Annlouise R. Assaf

Design of the Women's Health Initiative clinical trial and observational study

The Womens Health Initiative (WHI) is a large and complex clinical investigation of strategies for the prevention and control of some of the most common causes of morbidity and mortality among postmenopausal women, including cancer, cardiovascular disease, and osteoporotic fractures. The WHI was initiated in 1992, with a planned completion date of 2007. Postmenopausal women ranging in age from 50 to 79 are enrolled at one of 40 WHI clinical centers nationwide into either a clinical trial (CT) that will include about 64,500 women or an observational study (OS) that will include about 100,000 women. The CT is designed to allow randomized controlled evaluation of three distinct interventions: a low-fat eating pattern, hypothesized to prevent breast cancer and colorectal cancer and, secondarily, coronary heart disease; hormone replacement therapy, hypothesized to reduce the risk of coronary heart disease and other cardiovascular diseases and, secondarily, to reduce the risk of hip and other fractures, with increased breast cancer risk as a possible adverse outcome; and calcium and vitamin D supplementation, hypothesized to prevent hip fractures and, secondarily, other fractures and colorectal cancer. Overall benefit-versus-risk assessment is a central focus in each of the three CT components. Women are screened for participation in one or both of the components--dietary modification (DM) or hormone replacement therapy (HRT)--of the CT, which will randomize 48,000 and 27,500 women, respectively. Women who prove to be ineligible for, or who are unwilling to enroll in, these CT components are invited to enroll in the OS. At their 1-year anniversary of randomization, CT women are invited to be further randomized into the calcium and vitamin D (CaD) trial component, which is projected to include 45,000 women. The average follow-up for women in either CT or OS is approximately 9 years. Concerted efforts are made to enroll women of racial and ethnic minority groups, with a target of 20% of overall enrollment in both the CT and OS. This article gives a brief description of the rationale for the interventions being studied in each of the CT components and for the inclusion of the OS component. Some detail is provided on specific study design choices, including eligibility criteria, recruitment strategy, and sample size, with attention to the partial factorial design of the CT. Some aspects of the CT monitoring approach are also outlined. The scientific and logistic complexity of the WHI implies particular leadership and management challenges. The WHI organization and committee structure employed to respond to these challenges is also briefly described.

The Pawtucket Heart Health Program: community changes in cardiovascular risk factors and projected disease risk.

OBJECTIVES Whether community-wide education changed cardiovascular risk factors and disease risk in Pawtucket, RI, relative to a comparison community was assessed. METHODS Random-sample, cross-sectional surveys were done of people aged 18 through 64 years at baseline, during, and after education. Baseline cohorts were reexamined. Pawtucket citizens of all ages participated in multilevel education, screening, and counseling programs. RESULTS The downward trend in smoking was slightly greater in the comparison city. Small, insignificant differences favored Pawtucket in blood cholesterol and blood pressure. In the cross-sectional surveys, body mass index increased significantly in the comparison community; a similar change was not seen in cohort surveys. Projected cardiovascular disease rates were significantly (16%) less in Pawtucket during the education program. This difference lessened to 8% posteducation. CONCLUSIONS The hypothesis that projected cardiovascular disease risk can be altered by community-based education gains limited support from these data. Achieving cardiovascular risk reduction at the community level was feasible, but maintaining statistically significant differences between cities was not. Accelerating risk factor changes will likely require a sustained community effort with reinforcement from state, regional, and national policies and programs.

Use of Recovery Biomarkers to Calibrate Nutrient Consumption Self-Reports in the Women's Health Initiative

Underreporting of energy consumption by self-report is well-recognized, but previous studies using recovery biomarkers have not been sufficiently large to establish whether participant characteristics predict misreporting. In 2004-2005, 544 participants in the Womens Health Initiative Dietary Modification Trial completed a doubly labeled water protocol (energy biomarker), 24-hour urine collection (protein biomarker), and self-reports of diet (assessed by food frequency questionnaire (FFQ)), exercise, and lifestyle habits; 111 women repeated all procedures after 6 months. Using linear regression, the authors estimated associations of participant characteristics with misreporting, defined as the extent to which the log ratio (self-reported FFQ/nutritional biomarker) was less than zero. Intervention women in the trial underreported energy intake by 32% (vs. 27% in the comparison arm) and protein intake by 15% (vs. 10%). Younger women had more underreporting of energy (p = 0.02) and protein (p = 0.001), while increasing body mass index predicted increased underreporting of energy and overreporting of percentage of energy derived from protein (p = 0.001 and p = 0.004, respectively). Blacks and Hispanics underreported more than did Caucasians. Correlations of initial measures with repeat measures (n = 111) were 0.72, 0.70, 0.46, and 0.64 for biomarker energy, FFQ energy, biomarker protein, and FFQ protein, respectively. Recovery biomarker data were used in regression equations to calibrate self-reports; the potential application of these equations to disease risk modeling is presented. The authors confirm the existence of systematic bias in dietary self-reports and provide methods of correcting for measurement error.

