This preview shows page 1. Sign up to view the full content.
Unformatted text preview: CRAIG R. JANES
Departments of Anthropology and Health and Behavioral Sciences
University of Colorado at Denver
Department of Health and Behavioral Sciences
University of Colorado at Denver Free Markets and Dead Mothers: The Social
Ecology of Maternal Mortality in Post-Socialist
Beginning in 1990, Mongolia, a former client state of what was then
the Soviet Union, undertook liberal economic reforms. These came as a
great shock to Mongolia and Mongolians, and resulted in food shortages,
reports of famine, widespread unemployment, and a collapse of public
health and health care. Although economic conditions have stabilized
in recent years, unemployment and poverty are still at disturbingly high
levels. One important consequence of the transition has been the transformation of the rural, primarily pastoral, economy. With de-collectivization,
herding households have been thrown into a highly insecure subsistence
mode of production, and, as a consequence, have become vulnerable
to local ﬂuctuations in rainfall and availability and quality of forage,
and many now lack access to traded staples and essential commodities.
Household food insecurity, malnutrition, and migration of impoverished
households to provincial centers and the capital of Ulaanbaatar are one
result. Reductions to investments in the health sector have also eroded the
quality of services in rural areas, and restricted access to those services
still functioning. Evidence suggests that women are particularly vulnerable to these political-ecological changes, and that this vulnerability is
manifested in increasing rates of poor reproductive health and maternal mortality. Drawing on case-study ethnographic and epidemiological
data, this article explores the links between neoliberal economic reform
and maternal mortality in Mongolia. [Mongolia, maternal mortality, neoliberal reform, post-socialist societies, political ecology of pastoralism] Medical Anthropology Quarterly, Vol. 18, Issue 2, pp. 230–257, ISSN 0745-5194, online ISSN 15481387. C 2004 by the American Anthropological Association. All rights reserved. Send requests for
permission to reprint to: Rights and Permissions, University of California Press, Journals Division,
2000 Center Street, Suite 303, Berkeley, CA 94704-1223. 230 MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 231 Introduction M aternal mortality, morbidity, and, by extension, perinatal mortality and
the health experience of surviving children, are the results of a complex interaction of biological, medical, and social factors (AbouZahr
et al. 1996; Figa-Talamanca 1996; Stokoe 1991). Although the proximal causes
of maternal mortality, pregnancy complications, and the impact on the fetus and
neonate have been the subject of substantial biomedical research, the association
between individual-level factors and those operating at the level of society and
community—political-economic factors, social behaviors, cultural beliefs, and accessibility (perceived and actual) of appropriate health care services—are as yet
poorly understood (Miller et al. 2003; Okonofua et al. 1992; Okolocha et al. 1998;
The purpose of this article is to assess the impact of Mongolia’s transition from
a socialist to a capitalist market economy on the multi-level “causal assemblages”
that affect women’s health (Dunn and Janes 1986). In this analysis, we use an
analytic strategy that draws on the social ecological models employed increasingly
in public health and biocultural anthropological research (e.g., Krieger 1994, 2000;
Stokols 1996; Thomas 1998). These models assess how events, processes, and
factors occurring across the social scale “get into” the body (Krieger 1994). Like
Krieger and others (e.g., Farmer 1997, 2003; Thomas 1998), we attempt to identify
the “spider” in the web of causation, focusing here primarily on the links between
neoliberal economic reform, consequent transformations to the rural economic
system, and the differential impact of this transformation on certain groups of
poor, rural women.
One of the contributions that medical anthropology can make to research on
critical global health problems such as this is our ability to work ethnographically
across multiple levels of analysis, moving from the tragic experiences of a single
family to the perspectives of health ofﬁcials and development experts. Like other
researchers who work ethnographically on the problem of maternal mortality (e.g.,
Barnes-Josiah et al. 1998; Hay 1999; Makhlouf-Obermayer 1999; Wall 1998),
we have found that each case is a unique tragedy, resulting from a particular
constellation or “cascade” of factors and events. Assembled, these cases suggest
a pattern, and this pattern points to the causal salience of social, political, and
economic forces not in the control of victims, their families, or their children. Yet
it is important to our understanding of the problem, particularly to understanding
how death and disability ramify well beyond the individual victims to affect the
lives and life chances of surviving family and children, that the individual stories,
and the suffering they articulate, be told (Kleinman and Kleinman 1997). We
draw on the model provided to us in the work of Joan Ablon (e.g., 1971, 1999,
2002; Shuttleworth and Kasnitz this issue). A consummate ﬁeldworker, Ablon
contributed greatly to our understanding of the consequences of stigma, disability,
and economic distress for individuals, their families, and their community. Inspired
by Ablon’s exemplary ethnographic skills, we endeavor here to illuminate the social
suffering of Mongolian women.
Determinants of Maternal Mortality
Maternal deaths1 comprise about 25 percent of the deaths of women between
the ages of 20 and 30 years worldwide (Walsh et al. 1994). In 1990, the World 232 MEDICAL ANTHROPOLOGY QUARTERLY Health Organization estimated that 585,000 women died of complications related
to pregnancy and delivery (WHO/UNICEF 1996),2 with 99 percent of these deaths
occurring in developing countries (Walraven et al. 2000). Furthermore, death and
disability from obstetric causes comprise 18 percent of the total disease burden
(in terms of “disability-adjusted life years” or DALYs) among women aged 15–
44 in poor economies (World Bank 1993), although these ﬁgures undoubtedly
underestimate the depth and extent of suffering among affected families and kin
groups. Maternal mortality ratios (MMRs) thus vary widely between poor and
middle-/high-income countries. The MMRs for many countries in Africa and South
Asia are as much as 200 times (1,000–1,500/100,000) those of the industrialized
countries of North America, Europe, and East Asia (4–10/100,000) (Koblinsky
1995; WHO/UNICEF 1996). The lifetime risk of maternal mortality is many times
greater insofar as each pregnancy adds to the total lifetime risk. In poor countries
where fertility is high, the average woman faces a lifetime risk of one in 33 that a
pregnancy will result in her death (Walsh et al. 1994).
The number of obstetric deaths represents only a small fraction of the total
burden of disease associated with pregnancy and childbirth. WHO estimates that
there are as many as 20 million pregnancy- and childbirth-related morbidities and
disabilities each year; some one-third of all women in the developing world have
suffered from injuries such as obstetric ﬁstula to chronic conditions such as severe
anemia, reproductive tract infections, and uterine prolapse (WHO/UNICEF 1996).
Some researchers have indicated that this number may be a vast undercount of the
true burden of disease (Koblinsky 1995; Kwast 1998). Studies suggest that between
3 and 15 percent of pregnant women will experience an obstetric complication
resulting from pregnancy (e.g., Prual et al 2000; Stewart et al. 1996).
Although there is a documented relationship between maternal health and
underlying risk factors such as maternal age, parity, socioeconomic status, education, women’s social status, and general level of health (Maine and Rosenﬁeld
1999), maternal mortality and morbidity have not been subject to careful socialepidemiological or contextual analyses. As a result, the global public health response to maternal mortality and morbidity has been seriously impaired. Write
Miller and colleagues (2003:10):
the most common approach to reduce maternal mortality has been the identiﬁcation of hypothetically useful “good ideas” program interventions, with advocacy
for their large scale implementation. . . . Since 1987, international agencies and
government working together have dedicated time and funds [to interventions]
without producing documented reductions in maternal mortality. Three interventions have been promoted: the training of traditional birth attendants; the implementation of antenatal “risk screening”; and, more recently,
promotion of skilled emergency obstetric care at delivery (Liljestrand 1996; Maine
and Rosenﬁeld 1999; Miller et al. 2003). Although large amounts of money and
effort have been put into these initiatives, none has yet been demonstrated to affect
maternal mortality. It is likely that this failure is due to the fact that these “good
ideas” approaches do not consider the fact that maternal mortality, “occurs in a
speciﬁc social, cultural and political context . . . [that] greatly affects maternal
individual and community health, even within developed countries” (Miller et al.
2003:12). MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 233 Women who are poorer, less educated, who have lower social status, and are
of poorer overall health are more likely to die than those who are more fortunate.
It is also the case that such women are less likely to have access to appropriate
and effective health care (Abou-Zahr and Royston 1991; AbouZahr et al. 1996;
Barnes-Josiah et al. 1998; Israr 2000; Miller et al. 2003; Rizvi et al. 1999; Shen
and Williamson 1999; Stokoe 1991; Sundari 1992; Thaddeus and Maine 1994;
Turshen 1999; Wall 1998).
