Re: New models for large prospective studies: Is there a better way?

摘要

Manolio et al. (1) proposed a "better way" for large prospective cohort studies based on the UK Biobank protocol, which used a novel centralized model rather than a traditional distributed model to recruit a population cohort of 500,000 participants.Cost was the stated motivation for identifying a better way despite possible selection biases. Adopting the alternative implied "...large anticipated costs [that] present a substantial hurdle" (1, p. 859) for mounting a large cohort study in the United States. The stated contrast was dramatic: The $100 million for the UK Biobank "...stands in stark contrast to the roughly $400 million per year for 10 years estimated for traditional models of conducting a large cohort study in the United States" (1, p. 864).