CERGAS Seminar "Measuring the impact of weight loss on Health-Related Quality of Life (HRQoL)"

Measuring the impact of weight loss on Health-Related Quality of Life (HRQoL)

Abstract

Understanding the relationship between weight and quality of life is critical to both clinical practice and economic evaluation. Indeed, a major determinant of the cost-effectiveness of weight management interventions is the assumption made about how weight loss impacts on health-related quality of life (HRQoL). Many economic evaluations have assumed a direct relationship between weight loss and improvements in health-related quality of life, outwith or in addition to that derived from reduced disease incidence. For instance, the study which informed the recommendation of orlistat by the National Institute for Health and Care Excellence (NICE) assumed that every kilogram of weight loss was associated with 0.017 higher utility (i.e. higher HRQoL). Currently, systematic reviews and meta-analyses of secondary data have not found consistent evidence that weight loss [derived from non-surgical weight management interventions] was associated with HRQoL in RCTs.

The current evidence base is, however, limited insofar as studies have only analysed cross-sectional data. We plan to add to the evidence base using panel data and the application of appropriate statistical methods. Further, we will use meta-analyses to compare results across trials. Given these advances, the results of this study would be expected to have major implications for conducting economic evaluations in weight management in the future, and may require the reconsideration of previous cost-effectiveness decisions.

We plan to collate individual patient level data from large, randomized controlled trials, in which individuals have both weight and HRQoL measured at multiple time points. We will then apply panel data techniques to derive the impact(s) of weight variation on HRQoL.

In this study, the beginnings of this process are presented using a single exemplar: Look AHEAD. We estimate the HRQoL-weight relationship using a range of econometric models and set of 5 HRQoL measures that were collected in this trial. Specifically, changes in individuals' HRQoL are regressed on changes in weight and a rich set of individual characteristics, clinical indicators and comorbidities.

Initial findings suggest a modest association between weight and HRQoL. Further, the results suggest that this association presents in the physical, rather than the mental, domain of HRQoL. We find evidence of important non-linearities in the HRQoL-weight relationship. The limitations of these analyses are discussed, along with directions for future research.