Lost productivity due to premature mortality in developed and emerging countries: an application to smoking cessation

Abstract

Background

Researchers and policy makers have determined that accounting for productivity costs, or “indirect costs,” may be as important as including direct medical expenditures when evaluating the societal value of health interventions. These costs are also important when estimating the global burden of disease. The estimation of indirect costs is commonly done on a country-specific basis. However, there are few studies that evaluate indirect costs across countries using a consistent methodology.

Methods

Using the human capital approach, we developed a model that estimates productivity costs as the present value of lifetime earnings (PVLE) lost due to premature mortality. Applying this methodology, the model estimates productivity costs for 29 selected countries, both developed and emerging. We also provide an illustration of how the inclusion of productivity costs contributes to an analysis of the societal burden of smoking. A sensitivity analysis is undertaken to assess productivity costs on the basis of the friction cost approach.

Results

PVLE estimates were higher for certain subpopulations, such as men, younger people, and people in developed countries. In the case study, productivity cost estimates from our model showed that productivity loss was a substantial share of the total cost burden of premature mortality due to smoking, accounting for over 75 % of total lifetime costs in the United States and 67 % of total lifetime costs in Brazil. Productivity costs were much lower using the friction cost approach among those of working age.

Conclusions

Our PVLE model is a novel tool allowing researchers to incorporate the value of lost productivity due to premature mortality into economic analyses of treatments for diseases or health interventions. We provide PVLE estimates for a number of emerging and developed countries. Including productivity costs in a health economics study allows for a more comprehensive analysis, and, as demonstrated by our illustration, can have important effects on the results and conclusions.