Hazards models for human population biology

An epidemiological approach to the analysis of cribra orbitalia as an indicator of health status and mortality in medieval and post-medieval London under a model of parasitic infection

OBJECTIVES

Many individuals living in medieval and post-medieval London suffered issues with sanitation, food insecurity, infectious disease, and widespread exposure to parasites from a multitude of sources, causing increased risk of death for many inhabitants. We examine this stressful environment and its relationship with various demographic and temporal dimensions, using cribra orbitalia (CO) as an indicator of stress, to model an increased risk of dying under the expectations of our proposed parasitic model of infection.

MATERIALS AND METHODS

We analyze the relationship between CO and mortality across seven medieval and post-medieval cemeteries from London by the covariates of sex, status, and age-at-death. A survival analysis (Cox regression) and a binomial logit estimated hazard and odds ratios of dying with CO across age-at-death, sex, status, and time-period within single statistical models. In addition, we provide new Bayesian age-at-death estimates for post-medieval samples.

RESULTS

The models show the rate of CO decreased over time and age-at-death, regardless of sex or status; post-medieval individuals were ~72% less likely to die with lesions than their medieval counterparts. Further, individuals with CO had ~1% decrease in risk of dying with CO per year of age.

DISCUSSION

These results suggest increased mortality risk for those with lesions indicative of anemia (CO), and selective mortality of younger individuals during the medieval period. Despite sex-specific nutritional and occupational hazards, and status-based access to resources, the prevalence of CO was similar across sex and status, which suggests living with parasitic infection that caused anemia was an everyday reality for medieval and post-medieval Londoners.

Survival analysis of the Black Death: Social inequality of women and the perils of life and death in Medieval London

OBJECTIVES

Described as an indiscriminate killer by many chroniclers, the Black Death descended on London during the 14th century. To best understand the pattern of transmission among demographic groups, models should include multiple demographic and health covariates concurrently, something rarely done when examining Black Death, but implemented in this study to identify which demographic groups had a higher susceptibility. Female predisposition to the Black Death was also explored, focusing on whether social inequality added to vulnerability.

MATERIALS AND METHODS

Three attritional cemeteries from the Wellcome Osteological Research Database were compared with the Black Death cemetery, East Smithfield. A Cox proportional hazards regression estimated hazards ratios of dying of the Black Death, using transition analysis ages-at-death as the time variable, and sex and frailty as covariates. Additionally, a binomial logistic regression generated odds ratios for age-at-death, sex, and frailty.

RESULTS

The Cox model produced a significant hazards ratio for individuals deemed frail. Similarly, the logit model calculated significantly increased odds ratios for frail individuals, and decreased odds for individuals aged 65+.

DISCUSSION

The older individuals were not undergoing growth during times of great stress in London pre-dating the Black Death epidemic, which may explain the decreased odds of contracting the Black Death. Further, although women dealt with social inequality, which partially led to the demographic puzzle of the Medieval "missing" women, women's susceptibility to infection by the Black Death was not increased. The phenomenon of the missing women may be due to a combination of factors, including infant and child mortality and preservation.

Iron deficiency anemia, population health and frailty in a modern Portuguese skeletal sample

Introduction

Portugal underwent significant political, demographic and epidemiological transitions during the 20^th century resulting in migration to urban areas with subsequent overcrowding and issues with water sanitation. This study investigates population health during these transitions and interprets results within a framework of recent history and present-day public health information. We investigate skeletal evidence for anemia (cribra orbitalia and porotic hyperostosis) as indicators of stress and frailty–i.e., whether the lesions contribute to susceptibility for disease or increased risk of death.

Methods

The presence and severity of skeletal lesions were compared against known sex and cause of death data to investigate potential heterogeneity in frailty and the relationship between lesions and risk of dying over time. Additionally, we tested for the presence of selective mortality in our data (i.e., whether or not the sample is biased for individuals with higher frailty). Our sample derives from a large, documented, modern Portuguese collection from Lisbon and is the first study of its kind using a documented collection. The collection represents primarily middle-class individuals.

