Skip to main content

Regression to the Mean

A Threat to Exercise Science?

Sports Medicine volume 33, pages 575–584 (2003)Cite this article

Abstract

Regression to the mean (RTM) can bias any investigation where the response to treatment is classified relative to initial values for a given variable without the use of an appropriate control group. The phenomenon and resulting errors of interpretation have been recognised by clinicians in a number of disciplines. The causes of RTM include both intra-individual variance and measurement error. The magnitude of RTM can be estimated quite simply, given a knowledge of intra- and inter-individual variance. RTM can be avoided by using a fully controlled experimental design. Difficulties can also be minimised by making duplicate measurements prior to the experimental manipulation, the first measurement serving for classification, and the second (with randomly distributed variance) allowing an assessment of the response to treatment. Less satisfactorily, surrogate measurements (for example, plasma volume for maximal oxygen intake [V̇O2max]) can assess the bias introduced by an initial non-random sorting of study participants. The impact of RTM on the design and interpretation of investigations has as yet received little consideration by exercise scientists and sports physicians. The response to training is often related to initial measurements of a dependent variable such as heart size, ST segmental depression, fitness or level of physical activity. In particular, analyses of this type have been adduced to support the belief that the response to aerobic training is inversely related to an individual’s V̇O2max. In fact, RTM may account for a major part of this apparent relationship.

This is a preview of subscription content, access via your institution.

Access options

Buy single article

Instant access to the full article PDF.

49,54 €

Price includes VAT (Finland)
Tax calculation will be finalised during checkout.

Table I
Fig. 1

References

  1. Galton F. Regression towards mediocrity in hereditary stature. J R Anthropol Inst 1886; 15: 246–63

    Google Scholar 

  2. Bland JM, Altman DG. Regression towards the mean. BMJ 1994; 308: 1499

    PubMed  Article  CAS  Google Scholar 

  3. Bland JM, Altman DG. Some examples of regression to the mean. BMJ 1994; 309: 780

    PubMed  Article  CAS  Google Scholar 

  4. Chen S, Cox C. Use of baseline data for estimation of treatment effects in the presence of regression to the mean. Biometrics 1992; 48: 593–8

    PubMed  Article  CAS  Google Scholar 

  5. Chen S, Cox C, Cui L. A more flexible regression-to-the-mean model with possible stratification. Biometrics 1998; 54: 939–47

    PubMed  Article  CAS  Google Scholar 

  6. Chesher A. Non-normal variation and regression to the mean. Stat Methods Med Res 1997; 6: 147–66

    PubMed  Article  CAS  Google Scholar 

  7. Chester MR, Bland M, Chen L, et al. The relationship between change and initial value: the continuing problem of regression to the mean. Eur Heart J 1995; 16: 289–90

    PubMed  CAS  Google Scholar 

  8. Chinn S, Heller RF. Some further results concerning regression to the mean. Am J Epidemiol 1981; 114: 902–5

    PubMed  CAS  Google Scholar 

  9. Chuang-Stein C, Tong DM. The impact and the implication of regression to the mean on the design and analysis of medical investigations. Stat Methods Med Res 1997; 6: 115–28

    PubMed  Article  CAS  Google Scholar 

  10. Cole TJ. Commentary: beware regression to the mean. BMJ 2000; 321: 276–81

    Article  Google Scholar 

  11. Copas JB. Using regression models for prediction: shrinkage and regression to the mean. Stat Methods Med Res 1997; 6: 167–83

    PubMed  Article  CAS  Google Scholar 

  12. Curnow RN. Correcting for regression in assessing the response to treatment in a selected population. Stat Med 1987; 6: 113–7

    PubMed  Article  CAS  Google Scholar 

  13. Davis CE. The effect of regression to the mean in epidemiologic and clinical studies. Am J Epidemiol 1976; 104: 493–8