Indeed, maternal mortality is considered by many global health scholars to be
an extremely sensitive indicator of social inequality insofar as it offers a “litmus test
of the status of women, their access to health care and the adequacy of the health
care system in responding to their needs” (AbouZahr et al. 1996:77). As such, the
economic conditions of a country and its commitment to health and social welfare,
particularly of women, are important structural determinants of maternal mortality
(e.g., Okonofua et al. 1992; Sundari 1992; Turshen 1999; Wall 1998). Attention
to the causes and effective prevention of maternal mortality requires a multiplelevel focus on the social, economic, and political determinants of women’s social,
economic, and health status, and access to effective health services. In the pages
that follow, we examine the multiple relationships between the collapse of the
Soviet-style, socialist “command” economy, the introduction of liberal economic
reforms, and patterns of poor pregnancy outcome in rural Mongolia.
Mongolia is a particularly useful context for conducting such an analysis.
Prior to the dismantling of the socialist system, which began in 1990, Mongolia
had built a rationally distributed and accessible health care system and focused
considerable attention on women’s health and education—particularly given the
demands of delivering services to a largely rural and sparsely distributed population. Mongolia had established maternity waiting homes in nearly all of its counties
by the 1980s,3 girls were provided free education, and, as a result, rates of female
enrollment in secondary and post-secondary education equaled or exceeded those
of men (UNDP 2000). The collective nature of the socialist economy, particularly
in the rural countryside, ensured that women’s labor in the household could be
replaced so that they could seek health care, including lengthy residence in a maternity waiting home; thus nearly all births were attended by skilled professionals.
Essential commodities were generally available through the local collectives, and
rural families were protected from the vagaries of weather and markets that plague
pastoral production elsewhere in central Asia (Humphrey and Sneath 1999). As
we discuss in the following pages, much of this changed after 1990, providing,
in effect, a natural experiment of the impact of capitalist reform on women’s
We do not wish to overly romanticize the heavy-handed Soviet-style socialism
that characterized the Mongolian political economy prior to 1990. In the 1930s
and 1940s, the processes of collectivization resulted initially in the widespread
disruption of the traditional pastoral economy, leading to social instability and
economic hardship. Suppression of Mongolian religious and cultural traditions was
extreme (Baabar 1999). Finally, and most importantly, the development of social
and health institutions in Mongolia depended heavily on considerable ﬁnancial
inputs from the Soviet Union. One of the ﬁrst challenges faced by the political elite
in post-transition Mongolia was how to ensure adequate levels of investment in
an expensive and unsustainable health care system. However, as most Mongolians 234 MEDICAL ANTHROPOLOGY QUARTERLY now recognize, it is clear that the social protections afforded by the old socialist
regime provided households and individuals far more health, economic, and social
security than are now afforded under the new capitalist system.
Our approach joins ethnographic and epidemiologic methods in placing maternal health in a broader social, political, ecological, and economic context. We
organize our discussion around three topics: (1) a modern social and economic
history of pastoralism in Mongolian Inner Asia; (2) economic reform policy and
its impact on Mongolian pastoralism at the level of community and household;
and (3) a discussion of the potential causal relationships between economic and
health reforms and maternal health as mediated by household and communitylevel institutions. We focus on the impact of economic reform on rural women
for three reasons. First, for reasons we discuss below, rates of maternal mortality
are far higher in the countryside than they are in the towns and cities. Second, the
political-economic transition has had a particularly profound impact on the organization of pastoralism, which remains the largest sector of the Mongolia economy
and occupies a substantial number of its people. Third, the impact of neoliberal
macroeconomic reform—particularly through privatization and government disinvestment in public goods—on the traditional cooperative structures essential to
effective pastoral production in the Mongolian environment is particularly evident. As we argue, these affect women by raising the level of economic risk borne
by individual households, increasing rural poverty, escalating the labor demands
placed on women, and impairing the quality and accessibility of health care.
Our discussion is based on analysis of the following kinds and quantities of
data: surveys of household economics and health care utilization in rural northern
Mongolia (Janes 2003); structured, qualitative interviews with 15 high-level ofﬁcials in the Mongolia Ministry of Health; collection of records on all recorded
(n = 217) maternal deaths occurring in rural Mongolia between 1996 and 1998;
interviews of doctors and midwives in seven counties and three townships in the
western provinces (aimag) of Arhangai and Huvsgol (see Figure 1); and case studies of maternal mortality based on records and interviews with physicians and
Economic Reform and Poverty in Post-Socialist Mongolia
After 70 years of Soviet-style socialism, in 1990 Mongolia began sweeping
political and economic reforms. In response to pressure from global economic
reformers, Mongolia chose a transition strategy that is popularly and infamously
known as “shock therapy” (UNDP 2000). In Mongolia, shock therapy consisted of
the following elements: price liberalization; removal of restrictions on international
trade and foreign investment; privatization of state-owned enterprises, initially by
a free distribution of vouchers to the entire population and later through auction
to domestic and foreign buyers; and a marked reduction in the size of government
(Grifﬁn et al. 2001).4 As several documents and reports now make clear, Mongolia
undertook reforms championed by the most extreme of liberal economic reformers (e.g., UNDP 2000:28–31). In an insightful evaluation of neoliberal reform in
Mongolia, Grifﬁn and colleagues observed that: MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 235 Figure 1
Map of Mongolia showing research sites.
At the beginning of the transition Mongolia’s economic reforms were widely
applauded: they were expected to lead to a swift transformation of the economy,
an acceleration in the rate of growth of output and a substantial improvement
in the standard of living of the population. Even as late as 1999, Mongolia was
described as “the darling of ultra-liberals in the West,” and “the star pupil of liberal
development economics.” Yet an analysis of the results of a decade of economic
reform shows that the transition has been highly disappointing. . . . The transition
to a market oriented economy has been accompanied by a decline in the average
standard of living, a dramatic increase in poverty, greater economic insecurity and
a rise in inequality in the distribution of income and productive assets. [Grifﬁn
et al. 2001:1] Put simply, economic reform resulted in widespread social chaos and economic collapse (Grifﬁn et al. 2001; UNDP 2000). Aid to Mongolia from the former
Soviet Union and other socialist countries was immediately eliminated, contributing to a sudden 30 percent drop in GDP. Industrial production, which up to 1989
had been a growing sector of the economy, employing large numbers of urban and
town dwellers, declined sharply. Price liberalization, overdependence on imported
energy resources, shortage of skilled technicians, and the collapse of the Soviet
trade block (the Council for Mutual Economic Assistance, or COMECON) contributed to the downfall of the industrial sector. As a result, the value of industrial
production as a percentage of GDP declined from 35 percent to 20 percent between
1990 and 1996 (Grifﬁn et al. 2001; UNDP 2000). There was a rapid increase in unemployment in urban areas, leading to widespread poverty, homelessness, crime,
and alcohol abuse.
The shock of poverty in urban areas led to widespread emigration from the
cities to the countryside to take up herding, resulting in a transfer of population
from urban to rural areas that is perhaps unprecedented (Humphrey and Sneath 236 MEDICAL ANTHROPOLOGY QUARTERLY
Economic output, inﬂation, and personal incomes, Mongolia, 1989–2000. Year GDP
(Index: 1989 = 100) Per capita income
(US$ at 1993 prices) Rate of inﬂation (% change in
the consumer price index) 1989
8.1 Source: Grifﬁn et al. (2001:2). 1999). Between 1989 and 1998, the proportion of the population living in urban
areas declined by 13 percent, and the rural population grew by an astonishing 17
percent (Grifﬁn et al. 2001). Since 1998, due to a string of severe winters, which
may have led to the failure of many inexperienced herders, and by a stabilizing of
the urban economy, population has begun to return to the towns and cities.
As shown in Table 1, per capita income declined from US$1,643 in 1989
to US$374 in 1999 (in constant 1993 dollars; Government of Mongolia 2000).
The government now estimates that more than a third of Mongolians are living
below the poverty line of approximately US$16 per person, per month (Asian
Development Bank [ADB] 2002; Government of Mongolia 1998). The depth and
severity of poverty has increased, indicating a rise in social inequality and in the
numbers of the very poor.5 The GINI index, a measure of income inequality, grew
from .31 to .44 between 1995 and 2002 (UNDP 2003).
Advocates of neoliberal economic reform advocate a small and relatively
weak state (Turshen 1999). Unfortunately for Mongolia, as Grifﬁn et al. (2001:11)
note wryly, this is exactly what they got. Between 1989 and 1999 government
expenditures declined from 50.2 percent of GDP to 26.9 percent, substantially
outpacing the decline in GDP. This sharp decrease reﬂects a widespread disinvestment in public goods—social services, health care, and education. Health care
expenditures fell from 6 percent of GDP in 1989 to 3.3 percent of GDP in 1999.
Investment in education declined from 11.5 percent of GDP to 5.5 percent of GDP
over the same period. School enrollment of children aged 8–15 years fell from
nearly 100 percent in 1990 to 87 percent by 1998 (UNDP 2000). Today, the government’s ability to provide social services such as education and health care and
to combat poverty has been seriously weakened.