Risk of Cardiovascular Disease by Hysterectomy Status, With and Without Oophorectomy The Women’s Health Initiative Observational Study

Background—Cardiovascular disease (CVD) is a leading cause of morbidity and mortality in women and may vary by hysterectomy (or oophorectomy) status. This study compared CVD risk factors and rates between postmenopausal women who had and had not undergone hysterectomy, with or without oophorectomy. Methods and Results—This analysis was conducted on 89 914 women in the Women’s Health Initiative (WHI) Observational Study. Participants reported demographic characteristics, medical history, dietary habits, physical activity, medications, and previous hysterectomy (with or without oophorectomy). Baseline weight, height, waist circumference, and blood pressure were measured. CVD events were ascertained during 5.1 years of mean follow-up and adjudicated with standard criteria. Black, Hispanic, and American Indian women had higher rates of hysterectomy than white women (52.9%, 44.6%, and 49.2% versus 40.0%, respectively), and Asian/Pacific Islander women had lower rates (33.8%). Women with a hysterectomy (regardless of oophorectomy status) had an adverse risk profile at baseline compared with women with no hysterectomy, including a higher proportion of hypertension, diabetes, high cholesterol, obesity, and lower education, income, and physical activity (all P<0.01). Total mortality and fatal and nonfatal CVD were higher among women with a hysterectomy. Hysterectomy (regardless of oophorectomy status) was a significant predictor of CVD (HR: 1.26, P<0.001). After adjustment for demographic variables and CVD risk factors, the effect was reduced and nonsignificant. Conclusions—Women with a hysterectomy had a worse risk profile and higher prevalence and incidence of CVD in this cohort. Multivariate models suggest that hysterectomy is not the major determinant of this outcome; rather, CVD risk may be due to the more adverse initial risk profile of women who had undergone hysterectomy.

Possible Influence of the Prospective Payment System on the Assignment of Discharge Diagnoses for Coronary Heart Disease

BACKGROUND The prospective payment system, under which diagnosis-related groups (DRGs) are used to reimburse hospitals for the care of Medicare patients, replaced the fee-for-service method of payment in Rhode Island in 1983 and in Massachusetts in 1985. Changes in financial incentives resulting from the use of the DRG system may have influenced the assignment of discharge diagnostic codes away from those with lower reimbursement toward codes with higher reimbursement. METHODS We collected data from the hospital records of patients 35 through 74 years of age who were discharged with codes 410 through 414 (representing various categories of coronary heart disease) of the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM). The patients were discharged from seven hospitals in two New England communities (one in Rhode Island and one in Massachusetts) between 1980 and 1988. The rates of diagnosis of various forms of coronary heart disease were determined by studying ICD-9-CM hospital discharge codes (codes 410 and 411 for acute forms of coronary heart disease and codes 412, 413, and 414 for chronic forms) and by using a computerized diagnostic algorithm designed to detect definite myocardial infarction and fatal coronary heart disease. RESULTS The rates of definite coronary events diagnosed by the algorithm and by the study of ICD-9-CM codes 410 through 414 were constant or increased slightly during the study period. However, the frequency of assignment of codes for the acute forms of coronary heart disease (which entail higher reimbursement) rose from 35.2 percent to 48.4 percent among discharged patients with cardiac disease after the institution of DRGs. The majority of this increase was associated with the code for unstable angina pectoris. The frequency of assignment of codes for the chronic forms of coronary heart disease (which entail lower reimbursement) decreased reciprocally, from 64.8 percent to 51.6 percent (P < 0.001). CONCLUSIONS Our data are consistent with the hypothesis that the prospective reimbursement system has influenced the assignment of hospital discharge codes in a way that would increase payment to hospitals. However, the data do not permit us to distinguish whether hospitals began to assign more precise diagnoses with the advent of the DRG system, or whether they began to favor diagnoses of acute conditions solely for financial reasons.

Physical activity, physical fitness, and coronary heart disease risk factors.

The relationships between physical activity, physical fitness, and coronary heart disease risk factors measured in a large community sample were evaluated. Self-reported physical activity using a single question, maximal oxygen consumption estimates derived from the Pawtucket Heart Health Step Test, blood pressure, nonfasting lipids, and body mass index were cross-sectionally evaluated in 381 men and 556 women. The correlation of estimated maximal oxygen consumption and self-reported physical activity was modest but statistically significant (r = 0.13 in men and r = 0.19 in women). Blood pressure, body mass index, and HDL cholesterol were correlated with physical fitness (r = 0.24-0.65) and correlated to self-reported physical activity (r = 0.09-0.14). Evaluation of coronary heart disease risk factors using both physical activity and physical fitness revealed a complex relationship that generally showed a stronger relationship with measures of physical fitness than with physical activity. This study suggests that simultaneous measurement of physical activity and physical fitness may be useful in epidemiologic studies of habitual physical activity and chronic disease.

