Elsevier

The Lancet

Volume 347, Issue 8996, 27 January 1996, Pages 241-243
The Lancet

Statistics
How to deal with regression to the mean in intervention studies

https://doi.org/10.1016/S0140-6736(96)90410-9Get rights and content

First page preview

First page preview
Click to open first page preview

References (23)

  • Uk Prospective Diabetes Study Group

    UK Prospective Diabetes Study (UKPDS) VIII: study design, progress and performance

    Diabetologia

    (1991)
  • Cited by (169)

    • Personalized redox biology: Designs and concepts

      2023, Free Radical Biology and Medicine
    • Placebo response in Raynaud's Phenomenon clinical trials: The prominent role of regression towards the mean: Placebo response in Raynaud's Phenomenon

      2022, Seminars in Arthritis and Rheumatism
      Citation Excerpt :

      Although “true” placebo effect may occur, our results clearly show that RTM exists in RP trials, and simulations suggest that its magnitude may explain a large part of the placebo response in RP trials. If allocation to treatment and placebo arms (or periods in the case of a crossover trial) is properly randomized, RTM should equally affect both groups [12]. Yet, this finding may have an impact on the design of future clinical trials.

    • Pickwell’s Binocular Vision Anomalies

      2021, Pickwell's Binocular Vision Anomalies
    • Corroborative evidence for an association between initial hypothalamic-pituitary-adrenocortical axis reactivity and subsequent habituation in humans

      2020, Psychoneuroendocrinology
      Citation Excerpt :

      Moreover, by virtue of avoiding the use of change scores, such relationships cannot be attributed to mathematical coupling. Yet, analyses reported as part of Study 1 could still be the product of imperfect correlation between repeated measurements (i.e., regression to the mean; Barnett et al., 2005; Stigler, 1997; Yudkin and Stratton, 1996), and measurement error. Nevertheless, results from a previous simulation study (Chiolero et al., 2013) stand as evidence that the present analytic approach can yield an unbiased estimation of the true relationship between initial value and subsequent change.

    View all citing articles on Scopus
    View full text