Because of the loss of Soviet subsidies and declining public investment, the
health care system disintegrated rapidly after transition. Stocks of essential drugs
and medical supplies were exhausted. Without heat and electricity, many hospitals MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 237 and clinics were forced to close or curtail operations (WHO 1999). Most of the
maternity waiting homes were closed after 1992, many physicians left the health
service, particularly in rural areas, and shortages of fuel in rural areas impaired
the ability of county and township medical facilities to respond to emergencies
(WHO 1999). In response to this crisis in the health sector, and following the World
Bank’s prescription for health development (World Bank 1993), the Government of
Mongolia endeavored to shift from a focus on hospital-based secondary and tertiary
care to essential primary care (ADB 2002; Janes 2003). A compulsory national
health insurance system was initiated in 1994, and the ADB has provided sizable
development loans in support of a series of health sector reforms (ADB 2002).
By 2000, many of the maternity waiting homes had reopened; however, changing
economic conditions in the countryside, as we will discuss below, impede universal
access to these facilities. Rural health care continues to suffer from shortages of
staff, equipment, drugs, and funds for capital investments in infrastructure (Janes
2003; UNDP 2000).
The Political Ecology of Mongolian Pastoralism
Until the Soviet-supported Mongolian revolution of 1921, cooperating groups
of herding households (hot ail, or ail) were invariably part of, and supported by,
large institutional forms that controlled access to pasture and animals. These included Buddhist monasteries, princely “banners” (territories), and smaller units
controlled by landed nobility. The larger institutions functioned economically as
organizers of redistribution, instruments of specialization, and managers of collective labor and produce. These institutions also controlled and managed the
seasonal migration of herders and animals to maximize availability of desirable grasses and minimize the effects of a harsh climate (Humphrey and Sneath
After 1921, the Mongolian government broke up these “feudal” institutions
and redistributed animals to household groups. The old estate boundaries were
redrawn and organized into districts and subdistricts. Within these units, state ofﬁcials organized production and essential social services, and managed the exchange
of animal products for consumer goods and nonanimal food staples. Radical collectivization began in earnest after World War II and was largely completed by the
mid-1950s. Herding households were stripped of most of their privately owned
animals. Families were settled in semi-sedentary towns and district centers and
organized into state-controlled cooperative enterprises (negdel). Negdel members
provided labor for the collective in return for necessities as well as social and
health services. Education and health care were well developed in most regions;
high literacy levels and favorable health indicators suggest that such services were
of reasonably high quality and accessible to the rural population (Grifﬁn et al.
2001; UNDP 2000).
The negdel functioned to regulate access to the critical resources of water, fodder and shelters for winter use, and pasturage (Grifﬁn et al. 2001). The
negdel also provided veterinary care for animals and ensured that there were adequate supplies of food, shelters, and transport to protect herders and their animals
from the severe winter disasters (dzud) that periodically affect the central Asian
steppe.6 In short, and perhaps ironically, the negdel operated in a fashion somewhat 238 MEDICAL ANTHROPOLOGY QUARTERLY similar to the old feudal institutions, functioning to regulate access to and distribute essential resources and to spread risks among large numbers of households
(Grifﬁn et al. 2001; Humphrey and Sneath 1999). The negdel differed, of course,
in its organization of production and in its emphasis on the provision of health and
social services. Both feudal and socialist institutions, however, operated to provide a measure of security to individual households in this harsh and unpredictable
The Impact of the Political-Economic Transition on Rural Communities
Subsistence security was the ﬁrst casualty of economic reform. As a result of
privatization, all collectives were completely dissolved, member households divided up the moveable assets (principally livestock), and, sometimes with related
households, began to herd on an independent basis. Perhaps for the ﬁrst time since
the origins of the Mongolian state in the 13th century, Mongolian herders became
domestic-level subsistence producers, lacking access to cooperative institutions
that served to spread economic risks among numbers of households. Access to
land and water resources, which had been formerly under the control of feudal,
and then socialist institutions, were suddenly available to any who would use them.
Rural land in Mongolia is considered public land and is open to use by any citizen.
The new Mongolian herder thus has access to what is a large, open commons of
productive resources. Unfortunately, this new independence has come at a very
high price. Declining public investments in transportation, water resource development, veterinary services, and emergency land and shelter have made herders
highly vulnerable to unpredictable weather, disease, and market conditions. In the
absence of institutions that would regulate access to essential resources, Mongolian
herders may experience the “tragedy of the commons”; overgrazing has already
been noted in areas closest to markets and veterinary care (cities) and where water sources are reliable (Grifﬁn et al. 2001).7 Access by households to markets
has become a critical determinant of economic stability. Under the socialist system, the negdel served as a marketing cooperative and bore all responsibility for
moving animal products to urban and foreign markets and bringing in essential
commodities for sale to negdel households. With the elimination of the collective,
individual households now have to devise means to transport their products to
market in order to generate the cash now needed not only for basic commodities,
clothing, and tools, but for essential services as well: room and board fees for
children attending a county or provincial school, health and veterinary care, and
Herders with the easiest access to existing markets—either those living close
to urban centers or to Chinese or Russian towns—have an advantage over those
in the remote hinterland where market institutions have yet to develop. In their
research on the economics of pastoralism in post-socialist Central Asia, Humphrey
and Sneath (1999) found in the remote regions of Mongolia that:
general economic conditions render making an adequate living by herding extremely difﬁcult. Herders obtain low prices for their livestock products and are
confronted with abruptly rising prices for food, fuel, transport, and clothing.
There is a lack of market institutions (banks, shops, trading companies, marketplaces), and those that exist often do not reach to the herders located on distant MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 239 pastures. The return to an economy of subsistence alone is even more marked in
Mongolia than it is in Russia, and one consequence of this may be an increasing number of livestock kept by pastoralists, who do not or cannot sell them.
[p. 110] In this new, higher risk economic context, the social and health impacts
of decisions made by individual households loom large. Some households have
been relatively successful, and have managed to acquire large herds that enhance their long-term economic security and also permit them to invest greater
resources in the education and health of individual members. But the numbers
of these self-sufﬁcient households is small. In one study of household economics
among pastoralists in Mongolia (Shombodon n.d., quoted in Humphrey and Sneath
1999:275), it was found that about 45 percent of households were able to be selfsufﬁcient, 40 percent required cash inputs from sources other than herding (e.g.,
wage labor, trading activities), and only 15 percent were able to produce consumer
goods for the market.
Variable environmental conditions, access to markets, and success of household strategies have thus produced a high degree of inequality in the countryside.
This inequality compounds what happened at the very beginning of privatization.
The initial distribution of animals was not equal among individuals or households,
nor was the distribution of animals limited to herders (Grifﬁn et al. 2001). As
one would expect, experienced herders and members of prominent households did
well, while female-headed households and young herders did less well. “As early
as 1992, at the top end of the distribution, roughly 5 percent of households had
herds with more than 200 animals whereas at the bottom end of the distribution,
42 percent of households had herds with fewer than 31 animals” (Grifﬁn et al.
2001:21).8 The situation has worsened signiﬁcantly since then. Many younger and
inexperienced herders were unable to maintain their herds, particularly through
the severe winter disasters of the past decade. Increasingly, non-self-sufﬁcient or
asset-less rural households are forced to rely on wage labor opportunities provided
by other, wealthier herders, or they are forced to migrate into towns and cities
where they join a growing underclass of dependent poor (Janes 2003).9 Probably
as a result of this increasingly marked inequality, animal theft has emerged as a
serious problem in the countryside.
This emergent social inequality has likely impaired the reconstitution of traditional cooperative institutions. As described above, in Mongolia, the primary
cooperative institution is the hot ail. The core of a hot ail is commonly a patrilineal
extended family, although it is a highly ﬂexible institution (Humphrey and Sneath
1999). Hot ail members pool resources and labor, make joint decisions regarding
the movement of animals and people, and thus help buffer the economic risks to
which individual households might otherwise be exposed. The hot ail has a long
history, although its cooperative functions were superseded by those of the negdel
during the collective period.
Although the hot ail has survived, it is not yet clear whether it remains strong
enough or stable enough to take over some of the functions it undertook in the past,
or were more recently provided by the negdel. In our interviews of rural herders,
we found that even within the hot ail, individual households were very careful to
differentiate the ownership of livestock, and the social ethic of cooperation was 240 MEDICAL ANTHROPOLOGY QUARTERLY clearly secondary to a new principle of individual and household-based private
property. The size and existence of hot ail were also highly variable. We found that
economic uncertainty and failure have led to the disintegration or weakening of
these groups; in several interviews, households engaged independently in herding
activities with no help from neighboring households. Few other cooperative institutions have yet emerged that would help organize production so as to maximize
pasturage and protect individual households from disasters and disease (Humphrey
and Sneath 1999).
