Postmenopausal Health and Disease from the Perspective of Evolutionary Medicine

Anyone seeking to learn more about the subject of menopause1 will find standard clinical and public health policy interpretations readily and prominently available. A simple Internet search for the word “menopause” returns numerous websites from reputable online medical resources2, all of which provide roughly the same, narrow definition: menopause is the complete cessation of menstruation, and occurs twelve months after the final menstrual period, usually between 45-55 years of age. Along with these definitions, much of the accompanying health information addresses symptoms, risk for associated diseases, and treatment options. This information fits the overall medical paradigm that menopause is not simply ARTICLES


Introduction
Anyone seeking to learn more about the subject of menopause 1 will find standard clinical and public health policy interpretations readily and prominently available.A simple Internet search for the word "menopause" returns numerous websites from reputable online medical resources 2 , all of which provide roughly the same, narrow definition: menopause is the complete cessation of menstruation, and occurs twelve months after the final menstrual period, usually between 45-55 years of age.Along with these definitions, much of the accompanying health information addresses symptoms, risk for associated diseases, and treatment options.This information fits the overall medical paradigm that menopause is not simply United States, 39.8% of women aged 45-64 years are obese, 39.7% are hypertensive, and 11.3% are diabetic, whereas rates for the same conditions among women aged 20-44 years are 33.2%,6.9% and 3.2%, respectively(3) 3. Of all new type 2 diabetes diagnoses in U.S. adults over age 20 years, 55% occur in the 45-64 year age range, with an additional 20% occurring after age 65 years (Centers for Disease Control and Prevention (CDC) 2011).
The medical deficiency disease model attributes elevated disease prevalence to the substantial and permanent reduction in circulating levels of the reproductive steroid hormone 17β-estradiol (henceforth referred to as simply estrogen) that occurs within the four years surrounding the final menstrual period (Randolph et al. 2011;Sowers et al. 2008).Estrogen has multiple non-reproductive physiological functions including fat metabolism for energy, and is involved in numerous metabolic signaling pathways (Campbell and Febbraio 2001;Campbell et al. 2003;Spangenburg, Geiger et al. 2012; Sugiyama and Agellon 2012).Thus, increased rates of metabolic diseases in menopausal and postmenopausal women are thought to result in large part from the menopausal reduction in circulating estrogen and the ensuing hyposteroidal physiological environment (Carr 2003).In the medical model, the female body is poorly equipped to maintain normal metabolic function under reducedestrogen conditions, and the postmenopausal period represents a state of "…uncontrolled degenerative loss of homeostasis…" (Austad 1997).
The idea that the female body is ill-prepared to function in the low-estrogen postmenopausal physiological environment derives from a specific view of the human lifespan, which holds that developments in hygiene, sanitation, and medical technology have only recently extended the normal human lifespan into the 6th and 7th decades of life, beyond the evolutionarily programmed physiological lifespan.In other words, these technological developments have resulted in humans "living too long", and outlasting our intrinsic capacity for physiological self-maintenance.This perspective relies on the concept that somatic longevity, the lifespan of non-reproductive physiological systems, mirrors reproductive longevity, or the lifespan of germline cells that allow one to reproduce.Menopause represents the termination of reproductive longevity, which, implicit in the medical view, means that evolution has programmed the maintenance of somatic longevity to cease at this same point.In other words, somatic systems are expected not to have evolved to function past menopause, and are thus incapable of adapting to the hyposteroidal postmenopausal environment.This view lends a character of inevitability to the medical conception of postmenopausal disease, and suggests that women simply live longer than their bodies are built to handle.
The main consequence of the medical deficiency disease model of menopause has been the extensive prescription of hormone replacement therapy (HT), with the intention of replacing lost estrogen to reduce symptoms and prevent development of disease.The promotion of HT as treatment serves as a thorough illustration of the medicalization of menopause, as demonstrated by a recent survey of HT-related pharmaceutical literature.Websites for HT generally cast menopause as having unnatural, negative effects and leading to suffering, not only physically, but also psychologically and socially, and present the physician's perspective as privileged, versus women's own experiences (Charbonneau 2010).Widespread prescription of HT as the solution to the medical problem of menopause in and of itself provides ample reason to seek alternative conceptions of the relationship between menopause and women's health.This is because over the past decade, a series of papers reporting the results of the Women's Health Initiative (WHI) and the Million Women Study have demonstrated an association between taking HT and increased risk of developing cardiovascular disease and breast cancer (Narod 2011;Rossouw et al. 2002).Although these results have been debated since their publication (Tanko and Christiansen 2006; Utian 2012; Shapiro 2004), the rate of women taking HT has since declined (Ettinger et al. 2012).
