Combining data from primary and ancillary surveys to assess the association between neighborhood-level characteristics and health outcomes: the Multi-Ethnic Study of Artherosclerosis.

There is increasing interest in understanding the role of neighborhood-level factors on the health of individuals. Many large-scale epidemiological studies that accurately measure health status of individuals and individual risk factors exist. Sometimes these studies are linked to area-level databases (e.g. census) to assess the association between crude area-level characteristics and health. However, information from such databases may not measure the neighborhood-level constructs of interest. More recently, large-scale epidemiological studies have begun collecting data to measure specific features of neighborhoods using ancillary surveys. The ancillary surveys are composed of a separate, typically larger, set of individuals. The challenge is then to combine information from these two surveys to assess the role of neighborhood-level factors. We propose a method for combining information from the two data sources using a likelihood-based framework. We compare it with currently used ad hoc approaches via a simulation study. The simulation study shows that the proposed approach yields estimates with better sampling properties (less bias and better coverage probabilities) compared with the other approaches. However, there are cases where some ad hoc approaches may provide adequate estimates. We also compare the methods by applying them to the Multi-Ethnic Study of Atherosclerosis and its Neighborhood Ancillary Survey.

Self-efficacy as a predictor to PFMT adherence in a prevention of urinary incontinence clinical trial.

Past research suggests a positive correlation between self-efficacy (SE) and adherence to behavioral interventions. Less is known about SE and adherence in behavioral programs that are preventive in nature and specific to urinary incontinence (UI). Using treatment-group data from a previously reported randomized controlled trial, the authors assess the role of SE in predicting adherence to pelvic-floor muscle training (PFMT) for UI prevention in a sample of postmenopausal women. Results indicate that at 12 months follow-up, nearly 70% of participants reported medium or high adherence, performing the recommended PFMT regimen 2 to 3 times per week or more. Summary scores of both Task SE, beta = .25, SE (beta) = .08, p < .01, and Regulatory SE, beta = .43, SE (beta) = .06, p < .0001, predict adherence. Also, the authors found a modest decline in self-efficacy scores over time. These findings highlight the importance of SE in sustained behavioral change.

Adherence to a behavioral program to prevent incontinence.

This exploratory study assesses factors predicting adherence to a behavioral intervention to prevent urinary incontinence. Community-dwelling, post-menopausal women (N = 164) were taught pelvic floor muscle training (PFMT) and bladder training (BT) and followed with surveys for 1 year. Content analysis of open-ended responses coded descriptions of approaches participants used to incorporate PFMT into daily life. Exploratory bivariate and logistic regression analyses determined predictors of approach used and adherence. Results indicate women incorporated PFMT into their lives using either a routine or ad hoc approach. Those using a routine approach at 3 months were 12 times more likely to adhere (odds ratio [OR] = 12.4, confidence interval [CI] = 4.0-38.8, p < .001) at a high level at 3 months and significantly more likely to maintain that level 12 months post-intervention (OR = 2.7, CI = 1.2-6.0, p < .014). Practicing BT was related to high adherence.

Evaluation of a mass mailing recruitment strategy to obtain a community sample of women for a clinical trial of an incontinence prevention intervention.

OBJECTIVES: Questions exist about using mass mailings to recruit representative samples to participate in clinical trials. The MESA Prevention Study (Medical, Epidemiologic and Social Aspects of Aging), a randomized controlled clinical trial to prevent urinary incontinence (UI), utilized a mass mailing recruitment procedure to recruit a representative sample of women to participate in a behavioral modification program. This paper seeks to expand the literature of mass mailing recruitment strategies for prevention studies by describing the procedures used to recruit healthy, continent, post-menopausal women aged 55-80 years. METHODS: Sociodemographic data collected from recruited subjects is compared with on-line national census data to evaluate the representativeness of the sample recruited from a purchased mailing list. RESULTS: The mass mailing procedure resulted in 3.3% positive response. Of those that returned a positive response, 37.6% were deemed eligible at first screening. Comparisons of study demographic data with state and county census data indicate that the sample obtained was representative of the communities. CONCLUSIONS: The mass mailing strategy was an effective means of recruiting a representative sample of women, aged 55-80. Short falls and recommendations for successful community sample recruitment strategies for clinical trials in older adult women are elaborated upon.