In the absence of strong cooperative institutions, and with the disinvestment
by government in rural development, the social and health services that herders
were provided during the socialist period, and before, have become less accessible. It is clear that the socialist ethic of social solidarity and justice, where all
individuals were provided access to essential services and buffered from the inherent risks of pastoralism in a harsh environment, has been replaced by an ethic
of market justice, where services and beneﬁts accrue only to those judged deserving by virtue of economic success. Educational opportunities for rural youth
have declined. The quality of health care has been affected by declining public
investment, although external inputs from development grants and activities of
organizations such as UNICEF, UNFPA, and several bilateral development agencies have protected the rural health sector from complete collapse. However, rural
clinics have been clearly affected by lack of funding. Many rural doctors left
government service during the transition period when the incentives designed to
attract physicians to practice in rural areas were eliminated and clinic funding dried
up. Today, rural clinics still suffer from lack of essential equipment and medications and must engage in cash-producing activities to fund salaries and medical
In summary, liberal economic reform has radically transformed the structure
and organization of the rural economy, exposing individual households to a high
level of risk. The collapse of the negdels and the privatization of herds have eliminated the formal institutions that buffered this risk and protected the environment
by regulating access to and maintaining land, pasturage, and water resources; providing veterinary services; and establishing reserves for emergencies. The shift
in ethics from social justice to market justice has resulted in a substantial disinvestment in human capital. What rural Mongolians have experienced over the
past decade is a process of underdevelopment whereby they ﬁnd themselves—
for the ﬁrst time in hundreds of years—practicing a household-based system of
subsistence production in the absence of the formal state institutions (feudal and
socialist) that spread the risk of that system across social collectives. In the absence
of state support, households attempt to manage risk in a number of ways. They
diversify sources of income within the household (e.g., combining wage labor with
herding activities). Some family members may move away for entire seasons to
work in towns or cities. In some households, the elderly are sent to live in county
and provincial centers so that they can support grandchildren sent to attend school.
Write Grifﬁn et al. (2001:17–18):
Like survival strategies, these [risk management strategies] are largely responses
to a reduction in the number of risk reducing institutions. Almost always these
informal arrangements are inferior to the formal institutions they attempt to MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 241 replace. Privatization of risk management seldom works well, particularly for
low income groups in low income countries. Vulnerable Women in the New Pastoral Economy
Gender roles are central elements of the scheme by which tasks are allocated
in pastoral families. The female sphere of capability and responsibility is centered
on the household. Women prepare food products (cheese, butter, yogurt), cook,
make tea, make and repair clothes, care for children, wash and clean, undertake all
milking activities, care for animals close to the encampment (particularly young
animals in springtime), collect dried dung for fuel, and fetch water. Men are responsible for all herding activities, especially those at a distance from the encampment.
From a very young age, girls begin to help their mothers in the household. Lack
of economic stability and a general and pervasive sense of subsistence insecurity
have led many Mongolian families to increase the amount of self-provisioning production (Grifﬁn et al. 2001). Much of this work is production directed by women
(Humphrey and Sneath 1999).
In a harsh climate with long and cold winters and frequent summer drought,
substantial labor is required for raising animals. Except in winter, women, men,
and their older children engage in physically demanding work 15–18 hours per
day. Women’s work is particularly critical during spring and summer, when they
are responsible for all milking and milk processing chores along with the care of
the young animals.
In a subsistence-oriented mode of production, such chores may affect a
woman’s health, both directly and indirectly. She may reduce food consumption in
favor of other family members. If a problem seems minor, she may delay visiting a
hospital or clinic so as not to waste valuable time or leave her children unattended.
The call for a midwife may be delayed if she is in labor but the contractions are
weak. If she has few family or kin nearby, there may be no one to ride a horse
40–50 km to the district center to fetch help. And these days, even if her husband
makes it to the district center on his horse, he may ﬁnd that the ambulance is out
of fuel, and the driver reluctant to venture out.
In the past, the organization of the negdel would have ensured the receipt
of timely health care. Until 1992, all districts had maternity waiting homes that
provided accommodation close to medical care. Pregnant women were urged to
move to the homes well in advance of expected delivery and provided transportation. The cooperative nature of the collective ensured that a woman’s labor in the
home was replaced. She had little concern that her children might be unattended
or unfed, that the animals might go without milking, or that the young animals
might be neglected. In interviews with county-level medical staffs in Arhangai
and Huvsgol provinces, we found that maternity waiting homes are not being as
heavily or consistently used as they were in the past. We believe this more variable
use is one result of the weakening of cooperative institutions above the level of
the household. Women may not have access to the social supports they counted on
during the days of the collective. Kin and hot ail are themselves consumed with
their own subsistence activities and may not be available to care for a woman’s
children or replace her labor in the household. In many cases, the chaos of economic transition has led to the dispersal of kin to towns, province centers, and the 242 MEDICAL ANTHROPOLOGY QUARTERLY capital, thus eroding traditional, kin-based social networks. In a highly insecure
economy, women and their children are the most vulnerable.
Regional Socioeconomic Differences
Pastoralism in Mongolia requires the seasonal movement of households to
maximize the use of natural resources available in different ecological zones.
Generally speaking, Mongolia is divided into four geographical/ecological zones:
western high mountains, central mountains, eastern steppe, and Gobi desert regions. The movement pattern varies greatly from west to east. Herders in the
western region are the most mobile due to the great variability of seasonal productivity in the higher altitude environment, compared to central provinces where
the land productivity permits herders to make less extensive moves. Households in
the Gobi desert area migrate short distances depending on the availability of water
resources. On the other hand, households in eastern provinces often make significantly longer migrations of 100 kilometers from north to south to avoid the cold
northern winters, but take advantage of the productive northern grasslands in summer, with multiple stops in-between. The long and frequent movements in eastern
and western regions require greater labor investments and distance people from
their social ties, local administrative centers and markets, information sources, and
health care centers (Humphrey and Sneath 1999). These different environmental
pressures on livestock and herders explain in part the variable survival rates of
humans and animals across the country. Poverty and unemployment rates are the
highest in western provinces, as is the risk of losing the animals during droughts
and winter disasters.
Access to markets and cash incomes varies greatly from region to another.
Being closer to cities and market centers, herders in the central provinces, near
the borders with Russia and China, or along major road or railway lines have
a greater opportunity to sell their products (meat, skins, wool, cashmere). But
herders from eastern, and especially from the more remote western, provinces do
not have such opportunities, unless they have connections through intermediaries
(traders, kin who own trucks, etc.), and are thus forced to sell their products in
local markets where high supplies and low demand result in very low prices. Conversely, in local, rural markets, manufactured goods are prohibitively expensive.
Another important characteristic of the western provinces is that the percentage
of minority ethnic groups (mainly Kazakhs) is much higher compared to other
three regions (Humphrey and Sneath 1999:26). In these areas, the comparatively
lower social status of women, younger age at marriage, higher fertility rates, and
language difﬁculties that impair access to health services contribute to higher risks
for complications of pregnancy.
The above review of the impact of the neoliberal economic reform on women
in the context of the transformation of pastoral economy suggests that the following factors place women at risk of poor pregnancy outcome. First, rural households
are now largely independent, subsistence-oriented productive units, and, as such,
are highly vulnerable to region-speciﬁc economic risks. These include risks to
livestock from climate and disease, degradation of pasturage due to public disinvestment in water resources and absence of supra-household cooperative structures
that would mitigate the “tragedy of the commons,” and lack of access to the market MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 243 Figure 2
Gross domestic product per capita and maternal mortality rates, Mongolia
1985–1998. institutions from which households might derive cash incomes. Level and stability
of cash income, in turn, determine household access to essential social and health
services. Second, in this new subsistence economy, women’s labor is both critical
and highly demanding. Labor demands, especially in the spring and summer seasons, may affect women’s freedom to seek health care, particularly if they have no
kin close by to help. Third, these labor demands in the presence of severe poverty
may affect women’s health directly through a negative impact on the nutritional
adequacy of their diets. We suggest that these three kinds and levels of risk combine to produce the particular pattern of maternal mortality seen in Mongolia over
the transition period.
Maternal Mortality in Post-Socialist Mongolia
Maternal mortality in Mongolia peaked during the most chaotic years of the
transition (1992–1994) and remains higher than the pretransition rates. The degree
to which maternal mortality rates reﬂect general macroeconomic conditions can
be seen in Figure 2, where we plot maternal mortality rates (maternal deaths
per 100,000 live births) and per capita GDP from 1990 to 1999. In 1990, the
maternal mortality rate was 119. It increased dramatically to a high of 240 in
1994, and declined slightly to 175 in 1999. Rates during the socialist period must
be interpreted cautiously, but show that maternal mortality was lower than at
present, falling from 140/100,000 in 1970 to the rate of 119/100,000 in 1990.10
GDP fell dramatically from 1990 to 1994, then increased slowly, but remains much 244 MEDICAL ANTHROPOLOGY QUARTERLY lower than pretransition levels (the per capita GDP in 1985 was about US$1,800
in 1993 dollars; it is now about US$360). As Figure 2 suggests, there is a strong
inverse correlation of per capita GDP and maternal mortality. Controlling for the
total fertility rate—which is a measure of “risk” associated with each pregnancy—
GDP per capita explains 93 percent of the variation in maternal mortality between
1990 and 1999 (i.e., R2 = .93).
