Regional Health Forum WHO South-East Asia Region(Volume 6, Number 2)

Reproductive Health

No More Cradles in the Graveyards
Than Sein* and Uton Muchtar Rafei**

Several years ago in Myanmar, the graveyards in villages and small towns used to have many cradles scattered around the ground, some partially buried and some broken. These cradles were sad reminders of the death of mothers either during pregnancy, at the time of delivery or after. The baby too might have died with the mother or after a few days. In some cases, it meant a still-birth or the death of young infants, mainly neonates. It is the same story in most countries of the South-East Asia Region, where, compared to the global average four to ten times higher maternal deaths take place. According to a popular Myanmar saying, there are two risks to life, one for men who go rafting along fast flowing rivers; and the other for women during childbirth.

Why should pregnancy and childbirth pose such a risk to the precious life of women? Pregnancy and childbirth are normal physiological, reproductive pheno-menon for human existence and most pregnancies and childbirth are normal without any risk to health. Many causes for ill-health during pregnancy and childbirth can be prevented and good health practices

promoted during these periods. Advocacy for health interventions to reduce the burden of illness during pregnancy and childbirth must be founded on fact, not conviction. These interventions should be based on evidence and not on habits and tradition.

This paper highlights the risk to pregnant women, how to reduce ill-health and death during pregnancy and childbirth, and examines the reasons for different risk factors for women and mothers in different countries in South-East Asia.

Measure of Burden

The burden of women's health related to pregnancy and childbirth is usually measured through monitoring maternal deaths in a community. WHO defines¹ a maternal death as the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes. Maternal deaths are also indicative of the performance of the health systems.

The most commonly used indicators of maternal mortality² include:

     Maternal mortality ratio (MMR) is the most commonly used indicator and it is the ratio of the number of maternal deaths during a given time period per 100 000 live births during the same time period. It is not a true death rate in that it does not measure deaths per women-year of exposure to pregnancy and childbirth within a specified time period. An appropriate denomi-nator would be the total number of pregnancies (live-births, foetal deaths (still-births), induced or spontaneous abortions, ectopic and molar pregnancies), but this figure is seldom available. Hence, the general use is live births for practical reasons. It is the ratio which indicates the risk of death once a woman has become pregnant.

     Maternal mortality rate (MMR) is the measurement of true rate and it is the total number of maternal deaths in a given period (usually a calendar year) per 100 000 women of reproductive age (aged 15-49) during the same time period. It reflects both the risk of death among pregnant women, and the frequency with which women are exposed to this risk. Different countries may have used the age-groups of women of reproductive age (15-45) or (15-49). The rate could be reduced either by making childbearing safer (as is true for the ratio mentioned above) and/or by reducing the number of pregnancies.

     Lifetime risk is a new measurement for maternal mortality, taking into account both the probability of becoming pregnant and the probability of dying as a result of that pregnancy cumulated across a woman's reproductive years. It is a cohort measure of maternal mor-tality. It can be approximated by multiplying the maternal mortality rate (MMR) by the length of the reproductive period (around 35 years). Thus, the lifetime risk is calculated as:1-(1-MMR)35.

The lifetime risk can also be approximated by the product of the total fertility rate (TFR) and the maternal mortality ratio (MM Ratio) with an adjustment factor of 1.2 or 1.5 in order to compensate for pregnancy loss (stillbirths or ectopic pregnancy, etc.). Thus, the lifetime risk is 1.2 X TFR X MM Ratio).

The global estimates for 1995³ by WHO/UNICEF/UNFPA showed that while Thailand, Sri Lanka and Myanmar have the lifetime risk of 1 in 1100, 610 and 190 respec-tively, Bangladesh, Bhutan, Maldives and Nepal have 1 in 42 or below. While DPR Korea has the risk of 1 in 1100, India and Indonesia have the risk of 1 in 55 and 65. The bigger the ratio, the higher the risk.

In practice, it is extremely difficult to measure maternal mortality, especially in countries where the system of recording vital events including maternal deaths is relatively incomplete. One of the difficulties encounte-red in measuring maternal mortality is related to the timing of death and the causes of death in relation to the pregnancy status of the women.

Vital registration systems usually account maternal deaths, but rarely record the cause of death. In some case, hospitals might have recorded the cause of death (e.g. kidney or heart failure or tuberculosis) but the fact of the pregnancy may not have been noted in the death certificate. Various studies show that underreporting of maternal deaths in official data ranges from 20-70%.