Results and conclusions

Analyses indicated that porotic hyperostosis became more common and severe over time, while cribra orbitalia severity increased over time. Neither process was linked to cause of death. However, there was a significant relationship to sex; males exhibited a higher prevalence and severity of lesions and increased mortality. A Gompertz function showed decreased survivorship in early life but increased survivorship over age 60. Using comorbidities of anemia, we were unable to detect selective mortality–i.e., in our sample, lesions do not represent a sign of poor health or increased frailty and are not significantly linked with a decreased mean age-at-death. However, lesion prevalence and severity do reflect the socioeconomic processes in urban Lisbon during the 1800s and 1900s and the possibility of water-borne parasites as the contributing factor for iron deficiency anemia.

Evolutionary theory of ageing and the problem of correlated Gompertz parameters

The Gompertz mortality model is often used to evaluate evolutionary theories of ageing, such as the Medawar-Williams' hypothesis that high extrinsic mortality leads to faster ageing. However, fits of the Gompertz mortality model to data often find the opposite result that mortality is negatively correlated with the rate of ageing. This negative correlation has been independently discovered in several taxa and is known in actuarial studies of ageing as the Strehler-Mildvan correlation. We examine the role of mortality selection in determining late-life variation in susceptibility to death, which has been suggested to be the cause of this negative correlation. We demonstrate that fixed-frailty models that account for heterogeneity in frailty do not remove the correlation and that the correlation is an inherent statistical property of the Gompertz distribution. Linking actuarial and biological rates of ageing will continue to be a pressing challenge, but the Strehler-Mildvan correlation itself should not be used to diagnose any biological, physiological, or evolutionary process. These findings resolve some key tensions between theory and data that affect evolutionary and biological studies of ageing and mortality. Tests of evolutionary theories of ageing should include direct measures of physiological performance or condition.

Index of opportunity for natural selection among the Gowdas of Kodagahalli village, Karnataka, India

BACKGROUND

In order to understand how selection is operating in the Gowda population, the index of opportunity for selection was calculated and the present findings were compared with some related findings from other South Indian (SI) populations.

MATERIALS AND METHODS

Crow (1958) and the modified method by Johnston and Kensinger (1971) were used for the present purpose.

RESULTS AND DISCUSSION

The index of total selection intensity (I) was found to be moderate taking into consideration the range for many Indian populations. Considering certain differences in fertility and mortality heritable, it appears that natural selection play an important role in shaping the genetic constitution of the Gowda population. Analysis of data indicates that the index due to fertility seems to contribute more towards selection than mortality. This trend might be because of better living condition and health-care system among the Gowdas which have a positive impact on the lower contribution of mortality for the evolution mechanism of the Gowda population through natural selection.

Infant mortality and intra-household competition in the Northern Islands of Orkney, Scotland, 1855-2001

This study applies principles from the theory of household life cycles to the study of early childhood mortality in the population of the Northern Orkney Islands, Scotland. The primary hypothesis is that unfavorable household economic conditions resulting from changes in household demographic composition increase the risk of death for children under the age of 5 years because of limited resources and intra-household competition. We apply Cox proportional hazards models to nearly 5,000 linked birth and death records from the Northern Orkney Islands, Scotland, from the period 1855 to 2001. The dependent variable is the child's risk of death before age 5. Findings suggest that children in households with unfavorable age compositions face higher risk of death. This elevated risk of death continues once heterogeneity among children, islands, and households is controlled. Results also show differential risk of death for male children, children of higher birth orders, and twin births. The analyses present evidence for intra-household competition in this historic setting. The most convincing evidence of competition is found in the effects of household consumer/producer ratios and twinning on child mortality risks.