In a global context, these data are not particularly surprising; maternal mortality rates tend to reﬂect the overall economic conditions within a country (AbouZahr et al. 1996). However, poor economic conditions in many of the countries
studied globally also correspond with high fertility rates, limited access to reproductive health care including legal abortion, lack of education, and a lower social
status for women (Abou-Zahr et al. 1996; Koblinsky 1995; Okonofua et al. 1992;
Sundari 1992; Turshen 1999; Wall 1998).
In Mongolia, all else being equal, maternal mortality occurs in a social context that should otherwise be favorable to women’s health. Although Mongolian
women’s access to education and careers outside the home has been compromised over the past decade due to rising levels of poverty and unemployment,
in comparison with other nations in central Asia indicators of education and employment rates of women are highly promising, especially when compared with
those of Mongolian men (ADB 2002). Mongolian women tend to marry and begin
childbearing in their mid-20s; birth spacing is culturally desirable; and abortion is
legally available, appears to be reasonably safe, and is widely sought even though
in Mongolian terms it may be costly (the ofﬁcial abortion rate in 1998, which
does not include many private abortions, was 20 percent of live births; Rak and
Janes 2004). As a result, the total fertility rate has dropped steadily over the past
several decades to the present rate of 2.3 births/woman (the TFR in 1990 was
4.3). As we discuss further below, the vast majority of women continue to receive prenatal care, and nearly all births take place in medical facilities. These
data suggest that the social, demographic, and institutional correlates of maternal mortality found in many poor and middle-income countries are not present in
Representatives of the Mongolia Ministry of Health and the World Health
Organization interviewed in 2000 attributed the problem of maternal mortality to
several interrelated factors that affect the quality and accessibility of health care,
particularly in rural areas. They point to the closure of many maternity waiting
homes, especially during the early years of the transition when government investments in health care declined rapidly. One consultant to the WHO suggested that
much of the blame for high maternal mortality can be attributed to the inadequate
training of rural physicians and midwives, coupled with the lack of emergency
drugs and an overall disintegration of the health system. All interviewees suggested that a weakened communications infrastructure, the disintegration of the
rural ambulance service, and a poor transportation system also contribute to the
problem. Finally, an ofﬁcial in the Ministry of Health noted that substantial and
increasing levels of underlying disease might contribute to the rates of complications and death. Women in Mongolia suffer from high rates of urinary tract
infections and anemia (see Table 3 below, also UNDP 2000; WHO 1999). High
rates of urinary tract infections may be explained in part by disturbingly high rates
of sexually transmitted disease (UNFPA/MMOH 2001). MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 245 Table 2
Attributed direct and indirect causes of death, cases of maternal mortality, Mongolia
Main causes: No. of cases % of cases 86
1 Contributing factors:
Poor antenatal care
“Inadequate skills” of doctor or midwife
Long distance from medical care facilities
Late referral to a doctor or midwife
“Irresponsibility” of doctors and medical staff
Shortage of drugs
Delay of ambulance
No clear contributing factor
No data 79
6.9 Underlying disease present in mortality cases (some
cases had more than one diagnosis)
Kidney and urinary tract
Neurological and psychiatric diseases
No underlying disease noted
No data 64
12.9 Postpartum hemorrhage
Indirect causes “extragenital diseases”
Rupture of the uterus
No data Ofﬁcials noted that rates have improved since 1994 as a result of rebuilding and reopening many maternity waiting homes, improved supplies of emergency drugs, and concerted efforts by the government, supported by UNICEF and
UNFPA, to train reproductive health care providers in managing reproductive and
obstetric health problems.
In our review of documents and case records, we sought to examine these
hypothesized determinants in greater detail. In Tables 2 and 3, we present information on all 217 maternal deaths recorded in Mongolia during the three-year period 246 MEDICAL ANTHROPOLOGY QUARTERLY
Description of maternal deaths, Mongolia, 1996–1998. Indicator Cases (217) % of Cases Mother’s residence
Aimag (province) centers
Soum (county or district) centers
Bhag (township or community) 19
43.3 Where delivery occurred:
Maternity houses in cities
Aimag hospital’s obstetrical department
15.2 Where death occurred:
Maternity houses in cities
Aimag hospital’s obstetrical department
1.8 Educational level of mothers:
General occupational status of mothers:
Age unknown MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 247 from 1996 to 1998. Attributions of cause of death (Table 2) were made in this
case by physicians in the Ministry of Health, and are reported here as they were
reported to us. Postpartum hemorrhage is the main direct cause of death, followed
by underlying or “indirect causes” and eclampsia. Our review of the 1998 records
suggests that underlying disease is increasing in importance as the attributable
cause of death. Analysis of rural records suggest that eclampsia is more prevalent
in rural than in urban areas. Physicians attribute these higher rates of eclampsia in
rural areas to inadequate antenatal care and underlying urinary tract infections (kidney disease), which they consider to be associated with “difﬁcult conditions” and
“cold” in the countryside.11 Other underlying causes include anemia, respiratory,
and cardiovascular disease.
Published analyses of the most recent data (1999; these ﬁgures are not shown
in Tables 2 or 3) by the Mongolian Ministry of Health indicates that 43 percent
of deaths were attributed to underlying or indirect causes. These 1999 data also
indicate high rates of anemia (47 percent) and urinary tract infections (32 percent)
(Demberelsuren and Dorjpurev 2000). Corresponding to seasonal demands on
women’s labor in a high-risk economic environment, maternal mortality is greater
during the highest stress period of February through July; 60 percent of all maternal
deaths in 1996–1998 occurred during this period (Demberelsuren and Dorjpurev
Ministry of Health physicians attempted to come to a consensus regarding
the relationship of inadequate health care and poor health care infrastructure to
maternal deaths. Though antenatal care rates are quite high (the great majority
of Mongolian women receive antenatal care by the second trimester), Ministry
of Health physicians attributed most deaths to poor antenatal care (36 percent of
Our review of these records indicate that they typically provide little information on the speciﬁcs of this care provided and a judgment that there was some
mistake made in assessing risk for later complications is, in our opinion, questionable. In about 20 percent of the cases, review of the medical records suggested
that the physician or midwife made mistakes in handling the emergency, and in 12
percent of the cases, it is believed that the death might have been prevented had
the doctor or midwife not delayed in referring the mother to a higher-level facility.
Sixteen percent of the cases were attributed to difﬁculties women experienced in
reaching a health care facility by virtue of the long distances she had to travel.
The larger points to be drawn from this review are that: (1) only about half of the
cases of death could be attributed to unanticipated emergent processes during labor
and delivery (e.g., hemorrhage and eclampsia); (2) Mongolian women who died
of maternal causes have high rates of underlying sickness, reﬂecting their overall
poor health status (only 18 percent were considered free of underlying disease);
and (3) medical system factors—primarily inadequate care provided by doctors
and midwives—contributed to the death.
Table 3 describes the setting in which the death occurred and the sociodemographic status of the mothers. Delivery was more likely to occur in lower-level and
rural facilities (county clinics, township clinics, or “feldsher” posts) or at home,
although most deaths occurred in county or provincial level facilities (reﬂecting
referral patterns from county to provincial hospitals in rural areas). Victims were
primarily herders (49 percent), working class (14 percent), or were considered 248 MEDICAL ANTHROPOLOGY QUARTERLY Table 4
Relationship of residence and sociodemographic factors to maternal mortality,
deaths giving birth incidence rate Rate ratio
Eastern & central provinces
Pastoral (most rural) zones
Living elsewhere 208
0.001027 1.53 (1.23, 1.90)
2.17 (1.71, 2.75)
1.0 (ref) Age
0.004548 1.33 (0.9, 1.96)
3.41 (2.66, 4.36) Occupation
0.00137 1.59 (1.29, 1.97)
1.0 (ref) The 1995 cases we include in this analysis lacked sufﬁcient information on the cause of
death to be included in Tables 2 and 3. However, there was sufﬁcient information present
to determine residence, age, and main occupation of the deceased to be included in this
ecological analysis of mortality incidence. by their physicians to be unemployed outside the home (“housewives,” 27 percent). Most victims had a middle school or higher level of education, reﬂecting
the overall high literacy rates in Mongolia. Finally, and consistent with the global
epidemiology of maternal mortality, the risk of dying is greatest for primigravids
and women who have four or more pregnancies; together these account for more
than half of all mortality.
