5. Resolving the Paradox of Effectiveness: When do Observational Studies Offer the Same Degree of Evidential Support as Randomized Trials?

  1. Jeremy Howick1,2

Published Online: 23 MAR 2011

DOI: 10.1002/9781444342673.ch5

The Philosophy of Evidence-Based Medicine

The Philosophy of Evidence-Based Medicine

How to Cite

Howick, J. (2011) Resolving the Paradox of Effectiveness: When do Observational Studies Offer the Same Degree of Evidential Support as Randomized Trials?, in The Philosophy of Evidence-Based Medicine, Wiley-Blackwell, Oxford, UK. doi: 10.1002/9781444342673.ch5

Author Information

  1. 1

    Centre for Evidence-Based Medicine, Department of Public Health and Primary Care, University of Oxford, Oxford, UK

  2. 2

    Department of Science and Technology Studies, University College London, London, UK

Publication History

  1. Published Online: 23 MAR 2011
  2. Published Print: 8 APR 2011

ISBN Information

Print ISBN: 9781405196673

Online ISBN: 9781444342673

SEARCH

Keywords:

  • resolving paradox of effectiveness - observational studies, same degree of evidential support as randomized trials;
  • paradox of effectiveness - EBM “hierarchies” of evidence, ranking randomized trials categorically above observational studies;
  • critiques of randomized trials, like straw men against EBM position - EBM view modified to overcome paradox of effectiveness;
  • observational designs, as case studies, case series, case–control studies, cohort studies and historically controlled studies;
  • careful adjusting, reducing confounding - quality of observational studies;
  • randomized trials, differing from observational studies - random allocation and experimental administration of treatments;
  • randomization, adhered to - ruling out self-selection bias and allocation bias;
  • trial, experimental design - and ruling out performance bias;
  • EBM view, randomized trials - better evidence than observational studies;
  • randomized trials, ruling out more confounding factors - than observational studies

Summary

This chapter contains sections titled:

  • The paradox of effectiveness

  • Observational studies: definition and problems

  • Randomized trials to the rescue

  • Defending the EBM view that randomized trials provide better evidence than observational studies

  • Overcoming the paradox of effectiveness

  • Conclusion: a more subtle way to distinguish between high- and low-quality comparative clinical studies

  • Appendix 1: types of restricted randomization

  • Appendix 2: Worrall's arguments that randomization is required for classical hypothesis testing and establishing probabilistic causes