The gender longevity gap: is it really biology?
Article Outline
In surveying nations on per capita income and rates of colon cancer, a consistent statistically significant positive association is found. Nations with higher per capita incomes tend to have higher rates of colon cancer. However, does the simultaneous coexistence of these two factors therefore mean that it is accurate to assume that the act of earning more money is the cause of colon cancer? The answer is emphatically no. A cardinal principle of the science of epidemiology is that association does not prove causation. Broad, sweeping ecologic measurements cannot reliably predict biologic effects at the individual level because they neither control for nor exclude competing variables. Uncontrolled variables in any study may increase, decrease or account wholly for the observed effect. When this occurs it is said that confounding exists, because we are not sure which variable or variables actually caused the observed effect. The observed increase in colon cancer is actually linked to more frequent consumption of red meat, owing to greater availability in nations where more people can readily afford it. The false association of income with rates of colon cancer is termed an ecological fallacy [1].
Two other coexisting ecological variables that appear to correlate consistently are gender and life expectancy. In Western nations men do not live as long, on average, as women. A long-accepted assumption has been that any observed excess of male mortality over female mortality may be dismissed as being of natural and biological origin. This view has been so long entrenched in our thinking that it may seem foolish to even think of questioning it. However, over the extensive course of a human lifetime it is likely for a person of either gender to be exposed to numerous non-biological determinants of mortality, some of which may be linked to gender. For example, culturally determined gender roles may determine significant differences in health habits, lifestyle, and exposure to environmental hazards. The assumption that the observed gender disparity in life expectancy exists chiefly, or entirely, because maleness causes earlier mortality might also involve an ecological fallacy. As per capita income served as a false causation for colon cancer, the true causal association being red meat consumption, the relationship between gender and other variables known to determine longevity may also involve confounding, misleadingly pointing to gender as the sole or principal cause of reduced male longevity.
There are a number of possible causal variables that may come between gender and life expectancy outcomes. In the developed nations, life expectancy increased substantially in the 20th century. Life expectancy at birth in the United States in 1901 was only 49 years, but by the end of the 20th century it was 77 years [2]. Life expectancy increased in this period by approximately 57%, a gain of nearly 30 years. Speculation about a biological upper limit of human life expectancy has generally increased by roughly 2½ years per decade. Mortality experts have repeatedly proven unable to demonstrate definitively that human life expectancy is close to any ceiling. It may be premature to speak of gender-based longevity differences at a time when neither gender appears to have reached any biological limits of life expectancy. In fact, men today in the U.S. live about 75.2 years [3], [4], surpassing women's life expectancy in the early 1950s (about 70 years) when the gender life expectancy difference was widely publicized by Ashley Montagu. How can it be argued logically that the gender life expectancy difference is biological when neither gender appears to have attained its full biological potential for longevity yet? An alternate evaluation of the gender longevity gap might be that, for some reason, the increase in life expectancy that took place over the course of the 20th century for males did not keep pace with the corresponding increase for females.
Furthermore, population gender differences in life expectancy have not remained stableover time. The current substantial gender differences in life expectancy have not always existed. In 1920 there was only one year's difference between the life expectancy of the genders in the U.S. [3], [4]. In 2004, that difference stood at 5.2 years, having narrowed from a peak of 7.6 years in 1970 [3], [4]. In France, a large life expectancy gap stopped increasing in the early 1980s and has decreased in recent years, as in most industrialized countries [5]. Decreases in male cardiovascular mortality and cancer mortality, especially lung cancer, appear to be primarily responsible among French males. In other European countries, e.g. England, Wales, Sweden, Switzerland, and Italy, reductions in cardiovascular mortality are principally responsible for the decreasing differences. The changes are not attributable to declining female health, but due to acceleration of health progress for males [5].
