Differences in Health Outcomes between Men and Women: Biological, Behavioral, and Societal Factors

In the US and almost all other high-income countries, women live longer than men. As we enter a new era in which biomedical data are increasingly ubiquitous, it will serve us well to reenvision the research needed to understand the fundamental issues that drive differences in longevity and other health outcomes between men and women. The “Fourth Industrial Revolution” in which biological, physical, and information sciences are merging opens up important possibilities. The ability to measure the entire spectrum of information about the human biological, environmental, and behavioral condition will become both routine and relatively inexpensive: from genes to genomics; from clinical data to the electronic health record and insurance claims; from digital information about behavior and social interaction to geospatial referencing. In addition, because of ubiquitous access to cell phones and the Internet, these streams of information can be accessed in real time. Amid this revolution in the global scientific and social fabric, some of the most fundamental issues in the human condition are the differences in biology, behavior, and outcomes as a function of sex and gender. Sex differences begin with a difference in a single chromosome that leads to multiple obvious physical differences, but the evolution of societies and the changing roles of individuals within societies are also important. In parallel with variation in physical and biological function, gender identity provides a related but often complex construct with distinct biological and social influences. At the beginning of any such discussion, it is useful to acknowledge that beyond the X and Y chromosomes and direct biological structural manifestations, men and women overlap much more than they differ, so that differences should be understood as occurring within a broad spectrum rather than pointing to specific distinct differences. As far as we can tell, for most of human history, men have had life spans equal to or longer than women. For example, in 1812, the estimated mean life expectancy at birth in the city of Boston was 28 years for men, and the life expectancy for women was 25 years (1). The lower life expectancy for women was attributed to the high rate of death during childbirth. The past 2 centuries, though, have seen immense progress in biomedical science and public health that has led to unprecedented extensions in life span for both men and women. Major reductions in deaths during childbirth, especially, have resulted in women having consistently higher life expectancy than men, especially in high-income countries. This trend is clearly demonstrated in the US according to data collected by the National Center for Health Statistics (Fig. 1). (A), Trends in life expectancy at birth for US individuals from 1946 to 2016. Data sourced from the National Health Statistics Center by sex. (B), Age-adjusted death rates for US individuals in 2016 by sex. Data sourced from the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research (WONDER) Underlying Cause of Death database. CKD, chronic kidney disease; CLD, chronic lung disease; CVA, cerebrovascular accident. Recently, the US has seen a troubling reversal in the overall gains in life expectancy that have been consistently noted since the end of the Second World War. US life expectancy dropped for the second year in a row in 2016, a trend primarily driven by increases in deaths among men because of opioid overdose, suicide, and cardiometabolic disease. In addition, differences in life expectancy as a function of sex over the past decade shrank to a historical low of 4.8 years from 2010 to 2015 before widening to 5.0 in 2016 (Fig. 1). It is important to note that differences in mortality between men and women are not uniformly distributed. The pioneering “Eight Americas” work by Murray and colleagues demonstrated that when simply divided into categories by sex, race, and ethnicity, 8 distinct patterns of survival are observed, with more than a decade's difference in life expectancy between Asian women and African American men (2). More recently, distinct geographical patterns in the US have been observed, driven by differences in wealth, educational level, sex, race, and ethnicity (3). These in turn seem to drive differences in key biological and behavioral measures, including blood pressure, lipid concentrations, tobacco use, and obesity (with resulting diabetes mellitus). This sobering reversal in mean life expectancy—one driven by excess deaths among men—and exacerbation of lifestyle and biological interactions calls for a critical appraisal of the state of men's health and the differences between men and women in the US and around the world. The longer life span of women seen in wealthy countries in the modern era has been frequently attributed to fixed genetic or hormonal causes (4). Several theories have postulated that a higher male birth rate among humans might contribute to unspecified evolutionary pressure for shorter male life spans. It has also been suggested that the additional X chromosome in women may represent a “backup” in case genetic abnormalities occur. Many believe that endogenous estrogen confers a beneficial cardiovascular effect, leading to lower cardiovascular mortality among younger women. A higher risk for infections among men has also been postulated as an additional