A REVIEW AND META-ANALYSIS OF LUNG CANCER UTILITIES

Purpose: There are numerous published health utility estimates for lung cancer, which span a very broad range, even for the same level of severity.  The reason for this wide range is not completely clear, but this variability may be the result of the methods used to elicit each utility.  Depending on which published utility value is selected for a cost effectiveness or other type of secondary analysis, the results can be drastically different.  The purpose of this study was to determine a set of pooled lung cancer utility estimates which summarize the available literature and to determine which methodological factors significantly influence the value of lung cancer utility.

Methods:  A meta-regression was performed to obtain pooled health utility estimates for lung cancer and to assess which methodological factors significantly influence the value of utility.  A systematic review identified twenty three articles containing 223 unique utility values which were included in the analysis.  A hierarchical linear model (HLM) was used to perform the meta-regression with cancer stage, lower bound of scale, upper bound of scale, respondent, elicitation method, and lung cancer subtype as the explanatory variables. 

Results: Lung cancer stage, lung cancer subtype, and respondent were all significant predictors of utility (p<.05).  Elicitation method was a near significant predictor of utility.  Neither the upper nor lower bound label of utility scale significantly influenced the value of lung cancer utility.  The HLM provided a set of pooled utility values: 0.57 for metastatic lung cancer, 0.77 for mixed or nonspecified stage lung cancer, and 0.82 for nonmetastatic lung cancer when the standard gamble method is used to elicit utilities from patients with non-small-cell lung cancer and the scale ranges from death to perfect health. 

Conclusions:  This study represents the first and only known effort to systematically review and synthesize the results of available studies containing lung cancer utility estimates and as such, it is a valuable resource for researchers and policy makers in need of reliable lung cancer utility estimates.  This study provides evidence that certain methodological factors do significantly affect lung cancer utilities and as a result, analysts should avoid direct comparisons of lung cancer utility values elicited with dissimilar methods.