A Review and Meta-Analysis of Utility Values for Lung Cancer

Abstract

BACKGROUND: Published utility estimates for lung cancer are plentiful and vary greatly. The reason for this variability is unclear, but may result from differences in the methods used to elicit each utility. PURPOSE: To identify a set of pooled lung cancer utility estimates reflective of the available literature and determine which methodological factors significantly influence the value of lung cancer utility. DATA SOURCES: Searches of PubMed, the NHS Economic Evaluation Database, and the Cost Effectiveness Analysis Registry from the Center for the Evaluation of Value and Risk in Health. STUDY SELECTION: English-language studies were included if they presented at least one previously unpublished lung cancer utility value, noted the elicitation technique and utility value provider. DATA EXTRACTION AND ANALYSIS: Two trained readers independently reviewed each article and extracted information for analysis. A hierarchical linear model (HLM) was used to perform a meta-regression with cancer stage, lower bound of scale, upper bound of scale, respondent, elicitation method, and lung cancer subtype as explanatory variables. DATA SYNTHESIS: . Twenty-three articles containing 223 unique utility values were included. Lung cancer stage and subtype, the upper bound label of the utility scale, and respondent identity were significant predictors of utility (P < 0.05), while the lower bound label of utility scale was not. The HLM provided a set of pooled utility values for metastatic (0.57), mixed or nonspecified stage (0.77), and nonmetastatic lung cancer (0.87)-for the case of standard gamble as method, patients as respondents, non-small-cell lung cancer and scale labeled death to perfect health. CONCLUSION: Methodological factors significantly affect lung cancer utilities; therefore, analysts should avoid direct comparisons of lung cancer utility values elicited with dissimilar methods.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 78 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.