In many communities, maternal death is a rare event, especially in countries with the MM Ratio not exceeding 1 per 1 000 live births. Moreover, to get statistically reliable maternal mortality data, the community surveys require a very large sample and are thus very expensive to conduct. Inadequate information on maternal deaths has long been an obstacle in alerting health policy-makers to the magnitude and causes of the problem.

It is more relevant today as the global target has been set for a significant reduction in maternal mortality ratio, i.e. reduction by three-quarters, between 1990 and 2015, as an international Millennium Development Goal (MDG) (Goal 5 Target 6), by the United Nations at the Millennium Summit in 2000.

In order to determine the extent of the problem, WHO and UNICEF had introduced a new methodology to estimate global maternal mortality in the early 1990s. Data from various sources were taken into account when computing these estimates. The results based on the model-based estimates for a large group of developed and developing countries were issued by WHO/UNICEF in 1996. This report had generated consider-able discussion around a number of key issues, particularly the way in which countries were classified, the use of survey data and its adjustment, and the use and manner of presentation of the estimates.

However, the results drew the attention of many countries and scientists around the world and resulted in (a) increased awareness of the problems of measuring maternal mortality; (b) identification of various weaknesses and strengths of the different measurement approaches; and (c) getting appropriate national and sub-national data on maternal mortality which were hitherto unavailable. It also helped stimulate new data collection.

After a series of consultations, WHO together with UNICEF and UNFPA made efforts to carry out a similar exercise using revised methodologies, and incorporating new data sets. The revised estimates for the MM Ratio by countries and regions for 1995 were presented. The wide uncertainty levels of the estimates indicated that there could be a considerable range (Figure 1).

According to these estimates, around half-a-million maternal deaths occurred globally in 1995, of which 31% were in WHO’s South-East Asia Region. Twelve countries including Bangladesh, India and Indonesia contributed to nearly two-thirds of all maternal deaths. India, with an estimated 110 000 deaths, has the highest number in the world. While estimating the 1995 maternal deaths, it was stressed that such estimates should not be used for monitoring, but for drawing attention to the existence and likely dimensions of the problem.

In addition to using MM Ratios as an outcome indicator for measuring maternal mortality, countries have started using other process indicators to monitor and evaluate health systems development, most likely to reduce maternal mortality. Such common indicators include:

     Proportion of pregnant women receiving antenatal care at least once by skilled health personnel;

     Proportion of childbirths attended by skilled health personnel;

     Proportion of women receiving postpartum care by skilled health personnel;

     Proportion of complicated obstetric admissions compared with all deliveries at health institutions;

     Proportion of births by Caesarean section, and

     Proportion of women or pregnant women immunized with tetanus toxoid.

These process indicators are more relevant, since the appropriate health information related to them is generally easier and less expensive to record, collect and analyze, and also more sensitive to health system changes. According to WHO estimates, only 70% of births in the developing world in any given period are preceded by a single antenatal visit. Nearly 38 million women receive no antenatal care.

The universal challenge in maternal care is its unpredictability, especially at the time of delivery. Since life-threatening complications can arise suddenly and without warning, the presence of skilled, knowled-geable personnel at delivery is essential to ensure appropriate management of compli-cations. Skilled health personnel usually attend an average of 53% of births in the developing world. There is evidence of a strong relationship between the level of maternal mortality and the proportion of births attended by skilled health personnel, as clearly indicated in Figure 2.

There is, of course, a clear clinical justification for the presence of a skilled attendant during pregnancy and childbirth, as this may reduce both the incidence of complications of pregnancy and births. As shown in Figure 2 (as of data from 1998 estimates), in DPR Korea where skilled health personnel (doctors and nurses) attended all deliveries either at home, community health clinics or hospitals, the estimated MM Ratio was 35 per 100 000 live births. In contrast, Nepal with 9% of births attended by skilled health personnel (mostly auxiliary midwifery personnel), and usually at home, had a ratio of 830. The estimated MM Ratio for Maldives was 390, where trained health staff (mostly midwifery personnel including trained traditional birth attendants) provided 90% of deliveries. This paradoxical situation needs further investigation.