While decreased HT use represents a modest shift away from the emphasis on pharmaceutical responses to menopause, it appears mainly to be due to women's choices rather than a change in the medical model (Marnocha et al. 2011).Instead, the medical model appears to remain intact, as evidenced by the HT advertising discussed above (Charbonneau 2010), by the reflexive readiness of physicians to prescribe HT (Marnocha et al. 2011), and by efforts in the scientific community to promote HT by citing criticism of the WHI's results (Nedergaard et al. 2013).Whereas women can reduce the physical consequences of the medical model by choosing not to fill HT prescriptions, the continued medicalization of menopause still has important and often negative social and psychological effects on women (Cimons 2008;Charbonneau 2010;Marnocha et al. 2011).This paper presents an alternative approach to understanding menopause and postmenopausal health, that of evolutionary medicine, which follows from the principle that modern human health and disease are at least partially products of the evolutionary forces that have shaped modern human biology and variation (Nesse and Williams 1994;Williams and Nesse 1991;Gammelgaard 2000).As such, the following evidence serves as a critique of the medical model of menopause as a deficiency disease, supporting a different, evolutionary view that situates human menopause and the postmenopausal lifespan in its broader ecological and life history contexts.The evolutionary view derives from a wealth of comparative anthropological and biological evidence, incorporating cross-cultural and cross-species comparisons to more holistically understand the process of menopause and health expectations for the postmenopausal lifespan.Adopting the evolutionary perspective has the potential to change clinical practice in two key ways.The first is to further obviate the routine prescription of HT by establishing that extended maintenance of sound physiological function under low-estrogen postmenopausal conditions is the human norm.In other words, somatic and reproductive longevity are divorced, so that the human female body has evolved to function and remain healthy for a substantial period of time after menopause, despite the accompanying reduction in circulating estrogen.
Second, comparisons of menopausal physiology, postmenopausal survival, and health across species and between human cultures allows for the identification of lifestyle and behavioral factors that may play a role in the etiology of metabolic diseases among women living in industrialized societies, presenting non-pharmacological and non-hormonal avenues for disease prevention.Although there is a vast literature on the physiology of metabolic disease, detailing the relevant physiochemical pathways and mechanisms of preventive strategies is beyond the scope of this paper.Instead, the following sections review the evidence for evolutionary medicine's interpretation of menopause, which can point to promising areas of research into the specific physiological mechanisms that promote postmenopausal health.Overall, this paper contends that the medical paradigm of deficiency disease is out of step with the biological reality of human menopause, and that an evolutionary medicine approach has the potential to enhance medical, cultural, and individual understandings of menopause and the postmenopausal lifespan.

The Process of Menopause and its Relationship to Postmenopausal Health
An important aspect of the evolutionary medicine approach is the recognition that menopause is the culmination of a long-term, highly-variable process, rather than a brief transition or event, as sometimes characterized in the medical model.The length of the menopausal process contributes to its variability and also to variability in postmenopausal physiology, providing the raw material upon which natural selection acts.Also important to recognize is that menopause, defined as the end of menstruation, is better seen as the culmination of a larger and longer process wherein ovulation ceases and reduces fecundity, or reproductive potential, to zero.This larger process of the cessation of ovulation, in which menopause is the most readily observable outcome, places a limit on lifetime reproductive success, which is the key variable in natural selection.Evolutionary interpretations of menopause must therefore incorporate an understanding of variation in the process of menopause and its relationship to reproductive potential.
The capacity to experience menopause is dependent on the property of semelgametogenesis, where all reproductive cells (in this case oocytes, or egg cells) are overwhelmingly produced only early in life, and are not replenished at any point later in the lifespan (Finch 1990;Peccei 2001b;Ellison 2010).Human females are semelgametogenic, a trait they share with all other female mammals and birds (Bribiescas 2006), and human oocyte production occurs largely prior to birth (but see (White et al. 2012) with the number of egg cells peaking during the fifth month of gestation.Thereafter, due to atresia (a form of apoptosis, or programmed cell death) the population of oocytes decreases to roughly two million at birth, and 400,000 at puberty.After puberty, ovulation and continued atresia cooperate to deplete the remaining store (Faddy et al. 1992;Hansen et al. 2008), eventually reaching a threshold of ~1000 surviving ova in the years leading up to menopause.Below this threshold, the remaining oocytes produce very little estrogen, which is required to stimulate uterine wall preparation in anticipation of implantation of the mature egg (Clancy 2009).
Among other hormonal changes, the menopausal reduction in estrogen is thought to disrupt endocrine pathways responsible for ovulation and thus fecundity,  Walker et al. 1984).The age at menopause does, however, vary on population and individual scales, owing to a combination of genetic, environmental and behavioral factors.On the whole, data from diverse human populations consistently fall within the aforementioned age range (see Figure 1).The three lowest population average ages at menopause (≤47 years), however, all come from subsistence farming or herding populations practicing natural fertility (i.e.non-contracepting), suggesting that nutritional and reproductive factors influence menopausal timing.