    PubMed  CAS  Google Scholar 

  14. Fitzmaurice G. Regression to the mean. Nutrition 2000; 16: 80–1

    PubMed  CAS  Google Scholar 

  15. Gardner MJ, Hardy JA. Some effects of within person variability in epidemiological studies. J Chronic Dis 1973; 26: 781–95

    Article  Google Scholar 

  16. Hayes RJ. Methods for assessing whether change depends on initial value. Stat Med 1988; 7: 915–27

    PubMed  Article  CAS  Google Scholar 

  17. Healy MJ, Goldstein H. Regression to the mean. Ann Hum Biol 1978; 5: 277–80

    PubMed  Article  CAS  Google Scholar 

  18. Jamieson J. The law of initial values: five factors or two? Int J Psychophysiol 1993; 14: 233–9

    PubMed  Article  CAS  Google Scholar 

  19. Johnson WD, George VT. Effect of regression to the mean in the presence of within-subject variability. Stat Med 1991; 10: 1295–302

    PubMed  Article  CAS  Google Scholar 

  20. McDonald CJ, Mazzuca SA, McCabe GP. How much of the ‘placebo’ effect is really statistical regression? Stat Med 1983; 2: 417–27

    PubMed  Article  CAS  Google Scholar 

  21. Ross DC. Using severity rather than randomization for assignment in clinical trials: Monte Carlo tests and practical illustrations of the Robbins-Zhang method. J Psychiatr Res 1995; 29: 315–32

    PubMed  Article  CAS  Google Scholar 

  22. Senn SJ, Brown RA. Estimating treatment effects in clinical trials subject to regression to the mean. Biometrics 1985; 41: 555–60

    PubMed  Article  CAS  Google Scholar 

  23. Senn S. Regression to the mean and crossover trials revisited. Stat Med 1994; 13: 1181–6

    PubMed  Article  CAS  Google Scholar 

  24. Sharp SJ, Thompson SG, Altman DG. The relation between treatment benefit and underlying risk in meta-analysis. BMJ 1996; 313: 735–8

    PubMed  Article  CAS  Google Scholar 

  25. McDonald IG, Daly J, Jelinek VM, et al. Opening Pandora’s box: the unpredictability of reassurance by a normal test result. BMJ 1996; 313: 329–32

    PubMed  Article  CAS  Google Scholar 

  26. Asmar R, Safar M, Oueneau P. Evaluation of the placebo effect and reproducibility of blood pressure measurement in hypertension. Am J Hypertens 2001; 14 (6 Pt 1): 546–52

    PubMed  Article  CAS  Google Scholar 

  27. Wolfson M, Kaplan G, Lynch J, et al. Relation between income equality and mortality: empirical demonstration. BMJ 1999; 319: 953–7

    PubMed  Article  CAS  Google Scholar 

  28. Beebe JC. Medicare reimbursement and regression to the mean. Health Care Financ Rev 1988; 9: 9–22

    PubMed  CAS  Google Scholar 

  29. Wright CC, Abbess CR, Jarrett DF. Estimating the regression-to-the-mean effect associated with road accident black spot treatment: towards a more realistic approach. Accid Anal Prev 1988; 20: 199–214

    PubMed  Article  CAS  Google Scholar 

  30. Lambrechtsen J, Rasmussen F, Hansen HS, et al. Tracking and factors predicting rising in ‘tracking quartile’ in blood pressure from childhood to adulthood: Odense Schoolchild Study. J Hum Hypertens 1999; 13: 385–91

    PubMed  Article  CAS  Google Scholar 

  31. Pitts SR, Adams RP. Emergency department hypertension and regression to the mean. Ann Emerg Med 1998; 31: 214–8

    PubMed  Article  CAS  Google Scholar 

  32. Shepard DS, Finison LJ. Blood pressure reductions: correcting for regression to the mean. Prev Med 1983; 12: 304–17

    PubMed  Article  CAS  Google Scholar 

  33. Zaborskis AA, Shachkute AA, Piatkiavichius RV. Value of the regression to the mean effect in the evaluation of arterial blood pressure in adolescents in epidemiological studies [in Russian]. Kardiologia 1985; 25: 56–61

