Design of the Women’s Health Initiative Clinical Trial and Observational Study
Introduction
The WHI clinical trial includes three overlapping components, each a randomized controlled comparison among women who are postmenopausal and in the age range of 50 to 79 at randomization. The dietary modification (DM) component randomly assigns 48,000 eligible women to either a sustained low-fat eating pattern (40%) or self-selected dietary behavior (60%), with breast cancer and colorectal cancer as designated primary outcomes and coronary heart disease as a secondary outcome. The nutrition goals for women assigned to the DM intervention group are to reduce total dietary fat to 20% and saturated fat to less than 7% of corresponding daily calories and, secondarily, to increase daily servings of vegetables and fruits to at least five and of grain products to at least six. The hormone replacement therapy (HRT) component is a randomized, double-blind comparison among 27,500 women, with coronary heart disease as the primary outcome, with hip and other bone fractures as secondary outcomes, and with breast cancer as a potential adverse outcome. The design of the HRT component assumes that 45% of women will be post-hysterectomy at randomization, in which case there is a 1:1 randomized double-blind allocation between conjugated equine estrogen (ERT) 0.625 mg/day or placebo. The remaining 55% of women, each having a uterus at baseline, are randomized to 1:1 to this same preparation of estrogen plus continuous 2.5 mg/day of medroxyprogesterone (PERT) or placebo. Eligible women can be randomized into one or both of the DM and HRT components. At the 1-year anniversary of randomization, all women are further screened for possible randomization into the calcium and vitamin D (CaD) component, which plans to enroll 45,000 women, with hip fracture as the primary outcome and with other fractures and colorectal cancer as secondary outcomes. The CaD component is a 1:1 randomized double-blind trial of 1000 mg elemental calcium plus 400 international units of vitamin D3 daily, versus placebo. Table 1 shows the anticipated numbers of women in various cells of the CT.
Postmenopausal women ages 50 to 79 who are screened for the CT but prove to be ineligible or unwilling to be randomized are offered the opportunity to be one of 100,000 women enrolled in an observational study (OS). The OS is intended to provide additional knowledge about risk factors for a range of diseases, including cancer, cardiovascular disease, and fractures. It has an emphasis on biological markers of disease risk and on risk factor changes as modifiers of risk.
There is also an emphasis on the inclusion of women of racial/ethnic minority groups, with an overall target of 20% in both the CT and OS. Such a fraction will allow meaningful study of disease risk factors within minority groups in the OS, while certain CT subsamples are weighted heavily in favor of the inclusion of minority women in order to strengthen the study of intervention effects on specific intermediate outcomes (e.g., changes in blood lipids or micronutrients) within minority groups.
Age distribution goals are also specified for the CT as follows: 10%, ages 50–54; 20%, ages 55–59; 45%, ages 60–69, and 25%, ages 70–79. While the age range of 55 to 69 may be regarded as most natural for the “treatments” to be tested, there was also interest in having a sufficient representation of younger (50–54) postmenopausal women for intermediate outcome (biomarker) studies and of older (70–79) women for studies of treatment effects on quality of life measures, including aspects of physical and cognitive function.
With a projected 164,500 women enrolled in a complex program at 40 clinical centers (CCs) in the United States, the WHI is perhaps the most massive and challenging program ever undertaken. It has a budget of $628 million over the 15-year period, 1992–2007. Not surprisingly, a good deal of preliminary research and advance planning were necessary to provide sufficient bases for the development of such a program. An account of the evolution of the WHI has previously been provided [1]. Key preliminary studies included the National Cancer Institute-sponsored Women’s Health Trial 2, 3 and the subsequent Women’s Health Trial: Feasibility Study in Minority Populations [4], which demonstrated the feasibility of recruiting postmenopausal women to a dietary intervention trial and the feasibility of such women making a major dietary change toward a low-fat eating pattern, and the National Heart, Lung and Blood Institute-sponsored Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial, which examined the effects of various postmenopausal hormone replacement therapy regimens on heart disease risk factors [5], endometrial pathology [6], and bone mineral density [7].
The next section provides a description of the rationale behind the choice of the specific interventions used in each of the three CT components and of the designated clinical outcomes in both the CT and OS. The presentation then turns to a discussion of study design choices and statistical aspects of the design, followed by a brief description of CT monitoring issues. Finally, a description is given of the organizational and management approach to this complex undertaking, including mention of a committee structure designed to ensure adequate communications and decision-making procedures and to preserve investigator interest.
