Catastrophic healthcare expenditures (CHE) are highly problematic for families are are unequally distributed throughout society. However, how can we quantify the incidence, intensity and inequality of CHE in a society?
A paper by Ogwang and Mwabu (2025) provide one methodology by using the Watts poverty measure and adapting it to measure CHE. We first describe what the Watts poverty measure is. Next we explain how to calculate the Ogwang and Mwabu Watts catastrophic health expenditure (WCHE) metric both in aggregate and by subgroup. Finally, I have provided a simple Excel spreadsheet you can download to see how WCHE is calculated.
As described above, the metric is based on the Watts poverty measure, developed by Harold W. Watts in 1964. Perplexity describes the Watts Poverty Measure as follows:
What is the Watts Poverty Measure?
The Watts Poverty Measure (or Watts Index) is a poverty metric that captures not just how many people are poor, but also how far below the poverty line they are. It is especially sensitive to the poorest individuals.
How is it Calculated?
The formula is:
W = (1/N) ∑ ln(z/yi)
where N is the total population, z is the poverty line, and yi is the income of each poor person (those with income below z).
What Does the Number Mean?
A higher Watts Index means greater poverty and deeper shortfalls from the poverty line.
The index is always zero or positive (never negative).
If no one is poor, the index equals zero.
Whereas the Watts poverty measure is calculated based on whether individual’s income falls below the poverty line, the WCHE metric proposed examines the share (and degree) to which health expenditures are above a given CHE threshold. Let’s examine how WCHE is calculated, first at the aggregate level.
The formula for WCHE is:
WCHE = (1/N) ∑ ln(E*i/T)
where N is the total population, T is the CHE threshold and E*i is the health expenditures only for individuals who’s health expenditures exceed the CHE (T).
The interpretation of this index is similar to the Watts poverty index but applied to CHE:
A higher WCHE value means CHE is a larger problem in society WCHE is always zero or positive (never negative).If no one is has any CHE, the index equals zero
Now let us consider how to calculate WHCE this if we decompose this metric by group.
WCHE = (H/N) ∑j=1 to ksj βj)
where N is the total population, H is number of people in CHE overall, sj is the share of individuals in group j who have expenditures over the CHE threshold and βj is the intensity of the CHE, which is the average amount people are over the CHE threshold when it happens
The benefit of this approach is that you can decompose CHE into three categories: the overall incidence (H/N), the inequality of the incidence of CHE across groups (sj), and the typical amount over the threshold (by group) that CHE are when they occur βj.
You may still be wondering how to calculate WCHE in practice. I encourage you to read the full paper here and then check out my simple spreadsheet calculation example here.
