Skip Navigation


American Journal of Epidemiology Advance Access originally published online on July 13, 2005
American Journal of Epidemiology 2005 162(5):415-418; doi:10.1093/aje/kwi224
This Article
Right arrow Extract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
162/5/415    most recent
kwi224v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Petitti, D. B.
Right arrow Articles by Freedman, D. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Petitti, D. B.
Right arrow Articles by Freedman, D. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2005 by the Johns Hopkins Bloomberg School of Public Health All rights reserved

ORIGINAL CONTRIBUTIONS

Invited Commentary: How Far Can Epidemiologists Get with Statistical Adjustment?

Diana B. Petitti1 and David A. Freedman2

1 Kaiser Permanente Southern California, Pasadena, CA
2 Department of Statistics, University of California, Berkeley, CA

Correspondence to Dr. Diana B. Petitti, Kaiser Permanente Southern California, 393 E. Walnut Street, Pasadena, CA 91188 (e-mail: diana.b.petitti{at}kp.org).

Received for publication March 22, 2005. Accepted for publication April 29, 2005.

Abbreviations: WHI, Women's Health Initiative


   INTRODUCTION
TOP
 INTRODUCTION
TRADITIONAL CONFOUNDERS
 TIME COURSE OF EXPOSURE
 FURTHER INFORMATION ON STROKE...
 CONCLUSION
 References
 
In 2002, the Women's Health Initiative (WHI) clinical trial reported that combined estrogen-plus-progestin hormone therapy did not prevent coronary heart disease in women (1Go). Combined estrogen-plus-progestin therapy increased the risk of stroke by a factor of 1.4 on average and doubled the risk of venous thromboembolism.

Observational research up to the time of the WHI suggested that the relative risk of coronary heart disease was 0.50–0.65 in hormone users compared with nonusers (2Go, 3Go). Based on observational research, the relative risk of stroke was considered less than or near 1.0 for hormone therapy (2Go). Both observational studies (4Go) and a randomized trial, the Heart Estrogen/progestin Replacement Study (HERS) (5Go), showed a relative risk of 1–2 for venous thromboembolism in hormone users.

The WHI clinical trial joined a body of experimental evidence showing no effect of postmenopausal hormone therapy on coronary heart disease clinical endpoints or measures of subclinical coronary atherosclerosis (6Go–16Go). In 1998, the Heart Estrogen/progestin Replacement Study reported no beneficial effect of hormone therapy on morbidity or mortality from coronary heart disease in women with established coronary disease (6Go). Seven other randomized trials of the effect of hormone therapy on arterial disease endpoints published in 2000–2002 found no effect of hormone therapy (7Go–13Go). A 1997 meta-analysis of small randomized trials of hormone replacement found that hormone therapy did not prevent cardiovascular disease (14Go). Follow-up of participants in the Heart Estrogen/progestin Replacement Study (15Go) showed no benefit of hormone therapy for any cardiovascular endpoint. The estrogen-only arm of the WHI clinical trial was terminated early after showing that hormones did not prevent coronary heart disease (16Go).

In this issue, Prentice et al. (17Go) report an analysis of data from the WHI clinical trial and the WHI observational study. The WHI clinical trial and observational study were conducted by the same group of researchers, in the same time frame, and had similar procedures for ascertaining and monitoring events. Prentice et al. explore the reasons for the differences in estimates of coronary heart disease, stroke, and venous thromboembolism risk for hormones between the WHI clinical trial and the observational study. They attempt to resolve the discrepancies by statistical adjustment.


   TRADITIONAL CONFOUNDERS
TOP
 INTRODUCTION
 TRADITIONAL CONFOUNDERS
TIME COURSE OF EXPOSURE
 FURTHER INFORMATION ON STROKE...
 CONCLUSION
 References
 
Table 1 below shows how adjustments of the WHI clinical trial data and observational study data affect hazard ratios for coronary heart disease, stroke, and venous thromboembolism. Given the differences in age between hormone users and nonusers in the WHI observational study, it is not surprising that age adjustment has a large effect on hazard ratios for all three endpoints. Adjustment for age is standard in observational research. After adjustment for other confounders, including known cardiovascular risk factors and education, the hazard ratios for coronary heart disease, stroke, and venous thromboembolism for the WHI observational study all move closer to those from the WHI clinical trial.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Unadjusted ratio of incidence rates and adjusted hazard ratios for coronary heart disease, stroke, and venous thromboembolism in the Women's Health Initiative clinical trial (1994–2002) and the Women's Health Initiative observational study (1994–2003)

