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

Teri A. Manolio, Brenda K. Weis, Catherine C. Cowie, Robert N. Hoover, Kathy Hudson, Barnett S. Kramer, Chris Berg, Rory Collins, Wendy Ewart, J. Michael Gaziano, Steven Hirschfeld, Pamela M. Marcus, Daniel Masys, Catherine A. McCarty, John McLaughlin, Alpa V. Patel, Tim Peakman, Nancy L. Pedersen, Catherine Schaefer, Joan A. ScottTimothy Sprosen, Mark Walport, Francis S. Collins

Research output: Contribution to journalReview articlepeer-review

98 Scopus citations

Abstract

Large prospective cohort studies are critical for identifying etiologic factors for disease, but they require substantial long-term research investment. Such studies can be conducted as multisite consortia of academic medical centers, combinations of smaller ongoing studies, or a single large site such as a dominant regional health-care provider. Still another strategy relies upon centralized conduct of most or all aspects, recruiting through multiple temporary assessment centers. This is the approach used by a large-scale national resource in the United Kingdom known as the "UK Biobank," which completed recruitment/examination of 503,000 participants between 2007 and 2010 within budget and ahead of schedule. A key lesson from UK Biobank and similar studies is that large studies are not simply small studies made large but, rather, require fundamentally different approaches in which "process" expertise is as important as scientific rigor. Embedding recruitment in a structure that facilitates outcome determination, utilizing comprehensive and flexible information technology, automating biospecimen processing, ensuring broad consent, and establishing essentially autonomous leadership with appropriate oversight are all critical to success. Whether and how these approaches may be transportable to the United States remain to be explored, but their success in studies such as UK Biobank makes a compelling case for such explorations to begin.

Original languageEnglish (US)
Pages (from-to)859-866
Number of pages8
JournalAmerican journal of epidemiology
Volume175
Issue number9
DOIs
StatePublished - May 1 2012

Keywords

  • cohort studies
  • epidemiology
  • prospective studies

Fingerprint

Dive into the research topics of 'New models for large prospective studies: Is there a better way?'. Together they form a unique fingerprint.

Cite this