In addition, if a true gender difference in life expectancy determined by natural factors does exist, no one knows its precise magnitude. Great variations in the gender gap exist regionally. However, no evidence exists for any corresponding regional variation in the biology of the genders. In some countries, the difference is only a few years, while in others it may reach well over 10 years. Eastern European communist countries (such as the Soviet Union, Poland, Czechoslovakia, and Hungary) experienced decreasing life expectancy and increasing mortality from the late 1960s, especially among adult men. In Russia and other former USSR republics after the collapse of the Soviet Union, the life expectancy of men dropped to 59.9 years and that of women to 72.43 years in 1999 [6]. This gender difference was largely attributed to structural conditions and changes in Russian society and, to a lesser extent, to behavioral factors [7]. Another major exception to the general international improvement in life expectancy has occurred in those countries worst affected by AIDS. In Sub-Saharan Africa, AIDS-affected nations have experienced significant declines in life expectancy with a corresponding reduction or elimination of the life expectancy gender gap [8].
Longevity does not appear to be fixed within a gender even within a given region. Socioeconomic status appears to be increasingly related to longevity for both genders [9]. When education, occupation, poverty, income distribution, wealth, and housing quality are used to define levels of deprivation, the life expectancy difference between the most deprived and least deprived increased from 2.8 years (73.0 vs. 75.8 years from 1980 to 1982) to 4.5 years (74.7 vs. 79.2 years between 1998 and 2000). Disproportionate gains in life expectancy by those in higher socioeconomic groups contributed to the widening gap [9]. Life expectancy also varies considerably by race, with no explanatory hormonal or chromosomal differences similar to those that exist for gender. The average life expectancy of an African-American male in Fulton County (which comprises most of the City of Atlanta) remains only about 64 years today. The health concerns facing African-American males are especially grave.
Also, the gender mortality ratio varies prominently by chronological age [10]. From infancy to age four, the relative risk of mortality is 1.26 male: 1.00 female. At age 85 and above the relative risk is only 1.12 male: 1.00 female. However, the greatest disparity occurs between the ages of 15 and 24, when the mortality ratio is many times higher, a striking 2.62 male: 1.00 female [10]. Regarding the hypothesis of biological causation, the observed peak in mortality ratio at this time in the life cycle appears counter-intuitive. At this age, biologically determined causes of death such as cancer, stroke, heart disease, and other major killers are not at all common. The time of greatest biological vulnerability is widely held to be the extremes of age, when mortality caused by biological differences would be expected to be greatest. Paradoxically, this is also the time of least disparity in mortality between the genders.
If biology does not fully explain the observed differences, what other variables might be causing or contributing significantly to it? Could differences in the roles, habits, and treatment of the genders be responsible for much of the gap? It is well known that as a gender, men participate in preventive healthcare considerably less frequently than women. Excluding pregnancy-related visits, the frequency of doctor visits for preventive services such as annual examinations is twice as high for women as for men [11]. Over all age groups, 33% of men, as contrasted with only 19% of women have no regular physician [12]. Notably, the difference is greatest in the 18–29 year age group in which 53% of men, compared with 33% of women, have no regular physician. In the 65-year age group, the difference declines to 10% of men and 6% of women.
Twenty-four percent of men (compared with only 8% of women) have not seen a physician in the past year [12]. Once more, the gender disparity is most pronounced in the 18–29 year old age group, with 33% of males but only 7% of females not having seen a doctor in the past year [12].