factor contributing to male–female life span discrepancies (5). However, there is a growing consensus that biological mechanisms alone explain only a part of the difference in life spans. Wide variation occurs in the difference in outcomes between men and women, not just between countries but within countries; these variations point to additional factors being at play. Increasingly, male behavior rather than male biology is seen as a major reason for the relatively poor health of men, although at least some deleterious male behaviors may have biological origins. For instance, the male sex hormone testosterone increases the propensity for risk-taking and aggression. According to 2016 data from the Centers for Disease Control and Prevention's Wide-ranging Online Data for Epidemiologic Research (WONDER) Underlying Cause of Death database, men were far more likely than women to die in accidents such as motor vehicle accidents or drownings (Fig. 1). Furthermore, although men are less likely than women to attempt suicide, the means they choose are usually violent and frequently involve firearms (6), another reason why the age-adjusted rate of suicide is much higher in men than women. Poor health-seeking behaviors play a significant role in men failing to access potentially lifesaving healthcare in a timely fashion (7). Men are less likely than women to seek medical or mental healthcare (7). On average, men also exhibit more “restricted emotional expression,” which limits their expression of negative emotions (7). Therefore, even as men may be experiencing distress from psychological illness, they may be less likely to disclose this to friends, family, or healthcare providers. A wide spectrum of adverse psychological reactions to sickness deters men from being seen as needing assistance or care, including embarrassment, denial/minimization of symptoms, negative attitudes toward help-seeking, and poor communication and understanding of medical issues (7). Adverse health-seeking behaviors among men are most likely not fixed but rather actively developed and molded by societal factors in combination with genetic predispositions. This is reflected in a content analysis of men's magazines, which are noted to reinforce traditional hegemonic masculine ideals (8). In men's magazines, men are more often positioned as “self-reliant, independently knowledgeable, stoic and avoiding associations with femininity” (8). In another analysis, older men were largely absent from men's magazines (9). When older men were depicted, they were usually famous and successful celebrities, and there were few discussions about the health challenges men face as they age. However, there is some evidence that the masculine identity can be effectively molded into one that is more likely to practice healthier health habits. The rejection of help and the denial of weakness has been noted to be a core part of Western male cultural identity and is largely ubiquitous in other cultures, albeit with some notable exceptions (10). Even among men, significant disparities result in widely discordant health outcomes. In the US, black men have much shorter life spans than white men and are much more likely to die of cardiovascular disease, cancer, diabetes, cerebrovascular disease, chronic kidney disease, and sepsis (Fig. 2). Much of this increased burden of disease is directly related to socioeconomic, environmental, and behavioral factors. These factors also shape the ideas of masculinity affecting black men, who are much more likely to die of assault and homicide than white men. (A), Trends in life expectancy at birth for US males from 1946 to 2016. Data sourced from the National Health Statistics Center by race. (B), Age-adjusted death rates for US males in 2016 by race. Data sourced from the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research (WONDER) Underlying Cause of Death database. CKD, chronic kidney disease; CLD, chronic lung disease; CVA, cerebrovascular accident. The poor health-seeking behaviors that men exhibit at an individual scale are also visible at a macroscopic scale, leading to relatively poor recognition and advocacy of men's health. Importantly, women's health research advocacy has migrated to focus much more on sex and gender differences (11). Initiatives to improve men's health, however, are much less visible than women's health efforts. We believe such initiatives should focus on modifiable behaviors and should span the entire spectrum of male lives, starting with childhood upbringing and conceptions of what constitutes appropriate expressions of masculinity. This is an area ripe for public–private partnerships aimed at helping men become more skilled at taking care of others and better at taking care of themselves as well. Such an initiative could be part of a research enterprise that addresses the sex survival gap as a multifactorial issue that includes biological, psychological, and psychobiological issues. The complexity of outcomes related to sex is exemplified by differences in outcomes for cardiovascular disease. On average, men develop vascular disease a decade sooner than women and die at a younger age from heart disease, but when women develop ischemic heart disease, their risk of death is similar to or greater than men's (12). As they age, men tend to develop atherosclerosis and dilated hearts, whereas women are much more likely to develop heart failure with preserved systolic function (12). Women