Empirical studies in many parts of the world indicate that the majority of maternal deaths occur due to five main causes:

     Haemorrhages (25%);

     Sepsis (15%);

     Unsafe abortion (13%);

     Hypertensive disorders of preg-nancy and eclampsia (12%);

     Obstructive labour (8%),

     Other direct causes (8%), such as ectopic and molar pregnancies, anaesthetic complications, cere-brovascular accidents, and embo-lisms, and

     Balance (around 20%) due to other indirect obstetric causes like anaemia, malaria, tuberculosis, jaundice, and heart disease, which are aggravated during pregnancy or delivery.

Prolonged and obstructed deliveries are more common in humans than in other primates, because the birth canal of a woman is not straight and wide, but varies in width. If obstructed labour cannot be overcome by manipulation or instrumental delivery, Caesarean delivery is needed. Neglected obstructed labour is a major cause of both maternal and newborn mortality and morbidity. Estimates from surveys showed that almost one-sixth to one-quarter of all women develop complications of pregnancy and childbirth serious enough to require rapid and skilled intervention if the woman is to survive without lifelong disabilities.

Maternal Morbidity

Information on maternal morbidity is very scanty in the countries of the Region, mainly due to inconsistent and unclear definition on what constitutes maternal morbidity. In general, maternal morbidity is categorized into three types: (a) direct obstetric morbidity resulting from complications of pregnancy during antenatal, natal and postnatal periods; (b) indirect obstetric morbidity resulting from conditions and diseases aggravated during pregnancy like anaemia, malaria, jaundice, tuberculosis, hypertensive disorders and other cardiovascular diseases or a combination of conditions/diseases; and (c) psychological obstetric morbidity that includes postpartum psychosis or depression and other mental health problems related to pregnancy and childbirth.

During pregnancy, immunity is reduced for various reasons and there is a greater risk of infections than during the non-pregnant period. Severe anaemia in pregnancy is a major obstetric problem in areas with high prevalence of malaria and/or other soil-transmitted helminthic infection. Viral hepa-titis is another disease that pregnant women are susceptible to, and also more likely to die due to infulminating form. The management of malaria in pregnant mothers is more difficult as most of the anti-malarial drugs like sulphonamides, pyrimethamine, and mefloquine are contraindicated during pregnancy, especially during the first trimester, to avoid teratogenic effects. Similarly, pregnant mothers are more prone to infections, especially tuberculosis, viral hepatitis, HIV/AIDS and other sexually transmitted diseases.

Infections due to sexually transmitted diseases (STD) during pregnancy have been a major issue, as the diseases affect both the mother and the unborn child. Many babies have been born with diseases infected while they are in their mothers' wombs or just after delivery. HIV infection is on the rise in the countries of the Region and HIV-infected mothers transfer the virus to the child, mostly at the time of delivery and breastfeeding. Voluntary testing and counselling during the antenatal period for HIV and other STDs would reduce the disease burden of both mothers and children. Short-course ARV therapy in the last trimester of pregnancy has proved an effective measure to protect mother-to-child-transmission.

Hypertensive disorders of pregnancy and eclampsia are easily diagnosed by measuring blood pressure regularly during the pregnancy, urine testing and clinical observation of seizures or convulsions. Eclampsia is more common among the young women during the first pregnancy, or those with previous abortions, molar pregnancy and family history of high-blood pressure. Adequate antenatal monitoring and timely referral for signs and symptoms of eclampsia could result in a higher percentage (around 65%) of reduction of maternal mortality due to eclampsia.

Low stature of women alone is not responsible for obstructed labour, but mal-position and mal-presentation of foetus, maternal pelvis abnormality and prolonged labour are also other causes. Thus, in countries where women live far from health care facilities that can handle obstetric emergencies, they are advised to await the delivery nearer such facilities.

Many mothers suffer from infections of the reproductive tract due to unclean environment and inappropriate care during delivery, which, if not properly managed, can lead to chronic complications and consequences like infertility. Puerperal sepsis or childbed fever is a common occurrence, greatly feared by women. With the wide availability of antibiotics, many lives have been saved, mostly those with access to appropriate health care.

Uterine infections and their compli-cations are responsible for the majority of deaths from puerperal infection. Unsafe deliveries, delivery by an unskilled birth attendant, and/or the traditional practice of inserting foreign objects and substances during delivery, or lack of appropriate postnatal care are major causes of sepsis and death. Though the actual incidence of puerperal infections in developing countries is not known, the number may be vast with estimates ranging from 5-15 percent. A few mothers getting infections during the post-partum period may die, but most of them suffer complications such as ectopic pregnancy, chronic or acute pelvic inflammatory disease (PID), chronic pelvic pain and low backache, and infertility.