Common policy-driven structural differences in the health care system may contribute to this time of greatest gender disparity in mortality ratio and participation in health care. Most children of both genders have their health care provided by pediatricians. At the threshold of adulthood, the pediatrician typically hands off the health care of the young woman to the gynecologist/obstetrician, thereby maintaining continuity of care. By contrast, with no corresponding health specialty for males, when the boy becomes a young man, his health care may simply drop off instead of being handed off to another specialist. Failure to maintain connection with the healthcare system likely contributes to the significantly greater male to female mortality ratio in young adulthood, and fails to habituate young adult males into regularly seeking health care. The opportunity is lost to screen early in life for disease processes and lifestyle risk factors such as hypertension, overweight and obesity, dyslipidemia, glucose intolerance, poor diet, lack of exercise, cigarette smoking and substance abuse, and mental health disorders such as depression. Lack of early identification and intervention concerning these conditions may lead to poor health outcomes, including increased mortality in youth, a predisposition to more serious health conditions in mid and late life, and likely contribute to ultimately limiting male life expectancy overall. In addition, men's work hours frequently conflict with the hours that health care is typically available. There are fewer programs that target men as specifically as women in the health care system. Men's health concerns tend to receive more limited publicity, such as less significant attention and funding for prostate cancer than breast cancer in the U.S. This often results in a lack of public information and awareness as well as less frequent intervention for men's health challenges.
The workplace, as presently structured, is a factor that preferentially keeps men in harm's way. There is an overwhelming male preponderance in workplace mortality related to exposure to workplace hazards. In 2005, 1,097 Canadians died in the workplace, 97% or 1,069 of them were male, with Canadian workplace mortality trends indicating a rising rate for men and a falling rate for women [13].
In the U.S., data on fatal work-related injuries does not include combat deaths. If included, combat would have been America's single greatest source of occupational deaths in 2005. In the non-military sector, traffic accidents, violent acts, and construction accidents all contribute to greater male mortality in the workplace [14]. Over 95% of workers in the 10 most hazardous jobs are men, and over 90% of occupational deaths occur among men [15]. The top 10 most dangerous occupations were fishers and fishing workers, logging workers, aircraft pilots, structural iron and steel workers, refuse and recyclable material collectors, farmers and ranchers, electrical power line installers/repairers, driver/sales workers and truck drivers, agricultural workers, and construction laborers [14]. Many of these occupations involve hazardous environments, working with heavy equipment, working in inclement weather, and other unsafe conditions. Workers may sustain injuries in areas far removed from medical help, and may therefore succumb to injuries that might not otherwise have proven fatal.
Specific hazardous material and environmental exposures occur commonly in the workplace [16]. Asbestos, solvents, welding fumes, pesticides, metals (including lead), noise exposure, allergic and irritant contact dermatitis, ionizing and non-ionizing radiation injury, respiratory toxins, extremes of temperature, and many other potentially injurious exposures may occur [16]. These exposures can lead to immediate and short-term illness and death, or death and disability years or even decades down the road.
Men's lower rate of involvement in preventive health care appears to be related, in part, to traditional male gender role socialization, which typically promotes stoic attitudes towards pain, fear, and illness. Males are often raised from childhood to minimize the signals of their own bodies, taught to expect that ignoring pain will make it simply go away eventually. Perhaps it is noteworthy that very similar expressions of male stoicism appear to cross cultural, political, and socioeconomic boundaries. A major study found that many men would not respond to pain by seeking health care. Less than 20% of males stated that they would seek help promptly even if they were in pain or sick, with 24% stating that they would delay seeking health care as long as possible, and 17% stating that they would delay going to a doctor for a week or more [12]. Other male cultural characteristics that likely impede participation in therapy and other forms of health care include achievement orientation, restricted emotional expression, instrumental nature, self-reliance, and restricted expression of same-sex affection [17]. In being taught to prize self-reliance, many men may chose not to seek outside help even when a problem is recognized, believing that a man is expected to take care of his own problems. Men are less likely to carry health insurance and may be less eligible for public programs specifically subsidizing health care, such as Medicaid in the U.S.
Undoubtedly, there are many other social, cultural, structural, and environmental variables that could potentially impact male longevity negatively, far more than it would be possible to discuss in a single article. In addition, there are unquestionably biological differences that may facilitate or impair longevity for both genders. However, at present, a deterministic view has dominated our thinking, holding that the observed gender differences in longevity are due exclusively to unchangeable biological factors. This assumption has prevented us from searching for and examining specific cultural, environmental, and behavioral factors that might be amenable to modification through targeted preventive health care efforts. Longevity differences between the races are widely regarded as health disparities and considered unacceptable. As a result, there has been dedicated funding, ongoing research into potential causes, and active proposals for policy changes aimed at reducing or eliminating these disparities. In my view, the time has come to pose the following challenge: why should anything less be done for the similarly compelling disparities associated with gender?
