Pre-print, Working paper: This paper investigates the impact of mortality in the districts/states represented in key congressional groups (i.e. committees, subcommittees, and parties) on the public investment in medical research in the US. I focus on National Institutes of Health (NIH) R01 grants awarded between 1985-2002. Exploiting the recomposition of any group after congressional elections, I estimate that the composition of the House Subcommittee on Labor, Health and Human Services, Education and Related Agencies (HouS), impacts the NIH budget: a 1% increase of life-years lost because of a disease in the districts represented in HouS increases the funds for clinical research on that disease by 1.2-3.2%. I also find that this impact results from the larger bargaining power of HouS or the House majority, or both groups, in the budget process. No group significantly impacts the allocation of funds for basic research, or the allocation of funds across states.