Study Analyzes Relationship Between Disease Burden and NIH Funding

A study designed to find out if the amount of money the National Institutes of Health (NIH) spends on research for a set of specific diseases is related to how much of a burden that disease is on society has shown that some measures of burden of disease (BOD) are related to funding levels and others are not. The study was prompted by a proposal from the Institute of Medicine that NIH explicitly compare disease burden and funding. Researchers at the Johns Hopkins School of Public Health and Johns Hopkins School of Medicine published their study in the June 17, 1999 issue of the New England Journal of Medicine.

Lead author Cary P. Gross, MD, a Robert Wood Johnson Clinical Scholar at the Johns Hopkins School of Medicine said, "Critics have long contended that NIH's funding is arbitrary and politically weighted toward heavily lobbied diseases. However, our analysis showed that there is a strong relationship between NIH funding and some measures of disease burden." The researchers found a strong correlation between disability-adjusted life years (DALY) and the amount of funding given to a particular disease. DALY, a recently developed burden of disease measure, expresses the balance between duration and quality of life by converting time spent in various conditions of health to "healthy-year equivalents." In contrast they found no relationship between NIH funding levels and the cost of disease to society as measured by days spent in the hospital; the prevalence of a disease (the total number of people who have it); or the incidence, how often new cases arise. There was a significant relationship between both mortality and years of life lost (YLL) and funding levels.

Dr. Gross said, "We were particularly interested in DALYs, because funding relationship shows that both death and disability have an impact on how money is allocated." The investigators looked at 29 diseases and then compared levels of funding predicted by three categories of disease burden (number of deaths, years of life lost, and disability adjusted life years) with the actual funds allocated by NIH. For eight of the 29 diseases, conclusions drawn about over or under-funding depended on which BOD measure was used.

For instance, people with depression are often severely disabled, with high DALYs, but a low mortality rate. If mortality were the BOD used, then the researchers who studied depression appeared to get $53 million too much. When DALYs were used, depression research appeared to be shortchanged by $140 million. Coronary heart disease appeared "overfunded" by $70 million based on mortality, but appeared "underfunded" by $24 million based on DALYs. Dr. Gross said, "As you can see, advocates for a particular disease can select only the measures that paint a picture of underfunding."

Linking funding to a specific disease may be overly simplistic since many factors besides burden of disease affect funding decisions. These include the probability of success in a given area (scientific opportunity); scientific quality of research, maintaining a diverse portfolio, and ensuring an adequate scientific infrastructure. The value of research in one area contributing to knowledge in another must also be considered. Senior author Neil R. Powe, MD, MPH, professor, Epidemiology, Johns Hopkins School of Public Health said, "Burden of disease represents only one of the NIH's five criteria for priority setting and they should all be considered when allocating research funds. Using six different measures of disease burden to determine whether a disease is under or over-funded is not practical. A single, agreed-upon measure of disease burden is needed so that policy makers won't be misled."

This study was funded by the Robert Wood Johnson Foundation.