its research knowledgebase. An understanding of the dimensions of QOL and their relationships to each other is essential for future development of empirical knowledge for practice. A theoretical approach to measuring QOL will expand its clinical use as an outcome measure and increase its generalizability and relevance. A better understanding of the construct allows nurses to focus services that enhance QOL in clients with particular health problems.
Stewart, A. L., & Ware, J. E. (1992). Measuring functioning and well-being: The medical outcome study approach. Durham, N. C.: Duke University Press.
Acknowledgments: This project was partially funded by Grant #R01 NR04817 from the National Center for Nursing Research, National Institutes of Health.
Back to Issues in Quality of Life and the Needs of Caregivers
Back to The Advancing Nursing Practice Excellence: State of the Science