Whereas the medical model would tend to view such variation and deviations from the norm as pathological (Meyer 2001), the evolutionary perspective sees this variation as potentially adaptive (Gluckman and Beedle 2007; Ellison 2010).Given that the process of oocyte depletion occurs over decades and across several very distinct life history periods, it is possible that variation in age at menopause and postmenopausal physiology may be optimized for an individual in the mode of a predictive adaptive response (Gluckman and Beedle 2007;Gluckman et al. 2005).The "predictive adaptive response" concept suggests that individual and population-level phenotypic variation reflects an evolved flexibility that responds to cues during development, influencing subsequent life history and geared towards maximizing lifetime fertility in the face of an unstable, yet to some extent predictable, environment.This approach has been used extensively to explain variation in menstrual cycle characteristics (Vitzthum 2008;Vitzthum 2009), and in a similar manner may be applicable to menopause and the postmenopausal lifespan (Ellison 2010).In this view, variation in the process of menopause and postmenopausal physiology represents the normal range of phenotypic expression, within constraints set by the specific evolutionary history of the human species.
On top of population-level variation, there is also substantial inter-individual variation within populations in terms of the slowing of reproductive function and age at menopause (Treloar 1981 Although continued research is needed to precisely define the major factors that influence menopause, it is clear that the timing and process of menopause are quite variable, and receive inputs from both external and internal environmental factors.The variability and heritability of the timing of menopause provide the raw material from which adaptations evolve, and part of the menopausal adaptation in humans appears to have been the retention of flexibility as part of an overall ovulatory predictive adaptive response mechanism, aimed at maximizing reproductive success.This flexibility may predispose different women to varying metabolic profiles after menopause, differentiating between levels of function and disease risk in the low-estrogen postmenopausal environment.As seen below, however, the evolutionary evidence also points to selection for extended somatic maintenance well-into the postmenopausal period, which may have also evolved as a method of increasing lifetime reproductive success.
Evolutionary Views of the Postmenopausal Lifespan: Paradox or Adaptation?
Before discussing the postmenopausal lifespan specifically, it is important to understand how biologists conceptualize the more general issue of aging, defined not simply as the accrual of time spent living, but as the loss of function in physiological systems with increasing age (Gavrilov and Gavrilova 2006).Aging in and of itself has long been a problematic concept for evolutionary biology (Weismann 1889; Medawar 1952), since theoretically the most reproductively successful organisms should be those that live forever and never cease to reproduce.Why should selection allow for functional deterioration and the regular occurrence of intrinsic causes of mortality (i.e.death due to failure of internal physiological systems rather than succumbing to extrinsic sources of mortality such as predation), rather than favoring self-maintenance in perpetuity, so long as organisms can avoid extrinsic mortality factors?Reproductive decline, in particular, requires explanation from the standpoint of natural selection (Williams 1957), since reducing fecundity to zero curtails reproductive success and is of no apparent selective value.
The prevailing theory of aging posits that the rate of intrinsic senescence in physiological systems results from the interaction between the population-specific rate of extrinsic-cause mortality (e.g., predation) and age-specific fertility potential (Medawar 1952;Williams 1957;Hamilton 1966;Charlesworth 1994).For any age cohort, the number of individuals left alive at a particular age to pass on genes, and the offspring those remaining individuals will potentially produce as a fraction of total offspring production for that cohort, will determine the force of selection of any gene acting at that particular age.At ages where most individuals are still alive and only a small proportion of potential offspring have been produced (such as shortly after puberty), selection to minimize intrinsic mortality is very high.In contrast, at ages where many individuals from a cohort have fallen to accidental death, predation, disease, or other extrinsic factors, and when most of the likely offspring for that cohort have already been produced, then the strength of selection to maintain somatic integrity against intrinsic sources of mortality is much lower, since it has little effect on subsequent generations' ability to reproduce (Williams 1957;Hamilton 1966;Kirkwood 1977; Kirkwood 1980;Kirkwood and Holliday 1979).
Intrinsic challenges to somatic integrity arise as inborn biochemical byproducts of multiple physiological processes.For example, errors in cellular division, DNA replication, and protein synthesis (Kirkwood 1977; Kirkwood and Holliday 1979; Kirkwood 1980) can occur easily and challenge somatic viability, so that organisms invest considerable energy in mechanisms that increase accuracy in these processes.Such mechanisms are, however, only maintained as long as required to ensure that genes are successfully passed on to offspring at a population-specific rate.With increasing age, higher error rates in these processes lead to accumulation of dysfunction, disruption of physiological systems, and ultimately death (Kirkwood and Shanley 2010).