References
- . Modern Epidemiology. 2nd edition. Philadelphia, PA: Lippincott-Raven Publishers; 1998;
- Spiritus Temporis Web Ring Community. Life Expectancy over Human History. Available at: http://www.spiritus-temporis.com/life-expectancy/life-expectancy-over-human-history.html.
- Miniño AM, Heron M, Smith BL. Deaths: Preliminary Data for 2004. Health E-stat. Hyattsville, MD: National Centre for Health Statistics (NCHS), released April 2006. Available at: http://www.cdc.gov/nchs/products/pubs/pubd/hestats/prelimdeaths04/preliminarydeaths04.htm.
- Miniño AM, Heron M, Murphy SL, Kochanek KD. Deaths: Final Data for 2004. National Vital Statistics Reports 2007, vol. 55, no. 19. Hyattsville, MD: CDC and National Centre for Health Statistics (NCHS), Available at: http://www.cdc.gov/nchs/data/nvsr/nvsr55/nvsr55_19.pdf.
- . Ecart d’esperance de vie entre les sexes: les raisons du recul de l’avantage feminine. [Gender gap in life expectancy: the reasons for a reduction of female advantage]. Rev Epidemiol Sante Publique. 2004;52(4):333–352[In French]
- . Life expectancy in Central and Eastern European countries and newly independent states of the former Soviet Union: changes by gender. Croat Med J. 1999;40(2):237–243
- . ‘Health is not a man's domain’: lay accounts of gender difference in life-expectancy in Russia. Sociol Health Illn. 2008;30(7):1070–1085
- . Health and disease in southern Africa: a compariative and vulnerability perspective. Soc Sci Med. 2000;50(7-8):965–983
- . Widening socioeconomic inequalities in US life expectancy, 1980–2000. Int J Epidemiol. 2006;35(4):969–979
- Men's Health Network. Excess mortality in males by age group: 2003. In: The Silent Health Crisis ( 2006. Washington, DC: Men's Health Network. Available at: http://www.ok.gov/health/documents/silenthealthcrisis.pdf.
- Brett KM, Burt CW. Utilization of ambulatory medical care by women: United States, 1997–98, Vital Health Stat 13 2001;(149):1–46.
- Sandman D, Simantov E, An, C. Out of Touch: American Men and the Healthcare System. Commonwealth Fund Men's and Women's Health Survey Findings. New York: The Commonwealth Fund, 2000. Available at: http://www.cmwf.org/usr_doc/sandman_outoftouch_374.pdf.
- Centre for the Study of Living Standards. Jobs that can Kill: Workplace Fatalities in Canada by Gender. Ottowa: CSLS, 2005. Available at: http://www.kanetix.ca/ic_life_info_life_articles_30.
- U.S. Department of Labor. National Census of Fatal Occupational Injuries in 2005. Washington, DC: Bureau of Labor Statistics, 2006. Available at: http://www.bls.gov/news.release/archives/cfoi_08102006.pdf.
- . Health Issues in the Black Community. 2nd edition. Jossey-Bass Publishers: San Francisco; 2001;
- Rosenstock L, Cullen MR. Textbook of Clinical Occupational and Environmental Medicine. London: W.B. Saunders Company; 1994.
- . Traditional men in therapy: obstacles and recommendations. J Psychol Pract. 1996;2(3):40–45
PII: S1875-6867(09)00073-6
doi:10.1016/j.jomh.2009.06.004
© 2009 WPMH GmbH. Published by Elsevier Inc. All rights reserved.