    CAS  Google Scholar 

  34. Murakawa Y, Yamashita T, Ajiki K, et al. Ostensible day-night difference of QT prolongation during long-term treatment with antiarrhythmic drugs: reappraisal of the law of ‘regression to the mean’. J Cardiovasc Pharmacol 1998; 32: 62–5

    PubMed  Article  CAS  Google Scholar 

  35. Boissel JP, Duperat B, Leizorowicz A. The phenomenon of regression to the mean and clinical investigation of blood cholesterol lowering drugs. Eur J Clin Pharmacol 1980; 17: 227–30

    PubMed  Article  CAS  Google Scholar 

  36. Denke MA, Frantz ID. Responses to a cholesterol-lowering diet: efficacy is greater in hypercholesterolemic subjects even after adjustment for regression to the mean. Am J Med 1993; 94: 626–31

    PubMed  Article  CAS  Google Scholar 

  37. Forrow L, Calkins DR, Allshouse K, et al. Evaluating cholesterol screening: the importance of controlling for regression to the mean. Arch Intern Med 1995; 155: 2177–84

    PubMed  Article  CAS  Google Scholar 

  38. Takashima Y, Sumiya Y, Kokaze A, et al. Magnitude of the regression to the mean within one-year intra-individual changes in serum lipid levels among Japanese male workers. J Epidemiol 2001; 11: 61–9

    PubMed  Article  CAS  Google Scholar 

  39. Cummings SR, Palermo L, Browner W, et al. Monitoring osteoporosis therapy with bone densitometry: misleading changes and regression to the mean. JAMA 2000; 283: 1318–21

    PubMed  Article  CAS  Google Scholar 

  40. Prescott RJ, Garraway WM. Regression to the mean occurs in measuring peak urinary flow. Br J Urol 1995; 76: 611–3

    PubMed  Article  CAS  Google Scholar 

  41. Moller K, Dirksen A. Mathematical coupling and ‘regression to the mean’: a statistical case history [in Danish]. Ugesk Laeger 1999; 161: 5325–6

    CAS  Google Scholar 

  42. Andrews G, Harvey R. Regression to the mean in pretreatment measures of stuttering. J Speech Hear Disord 1981; 46: 204–7

    PubMed  CAS  Google Scholar 

  43. Spector R, Park GD. Regression to the mean: a potential source of error in clinical pharmacological studies. Drug Intell Clin Pharm 1985; 19: 916–9

    PubMed  CAS  Google Scholar 

  44. Herpin D, Demange J. Effect of regression to the mean in serial echocardiographic measurements of left ventricular mass: quantification and clinical implications. Am J Hypertens 1994; 7: 824–8

    PubMed  CAS  Google Scholar 

  45. Stewart RA, Kittelson J, Kay JP. Statistical methods to improve the precision of the treadmill exercise test. J Am Coll Cardiol 2000; 36: 1274–9

    PubMed  Article  CAS  Google Scholar 

  46. Trochim WMK. Regression to the mean [online]. Available from URL http://www.trochim.omni.cornell.edu/kb/regrmean.htm [Accessed 2002 Apr 17]

  47. Kuskowska-Wolk A, Karlsson P, Stolt M, et al. The predictive value of body mass index based on reported weight and height. Int J Obes 1989; 13: 441–3

    PubMed  CAS  Google Scholar 

  48. Schilchting P, Hoilund-Carlsen QF. Comparison of self-reported height and weight with controlled height and weight in women and men. Int J Obes 1981; 5: 67–76

    Google Scholar 

  49. Leon AS, Gaskill SE, Rice T, et al. Variability in the response of HDL cholesterol to exercise training in the HERITAGE Family Study. Int J Sports Med 2002; 23: 1–9

    PubMed  Article  CAS  Google Scholar 

  50. Huang Y, Macera CA, Blair SN, et al. Physical fitness, physical activity, and functional limitation in adults aged 40 and older. Med Sci Sports Exerc 1998; 30: 1430–5