Clinical depression and risk of out-of-hospital cardiac arrest.

BACKGROUND: The association of depression with coronary heart disease-related mortality has been widely recognized. This finding may partly reflect an association between depression and sudden death, in part because the imbalance between sympathetic and parasympathetic tone is altered in depressed subjects. We, thus, investigated whether the presence and severity of clinical depression was associated with a higher risk of sudden cardiac death. METHODS: We used data from a population-based case-control study of risk factors for incident out-of-hospital cardiac arrest (CA) conducted among enrollees of a health maintenance organization in western Washington State. Cases (n = 2228) were aged 40 to 79 years and experienced CA between January 1, 1980, and December 31, 1994. Controls (n = 4164) were a stratified random sample of enrollees defined by calendar year, age, sex, and prior heart disease. Clinical depression was defined as physician diagnosis of depression or use of antidepressant treatment within the year before the event. Referral to mental health clinics or hospitalization for depression defined severe depression. RESULTS: Clinically depressed patients had a higher odds ratio (OR) of CA (1.88; 95% confidence interval [CI], 1.59-2.23), which persisted after adjustment for confounders (OR, 1.43; 95% CI, 1.18-1.73). The association was observed in both sexes, in various age groups, and in subjects with prior physician-diagnosed heart disease (OR, 1.27; 95% CI, 1.01-1.60) and without prior physician-diagnosed heart disease (OR, 1.71; 95% CI, 1.22-2.41) (P = .13 for the interaction). Compared with nondepressed subjects, the risk of CA was increased in less severely depressed subjects (OR, 1.30; 95% CI, 1.04-1.63) and further increased in severely depressed subjects (OR, 1.77; 95% CI, 1.28-2.45) (P<.001 for trend). CONCLUSION: Clinical depression may be associated with a higher risk of CA independently of established coronary heart disease risk factors.

Learning outcomes of a group behavioral modification program to prevent urinary incontinence.

This study describes acquisition of knowledge and motor skill in bladder training (BT) and pelvic floor muscle training (PFMT) and adherence following a behavioral modification program (BMP). Essentially continent (0-5 episodes in past year) community-dwelling older women (n = 359) were randomized to treatment (n = 164), a 2-h group education session supplemented by one brief individualized session of approximately 10 min, or control (n = 195), no instruction, and followed for 12 months. Knowledge, motor skill, and adherence to the BMP were documented. Changes in pelvic muscle function and voiding interval were used to validate self-reported adherence. Following group instruction, mean BT and PFMT knowledge was 90 and 86%, respectively; 68% demonstrated correct PFMT technique without additional instruction, 29% required brief instruction, and 3% were unable to learn PFMT technique. Adherence ranged from 63 to 82% for PFMT and 58 to 67% for BT. Group instruction supplemented with brief individual instruction as needed is an effective teaching method for BT and PFMT.

Prevention of urinary incontinence by behavioral modification program: a randomized, controlled trial among older women in the community.

PURPOSE: We determined whether a behavioral modification program (BMP) taught to groups of continent older women would decrease the incidence of urinary incontinence, increase pelvic muscle strength and improve voiding control. MATERIALS AND METHODS: We performed a randomized, controlled trial comparing a BMP treatment group to a control, no treatment group in ambulatory, postmenopausal, continent women (0 to 5 days of incontinent episodes in the previous year) 55 years and older who were followed for 12 months. Qualified volunteers from 4 Michigan counties were randomly assigned to a control or a treatment group, consisting of a 2-hour classroom presentation on BMP followed 2 to 4 weeks later with individualized evaluation to test knowledge, adherence and skills in behavioral techniques, and brief reinforcement of the technique as needed. Followup was done by telephone and mail every 3 months except month 12, when all participants underwent final clinical evaluation. Outcome measures were continence status, pelvic floor muscle strength and voiding frequency/intervoid interval. RESULTS: A total of 195 control and 164 treated participants completed the study. Baseline data on the 2 groups were not statistically different. At 12 months the treatment group was statistically significantly better than the control group in continence status (p = 0.01), pelvic muscle strength (pressure score p = 0.0003 and displacement score p <0.0001), improved voiding frequency (p <0.0001) and intervoid interval (p <0.0001). CONCLUSIONS: To our knowledge we report the first randomized, controlled trial of BMP for preventing urinary incontinence in older women. It demonstrated feasibility and efficacy in improving continence status, pelvic muscle strength and voiding control as long as a year after treatment.

