Regression analysis of indicating multiple incremental cost-effectiveness ratios for non-small cell lung cancer treatment

Objectives:

The value of a health technology can be measured in terms of cost and benefit on two-dimensional co-ordinates. This study is to quantitatively analyze the correlation and to conduct a regression on the X-Y plane constituted by cost and QALYs (quality-adjusted life years) associated with the first line treatment, the maintenance treatment, and the second line treatment for non-small cell lung cancer (NSCLC).

Methods:

The cost-effectiveness data of the cost and QALYs were extracted, with respect to the three categories of the NSCLC treatment, from the CEA Registry at Tufts Medical Center, regarding the literature published from 2000–2011. As a result, 44 QALY-cost ratios were identified.

Results:

Based on those extracted data, the correlation and regression analyses were performed by mathematical model using log and square-root functions. The plotted ratios stratified by the three stages for the NSCLC treatment were visually grouped into three clusters. There were statistically significant differences among the correlation coefficients of the cluster. In regression, the log model was found to be better fitted than the square-root model; formulating QALY?=??1.12?+?0.16 log(Cost), ?1.99?+?0.28 log(Cost), and ?0.69?+?0.10 log(Cost) for the first line, the maintenance, and the second line treatment, respectively. Monetary units were standardized to 2008 US dollars.

Conclusion:

A good methodological potential was confirmed so as to assess the Incremental Cost Effectiveness Ratio (ICER) variations, considering stratification by multiple factors such as disease and treatment categories. This study has certain limitations, such as the small number of included articles and the stratification, not reflecting a factor of new genetic findings.