Mortality and fertility rates in humans and chimpanzees: How within-species variation complicates cross-species comparisons

A grandmother hypothesis may explain why humans evolved greater longevity while continuing to end female fertility at about the same age as do the other great apes. With that grandmother hypothesis in mind, we sought to compare age-specific mortality and fertility rates between humans and chimpanzees, our closest living relatives, and found two puzzles. First, we expected that lower adult mortality in humans would be associated with slower senescence, but the rate of chimpanzee demographic aging falls within the human range. Second, we expected declines in age-specific fertility to be similar in the two species but instead of falling in the thirties as it does in women, fertility remains high into the forties in some chimpanzee populations. We report these puzzles using data from nine human populations and both wild and captive chimpanzees, and suggest that systematic differences in the heterogeneity of surviving adults may explain them.

Paleodemography of a medieval population in Japan: analysis of human skeletal remains from the Yuigahama-minami site

The purpose of this study is to obtain demographic data regarding the medieval population buried at the Yuigahama-minami site in Kamakura, Japan, and to detect a secular trend in the life expectancy of Japanese population over the last several thousand years. The Yuigahama-minami skeletal sample consists of 260 individuals, including 98 subadults (under 20 years old) and 162 adults. A Yuigahama-minami abridged life-table analysis yielded a life expectancy at birth (e0) of 24.0 years for both sexes, a life expectancy at age 15 years (e15) of 15.8 years for males, and an e15 of 18.0 years for females. The reliability of the estimated e0 was confirmed by analysis of the juvenility index. Demographic profiles comparing the Yuigahama-minami series with other skeletal series indicated that both the survivorship curve and life expectancy of the Yuigahama-minami sample are similar to those of the Mesolithic-Neolithic Jomon population, but are far lower than those of the early modern Edo population. These comparisons strongly suggest that life expectancy changed little over the thousands of years between the Mesolithic-Neolithic Jomon and medieval periods, but then improved remarkably during the few hundred years between the medieval period and early modern Edo period. The short-lived tendency of the Yuigahama-minami sample does not contradict the archaeological hypothesis of unsanitary living conditions in medieval Kamakura. This is the first investigation to address the demographic features of a medieval population in Japan, and will help refine our understanding of long-term trends in the demographic profiles of inhabitants of Japan.

Distribution of postpartum amenorrhea in rural Bangladeshi women

Previous studies of postpartum amenorrhea (PPA) demonstrated distinct subgroups of women with short and long durations of amenorrhea. This phenomenon was attributed to cases where breastfeeding is absent because of pregnancy loss or infant death, or confusion of postpartum bleeding with resumption of menses. We explored these ideas using data from an 11-month prospective study in Bangladesh in which 858 women provided twice-weekly interviews and urine specimens for up to 9 months; 300 women were observed while experiencing PPA. The resulting exact, interval-censored, or right-censored durations were used to estimate parameters of two-component mixture models. A mixture of two Weibull distributions provided the best fit to the observations. The long-duration subgroup made up 84% (+/- 4% SE) of the population, with a mean duration of 457 (+/- 31) days. The short-duration subgroup had a mean duration of 94 (+/- 17) days. Three covariates were associated with the duration of PPA: women whose husbands had high-wage employment had a greater probability of falling in the short-duration subgroup; women in the long-duration subgroup whose husbands seasonally migrated had shorter periods of PPA within the subgroup; and mothers in the short-duration subgroup who gave birth during the monsoon season experienced a shortened duration of PPA within the subgroup. We conclude that the bimodal distribution of PPA reflects biological or behavioral heterogeneity rather than shortcomings of data collection.

Longitudinal analysis of deciduous tooth emergence: IV. Covariate effects in Japanese children

Emergence of the deciduous teeth is generally considered to be robust to moderate environmental insults, malnutrition, and disease. Consequently, deciduous tooth emergence has been used to assess growth and development and for age estimation in children. In this paper, we examine the way in which nutritional status and other covariates affect deciduous tooth emergence in a sample of 114 Japanese children born in Tokyo in 1914 and 1924. Parametric survival analysis was used to quantify the effects of nutritional status, breastfeeding behavior, and sex on the hazard of deciduous tooth emergence. Children of poor nutritional status exhibited significantly delayed emergence of all deciduous teeth, with effects that ranged from 14-29% increases in mean emergence times. Children of medium nutritional status exhibited increases in mean emergence times of 5-9% for the canines and lower molars, and 13-17% for the incisors. Partial breastfeeding had no effect on tooth emergence, but children who were not breastfed at all showed delayed emergence of the upper incisors. No significant sex differences in emergence were found. The findings contradict the idea that moderate malnutrition has little effect on deciduous tooth emergence. Furthermore, nutritional differences may account for some of the observed differences among populations in the timing of tooth emergence.