Sudden death and myocardial infarction in first degree relatives as predictors of primary cardiac arrest.

The hypothesis that family history (FH) of myocardial infarction (MI) and FH of sudden death (SD) are both independent risk factors for primary cardiac arrest (PCA) was examined in a case-control study. PCA cases were attended by paramedics (1988-1994) and community-based age and sex matched controls were identified. Subjects (25-74 years) were free of prior clinically-recognized heart disease and major co-morbidity. Interviewers obtained a detailed history of MI and SD in first-degree relatives from spouses of 235 cases and 374 control subjects. A parental history of early-onset SD (age <65) was associated with an increased risk of PCA (odds ratio (OR)=2.69, 95% CI=1.35-5.36), after adjustment for parental history of MI and other risk factors. A parental history of late-onset SD was not associated with PCA risk (OR=0.94, 95% CI=0.55-1.62). Additionally, parental history of SD was related to early-onset PCA (OR=1.89, 95% CI=1.08-3.30) but not to late-onset PCA (OR=0.89, 95% CI=0.49-1.61). In contrast, parental MI (early/late) was related to PCA (early/late), after adjustment for other risk factors and parental history of SD. Similar results were observed in first-degree relatives. Findings suggest a potential role of familial factors related to both MI and SD in PCA. Stronger findings for a familial patterning of PCA were noted for early onset disease in cases and their relatives.

Adolescent antecedents of high-risk driving behavior into young adulthood: substance use and parental influences.

Driver history data, in combination with previously collected tenth-grade questionnaire data, for 4403 subjects were analyzed by Poisson regression models to identify the significant substance use and parental characteristics predicting subsequent high-risk driving of new drivers (starting at age 16) through age 23-24 years. Substance use (cigarettes, marijuana, and alcohol) reported at age 15 was shown to be an important predictor of subsequent excess risk of serious offenses and serious crashes for both men and women. In addition, negative parental influences (lenient attitudes toward young people's drinking; low monitoring, nurturance, family connectedness), were also demonstrated to increase the risk of serious offenses and serious crashes for both men and women.

Low density lipoprotein particle size and risk of early-onset myocardial infarction in women.

Previous studies of middle-aged men have shown a univariate association between low density lipoprotein (LDL) particle diameter (size) and coronary heart disease, but this association has yet to be examined in younger women. Using a subsample from a population-based case-control study of women living in western Washington State, the authors examined the association between LDL particle size and risk of early-onset myocardial infarction (MI) in 1992-1995. Gradient gel electrophoresis was used to characterize LDL subclasses in nonfasting blood samples from 72 MI cases and 159 controls aged 20-44 years. Mean LDL particle size in cases was significantly smaller compared with controls (26.4 vs. 26.9 nm, p < 0.001), with an odds ratio of 2.3 (p < 0.0001) for a 1-nm smaller LDL particle size. These results were independent of age, menopausal status, smoking, diabetes, hypertension, and LDL cholesterol (odds ratios = 1.9-2.3 for a 1-nm smaller LDL particle size, all p < 0.02) but were not independent of body mass index, high density lipoprotein cholesterol, or triglyceride (odds ratios = 1.4, 1.4, and 1.1, respectively; all p > 0.05). Therefore, in age-adjusted analyses, smaller LDL particle size was associated with MI in young women, but the risk was attenuated after adjustments for metabolic factors related to both LDL particle size and MI.

Long-term follow-up of a high school alcohol misuse prevention program's effect on students' subsequent driving.

