A concern with using quality-adjusted life-years (QALYs) to measure health gains in cost-effectiveness studies is that QALYs “underestimate” the health gains that people with severe disease or chronic disabilities derive from extending life. To see this clearly, look at the QALY formula (first formula below) and how it breaks down into Quality of Life (Q) and Survival (S) for the new intervention (with an index of 1) compared to the old intervention ) (with index 0).
Note that the survival benefit was only assessed in terms of the quality of life achieved by the intervention (Q1t). Therefore, the improvement in survival of patients whose quality of life is still poor is devalued relative to those patients whose health status is good or even perfect.
a paper by Campbell, Whittington and Pearson (2023) Describes how the Institute for Clinical and Economic Review (ICER) calculates an alternative measure: Equivalent Life Years Earned (evLYG).I have described this approach before here. evLYG can be calculated as follows:
On the right, the second terms in the evLYG and QALY equations are the same, indicating that the first estimate of quality of life gain during the survival of the individual is the same. However, under evLYG, the survival gain is valued at other numbers (Qt) which may differ from the quality of life value for the new treatment.According to the originally proposed evLYG method North et al. (1999), the survival gain is evaluated at perfect health (i.e. Qt=1).but that Campbell et al. The method used by ICER sets Qt=0.851, which is equal to the average utility for the general US adult population. Note that the ICER method specifies a lower evLYG than the Nord method.
In the context of ICER’s implementation of evLYG, the differences in how evLYG and QALY measure health gains can be summarized as follows:
The key advantage of evLYG is that the survival benefit is more important in sicker patients (although the value under the ICER implementation is lower than under the Nord implementation).
In fact, ICER’s evaluation of 32 unique interventions using evLYG and QALY found that ΔevLYG > ΔQALY in 66% of cases and ΔevLYG = ΔQALY in 34% of cases.The mean percent change between incremental QALY and incremental evLY for the intervention dyad was 16%. In interventions where ΔevLY > ΔQALY, incremental evLY was 25% higher than incremental QALY.
A major limitation of evLYG is that evLYG does not reflect the improvement in quality of life during the “survival gain”.
You can read the full text here. It is a useful primer on what evLYG is and how ICER uses evLYG in assessments.