Challenges to somatic integrity can also arise as a result of antagonistic pleiotropy, where individual genes have positive effects early in life prior to and during the peak reproductive period, but have detrimental effects later in life (Williams 1957).This occurs because genes that promote viability and reproductive success early in life, when much reproductive potential remains, will be strongly selected for, but will be only weakly selected against if they exert negative effects later in life, when fewer individuals remain alive and little if any reproductive potential remains.In fact, menopause itself and the resultant low estrogen environment may be products of antagonistic pleiotropy (Crews 2003).Genes that promote higher circulating estrogen levels during the reproductive years have likely been selected for, since they play a major role in stimulating ovulation and thus raise fecundity.This same process, however, leads to more rapid depletion of oocyte stores, and therefore menopause, after which the ovary drastically reduces estrogen production and poses significant challenges to maintaining the somatic integrity of metabolic and cardiovascular systems.
This sharp increase in disease risk after menopause is in agreement with the overarching biological perspective on aging.Maintaining somatic viability after menopause is anathema to the evolutionary process, so much so that Hamilton (Hamilton 1966) predicted a so-called "wall of death", where intrinsic mortality would increase sharply upon the loss of reproductive capacity.In the same paper, however, Hamilton also noted that humans fail to meet the "wall of death" prediction, with much more gradual mortality rates past the age of menopause than expected from theory.Thus, the human postmenopausal lifespan and its relative rarity in the living world present a theoretical paradox for evolutionary biology.Nonetheless, hunter-gatherer survivorship and life expectancy patterns are remarkably similar to the pre-industrial United States (Figure 4).In both cases, high rates of infant and juvenile mortality result in life expectancy at birth of less than 40 years and 57-65% survivorship from birth to age 15.For those who live to sexual maturity, however, ~65% will also reach age 45 and on average survive for another two decades.The data therefore strongly suggest that even under divergent socioeconomic conditions, a substantial proportion of all females born, and an even greater proportion of those who live to reproductive adulthood, share the expectation of multi-decade postmenopausal survival, amounting to on average a third of the total lifespan.Although extrinsic mortality risk likely differs between wild panins and human hunter-gatherers, much of the inter-species difference in survivorship and longevity past menopause may result from differences in the rate of aging in somatic physiological systems.Critically, although overall mortality rates are lower in captive chimpanzees, they too experience a major increase in adult mortality at roughly the same age as wild populations, implicating intrinsic rather than extrinsic causes of death (Hill et al. 2001).While adult mortality rates increase exponentially and at roughly the same rate in both chimpanzees and hunter-gatherers, the increase in mortality due to intrinsic senescence begins ≥10 years earlier in chimpanzees (Gurven and Kaplan 2007).Wild chimpanzees begin to show signs of somatic aging beginning in their mid-30s, and though these outward signs of deterioration do not always correlate with incapacitation (Tarou et al. 2002;Finch and Stanford 2004), it seems plausible that they tend to succumb to intrinsic somatic senescence at earlier average ages than do humans.Possible mechanisms that explain these apparent species-level differences in intrinsic mortality and somatic longevity are discussed below.

Mechanisms for Extension of the Somatic Lifespan
In seeking explanations for the extension of somatic maintenance past menopause in humans, is it important to consider both the proximate physiological mechanisms by which cellular viability and functional integrity is maintained, as well as the ultimate evolutionary pressures most likely to have driven selection for an extended lifespan.The two main proximate causes of aging are the metabolic production of free radicals, which can oxidize and damage tissues and DNA (HARMAN 1956), and the accumulation of errors in DNA and its macromolecular end-products due to poor replicative control, leading to malfunctioning cells and the eventual interruption of physiological pathways (Orgel 1963; Orgel 1970;Kirkwood 1977; Kirkwood and Holliday 1979; Kirkwood 1980).Extending somatic longevity can be achieved by investing energy in mechanisms that reduce free radical production or activity, and that enhance the accuracy of protein synthesis and cellular replication.
Although somewhat limited, existing data on differences between humans and other primates in the expression of such mechanisms tends to point to greater maintenance of somatic integrity among humans.On a large scale, a reduced metabolic rate may limit free radical production and thus reduce oxidative damage (Ku and Sohal 1993; Barja et al. 1994), and humans have slightly lower than expected mass-specific metabolic rates compared to chimpanzees (Froehle and Schoeninger 2006).On a more molecular scale, blood levels of the free radical scavenger uric acid (Ames et al. 1981) tend to be higher in primates than in most mammals (Friedman et al. 1995), and are especially high in apes and humans (Wu et al. 1992;Oda et al. 2002).Humans and apes, to the exclusion of monkeys, also share mutations in the genes responsible for another group of free radical scavengers, the superoxide dismutases (Fukuhara et al. 2002), which are more active in longer-lived primate species and are most active in human organ tissues (Tolmasoff et al. 1980).
In terms of DNA and fidelity in macromolecular synthesis and cellular replication, humans have higher rates of DNA repair than apes (Cortopassi and Wang 1996).Overall, then, in keeping with a longer lifespan, humans appear to have a greater physiological capacity for somatic maintenance via free-radical management and accuracy in macromolecule transcription and translation compared to other primates.