As 61 percent of maternal deaths occur after delivery, appropriate and effective postpartum care is an important element in reducing maternal deaths. Immediate care during the first 4-6 hours after delivery is required to review any postpartum haemor-rhage due to uterine atony or retained products of conception. From the limited hospital statistics available, the gynae-cological wards of hospitals in developing countries are usually seen to be filled with patients suffering from urine incontinence, recto/vesico-vaginal fistulae, prolapsed vagina and uterus, vaginal discharge and lower abdominal discomfort, which are chronic direct obstetric morbidity conditions of women resulting from complications of childbirth. Such diseases create tremendous psychological trauma and social disability. Many women avoid medical intervention due to lack of self-confidence and embar-rassment. It is estimated that maternal morbidity is 4-8 times higher than maternal deaths.

A retrospective study carried out in 1996 in Southern India using a sample of 3 600 women with at least one child below 5 years, found that 10% reported a potentially life-threatening condition during pregnancy, while 23% had at least one postpartum complication which included excessive bleed-ing, fever, discharge and/or lower abdominal pain.

Infant Deaths

Poor maternal health not only affects the mother but also the infant. There are pronounced intergenerational effects. The same causal factors responsible for maternal mortality and morbidity also affect the infants they are carrying. Of nearly 8 million infant deaths each year, around two-thirds occur during the neonatal period (before the baby is one-month old). Every year there are 5 million neonatal deaths, of which 3.4 million die within the first week of life. The causes of early neonatal deaths are largely the con-sequences of inadequate or inappropriate care during pregnancy, delivery or the first critical hours after birth.

Pre-term birth (delivery of the baby before reaching the full-term pregnancy) is one of the main causes of perinatal mortality. Another major cause is birth asphyxia and trauma. Other causes of perinatal mortality include poor maternal health due to malnutrition and infections during pregnancy, inadequate and inappropriate management of complications during pregnancy and delivery, poor hygiene during delivery and the first critical hours after birth and lack of newborn care.

It is indeed tragic that when a mother dies either during or just after delivery, her children are also more likely to die. A study in Bangladesh found that if a woman dies after childbirth, her newborn baby is almost certain to die. Another study, also in Bangladesh, found that children up to 10 years whose mothers die are 3 to 10 times more likely to die within two years than children with living mothers. Nevertheless, a significant improvement in child survival has been observed in all countries (see Figure 3). The estimated infant mortality rate, in 1998, ranged from 18 per 1 000 live births in Sri Lanka to 83 in Nepal.

Many studies have shown that the major proportion of infant deaths occur in the early and late neonatal stages. In most countries, information on stillbirths and infant deaths in the perinatal and neonatal period is often not properly recorded and/or reported.

According to a global estimate of perinatal and neonatal mortality for 1999 by

WHO, there were 6.9 million perinatal deaths, giving a rate of 52 deaths per 1 000 total births. Neonatal deaths for the same period were around 4 million (with a neonatal death rate of 31 per 1 000 live births). Least developed countries accounted for 98% of perinatal and neonatal deaths. The largest number of these infant deaths took place in Asia, where almost 60% of all births occurred.

The figures given in Table 1 below are estimates and provide an indication of the problem. In the SEA Region, perinatal and neonatal death rates were estimated to be around 67 and 42 per 1 000 live births respectively in 1999. Estimated perinatal mortality rates in SEAR countries ranged from 85 in Myanmar to 20 in Thailand per 1 000 live births, while estimated neonatal mortality rates ranged from 60 in Myanmar to 10 in Thailand per 1 000 live births. Only three countries in the Region (Indonesia, Sri Lanka and Thailand) had perinatal and neonatal mortality rates below the global average.

Table 1.Estimated perinatal and neonatal mortality rates and numbers by WHO Regions, 1999

Source: WHO Geneva, Department of Reproductive Health and Research, Perinatal and Neonatal mortality: Global, Regional and Country estimates, second edition, Draft 5, November 2001

Many of these deaths can be prevented with simple maternal and child care interventions, like resuscitation of newborns, immediate breastfeeding, keeping the baby warm, cleanliness and hygienic practices during delivery, and early detection and management of newborn illnesses, including use of appropriate antibiotics. The availability of a skilled attendant at birth could not only reduce maternal morbidity and mortality, but also be able to help ensure the survival of infants during the perinatal and neonatal periods.