Primary data collection: what should well-trained epidemiology doctoral students be able to do?

The increasing size and complexity of epidemiologic studies is leading more and more doctoral students in epidemiology to base their thesis work on existing data. While the analysis of existing data provides useful experience in complex analyses, it gives trainees little or no hands-on experience in the actual design and conduct of an epidemiologic study. As these students pursue their careers, most will eventually want to collect original data. I discuss what hands-on experience a well-trained doctoral-level epidemiology student should receive, and argue that we short-change our students if we do not provide them with the opportunity for primary data collection during their doctoral training.