BACKGROUND: Alcohol-related injuries, particularly motor vehicle, are an important cause of adolescent mortality. School-based alcohol prevention programs have not been evaluated in terms of driving outcomes. This study examined the effects on subsequent driving of a high school-based alcohol prevention program. METHODS: The Alcohol Misuse Prevention Study included a randomized test of the effectiveness of an alcohol misuse prevention curriculum conducted among 4,635 10th-grade students. Students were assigned to intervention (n = 1,820) or control (n = 2,815) groups and were followed for an average of 7.6 years after licensure, which typically occurred during or shortly after 10th grade. Outcomes examined included alcohol-related and other serious offenses, and at-fault, single-vehicle, and alcohol-related crashes. RESULTS: Only serious offenses (which included alcohol-related) had a significant treatment effect (statistically marginal) after we adjusted for sex, age, race, alcohol use/misuse, family structure, presence of prelicense offenses, age of driver licensure, and parental attitudes toward teen drinking. The effect was found only during the first year of licensure (estimated adjusted relative risk = 0.80, confidence interval = 0.63-1.01). Two first-year serious offense interactions were found. The positive effect was strongest among the largest subgroup of students, those who were drinking less than one drink per week on average before the curriculum, compared with those who drank more than one drink per week (p = 0.009). The effect was also stronger for the small subgroup of students whose parents had not expressed disapproval of teens' drinking, compared with those whose parents had disapproved (p = 0.004). CONCLUSIONS: These findings suggest that a high school-based alcohol prevention program can positively affect subsequent driving, particularly that of students who do not use alcohol regularly. The results highlight the need to start prevention efforts early and extend them beyond the initial exposure to driving. Programs should incorporate the differing backgrounds of the students.

Changes in young adult offense and crash patterns over time.

A study of 13,809 young adult drivers in Michigan examined offenses and crashes ('incidents') for an average of 7 years after their original license date. During this period, 73% of subjects committed an offense that resulted in a conviction and 58% had a crash that was reported to the police. Forty-two percent had committed an offense classified as 'serious,' and 21% had an 'at-fault' crash. The odds of an offense being serious decreased approximately 8% per year of licensure, independent of gender or age at licensure. Similarly, the odds of a crash being at-fault decreased overall about 6% per year of licensure, but the decline was more than twice as fast for women as for men. Examining the empirical rates directly, it was found that the rate for minor offenses increased somewhat with time and then stabilized, while the rate for serious offenses declined. Also, offenses were less likely to be serious the later they occurred in the sequence of offenses for an individual. For crashes, the risk of having an at-fault crash declined more rapidly than the risk of a not-at-fault crash, although the rate of decrease began to equalize after approximately 5 years of licensure. The proportion of crashes that were at-fault did not decline over the sequence of crashes for an individual. Although crashes and offenses are positively correlated, they follow different trajectories over the early years of licensure.

RR interval variation, the QT interval index and risk of primary cardiac arrest among patients without clinically recognized heart disease.

AIMS: Autonomic tone influences RR interval variation (RRV) and the heart rate-corrected QT interval index (QTI). Together, QTI and RRV may improve characterization of sympathovagal control and estimation of risk of primary cardiac arrest. We therefore examined effects of QTI and short-term RRV from standard, 12-lead electrocardiograms on risk of primary cardiac arrest among persons without clinically recognized heart disease. METHODS AND RESULTS: We analysed data from a case-control study of risk factors for primary cardiac arrest among enrollees in a large health plan. Cases (n=505) were enrollees aged 18 to 79 years without history of heart disease who had primary cardiac arrest between 1980 and 1994. Controls (n=529) were a demographically similar, stratified random sample of enrollees. We determined enrollee characteristics from ambulatory medical records, QTI and RRV from standard, 12-lead electrocardiograms, and medication use from automated pharmacy files. Low and high values of QTI and RRV were designated as the first and fifth quintiles of QTI (96% and 107%) and RRV (35 ms and 120 ms) among controls. In a model adjusting for clinical predictors of primary cardiac arrest, RRV modified the association between QTI and risk of primary cardiac arrest (P=0.05). Compared to high RRV and low QTI, the risk of primary cardiac arrest (odds ratio [95% CI]) was 0.95 [0.73-1.23] at low RRV and QTI, 1.23 [0.97-1.57] at high RRV and QTI, and 1.55 [1.16-2.06] at low RRV and high QTI. Risk remained elevated after adjustment for other electrocardiographic predictors and medication use. CONCLUSION: Autonomic dysfunction, characterized by high QTI and low RRV on the standard, 12-lead electrocardiogram, is associated with an increased risk of primary cardiac arrest among persons without clinically recognized heart disease.

Influenza vaccination and the risk of primary cardiac arrest.

