This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Landow, L. Search for Related Content PubMed PubMed Citation Articles by Landow, L.

---

LETTERS TO THE EDITOR

Another Example of Regression to the Mean

Bethesda, MD

To the Editor:

Using multiple regression analysis, Robson et al. (1) were unable to show a significant correlation between jugular bulb oxyhemoglobin desaturation and neurologic/cognitive outcomes at 3 mo in patients undergoing coronary artery surgery. Although this was a noninterventional trial, the authors neglected to address a serious flaw in study design that has been identified previously in this (2) and other settings (3): regression to the mean (RTM).

RTM is a statistical finding masquerading as a real one (4). To illustrate, consider a population selected at random in whom systemic blood pressure is measured at time zero and again at 3 mo. If the individuals found to be hypertensive are prescribed medical treatment, their blood pressure at 3 mo will be closer to the mean of the entire population–even if they fail to take their medication–owing to RTM (5,6). Individuals with the highest values at time zero will show the greatest reductions, whereas the decline in those with values closer to the mean will be much smaller. The converse is observed when initial blood pressures are less than the population mean.

Yudkin and Stratton (4) propose some approaches to minimizing RTM, such as randomized controlled trials (in which any difference in response between the two groups will exclude RTM) and basing selection on the mean of several (e.g., four) measurements at each time point, rather than a single value.

Future studies exploring neurological outcome after open-heart surgery should address RTM in their design.

Footnotes

Dr. Robson did not respond.

References

1. Robson MJ, Alston RP, Deary IJ, et al. Jugular bulb oxyhemoglobin desaturation, S100beta, and neurologic and cognitive outcomes after coronary artery surgery. Anesth Analg 2001; 93: 839–45.[Abstract/Free Full Text]
2. Browne SM, Halligan PW, Wade DT, Taggart DP. Cognitive performance after cardiac operation: implications of regression toward the mean. J Thorac Cardiovasc Surg 1999; 117: 481–5.[Abstract/Free Full Text]
3. Bland JM, Altman DG. Statistic notes: regression towards the mean. BMJ 1994; 308: 1499.[Free Full Text]
4. Yudkin PL, Stratton IM. How to deal with regression to the mean in intervention studies. Lancet 1996; 347: 241–3.[Web of Science][Medline]
5. Hayes RJ. Methods for assessing whether change depends on initial value. Stat Med 1988; 7: 915–27.[Web of Science][Medline]
6. Bland JM, Altman DG. Statistics notes: some examples of regression towards the mean. BMJ 1994; 309: 780.[Free Full Text]

This article has been cited by other articles:

				R. P. Alston, I. J. Deary, M. J. Robson, P. J. Andrews, and M. J. Souter Another Example of Regression to the Mean (Not) Anesth. Analg., December 1, 2002; 95(6): 1823 - 1823. [Full Text] [PDF]

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Landow, L. Search for Related Content PubMed PubMed Citation Articles by Landow, L.