Distributional Cost-Effectiveness Analysis of Treatments for Non-Small Cell Lung Cancer: An Illustration of an Aggregate Analysis and its Key Drivers.

Background And Objective: Distributional cost-effectiveness analysis (DCEA) facilitates quantitative assessments of how health effects and costs are distributed among population subgroups, and of potential trade-offs between health maximisation and equity. Implementation of DCEA is currently explored by the National Institute for Health and Care Excellence (NICE) in England. Recent research conducted an aggregate DCEA on a selection of NICE appraisals; however, significant questions remain regarding the impact of the characteristics of the patient population (size, distribution by the equity measure of interest) and methodologic choices on DCEA outcomes. Cancer is the indication most appraised by NICE, and the relationship between lung cancer incidence and socioeconomic status is well established. We aimed to conduct an aggregate DCEA of two non-small cell lung cancer (NSCLC) treatments recommended by NICE, and identify key drivers of the analysis.

Methods: Subgroups were defined according to socioeconomic deprivation. Data on health benefits, costs, and target populations were extracted from two NICE appraisals (atezolizumab versus docetaxel [second-line treatment following chemotherapy to represent a broad NSCLC population] and alectinib versus crizotinib [targeted first-line treatment to represent a rarer mutation-positive NSCLC population]). Data on disease incidence were derived from national statistics. Distributions of population health and health opportunity costs were taken from the literature. A societal welfare analysis was conducted to assess potential trade-offs between health maximisation and equity. Sensitivity analyses were conducted, varying a range of parameters.

Results: At an opportunity cost threshold of £30,000 per quality-adjusted life-year (QALY), alectinib improved both health and equity, thereby increasing societal welfare. Second-line atezolizumab involved a trade-off between improving health equity and maximising health; it improved societal welfare at an opportunity cost threshold of £50,000/QALY. Increasing the value of the opportunity cost threshold improved the equity impact. The equity impact and societal welfare impact were small, driven by the size of the patient population and per-patient net health benefit. Other key drivers were the inequality aversion parameters and the distribution of patients by socioeconomic group; skewing the distribution to the most (least) deprived quintile improved (reduced) equity gains.

Conclusion: Using two illustrative examples and varying model parameters to simulate alternative decision problems, this study suggests that key drivers of an aggregate DCEA are the opportunity cost threshold, the characteristics of the patient population, and the level of inequality aversion. These drivers raise important questions in terms of the implications for decision making. Further research is warranted to examine the value of the opportunity cost threshold, capture the public's views on unfair differences in health, and estimate robust distributional weights incorporating the public's preferences. Finally, guidance from health technology assessment organisations, such as NICE, is needed regarding methods for DCEA construction and how they would interpret and incorporate those results in their decision making.