Epidemiology of cardiovascular disease in women - A long way from our “founding fathers”

The early discoveries of the epidemiology of cardiovascular disease (CVD) are fascinating. Through clinical observation, autopsies and experiments, scientists were able to correlate symptoms of chest pain and diaphoresis with autopsy findings of a fatty gritty substance in the arteries noted to be atherosclerosis. Over time it was also noted that one's environment or lifestyle could largely impact genetic risk seen as populations emigrated from one place to another, and when post war food shortages with a scarcity of high fat/calorie foods led to a reduction in heart disease. Observational data through registries and prospective cohort studies have helped define the specific risk factors that contribute to CVD and our preventive strategies, but interestingly, the early studies were predominantly done in white men, leading one to wonder what could have been missed in such a narrow, homogenous population.

In fact, heart disease was originally believed to be one specific to men. Women were educated on how to cook to reduce the risk of heart disease in their spouses with the initial presentation of the American Heart Association's (AHA) Prudent Diet being a public education pamphlet titled “The Way to a Man's Heart.” As described in the astute editorial written by Dr. Nanette Wegner in Circulation [1], this misconception was widely propagated and as recently as the 1960s, there was an American Heart Association conference in Oregon titled “How Can I Help My Husband Cope with Heart Disease?”

Based on this a priori assumption that most forms of coronary heart disease (CHD) existed only in men, many early epidemiologic studies in the 1940s to 1960s were done entirely in men. The Minnesota Business and Professional Men Study was the first study to systematically identify risk factors such as high cholesterol and high blood pressure in the development of CHD in a cohort of 300 men followed for 15 years [2]. The London Transport Workers study compared the rates of CHD between almost 26,000 male drivers and conductors noting that employees in positions with higher physical activity had lower rates of incident CHD [2]. Similar findings were confirmed in the California Longshoremen Study which examined CHD in 6351 men who performed strenuous, moderate, versus light physical activity [2]. The Helsinki Policemen Study was one of the first prospective epidemiological studies examining the relationship between blood glucose and risk for CHD in a cohort of 1326 men [2]. The Seven Countries Study was a massive international study investigating CVD risk differences among entire populations (Finland, Greece, Italy, Japan, The Netherlands, United States, and Yugoslavia) among others. The study noted the influence of modifiable risk factors such as lifestyle and diet on CHD and stroke in a population of 12,763 men [3].

Perhaps the most iconic of the early CVD epidemiologic cohort studies, the Framingham Heart Study provided a basis for the understanding of CHD risk factors in a large cohort that represented two thirds of the town of Framingham, Massachusetts. Interestingly, it was one of the first studies to recruit both men and women equally with the original cohort of 5209 participants consisting of 2336 men and 2873 women [4]. The subsequent Offspring and Omni cohorts were also equally balanced in representation of women and thankfully much more racially and ethnically diverse. The next generation of large cohort studies were truly landmark in that there was a concerted effort to assure equal representation of race, sex, age, socioeconomic status and even urban vs. rural habitats. In the 1980s and 90s, large National Institutes of Health (NIH) sponsored studies such as the Atherosclerosis Risk in Communities (ARIC) study, Multi-Ethnic Study of Atherosclerosis (MESA), Coronary Artery Risk Development in Young Adults (CARDIA), and the Cardiovascular Health Study (CHS) recruited fairly equal numbers of men and women from diverse US communities collecting medical, social, and demographic data from extensive examinations and regular follow-up visits over a number of years [4]. From these studies, even more information on the natural history and variation in CHD and CHD risk factors was obtained and could be compared not just by sex and age, but by other important factors such as race/ethnicity, medical care, and geographic location over time. The 1993 NIH Revitalization Act was instrumental in recognizing the need for better representation of populations that were diverse from the standpoint of sex, race and ethnicity.

On an international level, INTERHEART was a large global case control study of acute myocardial infarction (MI) in 52 countries, with 12,461 cases and 14,637 controls; 24% of the participants were women [5]. This study serves as the basis for the population attributable risk that is now appreciated from modifiable risk factors such as smoking, diabetes, abnormal lipids, abdominal obesity, psychosocial factors, consumption of fruits, vegetables, alcohol and physical activity. These were noted to be significant in both genders and in all ages and regions although the sex specific weight of these risk factors on risk of CVD has been better defined in further studies. Similarly, the Prospective Urban and Rural Epidemiological (PURE) study which represented more than 1000 urban and rural communities in 27 high, middle and low-income countries was also more balanced with a larger percentage of women [6].

Despite the lack of representation of women in early observational studies, their findings were instrumental in characterizing the major risk factors for CVD that appear to broadly apply across gender, race and ethnicity. What then was missed in these early cohorts? Some of the key questions that went unanswered in women's heart disease include the impact of hormones on cardiovascular disease with specific attention to the menstrual cycle, pregnancy and menopause, the unique impact of psychological stress in women, and differences in symptomatology. Specific disease states either unique to or more prevalent in women such as spontaneous coronary artery dissection (SCAD), Takatsubo's cardiomyopathy and microvascular disease were also not appreciated as early as classic atherosclerosis, perhaps also due to their decreased prevalence.

Fortunately, in addition to the diverse patient cohorts that currently exist, there are large studies specific to women that help to enhance our understanding of disease. The Nurses' Health Study (NHS) and the Nurses' Health Study II (NHS II) are among the largest prospective investigations into the risk factors for major chronic diseases in women [7]. NHS was established in 1976 with over 121,700 female nurse participants has broadened in scope and cumulatively all the prospective cohort studies have followed more than 275,000 participants and supported hundreds of publications. The cardiovascular component that had over 120,000 participants contributed to many important findings including risk of post-menopausal menopause replacement therapy (MRT), smoking, and trans-fat consumption as it relates to CVD. The observational cohort of the Women's Health Initiative has also been monumental with respect to understanding the physiologic effects of menopause. A lot has been learned about women's health and menopause-related factors, however key questions such as the safety of MRT still remain only partially answered. Another key study in women's CVD is the Study of Women's Health Across the Nation (SWAN) which brought to light the key correlation between adverse pregnancy outcomes and maternal cardiovascular disease leading to more refined risk algorithms [8]. With respect to pregnancy, a state where randomized controlled trials have been largely avoided due to potential risk in mother and fetus, studies such as the European Society of Cardiology Registry of Pregnancy and Cardiac disease (ROPAC) have served to fill the knowledge gaps which is greatly needed [9]. Additionally, registries such as the Mayo Clinic SCAD registry and others serve to elucidate the patterns that define that disease state as we gain insight into best therapeutic approaches [10].

Had the early epidemiologic studies included women, would we have had a better understanding of CVD in women? Could specific disease states have been as well defined as traditional atherosclerosis? Could we have learned more about the impact of various risk factors in women? Although this may be the case, in the world of observational research, we have come a long way in representing diversity in sex, race and ethnicity. Future efforts need to focus efforts on the specific knowledge gaps that remain. Current existing registries can be used to analyze sex differences if they have not previously been explored. Real world data from large health systems and clinical databases can also be used to examine gender differences in risk factors, treatment patterns and outcomes. Women-focused registries dedicated to disease states specific to women, as well as the use of therapies and interventions can help to guide clinical practice. Attention must also be paid to the barriers and challenges women may face in participation in studies, such as time constraints to volunteer due to caregiving needs, financial restraints, scheduling, and other social determinants. Such efforts in conjunction with a focus on enrollment of women in clinical trials should serve to improve care for women in generations to come.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.