 
Prentice et al. (17Go) point out that education was related strongly to coronary heart disease risk in the WHI. Education is a surrogate for socioeconomic status. Inclusion of a measure of socioeconomic status as a potential confounder in the analysis of observational data for virtually any exposure/disease relations seems fundamental. Humphrey et al. (18Go) reported that only three of 10 observational studies examining coronary heart disease risk in relation to current hormone therapy adjusted for a measure of socioeconomic status. A summary estimate of the relative risk of coronary heart disease in current hormone users based only on studies that adjusted for a measure of socioeconomic status was 0.97 (95 percent confidence interval: 0.82, 1.16); when based on studies that did not adjust for a measure of socioeconomic status, the relative risk estimate was 0.71 (95 percent confidence interval: 0.64, 0.78). Reports on coronary heart disease risk from the most influential observational study of hormone therapy—the Nurses' Health Study—did not adjust for socioeconomic status (19Go–25Go).

It has been argued that adjustment for socioeconomic status is not necessary in studies that enroll subjects who are homogeneous (26Go). Routine presentation of results adjusted for socioeconomic status would bolster this kind of argument. However, the argument raises important questions. Why did it take so long to recognize that many studies failed to adjust for socioeconomic status? Why were studies that adjusted for socioeconomic status disregarded by most observers?


   TIME COURSE OF EXPOSURE
TOP
 INTRODUCTION
 TRADITIONAL CONFOUNDERS
 TIME COURSE OF EXPOSURE
FURTHER INFORMATION ON STROKE...
 CONCLUSION
 References
 
Prentice et al. (17Go) developed a model that adjusts for traditional covariates and takes into account the time course of exposure, including the time since initiation of use in the most recent episode of use. The model brings the WHI observational data and the WHI clinical trial data into statistical agreement for two of the three endpoints considered, namely, coronary heart disease and venous thromboembolism. Although the hazard ratios are in statistical agreement, some point estimates are substantively different (table 1). Prentice et al. continued to try different adjustments, because the WHI clinical trial results showed that the WHI observational study results were still incorrect after the initial set of adjustments was complete. Without the clinical trial, it would not be clear when to stop adjusting or whether prior adjustments were successful. As stated by Prentice et al., after adjustment for traditional confounders (including education as a measure of socioeconomic status), the WHI clinical trial found a hazard ratio for stroke of 1.21, whereas the observational study found a hazard ratio of 0.86 for stroke. For stroke, Prentice et al. state that the hazard ratios in the observational study compared with the clinical trial differed significantly (p < 0.01) even after further adjustment for the time course of estrogen-plus-progestin hormone therapy.


   FURTHER INFORMATION ON STROKE AND CONCERN ABOUT DEMENTIA
TOP
 INTRODUCTION
 TRADITIONAL CONFOUNDERS
 TIME COURSE OF EXPOSURE
 FURTHER INFORMATION ON STROKE...
CONCLUSION
 References
 
Both a nonsystematic review and a recent systematic review and meta-analysis of clinical trial data concluded that hormone therapy increases the risk of stroke (27Go, 28Go). Stroke is overlooked as a major cause of morbidity and mortality in women, and its frequency relative to coronary heart disease is underappreciated. The relative contributions of coronary heart disease and stroke to arterial disease in the WHI—roughly equal—are typical for women aged 55–70 years living in the United States now.

In contrast with a number of observational studies (29Go–36Go), which found substantial reductions in the risk of dementia in relation to hormone therapy, the Women's Health Initiative Memory Study (WHIMS) found that hormone therapy increased the risk of dementia by a factor of about two (37Go, 38Go). This effect may be related to subclinical brain infarction (37Go–39Go). The long-term effects of hormones on the risk of dementia remain unknown.


   CONCLUSION
TOP
 INTRODUCTION
 TRADITIONAL CONFOUNDERS
 TIME COURSE OF EXPOSURE
 FURTHER INFORMATION ON STROKE...
 CONCLUSION
References
 
The analysis by Prentice et al. (17Go) confirms that there is no overall benefit of hormone therapy in preventing cardiovascular disease. Randomized trials of hormone therapy changed the practice of medicine. Hormone therapy should not be used to prevent cardiovascular disease. The conclusion that hormone therapy prevents cardiovascular disease, which was based on observational research, was wrong. Possible explanations for the error include the apparent consistency of the cohort studies and a belief that the estimated size of the effect was too large to be explained by confounding.

Most of what we know about causation from John Snow to the present comes from observational studies. Observational studies will remain a staple in the field. Prentice et al. remind us to pay more attention to fundamentals in observational research. They highlight some ways that observational studies could be better designed and better analyzed. However, observational studies are not a substitute for clinical trials no matter how sophisticated the statistical adjustments may seem.