Influenza epidemics are associated with an excess of mortality not only from respiratory diseases but also from other causes, and cardiovascular mortality increases abruptly during influenza epidemics, with little evidence of a lag period. In a population-based case-control study, the authors examined whether influenza vaccination was associated with a reduced risk of out-of-hospital primary cardiac arrest (PCA), a major contributor to cardiovascular mortality in the community. Cases of PCA (n = 342) without prior heart disease or life-threatening comorbidity that occurred in King County, Washington, were identified from paramedic incident reports from October 1988 to July 1994. Demographically similar controls (n = 549) were identified from the community by using random digit dialing. Spouses of subjects were interviewed to assess treatment with influenza vaccine during the previous year and other risk factors. After adjustment for demographic, clinical, and behavioral risk factors, influenza vaccination was associated with a reduced risk of PCA (odds ratio = 0.51, 95 percent confidence interval: 0.33, 0.79). The authors suggest that while the association of influenza vaccination with a reduced risk of PCA is consistent with cohort studies of influenza vaccination and total mortality, further studies are needed to determine whether the observed association reflects protection or selection.

Lipid-lowering medication and risk of injury.

Meta-analyses of early primary prevention trials of lipid-lowering therapies suggested increased risk of injury deaths among treated persons. Our population-based case-control study examined the association of lipid-lowering medication use with fatal and nonfatal injuries in 298 cases and 332 controls. No increased injury risk was observed among current (OR = 0.46, 95% CI 0.18-1.21) or past users (OR = .92, 95% CI 0.44-1.95), after adjustment for behavioral disorders, medical conditions, and health status. Stratified analyses did not reveal sub-groups at significantly increased risk. These results, consistent with recent clinical trials and meta-analyses, suggest no increased injury risk associated with lipid-lowering medications.

Prevalence and severity of urinary incontinence in older African American and Caucasian women.

BACKGROUND: Few studies have investigated the prevalence and severity of urinary incontinence in older African American women. Comparisons of findings with those for older Caucasian women could provide important clues to the etiology of urinary incontinence and be used in planning screening programs and treatment services. METHODS: Data are from the first wave of the Asset and Health Dynamics Among the Oldest Old (AHEAD) study. A nationally representative sample of noninstitutionalized adults 70 years of age and older was interviewed. African Americans were oversampled to ensure that there would be enough minority respondents to compare findings across racial groups. RESULTS: A statistically significant relationship was found between race and urinary incontinence in the previous year: 23.02% of the Caucasian women reported incontinence, compared with 16.17% of the African American women. Other factors that appear to increase the likelihood of incontinence include education, age, functional impairment, sensory impairment, stroke, body mass, and reporting by a proxy. Race was not related to the severity (as measured by frequency) of urine loss among incontinent older women. CONCLUSION: This study identifies or confirms important risk factors for self-reported urinary incontinence in a national context, and suggests factors leading to protection from incontinence. Race is found to relate to incontinence, with older African American women reporting a lower prevalence.

Leisure-time physical activity and the risk of primary cardiac arrest.

BACKGROUND: Because the risks of sudden cardiac death and myocardial infarction are transiently increased during acute bouts of high-intensity activity, it is an important question from the public health perspective whether regular participation in moderate-intensity activity confers overall protection from sudden cardiac death. PARTICIPANTS AND METHODS: We used data from a population-based case-control study to assess the associations of regular high-intensity and moderate-intensity leisure-time physical activity with primary cardiac arrest. Cases were patients with primary cardiac arrest, aged 25 to 74 years, attended by paramedics between 1988 and 1994 in King County, Washington (n = 333). Controls were randomly identified from the same community (n = 503), matched for age and sex. All case patients and controls were free of prior clinical heart disease, major comorbidity, and self-reported poor health. Spouses of case patients and controls were interviewed to assess participation in 15 high-intensity and 6 moderate-intensity physical activities during the prior year. RESULTS: Compared with subjects who performed none of the activities, the odds ratio for primary cardiac arrest from matched analyses was 0.34 (95% confidence interval, 0.13-0.89) among subjects who performed only gardening activities for more than 60 minutes per week; 0.27 (95% confidence interval, 0.11-0.67) among subjects who walked for exercise for more than 60 minutes per week; and 0.34 (95% confidence interval, 0.16-0.75) among subjects who engaged in any high-intensity activities, after adjustment for age, smoking, education, diabetes, hypertension, and health status. CONCLUSIONS: The results suggest that regular participation in moderate-intensity activities, such as walking and gardening, are associated with a reduced risk of PCA and support current exercise recommendations.