Are There Gender Differences in Self-Reported Smoking Practices? Correlation with Thiocyanate and Cotinine Levels in Smokers and Nonsmokers from the Pawtucket Heart Health Program

OBJECTIVES This study compared serum cotinine and thiocyanate in assessment of self-reported smoking behavior among 1400 men and 1809 women from two New England communities. METHODS Serum thiocyanate and serum cotinine levels were analyzed on 2411 and 798 survey respondents, respectively, in an attempt to provide an objective measurement for validation of self-reported smoking behaviors that were obtained through an in-home interviewer-administered questionnaire. Cross-sectional household surveys were conducted with randomly selected men and women, aged 18-65, between 1981 and 1993 as part of the evaluation of the Pawtucket Heart Health Program. RESULTS Among smokers, the thiocyanate test had similar rates of agreement for women(88.0%) and for men (89.3%). However, among nonsmokers, thiocyanate had higher rates of agreement for women (91.5%) than for men (85.2%). For cotinine, the rates of agreement among smokers were higher for women (91.6%) than for men (89.7%). Similarly, the rates of agreement among nonsmokers were also higher for women (93.9%) than for men (91.9%). Overall,serum cotinine had a higher concordance rate than serum thiocyanate for both men and women. CONCLUSIONS Although our results suggested that there were some differences in self-reporting of smoking status by gender, results were quite similar between self-reports of smoking and both biochemical tests. The results obtained from this large population-based study from two New England communities lend credibility to the use of self-reports as a low-cost accurate approach to obtaining information on smoking behaviors among both men and women in large population-based health surveys.

Arthritis Increases the Risk for Fractures---Results from the Women’s Health Initiative

Objective. To examine the relationship between arthritis and fracture. Methods. Women were classified into 3 self-reported groups at baseline: no arthritis (n = 83,295), osteoarthritis (OA; n = 63,402), and rheumatoid arthritis (RA; n = 960). Incident fractures were self-reported throughout followup. Age-adjusted fracture rates by arthritis category were generated, and the Cox proportional hazards model was used to test the association between arthritis and fracture. Results. After an average of 7.80 years, 24,137 total fractures were reported including 2559 self-reported clinical spinal fractures and 1698 adjudicated hip fractures. For each fracture type, age-adjusted fracture rates were highest in the RA group and lowest in the nonarthritic group. After adjustment for several covariates, report of arthritis was associated with increased risk for spine, hip, and any clinical fractures. Compared to the nonarthritis group, the risk of sustaining any clinical fracture in the OA group was HR 1.09 (95% CI 1.05, 1.13; p < 0.001) and HR 1.49 (95% CI 1.26, 1.75; p < 0.001) in the RA group. The risk of sustaining a hip fracture was not statistically increased in the OA group (HR 1.11; 95% CI 0.98, 1.25; p = 0.122) compared to the nonarthritis group; however, the risk of hip fracture increased significantly (HR 3.03; 95% CI 2.03, 4.51; p < 0.001) in the RA group compared to the nonarthritis group. Conclusion. The increase in fracture risk confirms the importance of fracture prevention in patients with RA and OA.

Methods in Program Evaluation The Process Evaluation System of the Pawtucket Heart Health Program

A methodology for a process evaluation system as applied to the Pawtucket Heart Health Program is described. The rationale, goals, and design of this system, which are suited to the documentation of community demonstration programs or other interventions, are discussed. Examples of data from the system are provided, including those from linked, computer-based files for the systematic tracking of program activities and the characteristics of program participants. Also discussed are approaches to maintaining such a system in a context of limited resources.

Comparison of Three Methods of Teaching Women How to Perform Breast Self-Examination

This paper presents results from an experimental study designed to evaluate the relative effectiveness of three methods of teaching women how to do breast self-examination (BSE). Frequency of BSE, confidence in examination performance, proficiency of BSE technique, and lump detection performance were the main outcome variables assessed. The three training methods compared in this study were provision of a pamphlet describing how to do BSE, having women view a videotape depicting proper performance of BSE, and having women practice doing BSE on a life-like breast model. Results showed that passive methods of BSE instruction such as the use of pamphlets or films were of little value in helping women develop the tactile skills necessary for proficient BSE. Three months after training, it was found that lump detection performance, as measured on silicone breast models, was significantly higher among those women who had been given an opportunity to practice doing the breast examination on a breast model with corrective feedback given by a BSE instructor. The opportunity to practice doing the examination with corrective feedback on performance appears to be a critical variable in the acquisition of BSE skill.