are also much more likely to have ischemia without obstructive coronary lesion, a condition that has been the subject of intense study recently (13). Many studies have observed differences in outcomes between men and women with an acute coronary syndrome or in the setting of revascularization. Although women are less likely to have an ischemic event when adjusted for age, once an acute event occurs, women have worse outcomes than men (12). Atypical symptoms are more common in women, an issue thought to lead to delays in administering key acute care therapies (14). The smaller arterial diameter and size of cardiac structures in women is thought to lead to the higher risk of acute complications after surgical and percutaneous coronary revascularization in women (15, 16). Women also have a higher risk of bleeding with the anticoagulant therapy that is often required in cardiovascular intervention (17). A clear example of the interaction between cultural factors and biology was observed in the PURSUIT trial, which enrolled patients with acute coronary syndrome (18). In this trial, a benefit of the drug eptifibatide was observed in men, but the effect in women was negligible, and no benefit was observed in South American women compared with North American women. Analysis of treatment patterns revealed that although eptifibatide was particularly effective in reducing events in the setting of coronary intervention, South American women were much less likely to be referred for revascularization than were North American women. Another interesting example was the early finding that men were more often referred for coronary artery bypass grafting than women (19). Although this was a manifestation of sex bias, deeper investigation found that the excess male referral may have led to unnecessary procedures that exposed men to excess risk. In both cases, the intersection of social and biological factors led to suboptimal clinical practice that was amenable to improvement. As we enter the era of digital ubiquity, we are no longer fundamentally limited in the measurement, storage, and analysis of complex data. Several key areas are now approachable because of these changes. Expanded computing power and the advent of cloud computing, coupled with the ability to use sensors that can integrate continuously captured information over time, open up the opportunity to understand these differences throughout the life span. Previous limitations in the ability to acquire, store, curate, and analyze information have meant that most studies of disparities focused on a limited population or data set. Now we can integrate biomolecular, clinical, and digital information, including geospatial orientation. The development of the fetus and child has been the topic of intense interest, but study has been limited because of the complexity and risk of measurement in the intrauterine environment, as well as the rapidly changing biology of the infant and child. The ability to measure intrauterine biology, use noninvasive imaging, and model the fetal–maternal interaction should provide a new era of understanding of the development of men and women at this critical phase of life. The second major area of interest is the interaction of the evolving biology of male and female gender roles in the context of social interactions at a young age that may shape self-image and societal perceptions. Previous limitations in the time dimension of the data available in this area of research have relegated studies in this arena to slices of information, pieced together through analyses that depend on extrapolations. However, new techniques such as machine learning and natural language processing can now be used to interrogate behavioral and social interactions on a large scale. A third area is that as the issue of gender identity has come under more objective study, it is clear a continuum exists but the parameters are poorly defined, as are the appropriate times and circumstances for medical and surgical interventions. Finally, the ways that men and women interact with the healthcare delivery system and the many cues about health in the context of a plethora of measurements may lead to the development of behavioral interventions designed to optimize how men and women access the health system. Furthermore, tailoring interventions to the individual—a hallmark of “personalized medicine”—will also necessarily include the elements of sex and gender differences. Differences in symptoms, diagnostic strategies, and intervention rates are amenable to objective study, and specific interventions to deal with differences known to occur in areas such as complication rates of revascularization and valvular intervention or incidence of bleeding from anticoagulant therapy may benefit from incorporating sex-specific approaches or analyses that enable a better understanding of differences. In summary, although men and women face many similar health concerns, they also face unique challenges. As we expand our awareness of and response to these issues, it will be important to study sex and gender differences not only in terms of biological differences but also in terms of behavioral and societal factors. The advent of big data, biosensors, global communications, and modern analytics provides an opportunity to accelerate the understanding of the biological, behavioral, and social differences between men and women and apply that understanding to improved care and prevention.