   ACKNOWLEDGMENTS
 
Conflict of interest: none declared.


   References
TOP
 INTRODUCTION
 TRADITIONAL CONFOUNDERS
 TIME COURSE OF EXPOSURE
 FURTHER INFORMATION ON STROKE...
 CONCLUSION
 References
 

  1. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321–33.[Abstract/Free Full Text]
  2. Grady D, Rubin SM, Petitti DB, et al. Hormone replacement to prevent disease and prolong life in postmenopausal women. Ann Intern Med 1992;117:1016–37.[ISI][Medline]
  3. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47–63.[CrossRef][ISI][Medline]
  4. Grady D, Wenger NK, Herrington D, et al. Postmenopausal hormone therapy increases risk for venous thromboembolic disease. The Heart and Estrogen/progestin Replacement Study. Ann Intern Med 2000;132:689–96.[Abstract/Free Full Text]
  5. Grodstein F, Stampfer MJ, Goldhaber SZ, et al. Prospective study of exogenous hormones and risk of pulmonary embolism in women. Lancet 1996;348:983–7.[CrossRef][ISI][Medline]
  6. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. JAMA 1998;280:605–13.[Abstract/Free Full Text]
  7. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N Engl J Med 2000;343:522–9.[Abstract/Free Full Text]
  8. Angerer P, Stork S, Kothny W, et al. Effect of oral postmenopausal hormone replacement on progression of atherosclerosis: a randomized, controlled trial. Arterioscler Thromb Vasc Biol 2001;21:262–8.[Abstract/Free Full Text]
  9. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 2001;135:939–53.[Abstract/Free Full Text]
  10. Cherry N, Gilmour K, Hannaford P, et al. Oestrogen therapy for prevention of reinfarction in postmenopausal women: a randomised placebo controlled trial. Lancet 2002;360:2001–8.[CrossRef][ISI][Medline]
  11. Byington RP, Furberg CD, Herrington DM, et al. Effect of estrogen plus progestin on progression of carotid atherosclerosis in postmenopausal women with heart disease: HERS B-mode substudy. Arterioscler Thromb Vasc Biol 2002;22:1692–7.[Abstract/Free Full Text]
  12. Waters DD, Alderman EL, Hsia J, et al. Effects of hormone replacement therapy and antioxidant vitamin supplements on coronary atherosclerosis in postmenopausal women: a randomized controlled trial. JAMA 2002;288:2432–40.[Abstract/Free Full Text]
  13. Clarke SC, Kelleher J, Lloyd-Jones H, et al. A study of hormone replacement therapy in postmenopausal women with ischaemic heart disease: the Papworth HRT atherosclerosis study. BJOG 2002;109:1056–62.[ISI][Medline]
  14. Hemminki E, McPherson K. Impact of postmenopausal hormone therapy on cardiovascular events and cancer: pooled data from clinical trials. BMJ 1997;315:149–53.[Abstract/Free Full Text]
  15. Grady D, Herrington D, Bittner V, et al. Cardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/progestin Replacement Study follow-up (HERS II). JAMA 2002;288:49–57.[Abstract/Free Full Text]
  16. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMA 2004;291:1701–12.[Abstract/Free Full Text]
  17. Prentice RL, Langer R, Stefanick ML, et al. Combined postmenopausal hormone therapy and cardiovascular disease: toward resolving the discrepancy between observational studies and the Women's Health Initiative clinical trial. Am J Epidemiol 2005;162:404–14.[Abstract/Free Full Text]
  18. Humphrey LL, Chan BK, Sox HC. Postmenopausal hormone replacement therapy and the primary prevention of cardiovascular disease. Ann Intern Med 2002;137:273–84.[Abstract/Free Full Text]
  19. Bain C, Willett W, Hennekens CH, et al. Use of postmenopausal hormones and risk of myocardial infarction. Circulation 1981;64:42–6.[Abstract/Free Full Text]
  20. Stampfer MJ, Willett WC, Colditz GA, et al. A prospective study of postmenopausal estrogen therapy and coronary heart disease. N Engl J Med 1985;313:1044–9.[Abstract]
  21. Stampfer MJ, Colditz GA, Willett WC, et al. Postmenopausal estrogen therapy and cardiovascular disease. Ten-year follow-up from the Nurses' Health Study. N Engl J Med 1991;325:756–62.[Abstract]
  22. Grodstein F, Stampfer MJ, Manson JE, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N Engl J Med 1996;335:453–61.[Abstract/Free Full Text]
