An Evolutionary Framework for Understanding Sex Differences in Croatian Mortality Rates

Being male is the strongest demographic predictor of early mortality in Croatia. For every woman who dies between the ages of 15 and 34, three men die. Between the ages of 15 and 54, men are four times as likely as women to die from behavioral causes of death, such as accidents, homicides, and suicides. A causal explanation for sex differences in mortality must be based on an understanding of how sex differences were shaped by natural selection, and how those differences interact with environmental factors to create observed patterns and variations. In brief, males have been selected for riskier behavioral and physiological strategies than women, because of the greater variance and skew in male reproductive success. This paper examines the sex difference in Croatian mortality in three parts. First, we quantify the Croatian Male to Female Mortality Ratio (M:F MR) for 9 major causes of death across age group to provide a richer understanding of the sex difference in mortality from a life history framework. Second, we compare the Croatian M:F MR from behavioral, internal, and all causes with that of the available world population to demonstrate how Croatian mortality can be understood as part of a universal pattern that is influenced by unique environmental context. Third, we investigate how the War of Independence in 1991-1995 affected mortality patterns though its impact on behavioral strategies and the physical embodiment of distress. Gardner (1993, p. 67) noted that young adult males form the front ranks of every nation’s military, and “lacking the opportunity for warfare, some [young adult men] will find other ways to place their lives at risk.” In Croatia and many other countries, being male is the strongest demographic predictor of early mortality. Men die at higher rates from both behavioral causes of death, such as accidents, and behaviorally mediated internal causes of death, such as PSYCHOLOGICAL TOPICS 15 (2006), 2, 351-364 352 cardiovascular disease (Kruger & Nesse, 2004). In recent decades, existing explanations of sex-based mortality differences based only on mechanistic factors have been augmented by explanations of how these differentials emerge from characteristics shaped by sexual selection that interact with environmental factors (including culture, for humans) (e.g., Wilson & Daly, 1993). In humans and most other animals, females are better adapted than males to the environment, as demonstrated by their superior longevity. Differences between males and females are usually a result of sexual selection. In most species, males compete more intensely for mating access to females than females do for males. This is because females usually invest more in offspring, and are thus selected to be choosier in selecting mates (Bateman, 1948; Trivers, 1972). Male competition for mates can include fighting other males for rank or territory, as well as elaborate traits and displays that are attractive to females (Darwin, 1871). Those who succeed in these competitions will have higher reproductive success, and this selects for traits that foster such success, even if those traits also result in behavioral and physiological differences that increase the chances of injury, sickness, and early death. This is a classic illustration of the principle that selection shapes traits not for the welfare of individuals or species, but to benefit their genes (see Williams, 1957; Dawkins, 1976). In highly polygynous species, a few males will have many offspring while many others will have none, generating powerful selection for traits that lead to success in mating competition (Betzig 1986). The results of this selection include elaborate ornaments (such as the peacock’s tail) and armaments (such as a deer’s antlers), all with substantial costs. Humans are far less polygynous than most other primates, but the variation and skew in male reproductive success is still substantially higher than that for females. In humans, displays of wealth and social status may literally be a costly signal analogue to the peacock’s tail. Crossculturally, men are evaluated by potential partners in terms of social status and economic power (e.g., Ardener, Ardener & Warmington, 1960; Buss, 1989, 1994; Feingold, 1992; Kenrick & Simpson, 1997; Townsend, 1989; Townsend & Roberts, 1993; Wiederman & Allgeier, 1992). Measures of male social status and economic power have a direct relationship to reproductive success across a wide variety of societies (see Hopcroft, 2006). During recent human evolution, males who do not have substantial resources or status may have been unable to establish long-term relationships. Thus, sexual selection helps to explain some sex differences in psychology and behavioral tendencies, including the stronger male tendencies for risk-taking, competitiveness, and sensitivity to hierarchy (Cronin, 1991). These attributes are related to competition for resources, social status, and mates (Daly and Wilson, 1985), competition which is hazardous and sometimes fatal (Betzig, 1986; Kaplan & Hill, 1985). Selection pressures unique to females may have also contributed to the divergence in tendencies for risky strategies. Risk-taking is more costly for women because offspring survival depends more on maternal than paternal care and PSYCHOLOGICAL TOPICS 15 (2006), 2, 351-364 353 defense (Campbell, 1999). This may be related to the results of recent psychological research showing sex differences in behavioral responses to stress. The tissue-damaging “fight or flight” responses described universally in psychological textbooks may correspond more closely with male behavioral reactions to adverse circumstances. Women appear to have a "tend-and-befriend" response where nurturing and tending activities help protect and reduce distress in oneself and offspring (Taylor, Klein, Lewis, Gruenewald, Gurung & Updegraff, 2000). Tending and befriending activities help develop and sustain social networks facilitating recovery from stressful situations. The shift in the male allocation of effort from mating to parenting over the life course helps to explain why the peak sex difference in mortality occurs in young adulthood, mostly from behavioral causes (Kruger & Nesse, 2006). The steep discounting of the future associated with risky behavior by young people could be a rational response to uncertain prospects for the future (e.g., Gardner 1993; Wilson & Daly, 1997). Male mating effort peaks in young adulthood, possibly in part because young men do not have outlets for parental