Statistical adjustment for a measure of healthy lifestyle doesnt yield the truth about hormone therapy

📝 Abstract

The Women’s Health Initiative randomized clinical trial of hormone therapy found no benefit of hormones in preventive cardiovascular disease, a finding in striking contrast with a large body of observational research. Understanding whether better methodology and/or statistical adjustment might have prevented the erroneous conclusions of observational research is important. This is a re-analysis of data from a case-control study examining the relationship of postmenopausal hormone therapy and the risks of myocardial infarction (MI) and ischemic stroke in which we reported no overall increase or decrease in the risk of either event. Variables measuring health behavior/lifestyle that are not likely to be causally with the risks of MI and stroke (e.g., sunscreen use) were included in multivariate analysis along with traditional confounders (age, hypertension, diabetes, smoking, body mass index, ethnicity, education, prior coronary heart disease for MI and prior stroke/TIA for stroke) to determine whether adjustment for the health behavior/lifestyle variables could reproduce or bring the results closer to the findings in a large and definitive randomized clinical trial of hormone therapy, the Women’s Health Initiative. For both MI and stroke, measures of health behavior/lifestyle were associated with odds ratios (ORs) less than 1.0. Adjustment for traditional cardiovascular disease confounders did not alter the magnitude of the ORs for MI or stroke. Addition of a subset of these variables selected using stepwise regression to the final MI or stroke models along with the traditional cardiovascular disease confounders moved the ORs for estrogen and estrogen/progestin use closer to values observed in the Women Health Initiative clinical trial, but did not reliably reproduce the clinical trial results for these two endpoints.

💡 Analysis

The Women’s Health Initiative randomized clinical trial of hormone therapy found no benefit of hormones in preventive cardiovascular disease, a finding in striking contrast with a large body of observational research. Understanding whether better methodology and/or statistical adjustment might have prevented the erroneous conclusions of observational research is important. This is a re-analysis of data from a case-control study examining the relationship of postmenopausal hormone therapy and the risks of myocardial infarction (MI) and ischemic stroke in which we reported no overall increase or decrease in the risk of either event. Variables measuring health behavior/lifestyle that are not likely to be causally with the risks of MI and stroke (e.g., sunscreen use) were included in multivariate analysis along with traditional confounders (age, hypertension, diabetes, smoking, body mass index, ethnicity, education, prior coronary heart disease for MI and prior stroke/TIA for stroke) to determine whether adjustment for the health behavior/lifestyle variables could reproduce or bring the results closer to the findings in a large and definitive randomized clinical trial of hormone therapy, the Women’s Health Initiative. For both MI and stroke, measures of health behavior/lifestyle were associated with odds ratios (ORs) less than 1.0. Adjustment for traditional cardiovascular disease confounders did not alter the magnitude of the ORs for MI or stroke. Addition of a subset of these variables selected using stepwise regression to the final MI or stroke models along with the traditional cardiovascular disease confounders moved the ORs for estrogen and estrogen/progestin use closer to values observed in the Women Health Initiative clinical trial, but did not reliably reproduce the clinical trial results for these two endpoints.

📄 Content

arXiv:0805.2845v1 [stat.AP] 19 May 2008 IMS Collections Probability and Statistics: Essays in Honor of David A. Freedman Vol. 2 (2008) 142–152 c⃝Institute of Mathematical Statistics, 2008 DOI: 10.1214/193940307000000437 Statistical adjustment for a measure of healthy lifestyle doesn’t yield the truth about hormone therapy Diana B. Petitti∗1 and Wansu Chen∗2 University of Southern California and Kaiser Permanente Southern California Abstract: The Women’s Health Initiative randomized clinical trial of hor- mone therapy found no benefit of hormones in preventive cardiovascular dis- ease, a finding in striking contrast with a large body of observational re- search. Understanding whether better methodology and/or statistical adjust- ment might have prevented the erroneous conclusions of observational research is important. This is a re-analysis of data from a case-control study examin- ing the relationship of postmenopausal hormone therapy and the risks of my- ocardial infarction (MI) and ischemic stroke in which we reported no overall increase or decrease in the risk of either event. Variables measuring health be- havior/lifestyle that are not likely to be causally with the risks of MI and stroke (e.g., sunscreen use) were included in multivariate analysis along with tradi- tional confounders (age, hypertension, diabetes, smoking, body mass index, ethnicity, education, prior coronary heart disease for MI and prior stroke/TIA for stroke) to determine whether adjustment for the health behavior/lifestyle variables could reproduce or bring the results closer to the findings in a large and definitive randomized clinical trial of hormone therapy, the Women’s Health Initiative. For both MI and stroke, measures of health behavior/lifestyle were associ- ated with odds ratios (ORs) less than 1.0. Adjustment for traditional cardio- vascular disease confounders did not alter the magnitude of the ORs for MI or stroke. Addition of a subset of these variables selected using stepwise regres- sion to the final MI or stroke models along with the traditional cardiovascular disease confounders moved the ORs for estrogen and estrogen/progestin use closer to values observed in the Women Health Initiative clinical trial, but did not reliably reproduce the clinical trial results for these two endpoints.

1. Background The Women’s Health Initiative (WHI) clinical trial of hormone therapy is a large randomized trial whose primary aim was determining whether post-menopausal hormone therapy prevents coronary heart disease (Rossouw et al. [10] and Ander- son et al. [1]). The study began in the early 1990’s and published main results in 2002–2004. It involved recruitment and randomization of more than 18,000 post- menopausal women to hormones or placebo. The WHI found no overall effect, or ∗Supported by grant number R01-HL-47043 from the National Heart Lung and Blood Institute. 1University of Southern California, Department of Preventive Medicine, Keck School of Medicine, 299 E. Laurel Avenue, Sierra Madre, CA 91024, USA, e-mail: dbpetitti@verizon.net 2Kaiser Permanente Southern California, Department of Research and Evaluation, 100 S. Los Robles, Pasadena, CA 91101, USA, e-mail: Wansu.Chen@kp.org AMS 2000 subject classifications: Primary 92C60; secondary 62P10, 00B30. Keywords and phrases: cerebrovascular disorders, cerebral infarction, stroke, coronary, epi- demiological methods, estrogen, heart disease, hormone replacement, myocardial infarction. 142 Statistical adjustment for a measure of healthy lifestyle 143 perhaps an increase, in the risk of myocardial infarction (MI) in women assigned to combined estrogen/progestin (E/P) therapy and no effect of estrogen alone (E) (Rossouw et al. [10] and Anderson et al. [1]). Prior observational research con- cluded that the risk of coronary heart disease was reduced by half (Stampfer and Colditz [12]). Moreover, the risk of stroke was increased for both E and E/P in WHI (Rossouw et al. [10] and Anderson et al. [1]). Prior observational research found no effect of hormone therapy on stroke (Psaty et al. [9]). Publications by Prentice et al. suggest statistical approaches that would have overcome the discrepancy between the observational research and the clinical trial (Prentice et al. [7] and Prentice et al. [8]). Reviews of these approaches are mixed (Petitti and Freedman [5], Freedman and Petitti [3], DeMets [2] and Greenland [4]). Understanding whether better methodology and/or statistical adjustment might have prevented the erroneous conclusions made based on the observational research is important. We previously published the results of a case-control study in which we esti- mated the relative risks of myocardial infarction (MI) and stroke in current users of E and E/P that adjusted for traditional cardiovascular disease confounders—age, ethnicity, education as a measure of socioeconomic status, and factors known to increase the risk of MI or stroke causally (smoking, diabetes, hypertension, body mas

This content is AI-processed based on ArXiv data.