In a separate analysis, we examined the geographic distribution of maternal
mortality (Table 4). Our intent was to discover whether the risk for mortality
correlated with regional differences in ecology and impact of economic reforms we
describe in the preceding section. As expected, rates of mortality are signiﬁcantly
higher in the western part of the country. Western Mongolia is more mountainous, is
more prone to severe winters, has the least developed transportation infrastructure,
and has the least favorable economic conditions. In the very far west, there are
Kazakh communities where the risks for maternal mortality are high. As indicated
in our epidemiologic analysis of mortality during the four-year period of 1995
to 1998, women at signiﬁcantly highest risk for death are those who live in the
western regions of the country, especially in the most rural regions; are under 19
years of age or over the age of 35; and are herders.
In summary, our epidemiologic analysis of maternal mortality records suggests that maternal death is the result of a web of factors, operating at different
levels, and are related to or consistent with macroeconomic changes wrought by
neoliberal reform. In our search for the nexus where macroeconomic factors affect MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 249 speciﬁc constellation of individual risks (Dunn and Janes 1986), we recognize two
possible causal pathways. The ﬁrst identiﬁes the consequences of declining government investments in health care, speciﬁcally as these relate to the degradation
of the health care system’s ability to provide adequately for women’s health. The
importance of this pathway is suggested by the large proportion of causes attributed
to health system factors: inadequate care, delays, poor skills, and compromised or
degraded infrastructure. The second operates through the more diffuse impact of
economic reform on the practice of pastoralism in the rural countryside, raising
the levels of economic risk borne by individual households, increasing poverty
and social inequality, and exacerbating the insecurity and vulnerability of women.
The salience of this pathway is reﬂected in the sociodemographic and geographic
patterning of maternal mortality and in the high rates of underlying disease among
the cases of death. Poverty is likely the common denominator here.
Not all maternal mortality is related to neoliberal reform, and we do not intend
to suggest such a unicausal relationship. Providing adequate health care in remote,
rural regions where there is a poor transportation and communication infrastructure
is a signiﬁcant challenge under the best of circumstances. Yet increases to maternal
mortality rates since 1990 and documented changes to the rural economy and
public health infrastructure suggest that economic reform at minimum exacerbates
or ampliﬁes the social factors that underlie maternal mortality in Mongolia and
perhaps more importantly, will seriously impede attempts to prevent it.
Epidemiologic data outline only the rough edges of the problem. Ethnographic exploration of cases of maternal mortality—as seen from the perspectives
of the spouse, other relatives, and health care providers—show how case-unique
constellations of risks, in the context of individual and family decision-making,
come together to produce this most tragic of events (e.g., Hay 1999). Such case
studies have the potential to ﬁll in the causal spaces left unﬁlled by crude nature of
epidemiologic analyses and are particularly important to understanding the critical contextual factors that contribute to maternal deaths (Miller et al. 2003). The
case studies of maternal mortality we have collected to date show how the causal
pathways we identify above operate to create greater risks for and vulnerability of
rural women. They also show how particular factors, operating at different social
ecologic levels, coalesce to place women in danger.
Dulamsuren, a very poor woman and migrant from the countryside to the
provincial capital of Huvsgol, died in 2002. At the time of her death, Dulamsuren
was 41 years old, and her occupation was indicated by her family doctor as being
that of “housewife.” Our assessment suggests that both she and her spouse were
unemployed, poor, and often homeless. Dulamsuren had had six previous children.
During the birth of the sixth child, there were several complications—edema and
preeclampsia were noted, and after delivery, manual extraction of the placenta was
required. After this birth, Dulamsuren said she wanted no more children, and was
given an IUD. She had a history of problems with the IUD and ﬁnally, after six 250 MEDICAL ANTHROPOLOGY QUARTERLY years of use, due primarily to pelvic inﬂammatory disease and associated pain, she
asked to have the IUD removed. At the time, the province hospital’s obstetrician
who removed the IUD urged her to use another form of birth control. Her primary
care physician gave her the same advice. For reasons that are unclear, Dulamsuren
did not follow this advice and was soon pregnant again. She sought no prenatal
care for this pregnancy. The family doctor discovered the pregnancy when she
went to conduct a prenatal exam for Dulamsuren’s 18-year old pregnant daughter.
Because of Dulamsuren’s history and age, and because Dulamsuren said that she
did not want the child, the family doctor urged her to go to the province hospital
for an abortion. However, by the time Dulamsuren collected sufﬁcient funds for
the abortion and returned to the hospital, she was told that her pregnancy was too
advanced and that abortion was no longer an option.12 Dulamsuren returned home
and received prenatal care from the family doctor. As she came closer to term, she
manifested symptoms of preeclampsia—high blood pressure and edema. Because
of her age, history of complications, and the presence of these serious symptoms,
the doctor urged her to go to the province hospital’s maternity waiting home. However, initially because she did not have any record of having health insurance,13
and then because the province maternity waiting home had no beds at the time,
Dulamsuren’s admission to the hospital was delayed for over a week while the
bureaucratic issues were resolved. Finally, she was admitted to the hospital about
two days prior to going into labor. She was delivered via caesarean section, but
suffered severe hemorrhaging. The hemorrhaging was eventually stopped by emergency surgery (there was a delay in ﬁnding the anesthesiologist), but hemorrhagic
shock worsened because there was no blood infusion available at the province
hospital. Dulamsuren died from hemorrhagic shock.
Enkhtuya was a herdswoman in the central province of Tov, within a day’s
travel of the capital, Ulaanbaatar. At the time of her death in 2001, she was 35 years
old and pregnant for the fourth time. During this pregnancy, she was diagnosed
by a county-level physician as suffering from a cardiovascular anomaly, probably
(though not deﬁnitively) of congenital origin. The midwife who supervised her
prenatal care advised her to give birth in a maternity house in Ulaanbaatar organized
to provide care to women experiencing high-risk pregnancies. However, Enkhtuya
refused to go because her husband had gone to Ulaanbaatar to try to sell some
animal fur for desperately needed cash, and she needed to stay at home with two
of her children, who at the time were eight and ten years old. When she went
into labor, she traveled to the county clinic where she had received prenatal care.
The county clinic was ill equipped do deal with cardiovascular complications.
Enkhtuya died suddenly during childbirth from heart failure.
Norjmaa was a herdswoman in Gov-Altai province. At age 33, she became
pregnant for the fourth time. In October 2001, pregnant about 30 weeks, she walked
some distance into the mountains to cut wood for the coming winter. According
to her husband, she collected and carried a large quantity of ﬁrewood wood back MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 251 to the family home. This effort apparently resulted in placental abruption and she
began to hemorrhage. The woman did not tell anyone, but her husband noticed
her hemorrhaging and sent his brother to get help from the midwife who worked
in the county hospital. The midwife arrived within four hours, but Norjmaa was
already in deep hemorrhagic shock, and nothing could be done to save her.
Tsetseg was a herdswoman from a very remote county in the western
province of Arkhangai. At the time of her death in 2001, Tsetseg was 24 years old
and experiencing her second pregnancy. When she was about 32 weeks pregnant,
she went into labor and gave birth prematurely at home. Her husband attempted
to make a call from the nearest township center to the county hospital at 11:30
p.m., but because the county post ofﬁce was closed the call did not go through.
The husband was eventually able to contact the hospital at 9:00 the next morning.
However, the hospital ambulance was out of fuel, and it took one and a half hours
to ﬁnd sufﬁcient money and fuel for the driver to make the trip. It took about ﬁve
hours of driving to reach the woman’s home. By the time the ambulance arrived,
Tsetseg had died from hemorrhage.
In each of these cases, and in the many others we have collected, it appears
that women die as a result of highly unique constellations of factors, some preventable, some not. Although a focus on individual cases does not always reveal
links to higher-order, contributing causes for example, economic transformation
to the rural economy, we believe that these cases are suggestive of the etiologic
importance of four processes, acting singly or in concert, that link these cases to
the macroeconomic and macrosocial factors identiﬁed in this article.
First, women who die in childbirth are economically marginal: they are invariably poor, often desperately so. While many rural women and their families
were probably economically disadvantaged during the socialist period, particularly
in comparison with urban and bureaucratic elites, it is doubtful that they suffered
the depth and severity of poverty that is now virtually commonplace throughout
Mongolia (ADB 2002; Grifﬁn et al. 2001; UNDP 2000).
Second, victims are often socially marginal. They lack an integrated and wellfunctioning social support system: their spouses may be absent, abusive, and/or
alcoholic; their living situations are characterized by instability; and their relatives
may be absent or unable and/or unwilling to provide any assistance, economic or
otherwise. We believe that this breakdown in social support systems is a casualty
of the disordering of traditional social relations, ﬁrst during the postwar period of
radical collectivization, and later during the period of economic and social chaos
that accompanied privatization (Grifﬁn et al. 2001; Humphrey and Sneath 1999).