investment, and they may be more attractive partners because they have not committed their current and future resources (Hill & Kaplan, 1999). In both ancestral and modern times, men who control more resources married younger women, married more women, and produced offspring earlier (Betzig, 1986; Low 1998). Even relatively egalitarian foraging societies have different levels of status, and higher status men have greater access to mates (Chagnon 1992; Hill & Hurtado 1996). Male and female mortality rates are not however, genetically determined. All aspects of phenotype, including tendencies for risky behaviors, are influenced by environmental factors that vary by time and culture. Long term historical changes have affected mortality rates since the Pleistocene, including: increased spread of infectious diseases through increasing population size, mobility, and the domestication of animals (Diamond, 1997); public health measures such as improved sanitation and vaccination (McKeown, 1979); the emergence of scientific medicine including antibiotics; the increased availability and consumption of fatty foods, alcohol, tobacco, and other drugs (Eaton, et al., 2002); and the widespread availability of automobiles and lethal weapons. These changes have resulted in both the recent dramatic decline in mortality from infectious diseases (Cutler & Meara, 2001) and the increasing prominence of mortality from causes directly or indirectly influenced by behavior, most of which disproportionately affect men. The decline in maternal mortality in recent decades has also dramatically decreased the female mortality rate and increased the divergence from the male mortality rate (Guyer, Freedman, Strobino & Sondik, 2000). Recent historical events have also influenced mortality rates. The economic and political changes in the former Soviet Union led to increased inflation, unemployment, and lower wages (Little, 1998). Physical hardships, social disruption, and social distress associated with the 44% decline in Russia’s GDP are believed to have caused 3.4 million pre-mature deaths (Rosefielde, 2001). The PSYCHOLOGICAL TOPICS 15 (2006), 2, 351-364 354 increase in mortality rates was more pronounced for men than for women (Little, 1998), although a portion of the increased mortality differential can be attributed to inadequacies in heath care (Andreev, Nolte, Shkolnikov, Varavikova & McKee, 2003). Male life expectancy in Russia declined by six years between 1991 and 1994 (Cockerham, 1997). Cultural factors such as social norms may increase or decrease risky behavioral tendencies that influence differential mortality risks (see Kraemer, 2000). For example, expectations for boys to be tough and not express emotions such as anxiety and shame may result in riskier behavior (Kindlon & Thompson, 1999). The (paradoxical) belief that men are inherently tougher may also result in discrimination in providing medical assistance in life-threatening situations (Moynihan 1998). Also, favoritism for male offspring in some cultures leads to higher female mortality rates in infancy and childhood owing to infanticide and neglect (Hrdy, 1999; Rahaman, Aziz, Munshi, Patwari & Rahman, 1982). Of course, social norms do not exist in a vacuum and may reflect a combination of the common heritage of adaptations to ancestral challenges and features of the current ecological environment. The largest differences between male and female mortality rates occur from behavioral causes during young adulthood, but the greatest proportion in excess male life years lost result from behaviorally moderated internal causes (Kruger & Nesse, 2004). Sexual selection has shaped a riskier physiological strategy in males, which includes greater vulnerability to infection, injury, stress, physical challenge, and degenerative diseases (Kraemer, 2000). The epidemic of coronary heart disease in industrialized countries, resulting from increased consumption of dietary fats, affects men more so than women (Lawlor, Ebrahim, & Smith, 2001). Physiological susceptibility is exacerbated by higher rates of health adverse behaviors in males, including smoking, heavy alcohol consumption, and work in hazardous occupations (Hazzard, 1986). Elevated rates of alcohol consumption in males leads to increased risk of chronic liver disease and cirrhosis (Zhang, Sasaki & Kesteloot, 1995). Our integrative evolutionary framework for understanding morality patterns can provide insight on sex differences in modern Croatian mortality. The Croatian Male:Female Mortality Ratio (M:F MR) is expected to share aspects of the universal human pattern, but also show variations based on unique local cultural and historical factors. Documenting the M:F MR across ages and causes will provide precise picture of the magnitude of mortality differences in Croatia and will provide a richer understanding beyond the acknowledgement of a general sex difference. Comparison of the Croatian population to other populations where high quality mortality data is also available can highlight distinctive aspects of Croatian demography that require cultural and historical explanations. The Croatian Homeland War or War of Independence in 1991-1995 created a naturalistic experiment in which we can examine the impact of war and cultural disruption with high quality mortality data. These events include the sometimes PSYCHOLOGICAL TOPICS 15 (2006), 2, 351-364 355 forcible internal displacement of hundreds of thousands of individuals, influx of hundreds of thousands of refugees from Bosnia and Herzegovina and other nations, mortar shelling of urban populations, and rise in unemployment rates. We expect that the civilian M:F MR will rise due to the shift towards riskier behavioral strategies induced by evolved facultative adaptations responding to adverse and unstable environments. The mortality differences for internal causes of death are also likely to increase, as males may be more physiologically susceptible to substantial stress. The M:F MRs for some internal causes of death are expected to rise more so than others. For example, physiological distress is expected to have a greater impact on mortality rates from diseases of the digestive system, such as stress exacerbated ulcers, than on malignant neoplasms (cancers). Infectious disease rates are expected to rise due to the mass movement of individuals and probable decline in sanitation in areas affected by combat and/or massive population influx. Males are expected to be more susceptible to the increase in prevalence of infectious agents, resulting in a greater divergence of male and female mortality rates.