    PubMed  CAS  Google Scholar 

  51. Blair SN, Kohl HW, Paffenbarger RS, et al. Physical fitness and all-cause mortality: a prospective study of healthy men and women. JAMA 1990; 262: 2395–401

    Article  Google Scholar 

  52. Blair SN, Kohl HW, Barlow CE. Physical activity, physical fitness, and all-cause mortality in women: do women need to be active? J Am Coll Nutr 1993; 12: 368–71

    PubMed  CAS  Google Scholar 

  53. Blair SN, Kampert JB, Kohl HW, et al. Influences of cardiorespiratory fitness and other precursors on cardiovascular disease and all-cause mortality in men and women. JAMA 1996; 276: 205–10

    PubMed  Article  CAS  Google Scholar 

  54. Church TS, Kampert JB, Gibbons LW, et al. Usefulness of cardiorespiratory fitness as a predictor of all-cause and cardiovascular disease mortality in men with systemic hypertension. Am J Cardiol 2001; 88: 651–6

    PubMed  Article  CAS  Google Scholar 

  55. Farrell SW, Kampert JB, Kohl HW, et al. Influences of cardiorespiratory fitness levels and other predictors on cardiovascular disease mortality in men. Med Sci Sports Exerc 1998; 30: 899–905

    PubMed  Article  CAS  Google Scholar 

  56. Lee CD, Blair SN, Jackson AS. Cardiorespiratory fitness, body composition, and all-cause and cardiovascular disease mortality in men. Am J Clin Nutr 1999; 69: 373–80

    PubMed  CAS  Google Scholar 

  57. Wei M, Kampert JB, Barlow CE, et al. Relationship between low cardiorespiratory fitness and mortality in normal-weight, overweight, and obese men. JAMA 1999; 282: 1547–53

    PubMed  Article  CAS  Google Scholar 

  58. Wei M, Gibbons LW, Kampert JB, et al. Low cardiorespiratory fitness and physical inactivity as predictors of mortality in men with type 2 diabetes. Ann Intern Med 2000; 132: 605–11

    PubMed  CAS  Google Scholar 

  59. Brodney S, McPherson RS, Carpenter RS, et al. Nutrient intake of physically fit and unfit men and women. Med Sci Sports Exerc 2001; 33: 459–67

    PubMed  Article  CAS  Google Scholar 

  60. Hootman JM, Macera CA, Ainsworth BE, et al. Association among physical activity level, cardiorespiratory fitness, and risk of musculoskeletal injury. Am J Epidemiol 2001; 154: 251–8

    PubMed  Article  CAS  Google Scholar 

  61. Kampert JB, Blair SN, Barlow CE, et al. Physical activity, physical fitness, and all-cause cancer mortality: a prospective study of men and women. Ann Epidemiol 1996; 6: 452–7

    PubMed  Article  CAS  Google Scholar 

  62. Barlow CE, Kohl HW, Gibbons LW, et al. Physical fitness, mortality and obesity. Int J Obes 1995; 19Suppl. 4: S41–4

    Google Scholar 

  63. Blair SN, Kohl HW, Barlow CE, et al. Changes in physical fitness and all-cause mortality: a prospective study of healthy and unhealthy men. JAMA 1995; 273: 1093–8

    PubMed  Article  CAS  Google Scholar 

  64. Shephard RJ. Aerobic fitness and health. Champaign (IL): Human Kinetics Publishers, 1994

    Google Scholar 

  65. Shephard RJ. Intensity, duration and frequency of exercise as determinants of the response to a training regimen. Int Z Angew Physiol 1968; 26: 272–8

    PubMed  CAS  Google Scholar 

  66. Wenger HA, Bell GJ. The interactions of intensity, frequency and duration of exercise training in altering cardiorespiratory fitness. Sports Med 1986; 3: 346–56