Recent research has expanded beyond the above domains to include comparative studies of other biomarkers of aging in primates.Perhaps the most extensive body of work has focused on dehydroepiandrosterone sulfate (DHEAS), an important androgen hormone with a wide variety of physiological roles, and protective effects against several metabolic and cardiovascular diseases (Lane et al. 1997) In terms of ultimate evolutionary mechanisms for the extension of the somatic lifespan past menopause in humans, multiple theorists (Williams 1957;Hamilton 1966) have proposed the existence of a critical breach in the so-called "wall of death".In species with relatively low lifetime fertility and high levels of long-term parental care, infant and childhood survivorship can be enhanced by traits that increase the delivery of energy to offspring.This reproductive strategy is common to most primates, and mother-child food sharing is not uncommon among panins (Ueno and Matsuzawa 2004).Humans, however, represent an extreme version of the primate pattern, with offspring that are slow-developing, energy-needy, and highly nutritionally-dependent for a much longer period of time than panins.Thus, successfully raising human offspring to sexual maturity is an exceptionally energyexpensive endeavor (Gurven and Walker 2006).To address this challenge, humans also diverge from other primates in the frequency of food sharing, the quantity of calories shared, and the breadth of regular, dyadic food sharing relationships (Kaplan and Gurven 2005).
This pattern of food sharing greatly expands the range and frequency of opportunities to indirectly boost reproductive success (i.e., through means other than producing more of one's own children), since a proportion of one's genes are shared with even distant kin.Because of these potential inclusive fitness benefits via broad-based food sharing, the human lifespan is socially selected rather than individually selected.In other words, the human lifespan serves not only to facilitate one's own direct reproductive success, but also facilitates the reproductive success of close relatives, opening up the possibility of lifespan extension via kin selection and inclusive fitness effects (Carey and Judge 2001).This is the crack evolution requires to pry open a hole in the "wall of death": in humans, long-term parental care as well as "…altruistic contributions due to post-reproductives…" (Hamilton 1966)  both propose that the postmenopausal lifespan evolved because it allowed women to boost lifetime reproductive success.In the former, postmenopausal mothers benefit directly by eschewing risky new pregnancies that could jeopardize the survival of existing dependent offspring.In the latter, postmenopausal women indirectly increase fitness by sharing food to promote grandchild production and survival.
Several lines of evidence support the positive effect of postmenopausal women on the reproductive success, growth, development, and survival of their children and grandchildren.In pre-industrial and modern nonindustrial agricultural societies, a maternal grandmother's presence promotes weight gain in grandchildren and increases lifetime fertility in daughters (Mayer 1981 Lee 2003), with the greatest contributions to grandchild survival occurring during the critical period of 1-2 years old when infants are still nursing and on the cusp of being weaned (Shanley et al. 2007).This suggests that provisioning of adult daughters to subsidize nursing energy needs may have served as the primary role for postmenopausal women that initially selected for a longer lifespan.
Though many of the specific tenets of these two hypotheses differ, they share one crucial argument: active foraging and food sharing in support of offspring production and viability have imbued humans with the potential for reproductive value at much older ages than expected from theory, or from comparative data on panins.By extension, these hypotheses predict that selection has favored human phenotypes that retain the capacity for somatic maintenance at those older ages.As such, the human postmenopausal period is interpreted as an evolved, species-typical life history stage, as integral to the human life course as adolescence or childhood, and characterized by low rates of intrinsic senescence in physiological systems, at least into the seventh decade of life (Gurven and Kaplan 2007).This life-history interpretation informs the study of menopause and the postmenopausal lifespan from the perspective of evolutionary medicine.Following these evolutionary hypotheses, it seems reasonable to expect that the evolutionary extension of somatic maintenance should foster low rates of chronic and degenerative diseases, and overall good health during the postmenopausal life history period, at least into the seventh decade of life.Given this expectation, it becomes critical to evaluate the potential lifestyle and behavioral patterns among women in industrial society that may predispose them to postmenopausal disease by diverging sharply from conditions under the natural fertility foraging regime that has likely dominated human evolutionary history.

Evolutionary expectations for postmenopausal health
In seeking lifestyle factors that might underlie postmenopausal disease risk in industrialized societies, it is important to first examine the prevalence of the same diseases among older women in hunter-gatherer societies.Overwhelmingly, hunter-gatherers without access to medical care appear not to experience the high rate of decline in early postmenopausal somatic maintenance seen in industrialized populations.Degenerative diseases in foraging populations (as far as they can be diagnosed) are extremely rare, accounting for less than 3% of deaths before age 60 (Howell 1979 Hill 2003), including the observation that older Hadza women tend to work longer and perform difficult foraging tasks more frequently than women of reproductive age (Hawkes et al. 1989;Hawkes et al. 1997).These data show that the metabolic syndrome is rare among huntergatherers, likely related to an absence of age-related body composition change and continued physical activity.Maintenance of somatic physiological systems well past menopause appears to be the hunter-gatherer norm, with an absence of the related high mortality rates that would be present in industrialized society without advanced medical treatment (e.g., insulin for diabetes).
