What is HIDI?

ICER recently released a white paper titled “Advancing Health Technology Assessment Methods that Support Health Equity“. In addition to discussing a variety of methods–e.g., distributional cost-effectiveness analysis (DCEA), extended cost effectiveness analysis, multi-criteria decision analysis (MCDA)–for quantifying the value of new treatments in terms of their ability to reduce health disparities, ICER also introduces the very simple health improvement distribution index (HIDI). ICER calculates HIDI as:

…the disease prevalence in the subpopulation of interest divided by the disease prevalence in the overall population. A HIDI above one suggests that more health may be gained on the relative scale in the subpopulation of interest when compared to the population as a whole. For example, if a disease has a prevalence of 10% among Black Americans whereas the disease
prevalence among all Americans is 4%, then the Health Improvement Distribution Index is 10%/4%= 2.5. In this example, a HIDI of 2.5 means that Black Americans as a subpopulation would benefit more on a relative basis (2.5 times more) from a new effective intervention compared with the overall population.
It is important to note that the HIDI does not represent a full distributional analysis, in that it does not consider potential prognostic, treatment effectiveness, uptake, and access differences that may exist between subpopulations. Further, unlike DCEA methods, the HIDI does not attempt to consider the opportunity costs of adopting a treatment at a given price. However, the HIDI is not computationally complex, has minimal data requirements, and can be easily interpreted by appraisal committees.

HIDI is useful as a basic diagnostic for stakeholders to understand relative disease prevalence. However, it is not helpful for providing quantitative guidance with respect to how health equity impacts would impact ICER’s recommended value based price. They write:

The HIDI is not a normative measure, in that it does not have specific thresholds at which certain levels of priority are suggested, and it is not a standalone measure that comprehensively measures the opportunity to reduce health disparities.

Other methods are able to provide this approach. DCEA in particular would be an attractive option. ICER claims that DCEA requires too much additional data, can be challenging to interpret, and is complex. They also mention about limiations about quality adjusted life expectancy by geography, but would could also apply DCEA across different racial/ethnic, income, education, or deprivation indices (e.g., social vulnerability index). Despite ICER’s protestations, DCEA could readily be ipmlemented if there was the desire, but the more fundamental questions still to be answered include (i) acrosss which groups should health disparities be reduced and (ii) how much efficacy are people willing to trade off to increase equity.

You can read the full white paper here.