Section snippets
Rationale for and Feasibility of a Clinical Trial of a Low-Fat Eating Pattern
Nutrition, in its broadest sense, very likely plays a fundamental role in breast and colon cancer incidence [8]. Documented breast cancer risk factors, including age at menarche, adult height, and postmenopausal body mass, suggest that adult and early nutrition each may exert its own influences. A breast cancer prevention hypothesis [9] motivated the series of NCI-sponsored feasibility studies mentioned above.
The extent to which modifications of the diet alone can reduce cancer incidence has
Eligibility Criteria
The WHI is designed to be as inclusive as it is practical to be of postmenopausal women, initially in the age range of 50 to 79, with a sufficient follow-up duration to address adequately questions of risks versus benefits and of the public health potential of the CT interventions. The diseases responsible for much of the morbidity and mortality in women are largely concentrated among women who are at least 50 years old. A focus on postmenopausal women is essential for the HRT trial component,
Screening
As mentioned previously, the WHI wanted to make the eligibility criteria as broad as was practical in order to enhance the generalizability of the results to the population of postmenopausal women. Therefore, women with prevalent cardiovascular disease or a past history of bone fractures may be included, allowing the study of both primary and secondary prevention in the CT. The study cohort at each clinic can be drawn from a population-based sample, for example, using residential mailing lists;
Sample size and power for the CT and os
Data from preceding observational studies and feasibility clinical trials were used to specify a series of design assumptions for each CT component. These assumptions were then used to determine the sample size necessary to yield study powers in the range 80–95% for the primary endpoint comparisons in each trial component. In each case, power was based on a two-sided 5% level test of significance using a weighted logrank or weighted odds ratio test statistic, with weights based on the
CT Monitoring
Many aspects of monitoring the WHI clinical trial are similar to those required for monitoring other randomized multicenter clinical trials. A Data and Safety Monitoring Board (DSMB), responsible for monitoring the integrity of the trial and the safety of its participants, meets at approximate 6-month intervals to review the progress of the trial. Members of the DSMB are independent scientists with expertise in women’s health, including gynecology, oncology, cardiology, and bone metabolism as
Aspects of the organization and early implementation of the whi
By its very nature, the WHI is one of the most complex studies ever mounted. The study investigates multiple diseases in a design that includes both a multicomponent clinical trial and an observational study to be carried out in two phases. In the first phase of WHI, the protocol and procedures were to be evaluated among 16 vanguard clinical centers before the second phase or entry of the final set of 24 CCs. The combined operational units include the 40 clinical centers and a clinical
Discussion
The Women’s Health Initiative Clinical Trial is a major undertaking that is designed to carefully assess interventions and treatments that have great potential for improving the health of American women. The diseases targeted for prevention, including coronary heart disease, breast cancer, colorectal cancer, and osteoporotic fractures, are among the most common causes of morbidity and mortality in middle-aged and older women. While the sample size for each component of the CT was selected to
Acknowledgements
This work was supported by NIH contracts for the WHI.
References (117)
- et al.
Feasibility of a randomized trial of a low-fat diet for the prevention of breast cancerdietary compliance in the Women’s Health Trial Vanguard Study
Prevent Med
(1990) - et al.
Types of dietary fat and the incidence of breast cancer at five sites
Prev Med
(1990) - et al.
Dietary fiber in the reduction of colon cancer risk
J Am Diet Assoc
(1987) - et al.
Epidemiologic studies in coronary heart disease and stroke in Japanese men living in Japan, Hawaii, and Californiaincidence of myocardial infarction and death from coronary heart disease
Am J Cardiol
(1977) - et al.
Changes in plasma lipids and lipoproteins after a modified fat diet
Lancet
(1980) - et al.
The effect of diet on plasma lipids, lipoproteins, and coronary heart disease
J Am Diet Assoc
(1988) - et al.
Fifteen-year mortality in Coronary Drug Project patientslong-term benefit with niacin
J Am Coll Cardio
(1986) - et al.
Sex, plasma lipoproteins, and atherosclerosisprevailing assumptions and outstanding questions
Am Heart J
(1987) - et al.
Estrogen replacement therapy and coronary heart diseasea quantitative assessment of the epidemiologic evidence
Prev Med
(1991) - et al.
Effects of post-menopausal oestrogen-progestogen replacement therapy on serum lipids and lipoproteinsa review
Maturitas
(1990)