The potential cause(s) of postmenopausal health differences between women living in industrialized vs. hunter-gatherer societies must be understood from a biocultural and evolutionary perspective as a consequence of the divergent conditions under which they live.As stated previously, probably the most important physiological change with menopause is the reduction in estrogen levels.This "hyposteroidal" physiological environment is thought to cause disease by undermining the proper functioning of various somatic physiological systems, especially those related to metabolic health.This hypothesis, however, is contrary to expectations if the postmenopausal lifespan indeed represents an evolved life history period.According to Austad (Austad 1997), "… if menopause is an adaptive physiological state molded by evolution… then natural selection would presumably have tailored postreproductive physiology to the hyposteroidal state…" [emphasis in the original].In other words, if selection has promoted the extension of somatic maintenance past menopause, then one would expect postmenopausal physiology to retain functional capacity despite reduced estrogen levels.
Rather than see this necessarily as a contradiction, it is possible that factors other than just a reduction in steroid hormone levels affect the ability of metabolic systems to maintain their operations after menopause.
Alternatively, if lifestyle factors elevate estrogen levels during the premenopausal period in women living under industrialized conditions, the magnitude of hormonal change across the menopausal transition may be greater than in women living in subsistence-level societies, with possibly important physiological consequences (Pollard 2008).Given that physiological systems act not in isolation, but interact and affect one another, it seems reasonable that the physiological effects of behavior and experiences during pre-menopausal life could extend into the postmenopausal period.In the same way that it is reasonable to suspect that selection would have molded human postmenopausal physiology to operate under hypersteroidal conditions, it is also reasonable to predict that the environment (both internal and external) under which individual metabolic physiological systems develop and operate across the reproductive lifespan might predispose those systems to function better or worse in the postmenopausal environment.If this is the case, we might then expect the prevailing conditions of hunter-gatherer life to be integral in preparing metabolic physiological systems for operation in the low-estrogen postmenopausal milieu.From this perspective, cultural variation in subsistence practices and reproduction may be particularly important.
The experience of women in industrialized countries differs in a variety of ways from women's lives in modern hunter-gatherer societies, and many of these same differences likely distinguish industrialized life from conditions during the initial evolution of the postmenopausal life history period.For one, decreased infant and childhood mortality may remove biological filters on survival, thereby promoting higher adult disease risk if phenotypes prone to pre-adult mortality also tend to be prone to chronic disease (Hawkes 2010;Forbes 1997).
Reproductive history also diverges between women in low-fertility industrialized populations vs. women practicing natural fertility.In the latter, higher birthrates, longer nursing, and more frequent lactational amenorrhea (Wood 1994)  Researchers have only recently begun to focus on how these lifestyle and behavioral factors affect metabolic disease risk and prevalence specifically in the low-estrogen postmenopausal physiological context.The combined evidence provides some interesting leads, though, and points to potentially promising avenues for continued research.A common feature of the reproductive, dietary, and activity differences between women in industrialized societies and women in subsistence-level, natural fertility societies, is their effects on premenopausal estrogen levels.High lifetime fertility and breastfeeding, a diet high in fiber and low in fat, and greater levels of activity all characterize the hunter-gatherer lifestyle, and all have the common effect of suppressing estrogen levels during the reproductive years.The opposite is true of the typical lifestyle in industrialized societies, where women are subject to much higher estrogen exposure over the premenopausal lifespan.The difference may be critical to postmenopausal physiological functioning given the key role of estrogen in metabolism.It is possible that low estrogen levels during the premenopausal period "prime" metabolic systems to function under that hormonal regime, so that after the menopausal reduction in estrogen the body maintains greater capacity to function normally, even in the presence of low circulating estrogen.
In contrast, if the body adapts to high estrogen levels prior to menopause, the magnitude of the menopausal drop in estrogen and subsequent low-estrogen postmenopausal environment may more greatly disrupt metabolic pathways, resulting in higher disease risk without substantial intervention.This hypothesis has already been presented with regard to bone disease (Pollard 2008;Galloway 1997)

Conclusion
Overwhelmingly, the evidence from comparative demography and comparative biology contradict the tenets of the medical model of menopause and the postmenopausal lifespan.Rather than being a recent technological development in industrialized societies that extends the lifespan past its evolutionarily intended endpoint, the postmenopausal period is shared broadly across a wide range of human societies living under widely varying nutritional and medical conditions.The shared age at menopause between humans and panins, coupled with greatly reduced mid-life adult mortality rates in the former vs. the latter, also suggest strongly that the postmenopausal lifespan has evolved not by early termination of the reproductive period, but instead by lengthening the lifespan of somatic physiological systems beyond a retained ancestral age range for menopause.