  23. Grodstein F, Stampfer MJ, Colditz GA, et al. Postmenopausal hormone therapy and mortality. N Engl J Med 1997;336:1769–75.[Abstract/Free Full Text]
  24. Grodstein F, Manson JE, Colditz GA, et al. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med 2000;133:933–41.[Abstract/Free Full Text]
  25. Grodstein F, Manson JE, Stampfer MJ. Postmenopausal hormone use and secondary prevention of coronary events in the Nurses' Health Study. A prospective, observational study. Ann Intern Med 2001;135:1–8.[Abstract/Free Full Text]
  26. Stampfer M. Commentary: hormones and heart disease: do trials and observational studies address different questions? Int J Epidemiol 2004;33:454–5.[Free Full Text]
  27. Beral V, Banks E, Reeves G. Evidence from randomized trials on long-term effects of hormone replacement therapy. Lancet 2002;360:942–4.[CrossRef][ISI][Medline]
  28. Bath PM, Gray LJ. Association between hormone replacement therapy and subsequent stroke: a meta-analysis. BMJ 2005;330:342–5.[Abstract/Free Full Text]
  29. Mortel KF, Meyer JS. Lack of postmenopausal estrogen replacement therapy and the risk of dementia. J Neuropsychiatry Clin Neurosci 1995;7:334–7.[Abstract/Free Full Text]
  30. Paganini-Hill A, Henderson VW. Estrogen replacement therapy and risk of Alzheimer disease. Arch Intern Med 1996;156:2213–17.[Abstract]
  31. Tang MX, Jacobs D, Stern Y, et al. Effect of oestrogen during menopause on risk and age at onset of Alzheimer's disease. Lancet 1996;348:429–32.[CrossRef][ISI][Medline]
  32. Kawas C, Resnick S, Morrison A, et al. A prospective study of estrogen replacement therapy and the risk of developing Alzheimer's disease: the Baltimore Longitudinal Study of Aging. Neurology 1997;48:1517–21.[Abstract]
  33. Lerner A, Koss E, Debanne S, et al. Smoking and oestrogen-replacement therapy as protective factors for Alzheimer's disease. Lancet 1997;349:403–4.[ISI][Medline]
  34. Baldereschi M, Di Carlo A, Lepore V, et al. Estrogen-replacement therapy and Alzheimer's disease in the Italian Longitudinal Study on Aging. Neurology 1998;50:996–1002.[Abstract/Free Full Text]
  35. Slooter AJ, Bronzova J, Witteman JC, et al. Estrogen use and early onset Alzheimer's disease: a population-based study. J Neurol Neurosurg Psychiatry 1999;67:779–81.[Abstract/Free Full Text]
  36. Waring SC, Rocca WA, Petersen RC, et al. Postmenopausal estrogen replacement therapy and risk of AD: a population-based study. Neurology 1999;52:965–70.[Abstract/Free Full Text]
  37. Shumaker SA, Legault C, Rapp SR, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651–62.[Abstract/Free Full Text]
  38. Shumaker SA, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women's Health Initiative Memory Study. JAMA 2004;291:2947–58.[Abstract/Free Full Text]
  39. Yaffe K. Hormone therapy and the brain: déjà vu all over again? JAMA 2003;289:2717–19.[Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
M.-N. Vercambre, A. Fournier, M.-C. Boutron-Ruault, F. Clavel-Chapelon, V. Ringa, and C. Berr
Differential Dietary Nutrient Intake according to Hormone Replacement Therapy Use: An Underestimated Confounding Factor in Epidemiologic Studies?
Am. J. Epidemiol., December 15, 2007; 166(12): 1451 - 1460.
[Abstract] [Full Text] [PDF]

Home page
 JAMAHome page
A. Tatsioni, N. G. Bonitsis, and J. P. A. Ioannidis
Persistence of Contradicted Claims in the Literature
JAMA, December 5, 2007; 298(21): 2517 - 2526.
[Abstract] [Full Text] [PDF]

Home page
 Am J EpidemiolHome page
R. L. Prentice, R. Langer, G. Anderson, and D. Barad
Prentice et al. Respond to "How Far Can Epidemiologists Get with Statistical Adjustment?"
Am. J. Epidemiol., September 1, 2005; 162(5): 419 - 420.
[Full Text] [PDF]

This Article
Right arrow Extract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
162/5/415    most recent
kwi224v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Petitti, D. B.
Right arrow Articles by Freedman, D. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Petitti, D. B.
Right arrow Articles by Freedman, D. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?