Demographic trends and biological status of historic populations from Central Poland: the Ostrów Lednicki microregion

The microregion of Ostrów Lednicki in the province of Wielkopolska was the center of the formation of the Polish State in Early Middle Ages. The analysis of skeletal remains and parish records from the region documented the biological status of inhabitants and its changes over a millennium. The study material comprised 424 human skeletons from an Early Medieval burial ground on Lake Lednica, records of 2,704 deaths from the registers of the Roman Catholic parish of Dziekanowice, made between 1818-1903, data on the deaths of 929,192 inhabitants of rural areas all over the province Wielkopolska obtained from Prussian statistical materials for the years 1865-1900, and comparative data from the literature. Assuming both a stationary population model and a stable population model with nonzero natural increase, parameters of life tables and measures of opportunity for natural selection (Crow's index I(m), potential gross reproductive rate R(pot), and the biological state index I(bs)) were calculated for the Early Middle Ages and for the two periods of the 19th century which were characterized by different laws of land ownership and thus different rural economies. In the first period, peasants were tenants, whereas in the second, they were given freehold of the land they cultivated. Causes of death were also analyzed. A distinct increase in longevity from the early Middle Ages to the end of the 19th and the beginning of the 20th century was found. This was related to a higher level of demographic development in the parish of Dziekanowice during the 19th century, which was achieved earlier than in other areas of Poland. This was confirmed by genetic measures: coefficients of exogamy and coefficients of kinship. The reasons were related to the historical prominence of this region and to its proximity to the first two capitals of the Polish state, Gniezno and Poznań.

Demography, life history, and social structure in Propithecus diadema edwardsi from 1986-2000 in Ranomafana National Park, Madagascar

Prosimian lemurs differ fundamentally from anthropoid primates in many traits related to social structure. By exploring the demography of Milne-Edwards' sifakas (Propithecus diadema edwardsi), and comparing it to other well-studied primates, we explore the effect of demographic and life-history factors on social structure. Specifically, we compare lemur survivorship and fertility patterns to two published composite models: one created for New World and another created for Old World monkeys. Using longitudinal data collected on individual Propithecus diadema edwardsi from four study groups from 1986-2000 in Ranomafana National Park, Madagascar, we quantify 1) group composition, 2) birth seasonality, 3) interbirth interval, 4) life-table values, and 5) population growth estimates. The mortality, survivorship, and life-expectancy schedules indicate high infant and juvenile mortality. Fertility remains high until death. The intrinsic rate of increase and net reproductive rate indicate a shrinking population. We suggest that high mortality rather than low fertility causes the observed population decline. While sifaka survivorship closely resembles New World patterns, fertility resembles Old World patterns, i.e., like New World monkeys, few sifakas survive to reproductive age, and those that do, reproduce at a slow rate resembling the Old World pattern. This necessarily impacts social structure. An adult sifaka at the end of her lifespan will have one only daughter who survives to reproductive age, compared to 3.4 for New World or 2.7 for Old World monkeys. Demography limits the formation of large kin-based groups for sifakas, and survivorship and fertility patterns do not easily permit sifakas to form large same-sex family groups.