    PubMed  Article  CAS  Google Scholar 

  67. Shephard RJ, Hamm L, Mertens DJ, et al. Influence of initial fitness on the response to cardiac rehabilitation [abstract]. Med Sci Sports Exerc 2002; 34Suppl. 5: S181

    Google Scholar 

  68. Mandigout S, Lecoq AM, Courteix D, et al. Effect of gender in response to an aerobic training programme in prepubertal children. Acta Paediatr 2001; 90: 9–15

    PubMed  Article  CAS  Google Scholar 

  69. Epstein LH, Wing RR, Valoski A, et al. Long-term relationship between weight and aerobic fitness in children. Health Psychol 1988; 7: 47–53

    PubMed  Article  CAS  Google Scholar 

  70. Andersen LB. Changes in physical activity are reflected in changes in fitness during late adolescence: a 2-year follow-up study. J Sports Med Phys Fitness 1994; 34: 390–7

    PubMed  CAS  Google Scholar 

  71. Skinner JS, Jaskolski A, Jaskolska A, et al. Age, sex, initial fitness, and response to training: the HERITAGE study. J Appl Physiol 2001; 90: 1770–6

    PubMed  CAS  Google Scholar 

  72. Kohrt WM, Malley MT, Coggan AR, et al. Effects of gender, age and fitness level on response of V̇O2max to training in 60–71 yr olds. J Appl Physiol 1991; 71: 2004–11

    PubMed  CAS  Google Scholar 

  73. Swain DP, Franklin BA. VO(2) reserve and the minimal intensity for improving cardiorespiratory fitness. Med Sci Sports Exerc 2002; 34: 152–7

    PubMed  Article  Google Scholar 

  74. Convertino VA. Changes in peak oxygen uptake and plasma volume in fit and unfit subjects following exposure to a simulation of microgravity. Acta Physiol Scand 1998; 164: 251–7

    PubMed  Article  CAS  Google Scholar 

  75. Snedecor GW. Statistical methods. Ames (IA): Iowa State University Press, 1956

    Google Scholar 

  76. Graham TE, Andrew G. The variability of repeated measurements of oxygen debt in man following a maximal treadmill exercise. Med Sci Sports 1973; 5: 73–8

    PubMed  CAS  Google Scholar 

  77. Taylor HL, Buskirk ER, Henschel A. Maximal oxygen intake as an objective measure of the cardiorespiratory performance. J Appl Physiol 1955; 8: 73–80

    PubMed  CAS  Google Scholar 

  78. Wright GR, Sidney K, Shephard RJ. Variance of direct and indirect measurements of aerobic power. J Sports Med Phys Fitness 1978; 18: 33–42

    PubMed  CAS  Google Scholar 

  79. Wyndham CH, Strydom NB, Maritz JS, et al. Maximum oxygen intake and maximum heart rate during strenuous work. J Appl Physiol 1959; 14: 927–36

    PubMed  CAS  Google Scholar 

  80. Shephard RJ. Ischemic heart disease and exercise. London: Croom Helm Publishers, 1981

    Google Scholar 

  81. Capocaccia R, Mariotti S, Menotti A, et al. Evaluation of risk factor variations in relation to their baseline values in a controlled preventive trial: application to the Rome Project of CHD Prevention. J Chronic Dis 1982; 35: 509–20

    PubMed  Article  CAS  Google Scholar 

Download references

Acknowledgements

No sources of funding were used to assist in the preparation of this manuscript. The author has no conflicts of interest that are directly relevant to the content of this manuscript.

Author information

Affiliations

  1. Department of Public Health Sciences, Faculty of Physical Education & Health, and Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada

    Roy J. Shephard

Authors
  1. Roy J. Shephard
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Roy J. Shephard.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Shephard, R.J. Regression to the Mean. Sports Med 33, 575–584 (2003). https://doi.org/10.2165/00007256-200333080-00003

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.2165/00007256-200333080-00003

Keywords

  • Aerobic Training
  • Aerobic Fitness
  • Isosorbide Dinitrate
  • Training Response
  • Maximal Aerobic Power