In short, multi-decade postmenopausal survival that accounts for roughly one-third of the total lifespan is an evolved, species-defining trait in humans, and should be considered a normal life history phase akin to childhood, adolescence, or reproductive adulthood.
Whereas the postmenopausal lifespan is universally experienced across societies, however, it does not universally represent the initiation of a period of metabolic disease risk and decline.Given evidence that the lifespan has been selected to last past menopause, the low-estrogen deficiency disease model that prevails in Western medicine is flawed.Rather than a period of uncontrolled disruption to metabolic homeostasis, the evolutionary approach proposes that metabolic physiological systems have evolved to maintain sound function after menopause, despite decreased circulating estrogen.Instead of focusing on prescribing estrogen to cure a perceived deficiency, more effective strategies to minimize postmenopausal metabolic disease among women in industrialized countries should follow from a better understanding of the conditions under which human physiology largely evolved.Four key areas of continued research can potentially contribute to developing preventive strategies by addressing the effects of lifestyle and behavioral factors on estrogen levels during premenopausal life, their effect on the magnitude of the menopausal reduction in circulating estrogen, and the consequences for metabolic function after menopause.These areas are: 1) the effects of decreased infant and child mortality on adult disease risk; 2) the effects of reproductive patterns on estrogen levels; 3) the effects of diet on body composition and metabolic function in relation to estrogen; and 4) the effects of physical activity on metabolic function, again with regard to estrogen physiology.
First, there are clearly ethical concerns with advocating greater childhood mortality or interfering with individual women's choices about reproduction, contraception, or family planning, and so preventive strategies deriving from the first two areas above are unlikely to be developed (nor should they be, in this author's opinion).Studying the effects of early life experiences on later life physiology, however, as well as genetic predisposition to certain conditions, may be fruitful in identifying women who may be at greater risk of developing metabolic diseases after menopause.Such information would be useful in tailoring preventive exercise and dietary programs early in life so as to maximize postmenopausal health.
Certainly, advocating diets higher in fiber and lower in many kinds of fat, along with greater levels of physical activity is important not only to general health, but also specifically for metabolic health after menopause.Initiating these habits as early as possible in the lifespan may be critical to developing a metabolic adaptation to lower levels of estrogen, which may then prime these physiological systems to maintain function after menopause.There is already some interesting thought in this direction (Pollard 2008;Spangenburg et al. 2012), and continued work in this vein has great promise in reducing the prevalence of midlife metabolic disease among women in industrialized societies.Overall, it is critical to understand menopause in its broader evolutionary and biological context so as to recognize that life past menopause is not a disease, but has likely been an integral part of the lifespan of most women for millennia (Blurton Jones et al. 2002;Caspari and Lee 2004).Using the framework of evolutionary medicine to study menopause offers the opportunity to develop lifelong preventive strategies that ensure postmenopausal health, and that have the potential to change societal, and perhaps more importantly for health, medical perceptions of menopause.
Notes 1 This paper focuses on "natural menopause, " the result of innate physiological processes, as opposed to surgical menopause, which is a consequence of various surgical procedures, including hysterectomy and oophorectomy.

Figure 4 :
Figure 4: Survivorship from birth to age 15 years (black bars) and birth to age 45 years (dark gray bars); conditional survivorship from 15-45 years (light gray bars); and PM/LS ratio, the average postmenopausal lifespan (PM) remaining at age 45 years as a percentage of total lifespan (LS), calculated as e45/(e45+45), where e45 is the life expectancy at age 45 years.United States (US) data for 1850(Haines 1994) and for 2000(Bell and Miller 2005); hunter-gatherer (H-G) and chimpanzee data(Gurven and Kaplan 2007).
females have evolved to maintain non-reproductive physiological function for decades past the termination of ovulation?The assumption that somatic and reproductive longevity are closely linked is at least implicit in the medical model of menopause and postmenopausal health.
(Bell and Miller 2005)e apparent disconnect between the reproductive lifespan and somatic longevity: if differential reproductive success is the key functional outcome of evolution, then why would human doubled over that same time period, the average amount of life remaining at age 45 has only increased from ~25 years to ~35 years, with most of the increase occurring after 1940 (Figure3).Thus, despite modest gains in the past 150 Figure 3: L.E. = Life expectancy, or average years of life remaining at a specific age.Life table data for 1850-1890(Haines 1994)and for 1900-2000(Bell and Miller 2005).years, even in 1850 prior to greater sanitation, hygiene and medical technology, over two-thirds of women fortunate enough to live to age 15 could expect to live to age 45, and to then survive an average of another 25 years, or 35% of the total lifespan.the vast majority of human evolutionary history.Huntergatherers obtain food from seasonally fluctuating wild plant and animal supplies, practice natural fertility, and lack reliable access to antibiotics, immunizations, water treatment, and other developments that increase life expectancy at birth in industrialized societies (Wood 1994; Panter-Brick 2002; Blurton Jones et al. 2002; Gurven and Kaplan 2007).As such, hunter-gatherers are expected to differ in terms of mortality and survivorship rates from populations living under agricultural and industrialized conditions.
