Susan Macran

Measuring population health: A comparison of three generic health status measures

Objectives. The relative performance of three widely used generic health status measures (EQ-5D, a modified HUI3 [mHUI3], and SF-12) was compared within a general population sample. Materials and methods. Data were taken from a cohort of persons identified from the patient list of a large general practice in York, UK. Two-way comparisons were made between EQ-5D and mHUI3 and EQ-5D and SF-12. The measures were assessed in terms of their practical viability, coverage, and discrimination. Practical viability was evaluated in terms of the extent of missing responses and the proportion indicating difficulty with a measure. Coverage examined the range of responses across the items in the measures. Discrimination examined the capacity of the measures to discriminate between persons according to their self-reported morbidity and socioeconomic status. Results. One thousand one hundred twenty-six persons completed a postal questionnaire containing EQ-5D and either mHUI3 (n = 593) or SF-12 (n = 533). Missing responses were low across all three instruments. SF-12 showed a broad distribution of responses across its items however, responses on the mHUI3 hearing, speech and dexterity dimensions and the EQ-5D self-care dimension were highly skewed, with few persons reporting problems. In terms of summary scores, mHUI3 identified more mild health states than EQ-5D. EQ-5D and mHUI3 showed slightly better discrimination than SF-12. Conclusions. Despite the inherent differences in their descriptive systems and scoring functions, no one instrument performed better or worse than the other with respect to the criteria applied in this study. Some of the issues to be considered when choosing a population health measure are discussed.

“death” and the Valuation of Health-related Quality of Life

Background and Objectives.Despite evidence to the contrary, a common assumption in the area of health status measurement is that the state “dead” is the worst possible health state and by definition should be assigned a value of 0. However, the value of the state “dead” and the notion of states worse than “dead” have never been fully addressed as a research topic. This article demonstrates the e-tent of the variation in the value given to the state “dead” by individuals and the effects of transformation on individual and aggregate values using data elicited with 2 methods (visual analog scale rating and ranking) that place no constraint on the value given to the state “dead.” Research Design.Face-to-face interviews were conducted with 253 adults in North Yorkshire, UK, in 1998. Each participant performed ranking and visual analog scale rating e-ercises for 19 EuroQol EQ-5D health states. Conclusions and Results.Data showed that there is a small group of individuals who, when given the option, choose to place relatively high value on the state “dead” compared with other health states. This did not appear to be due to artifact. Evidence also suggested that the usual assumptions underlying the transformation of health state values, for which the distance between full health and “dead” is used to define the denominator, may not hold for these individuals and may distort aggregate preference structures. The authors stress the need for more systematic inquiry in this field.

The development of a new measure of quality of life for children with congenital cardiac disease.

The purpose of the study was to develop a questionnaire measuring health-related R1 quality of life for children and adolescents with congenital heart disease, the ConQol, that would have both clinical and research applications. We describe here the process of construction of a questionnaire, the piloting and the development of a weighted scoring system, and data on the psychometric performance of the measure in a sample of 640 children and young people recruited via 6 regional centres for paediatric cardiology from across the United Kingdom. The ConQol has two versions, one designed for children aged from 8 to 11 years, and the other for young people aged from 12 to 16 years. Initial findings suggest that it is a valid and reliable instrument, is acceptable to respondents, and is simple to administer in both a research and clinical context.

Eliciting social preference weights for Functional Assessment of Cancer Therapy-Lung health states.

AbstractBackground: The Functional Assessment of Cancer Therapy-Lung (FACT-L) is a multidimensional measure of quality of life developed for use in the evaluation of interventions in lung cancer. Objective: To develop a set of utility weights that could be used to convert FACT-L into a single index capable of being used in the economic analysis of clinical trial data. Method: A core set of FACT-L items were valued in two versions of a 14-page postal survey of over 400 members of the UK general population. Respondents valued hypothetical FACT-L health states using a scale from 0 to 100 (worst to best health state). Respondents also valued their own health using the standard form of the EuroQol EQ-5D. Data were entered into an ordinary least squares regression model. Results: Item weights estimated in regression analysis yielded values for 10 items from the FACT-L. The summary index based on this selected set of FACT-L items has a maximum value of 0.703 and a minimum value of 0.111. Conclusion: This study demonstrates a practical method of converting a standard condition-specific measure into a form that has the requisite properties to legitimise its use in cost-utility analysis. The methodology used here is not unique to FACT-L and might be considered appropriate for use in converting similar instruments.

A ‘new and improved’ EQ-5D valuation questionnaire?

Notwithstanding its widespread use, the standard questionnaire used to elicit visual analogue scale valuations for EQ-5D states is well known to suffer from problems with missing values (particularly for the state ‘dead’) and logical inconsistencies. This contribution reports on efforts to redesign the questionnaire to overcome these problems and the results from its use in a pilot study. The redesigned questionnaire asks respondents to provide a numerical score for each state (instead of drawing lines to a visual analogue scale) and employs a new method for valuing ‘dead’. A pilot study was undertaken to gauge the effects of these innovations, over and above other cosmetic changes to the questionnaire. The redesigned questionnaire was found to result in fewer missing values, fewer exclusions and fewer logical inconsistencies. Mean re-scaled values for the health states remained virtually identical. The results suggest that scoring is a better way of eliciting self-completed valuations than scaling.