Chapter

Methodological issues in the design and analysis of community intervention trials

Allan Donner

in Oxford Textbook of Public Health

Fifth edition

Published on behalf of Oxford University Press

Published in print September 2009 | ISBN: 9780199218707
Published online March 2011 | e-ISBN: 9780199609673 | DOI: http://dx.doi.org/10.1093/med/9780199218707.003.0033

Series: Oxford Textbooks

Methodological issues in the design and analysis of community intervention trials

More Like This

Show all results sharing these subjects:

  • Public Health and Epidemiology
  • Medical Statistics and Methodology
  • Public Health

GO

Show Summary Details

Preview

With the literature on community intervention trials showing rapid growth over the last two decades, there is an increasing need to better understand their methodological foundation. A key feature of such trials is the allocation of intact communities or clusters of individuals rather than individuals themselves to different intervention groups. Examples include trials evaluating a mass education intervention delivered through the media or alterations in the hygiene of villages located in low- and middle-income countries. Only recently, however, has it been recognized that the application of standard approaches to the design and analysis of such trials can lead to serious problems of interpretation. This is because methods that are extensively discussed in the clinical trial literature tend to assume that the outcomes on individuals within the same cluster are statistically independent, when in fact responses on individuals in the same community invariably tend to be more similar than responses on individuals in different communities. Moreover, the development of methods that take into account within-cluster dependencies becomes particularly challenging when a relatively small number of large communities are enrolled in the trial. This has led to the popularity of designs not frequently seen in large-scale clinical trials, such as pair-matching and repeated cross-sectional surveys.

In this chapter, we discuss a range of such issues, including the advantages and disadvantages of different study designs, methods of assuring adequate statistical power, and choice of analytic approach. Ethical issues arising from the need to obtain informed consent at both the cluster level and at the level of the individual are also discussed.

Chapter.  16715 words. 

Subjects: Public Health and Epidemiology ; Medical Statistics and Methodology ; Public Health

Full text: subscription required

How to subscribe Recommend to my Librarian

Buy this work at Oxford University Press »

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.