Some problems with international comparisons of health spending – and a suggestion about how to quantify the size of the problems

This paper investigates the extent to which conclusions from international comparison of health spending depend on different adjustment methods. The analysis shows, first, that health spending figures differ significantly because of different accounting standards. More specifically, spending on long term nursing care is not treated the same way in all the countries. Next, the rankings differ depending on whether spending is adjusted for purchasing power parity and health specific purchasing power adjustment. Finally, the paper examines the problem of how to quantify the degree to which an adjustment method changes the outcome of a comparison. A rank based approach sum changes in rank and compare this to maximum rank change possible. A share based approach sum changes in each units share of the total. Both approaches create a measure that is bounded between zero and one, but the share based approach also captures changes that do not result in rank differences. <p>