Third, their interactions with the health care system illustrate a cascade of
failures—failure to act on medical advice (which may or may not be determined by
their poverty), failure to seek health care when confronted with serious symptoms,
and failure of the health system to provide adequate or timely emergency care. As
we have discussed, poor quality health care and a weakened communications and
transportation infrastructure are some of the casualties of the decline in government
support for rural development and health care. 252 MEDICAL ANTHROPOLOGY QUARTERLY Fourth, many cases of maternal mortality illustrate how heavy labor demands
in the rural economy underlie the risks previously cited. These demands have
increased with the decline of cooperative production and the need for now economically independent households to maximize production when and wherever
possible in order to minimize risk in a difﬁcult environment.
In just over ten years, neoliberal economic reform and the retreat from social
investment have had signiﬁcant consequences for Mongolian women and their
families. Maternal mortality has increased, particularly among rural populations
in the west where privatization, economic chaos, and eroding infrastructure have
combined to produce a large number of economically vulnerable households. For
the ﬁrst time in many generations, herding households have been forced to rely
exclusively on independent, subsistence-oriented production. They have become
increasingly vulnerable to disease and ecological perturbations and have suffered
marked reductions in access to quality health care, education, and essential goods
and commodities. A high degree of variability in access to resources, and hence,
social inequality, now marks the rural landscape.
Women, chief architects of the household production of health, particularly
their own, suffer under these circumstances. They may not have the resources in
either money or time to invest in their own health. As a result, their level of overall
health is compromised. Women who are at the greatest risk of dying in pregnancy
and childbirth share the following characteristics: they are poor, have lower levels
of education, are herdswomen or housewives, live in remote rural areas, are the
youngest mothers, or have had many children. They die from hemorrhage, from
many underlying or indirect causes, and from eclampsia. They die at home or in
lower-level facilities with limited emergency drugs or equipment. They die because many social, economic, cultural factors lead them to postpone the call to a
doctor, because poor infrastructure limits their ability to get health care on time,
and because available medical personnel cannot cope effectively with the emergency. And ﬁnally, they die because in the frenzy to produce the ideal neoliberal
state, the policy of economic reform implemented by government and its international allies did little but produce widespread poverty, inequality, and a radical
decline in investment in human capital. In Mongolia, the strategies devised to create “free” markets have also created a population of vulnerable, sick, and dying
1. The causes of maternal deaths fall into two groups: direct and indirect obstetric
causes. Direct obstetric deaths are those resulting from obstetric complications of the pregnant state and are the same worldwide: hemorrhage, sepsis, eclampsia, obstructed labor,
and complications of abortion. Indirect obstetric deaths are those resulting from previous
existing disease, or nonobstetric diseases that arise during pregnancy and are exacerbated
by the physiologic effects of pregnancy (AbouZahr et al. 1996). These include such things
as anemia, cardiovascular diseases, hepatitis, diabetes, and tuberculosis.
2. The tenth revision of the international statistical classiﬁcation of diseases (ICD-10)
deﬁnes a maternal death as the death of a woman while pregnant or within 42 days of MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 253 termination of pregnancy, irrespective of the duration and site of the pregnancy, from any
cause related to or aggravated by the pregnancy or its management, but not from accidental
or incidental causes.
3. Maternity waiting homes were residential facilities attached to county-level hospitals throughout rural Mongolia. Pregnant women would be brought to stay in the homes
weeks prior to term, and would stay for weeks after birth. This process ensured that mother
and child had access to basic medical care prior to, during, and after birth (Figa-Talamanca
4. Over the period of 1990–1992, about 3,000 enterprises of various sizes were privatized. Initially, privatization was accomplished via the distribution of vouchers, although
after 1996 it was largely via cash sales at public auctions. The voucher system was used to
ensure a relatively equitable distribution of assets and to avoid the concentration of wealth
in the hands of well-positioned elites (which came to be known to some as “crony capitalism”) that characterized economic reform in the states of the former Soviet Union and
some eastern European countries (see, e.g., the works by Wedel  and Verdery ).
However, due to a lack of public education, and some confusion in its implementation, the
voucher system was not completely successful in preventing inequitable privatization.
5. The “depth” of poverty is the percentage that the average poor person’s income is
below the poverty line. The “severity” takes into account income distribution within the
poor, giving more weight to those who are furthest from the poverty line. These increases
correspond to the increasing income inequality referred to in the text.
6. Winter disasters, or dzud, are abnormally cold and/or snowy winters that follow dry
summers. Livestock, already weakened by the scarcity of grass during the summer, may die
in great numbers as a consequence of extreme cold, or when snow covers what few grasses
7. The so-called tragedy of the commons is not inevitable if there are customary
social and economic mechanisms in place that regulate access to productive resources.
Unfortunately, these mechanisms were displaced by economic reform and have yet to be
8. Generally speaking, to be reasonably self-sufﬁcient, a rural household with four–six
members requires about 100 animals. This number provides a buffer for difﬁcult years and
sufﬁcient numbers to trade or sell for essential commodities when needed.
9. In the summer of 2002, we interviewed the heads of 24 rural households on the
subject of household economics and access to health care (Janes 2003). Most of these
households depended either solely, or in part, on herding. We discovered a high degree of
economic inequality even among households or hot ail in close proximity. In one area of
southern Huvsgol province, we interviewed a household that owned over 1,000 animals.
They employed members of several other much less prosperous households. The cash
income from the sale of cashmere permitted the family to buy several motor vehicles,
which they used to transport animal products to market in Ulaanbaatar and which they
hired out to less-fortunate households. A few kilometers away, we interviewed a female
head of household that had lost what few animals they had to the winter disaster of 2000.
They survived with the help of neighbors, but it did not seem likely that they would be able
to remain in the countryside.
10. Maternal mortality rates varied during the socialist period from a high of
140/100,000 in 1970 to 120/100,000 in 1989. However, Mongolia did not use the ICD10 methodology in calculating maternal mortality rates, so pre-transition rates are probably
not comparable to post-transition rates. These ﬁgures are thus not included in the tables or
ﬁgures presented in this article.
11. The reportedly high rates of kidney disease, and the attribution of this disease
to “cold,” may reﬂect Tibeto-Mongolian humoral thinking. Western researchers have been
puzzled by the purportedly high rates of nephritis, particularly in the absence of diagnostic
testing. 254 MEDICAL ANTHROPOLOGY QUARTERLY 12. Several physicians noted the serious consequences of restricted access to abortion
services by poor women. Reported one family doctor, “By the time a woman collects money
for an abortion, the safer, ﬁrst-trimester period for the procedure has already passed, and a
woman must then collect money for the pregnancy termination in the second trimester of
pregnancy, which is three times as expensive as the price of medical abortion in the ﬁrst two
months. And by the time she collects enough money for the later termination of pregnancy,
it is already time to deliver the baby.”
13. Mongolia adopted a national health insurance system in 1994 (see Janes 2003).
Dulamsuren, as a pregnant woman, is entitled under this system for free health insurance.
However, she is required to apply for such insurance and to present proof of insurance when
seeking health services. Health insurance requires proof of residence, which is sometimes
difﬁcult for the very poor who are either homeless, move from one relative’s residence to
another’s, and do not take the steps necessary to establish legal residency. REFERENCES CITED
1971 Retention of Cultural Values and Differential Urban Adaptation: Samoans and
Relocated American Indians in a West Coast City. Social Forces 49(3):385–393.
1999 Living with Genetic Disorder: The Impact of Neuroﬁbromatosis 1. Westport, CT:
Auburn House, Greenwood Press.
2002 The Nature of Stigma and Medical Conditions. Epilepsy and Behavior 3:6 Supplement 2.
AbouZahr, Carla, and E. Royston
1991 Maternal Mortality: A Global Factbook. Geneva: World Health Organization.
AbouZahr, Carla, Tessa Wardlaw, Cynthia Stanton, and Kenneth Hill
1996 Maternal Mortality. World Health Statistics Quarterly 49:77–87.
Asian Development Bank (ADB)
2002 Social Impact Assessment for the Second Health Sector Development Project. TA
No. 3750-MON. Ulaanbaatar: Asian Development Bank and the Mongolia Ministry
1999 History of Mongolia. Translated by D. Suhjargalmaa, S. Burenbayar, H. Hulan,
and N. Tuya. Cambridge: The Mongolia and Inner Asia Studies Unit, University of
Barnes-Josiah, Deborah, Cynthia Myntti, and Antoine Augustin
1998 The “Three Delays” as a Framework for Examining Maternal Mortality in Haiti.
Social Science and Medicine 46:981–993.
