1 Section of Biostatistics, Department of Public Health, Faculty of Health and Medical Sciences, Københavns Universitet2 The Staff-Student Committee of the Bachelor/Master of Science in Public Health, Faculty of Health and Medical Sciences, Københavns Universitet3 unknown4 The Staff-Student Committee of the Bachelor/Master of Science in Public Health, Faculty of Health and Medical Sciences, Københavns Universitet5 Section of Biostatistics, Department of Public Health, Faculty of Health and Medical Sciences, Københavns Universitet
Despite the widespread acceptance of quality of life (QOL) as the ideal guideline in healthcare and clinical research, serious conceptual and methodological problems continue to plague this area. In an attempt to remedy this situation, we propose seven criteria that a quality-of-life concept must meet to provide a sound basis for investigation by questionnaire. The seven criteria or desiderata are: (1) an explicit definition of quality of life; (2) a coherent philosophy of human life from which the definition is derived; (3) a theory that operationalizes the philosophy by specifying unambiguous, nonoverlapping, and jointly exhaustive questionnaire items; (4) response alternatives that permit a fraction-scale interpretation; (5) technical checks of reproducibility; (6) meaningfulness to investigators, respondents, and users; and (7) an overall aesthetic appeal of the questionnaire. These criteria have guided the design of a validated 5-item generic, global quality-of-life questionnaire (QOL5), and a validated 317-item generic, global quality-of-life questionnaire (SEQOL), administered to a well-documented birth cohort of 7,400 Danes born in 1959-1961, as well as to a reference sample of 2,500 Danes. Presented in outline, the underlying integrative quality-of-life (IQOL) theory is a meta-theory. To illustrate the seven criteria at work, we show the extent to which they are satisfied by one of the eight component theories. Next, two sample results of our investigation are presented: satisfaction with one's sex life has the expected covariation with one's quality of life, and so does mother's smoking during pregnancy, albeit to a much smaller extent. It is concluded that the methodological framework presented has proved helpful in designing a questionnaire that is capable of yielding acceptably valid and reliable measurements of global and generic quality of life.