Stroke and use of low-dose oral contraceptives in young women: a pooled analysis of two US studies.

BACKGROUND AND PURPOSE: The available data on low-dose oral contraceptive pill (OCP) use and stroke risk in US women are limited by small numbers. We sought more precise estimates by conducting a pooled analysis of data from 2 US population-based case-control studies. METHODS: We analyzed interview data from 175 ischemic stroke cases, 198 hemorrhagic stroke cases, and 1191 control subjects 18 to 44 years of age. RESULTS: For ischemic stroke, the pooled odds ratio (pOR) adjusted for stroke risk factors for current use of low-dose OCPs compared with women who had never used OCP (never users) was 0.66 (95% confidence interval [CI], 0.29 to 1.47) and compared with women not currently using OCPs (nonusers) the pOR was 1.09 (95% CI, 0.54 to 2.21). For hemorrhagic stroke, the pOR for current use of low-dose OCPs compared with never users was 0.95 (95% CI, 0.46 to 1.93) and compared with nonusers the pOR was 1.11 (95% CI, 0.61 to 2.01). The pORs for current low-dose OCP use and either stroke type were not elevated among women who were >/=35 years, cigarette smokers, obese, or not receiving medical therapy for hypertension. pORs for current low-dose OCP use were 2.08 (95% CI, 1. 19 to 3.65) for ischemic stroke and 2.15 (95% CI, 0.85 to 5.45) for hemorrhagic stroke among women reporting a history of migraine but were not elevated among women without such a history. Past OCP use (irrespective of formulation) was inversely related to ischemic stroke but unrelated to hemorrhagic stroke. CONCLUSIONS: Women who use low-dose OCPs are, in the aggregate, not at increased risk of stroke. Studies are needed to clarify the risk of stroke among users who may be susceptible on the basis of age, smoking, obesity, hypertension, or migraine history.

The contribution of baseline weight and weight gain to blood pressure change in African Americans: the Pitt County Study.

PURPOSE: The positive association between obesity and blood pressure has been less consistent in African Americans than whites. This is especially true for African American men. This study investigated the sex-specific associations between baseline body mass index (BMI), weight change (kilograms), and five-year hypertension incidence and changes in blood pressure in a cohort of African Americans ages 25-50 years at baseline. METHODS: The Pitt County Study is a longitudinal investigation of anthropometric, psychosocial, and behavioral predictors of hypertension in African Americans. Data were obtained through household interviews and physical examinations in 1988 and 1993. RESULTS: Baseline BMI was positively and independently associated with changes in blood pressure after controlling for weight change and other covariates. When participants were stratified by sex-specific overweight vs. nonoverweight status at baseline, weight gain was significantly associated with increases in blood pressure only among the initially nonoverweight. CONCLUSIONS: Baseline weight for all respondents, and weight gain among the nonoverweight at baseline, were independent predictors of blood pressure increases in this cohort of African Americans.

Myocardial infarction and use of low-dose oral contraceptives: a pooled analysis of 2 US studies.

BACKGROUND: Population-based case-control studies to assess the relationship of low-dose oral contraceptive (OC) use with myocardial infarction (MI) were performed at 2 sites in the United States (California and Washington state). The purpose of the present study was to estimate risk of MI in relation to use of low-dose OCs in a pooled analysis combining results from the 2 sites. METHODS AND RESULTS: The study included as cases women aged 18 to 44 years with incident MI who had no prior history of ischemic heart disease or cerebrovascular disease. Women in the case and control groups were interviewed in person regarding OC use and cardiovascular risk factors. The analysis included 271 MI cases and 993 controls. Compared with noncurrent users, the adjusted pooled odds ratio for MI in current OC users was 0.94 (95% CI, 0.44, 2.20) after adjustment for major risk factors and sociodemographic factors. Compared with never users, the adjusted pooled odds ratio for MI was 0.56 (0.21, 1.49) in current OC users and 0.54 (0.31, 0.95) in past OC users. Among past OC users, duration and recency of use were unrelated to MI risk as was current hormone replacement therapy. There was no evidence of interaction between OC use and age, presence of cardiovascular risk factors (hypercholesterolemia, hypertension, diabetes), obesity, or smoking. CONCLUSIONS: We conclude that low-dose OCs as used in these populations are safe with respect to risk of MI in women.