(Graham 1979;Gould et al. 1981;Jurke et al. 2000;Videan et al. 2006;Lacreuse et al. 2008l women of the same age(Gould et al. 1981;Jurke et al. 2000;Videan et al. 2006).In terms of the effects of oocyte depletion on estrogenic stimulation of ovulation and fecundity, panin menstrual bleeding and sexual swellings, which reach peak frequency, regularity, and duration of maximal tumescence in the 20s, gradually become less frequent, more irregular, and shorter until eventually ceasing altogether in the 40s and 50s(Graham 1979;Gould et al. 1981;Jurke et al. 2000;Videan et al. 2006;Lacreuse et al. 2008).Thus, it appears that humans and panins share an ancestral pattern of oocyte reduction and a similar ovulatory "dose-response" to the small remaining population of sex cells during the fifth and sixth decades of life (Hawkes and Smith 2010).
similar to humans (Kasuya and Marsh 1984; Marsh and Kasuya 1984; Marsh and Kasuya 1986; Olesiuk et al. 1990; Bloch et al. 1993; Martin and Rothery 1993; McAuliffe and Whitehead 2005; Foote 2008), overall humans are unique among mammals, including primates, and are evolutionarily derived relative to panins in having specieswide, common, multi-decade survival past menopause (Caro et al. 1995; Packer et al. 1998; Bronikowski et al. 2002; Pavelka and Fedigan 1999; Bellino and Wise 2003; Cohen 2004; Fedigan and Pavelka 2006).et al. 1998; Hansen et al. 2008; Videan et al. 2008; Jones et al. 2007) that proceeds at essentially the same rate in both taxa (Jones et al. 2007).As in humans, the reduced oocyte population is related to hormonal changes in panins, and reproductive hormone (Bernstein et al. 2012)et al. 1997002)Macaca mulatta) and humans have demonstrated a decline in serum concentrations of DHEAS with age(Kemnitz et al. 2000;Roth et al. 2002), and an association between higher levels of circulating DHEAS and slower rates of aging(Roth et al. 2002;Lane et al. 1997).Compared to chimpanzees, human serum levels of DHEAS decline more rapidly with age, but overall concentrations are two-to-three times higher in humans at any age, and only begin to drop into the high end of the chimpanzee range at ages older than 65 years(Blevins et al. 2013).Other great ape species exhibit even lower DHEAS concentrations(Bernstein et al. 2012), consistent with the idea that higher human levels have evolved since our split with the apes and promote a unique, extended period of somatic viability.Additional research in these areas should help to clarify additional proximate physiological mechanisms by which the human lifespan is extended compared to our close primate relatives.

;
Mayer 1982; Turke 1988; Sear et al. 2000; Sear et al. 2002; Voland and Beise 2002; Lahdenpera et al. 2004; Fox et al. 2010).Hadza hunter-gatherer grandmothers target hardto-obtain, valuable resources that their reproductivelyactive daughters are not as able to procure due to the constraints of pregnancy, nursing, and childcare (Hawkes et al. 1989; Hawkes et al. 1997; Hawkes et al. 1995; Crittenden et al. 2009).As such, childhood weight gain is positively correlated with the amount of time maternal grandmothers living in the same camp spend foraging ; Hill and Hurtado 1996; Gurven and Kaplan 2007).Although degenerative diseases become more common as causes of death after age 60, obesity, hypertension, heart attack and stroke are still very much the exception (Eaton et al. 1988; Gurven and Kaplan 2007).Anthropometric work among the Hadza demonstrates that average body fat percentage remains constant at about 19% in women from age 18-75 (Sherry and Marlowe 2007), in stark contrast to the increase in body fat with age in postmenopausal women from industrialized societies (Heymsfield et al. 1994).Anecdotal and empirical evidence also points to the maintenance of physical vigor in old age in foragers (Blurton Jones et al. 2002; Hawkes et al. 1989; Walker and mean fewer lifetime ovulatory cycles and lower exposure to estrogens, possibly related to decreased risk for reproductive cancers (Henderson et al. 1985; Eaton et al. 1994; Bernstein 2002; Yang and Wu 2005; Yang and Jacobsen 2008).The same factors, given their relationship to estrogen levels, could also relate to metabolic function in the low-estrogen, postmenopausal physiological environment (Xue and Michels 2007).