Demberelsuren, C., and T. Dorjpurev
2000 The Causes and Determinants of Maternal Mortality in Mongolia and the Possible
Solutions. Ulaanbaatar, Mongolia: Mongolia Ministry of Health and Social Welfare,
and UNICEF: The Maternal and Children’s Research Center.
Dunn, Frederick L., and Craig R. Janes
1986 Introduction to Anthropology and Epidemiology. Craig R. Janes, Ron Stall, and
Sandra Gifford, eds. Dordrecht, The Netherlands: Kluwer.
1997 On Suffering and Structural Violence: A View from Below. In Social Suffering.
Arthur Kleinman, Veena Das, and Margaret Lock, eds.Pp. 261–284. Berkeley: University of California Press. MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 255 2003 Pathologies of Power: Health, Human Rights, and the New War on the Poor.
Berkeley: University of California Press.
1996 Maternal Mortality and the Problem of Accessibility to Obstetric Care: The Strategy of Maternity Waiting Homes. Social Science and Medicine 42:1381–1390.
Government of Mongolia
1998 Living Standards Measurement Survey, 1998. Ulaanbaatar, Mongolia: National
2000 National Statistical Report, 2000. Ulaanbaatar, Mongolia: National Statistics Ofﬁce.
Grifﬁn, Keith, Mark D. Brenner, Takayoshi Kusago, Amy Ickowitz, and Terry McKinley
2001 A Strategy for Poverty Reduction in Mongolia. Report of a UNDP mission on
the Integration of Equity and Poverty Reduction Concerns into Development Strategy.
Ulaanbaatar, Mongolia: United Nations Development Programme, July.
1999 Dying Mothers: Maternal Mortality in Rural Indonesia. Medical Anthropology
Humphrey, Caroline, and David Sneath
1999 The End of Nomadism? Society, State and the Environment in Inner Asia. Durham,
NC: Duke University Press.
Israr, S. M.
2000 Coping Strategies of Health Personnel during Economic Crisis: A Case Study
from Cameroon. Tropical Medicine and International Health 5:288–292.
Janes, Craig R.
2003 Market Fetishism and Attenuated Primary Care: Producing Poor Medicine for
Poor People in Post-Socialist Mongolia. Bulletin of the Royal Institute for Interfaith
Kleinman, Arthur, and Joan Kleinman
1997 The Appeal of Experience; the Dismay of Images: Cultural Appropriations of
Suffering in Our Time. In Social Suffering. Arthur Kleinman, Veena Das, and Margaret
Lock, eds. Pp. 1–24. Berkeley: University of California Press.
Koblinsky, M. A.
1995 Beyond Maternal Mortality: Magnitude, Interrelationship, and Consequences of
Women’s Health, Pregnancy-Related Complications and Nutritional Status on Pregnancy Outcomes. International Journal of Gynaecology and Obstetrics 48(Supp.):S21–
1994 Epidemiology and the Web of Causation: Has Anyone Seen the Spider? Social
Science and Medicine 39:887–903.
2000 Discrimination and Health. In Social Epidemiology. Lisa F. Berkman and Ichiro
Kawachi, eds. Pp. 36–75. New York: Oxford University Press.
Kwast, Barbara E.
1998 Building a Community-Based Maternity Program. International Journal of Gynaecology and Obstetrics 48(Supp.):S67–S82.
2000 Reducing Perinatal and Maternal Mortality in the World: The Major Challenges.
British Journal of Obstetrics and Gynaecology 106:877–880. 256 MEDICAL ANTHROPOLOGY QUARTERLY Maine, Deborah, and Allan Rosenﬁeld
1999 The Safe Motherhood Initiative: Why Has It Stalled? American Journal of Public
1999 Risk, Uncertainty, and Agency: Culture and Safe Motherhood in Morocco. Medical Anthropology 19:173–201.
Miller, Suellen, Nancy L. Sloan, Beverly Winikoff, et al.
2003 Where Is the “E” in MCH? The Need for an Evidence-Based Approach in Safe
Motherhood. Journal of Midwifery and Women’s Health 48:10–18.
Okolocha, Chike, Jasper Chiwuzie, Suleiman Braimoh, Jacob Unuigbe, and Patience Olumeko
1998 Socio-Cultural Factors in Maternal Morbidity and Mortality: A Study of a SemiUrban Community in Southern Nigeria. Journal of Epidemiology and Community
Okonofua, Friday E., A. Abejide, and Roger A. Makanjuola
1992 Maternal Mortality in Ile-Ife, Nigeria: A Study of Risk Factors. Studies in Family
PMMN (Prevention of Maternal Mortality Network)
1992 Barriers to Treatment of Obstetric Emergencies in Rural Communities of West
Africa. Studies in Family Planning 23:279–291.
Prual, A., M. H. Bouvier-Colle, L. de Bernis, and G. Breart
2000 Severe Maternal Morbidity from Direct Obstetric Causes in West Africa: Incidence
and Case Fatality Rates. Bulletin of the World Health Organization 78:593–602.
Rak, Kimberly, and Craig R. Janes
2004 Reproductive Health in Post-Transition Mongolia: Global Discourses and Local
Realities. Global Technology and Development, In Press.
Rizvi, Javaid H., Khalid S. Khan, Chika R. Nwosu, and Odenefe J. Isiavwe
1999 Maternal Mortality in the Developing World: Many Unanswered Questions.
Journal of Obstetrics and Gynaecology Research 25:149–151.
Shen, Ce, and John B. Williamson
1999 Maternal Mortality, Women’s Status, and Economic Dependency in Less Developed Countries: A Cross-National Analysis. Social Science and Medicine 49:197–214.
N.d. Development Trends and Distribution of Livestock in Mongolia. InnerAsia 5(3),
Stewart, M. Kathryn, Cynthia K. Stanton, Mario Festin, and Nora Jacobson
1996 Issues in Measuring Maternal Morbidity: Lessons from the Philippines Safe Motherhood Survey Project. Studies in Family Planning 27:29–35.
1991 Determinants of Maternal Mortality in the Developing World. Australian and New
Zealand Journal of Obstetrics and Gynaecology 31:8–16.
1996 Translating Social Ecological Theory into Guidelines for Community Health Promotion. American Journal of Health Promotion 10:282–298.
Sundari, T. K.
1992 The Untold Story: How the Health Care Systems in Developing Countries Contribute to Maternal Mortality. International Journal of Health Services 22:513–528. MATERNAL MORTALITY IN POST-SOCIALIST MONGOLIA 257 Thaddeus, Serena, and Deborah Maine
1994 Too Far to Walk: Maternal Mortality in Context. Social Science and Medicine
Thomas, R. Brooke
1998 The Evolution of Human Adaptability Paradigms: Toward a Biology of Poverty.
In Building a New Biocultural Synthesis: Political-Economic Perspectives on Human
Biology. Alan H. Goodman and Thomas L. Leatherman, eds. Pp. 43–74. Ann Arbor:
University of Michigan Press.
1999 Privatizing Health Services in Africa. New Brunswick, NJ: Rutgers University
UNDP (United Nations Development Programme)
2000 Human Development Report, Mongolia, 2000. Ulaanbaatar, Mongolia: UNDP.
2003 Human Development Report, Mongolia, 2003. Ulaanbaatar, Mongolia: UNDP.
UNFPA/MMOH (United Nations Family Planning Agency/Mongolian Ministry of Health)
2001 Population and Reproductive Health in Mongolia. Ulaanbaatar, Mongolia:
UNFPA/National Statistical Ofﬁce of Mongolia.
2001 What Was Socialism, and What Comes Next? Princeton: Princeton University
Wall, L. L.
1998 Dead Mothers and Injured Wives: The Social Context of Maternal Morbidity and
Mortality among the Hausa of Northern Nigeria. Studies in Family Planning 29:341–
Walraven, G., M. Telfer, J. Rowley, and C. Ronsmans
2000 Maternal Mortality in Rural Gambia: Levels, Causes and Contributing Factors.
Bulletin of the World Health Organization 78:603–613.
Walsh, Julia A., Anthony R. Measham, Christine N. Feifer, and Paul J. Gertler
1994 The Impact of Maternal Health Improvement on Perinatal Survival: Cost-Effective
Alternatives. International Journal of Health Planning and Management 9:131–149.
2000 Collision and Collusion: The Strange Case of Western Aid to Eastern Europe.
New York: Palgrave.
1993 World Development Report 1993: Investing in Health. New York: Oxford University Press.
World Health Organization
1999 Mongolia Health Sector Review. WHO-Mongolia and the Mongolia Ministry of
Health and Social Welfare, Ulaanbaatar, Mongolia, June 1999.
World Health Organization and UNICEF
1996 Revised 1990 Estimates of Maternal Mortality: A New Approach. Geneva: WHO. ...
View Full Document
This note was uploaded on 11/27/2011 for the course ECON 101 taught by Professor Robert during the Fall '08 term at Montgomery College.
- Fall '08