Patterns for the initiation of breastfeeding in humans

Establishment of lactation has important biological and emotional health consequences for the newborn. Even so, substantial variation within a culture and among different cultures is seen in the onset of breastfeeding. Parametric mixture models are used to explore this variation and to uncover general human patterns for the initiation of breastfeeding. The model components reflect two hypothesized patterns of behavior. The first component is a "natural" pattern of breastfeeding that reflects, to some extent, a general mammalian behavior. The second component arises through culturally mediated behaviors that affect the initiation of breastfeeding. The model was fit by maximum likelihood to interval- and right-censored observations on 26220 mother-infant pairs collected from 25 previously published studies of breastfeeding behavior. Both model components were clearly statistically identified. Effects of cultural and geographic covariates were found to have significant effects on all components of the model. Although there is clear evidence for two distinct patterns of behavior in the initiation of breastfeeding, the results suggest that learned behaviors play an important role in mediating even "natural" breastfeeding behavior.

Predictors of fecundability and conception waits among the Dogon of Mali

Surprisingly little is known about the mechanisms that underlie variation in female fertility in humans. Data on this topic are nonetheless vital to a number of pragmatic and theoretical enterprises, including population planning, infertility treatment and prevention, and evolutionary ecology. Here we study female fertility by focusing on one component of the interbirth interval: the waiting time to conception during menstrual cycling. Our study population is a Dogon village of 460 people in Mali, West Africa. This population is pronatalist and noncontracepting. In accordance with animist beliefs, the women spend five nights sleeping at a menstrual hut during menses. By censusing the women present at the menstrual huts in the study village on each of 736 consecutive nights, we were able to monitor women's conception waits prospectively. Hormonal profiles confirm the accuracy of the data on conception waits obtained from the menstrual hut census (Strassmann [1996], Behavioral Ecology 7:304-315). Using survival analysis, we identified significant predictors of the waiting time to conception: wife's age (years), husband's age (< 35, 35-49, > 49 years), marital duration (years), gravidity (number of prior pregnancies), and breast-feeding status. Additional variables were not significant, including duration of postpartum amenorrhea, sex of the last child, nutritional status, economic status, polygyny, and marital status (fiancée vs. married). We fit both continuous and discrete time survival models, but the former appeared to be a better choice for these data.

Longitudinal analysis of deciduous tooth emergence: II. Parametric survival analysis in Bangladeshi, Guatemalan, Japanese, and Javanese children

We present a form of parametric survival analysis that incorporates exact, interval-censored, and right-censored times to deciduous tooth emergence. The method is an extension of common cross-sectional procedures such as logit and probit analysis, so that data arising from mixed longitudinal and cross-sectional studies can be properly combined. We extended the method to incorporate and estimate a proportion of agenic teeth. While we concentrate on deciduous tooth emergence, the method is relevant to studies of permanent tooth emergence and other developmental events. Deciduous tooth emergence data were analyzed from four longitudinal studies. The samples are 1,271 rural Guatemalan children examined every three months up to age two and every six months thereafter as part of the INCAP study; 397 rural Bangladeshi children examined monthly to age one and quarterly thereafter as part of the Meheran Growth and Development Study; 468 rural Indonesian children examined monthly as part of the Ngaglik study; and 114 urban Japanese children examined monthly in studies from 1910 and 1920. Although all four studies were longitudinal, many observations from the Guatemala and Bangladesh studies were effectively cross-sectionally observed. Three different parametric forms were used to model the eruption process: a normal distribution, a lognormal distribution, and a lognormal distribution with age shifted to shortly after conception. All three distributions produced reliable estimates of central tendencies, but the shifted lognormal distribution produced the best overall estimates of shape (variance) parameters. Estimates of emergence were compared to other studies that used similar methods. Japanese children showed relatively fast emergence times for all teeth. Bangladeshi and Javanese children showed emergence times that were slower than are found in most previous studies. Estimates of agenesis were not significantly different from zero for most teeth. One or two central incisors showed significant agenesis that ranged from 0.1 to 0.8% in three of the samples; even so, failure to model the agenic proportion did not seriously bias the estimates.

Analysis of completed parity using microsimulation modeling

"This paper aims to construct a demographic microsimulation model, which is easily applicable to gene transmission..., and to examine the distribution of completed parity as a fertility measure at a population level according to the schedule of individual life course....The simulation model was applied to the data of the Gidra-speaking people living in the lowlands of Papua New Guinea."