A rich economic literature has examined the human capital impacts of disease-eliminating health interventions, such as the rollout of new vaccines. This literature is based on reduced-form approaches which exploit proxies for disease burden, such as mortality, instead of actual infection counts, which are difficult to measure. We develop an epidemiological dynamic accounting model based on the susceptible-infected-recovered (SIR) framework to derive precise measles infection shares across U.S. cohorts born around the introduction of the measles vaccine. Measles is highly infectious and fully immunizing which makes the disease an ideal candidate for epidemiological modeling. Our epidemiological model is strongly predictive of future measles outbreaks but the derived measles infection shares are not systematically related to cohorts' later educational, economic, or health outcomes. The reduced-form approach, on the other hand, shows that these long-term outcomes strongly improved among vaccinated cohorts in states with high pre-vaccine measles mortality. Our results suggest that differences in disease severity are more relevant for long-term human capital impacts than raw differences in actual infection rates, supporting the reduced-form approach used in the economic literature.