Notes
Maternal Deaths in India: An Exploration
The latest estimate of the maternal mortality ratio for India stands at 301 for every 1,00,000 live births. Though there has been a decline, reaching the set goal of 109 by 2015 remains a challenge. Maternal mortality is strongly associated with rural areas and low standards of living. For these women, access to quality healthcare services is difficult. In general, efficient antenatal services would help in identifying high-risk women
and providing referrals.
ANJALI RADKAR, SULABHA PARASURAMAN
D
After the ICPD in Cairo, India adopted the reproductive and child health programme at the national level. Maternal health is a problem of serious proportions in India, where an estimated 1,36,000 women die needlessly each year from causes related to pregnancy, childbirth and abortion. The three countries, India, Pakistan and Bangladesh account for 28 per cent of the world’s births and 46 per cent of its maternal deaths [Motashaw 1997].
One needs to understand the difficulties in calculating mortality indices in developing countries because of the scarcity of data. Sometimes the quality of available data is questionable. Though registration of births and deaths is compulsory, many a time a maternal death is not reported at all or wrongly classified [Motashaw 1997]. No one precisely knows how many women die as a result of pregnancy and childbirth each year. Many developing countries cannot even provide these statistics and in many countries less than half the maternal deaths that occur are actually reported [WHO 1991].
The maternal mortality scenario in India from various sources for the last 25 years has been presented in Table 1. A recent study conducted by the registrar general of India found that the maternal mortality ratio (MMR) from the 1997-98 retrospective survey is 398, from 1999-2000 survey it is 327 and from 2001-03 survey it stands at 301 [SRS 2006], which still is a far cry from 200, the MDG goal for 2007. During the period of 25 years, the MMR has come down from 753 to 301, a decline of 60 per cent. Though the decline seems to be large, we still have a long way to go.
Methodology
In the context of high levels of maternal mortality, very little information about the deceased women is available in India. One of the goals of all health and population policies aims at reducing maternal mortality. In order to take any specific action for its reduction, more information about the associated causes would help. The aim of this study is to explore more the background characteristics of the deceased women and the manner of death. An attempt is made also to compile symptoms associated with maternal deaths.
Information has been collected on the basis of a large sample in the Reproductive and Child Health Survey-2 (RCH-2, ministry of health and family welfare, 2004) during the period 2002-04 for the entire country. Maternal deaths have been recorded for two years preceding the survey. Information has been collected from a responsible adult in the family. Considering the need to explore more the nature of maternal deaths, information has been collected about the symptoms the woman had at the time of her death. It is not possible here to precisely classify the deaths by medical reasons as in various other hospital-based studies.
Results
Maternal death has been defined as the death that results from complications of pregnancy, childbirth or puerperium. Deaths due to puerperal causes are deaths that occur to mothers within 42 days of delivery due to causes associated with
Table 1: Maternal Mortality Ratio in India,1980-99
| Period | Maternal | Source |
|---|---|---|
| Mortality Ratio | ||
| 1980-82 | 753 | Bhaskar Rao |
| 1990 | 570 | WHO/UNICEF |
| 1992-94 | 572 | FOGSI |
| 1992-93 | 437 | NFHS-I |
| 1997 | 408 | SRS |
| 1998 | 407 | SRS |
| 1998-99 | 540 | NFHS-II |
Source: Rawal (2003).
Economic and Political Weekly August 4, 2007
Table 2: Distribution of Maternal Deaths by Selected Background Characteristics
| Background Characteristics | Reported Maternal Deaths | Total Women | |
|---|---|---|---|
| Number | Per- | Per | |
| centage centage |
Place of residence Rural 478 78.2 68.2 Urban 133 21.8 31.8
Religion Hindu 515 84.3 82.4 Muslim 62 10.1 12.2 Christian 24 3.9 2.3 Other 10 1.6 3.1
Caste Scheduled caste 163 26.7 18.9 Scheduled tribe 102 16.7 8.8 Other backward class 214 35 40.3 Other 119 19.5 30.8 Do not know 13 2.1 1.1
Standard of living index Low 379 62 43.3 Medium 151 24.7 31.5 High 81 13.3 25.3
Type of house
Kachcha 302 49.4 32.4 Semi-pucca 190 31.1 34.6 Pucca 119 19.5 32.9
Source of drinking water Inside the house 96 15.7 22.4 Outside the house 515 84.3 77.6
Type of toilet facility Own toilet 161 26.3 37.4 Shared toilet 11 1.8 5.4 No toilet 439 71.8 57.2
Total 611 100 100
Table 3: Distribution of Maternal Deaths by Selected DemographicCharacteristics
Demographic Reported Total Characteristics Maternal Deaths Women Number Per-Percentage centage
Age at death 15-19 111 18.2 8.8 20-24 195 31.9 20.7 25-29 161 26.4 21.8 30-34 84 13.7 19.5 35-39 35 5.7 16.3 40-44 25 4.1 12.8
Parity0 72 13.5 11.6 1 180 33.7 16.1 2 104 19.5 25 3 or more 178 33.3 47.3 Missing 77 --
Month of pregnancyFirst trimester 132 24.9 Second trimester 70 13.2 Third trimester 328 61.9 Not pregnant when dead 31 -Not reported 50 --
Total 611 100 100
Table 4: Distribution of Maternal Deaths by Timing of Death
Timing of Pregnancy Number Percentage
Was pregnant when dead 247 40.4
Died within six weeks of
abortion or at the time
of abortion 61 10.0 Died during childbirth 94 15.4 Died within six weeks of
childbirth or at the end
of pregnancy 209 34.2 Total 611 100.0
35 30 25 20 15 10 5 0
Per cent
pregnancy and childbirth. No attempt has been made here to compute either the maternal mortality ratio or the maternal mortality rate because the data does not allow us to do so directly.
In the sample of 6,20,107 households in India, 4,052 deaths occurred to women in the reproductive ages, out of this 611 deaths (15.1 per cent) were maternal deaths. The registrar general of India study shows that maternal deaths are not uniformly distributed over the states. Among the total maternal deaths, twothirds of the deaths are from empowered action group states namely – Bihar, Jharkhand, Madhya Pradesh, Chhattisgarh, Rajasthan, Uttar Pradesh, Uttaranchal and Orissa and Assam [SRS 2006]. The geographic distribution of maternal deaths in RCH-2, indicates that 67.8 per cent of deaths are from these regions, almost matching to SRS estimates.
The distribution of maternal deaths by selected background characteristics has been presented in Table 2 to identify vulnerable groups. The percentage of total women by background characteristics has also been presented here to understand the differentials in maternal deaths by background characteristics better. It is clear from Table 2 that the extent of maternal deaths is more among rural, backward women with low standards of living. Among those who died, 84 per cent fetched water from outside the house and 72 per cent had no toilets. Only 20 percent of them stayed in pucca houses. These facts are associated with poverty and rural residence. A similar finding has been recorded by Anandlakshmy et al (1993). The differentials in maternal deaths are observed by caste and standard
Chart: Maternal Deaths by Age at Death
15-19 20-24 25-29 30-34 35-39 40-44 Age at Death
of living of women along with the place of residence, indicating that lower socio-economic groups experienced maternal mortality more.
The data available on the demographic characteristics of the deceased women is limited. However, it is available for two important correlates of maternal health – age and parity.
The curve of maternal deaths by age looks exactly like the fertility curve. Maternal mortality is high when fertility is high. Most (32 per cent) maternal deaths are in the age group of 20-24 years, followed by 26 per cent in 25-29 years (see the chart). Fertility also is highest among women in the 20 to 29 age group. However in reality the risk of dying of pregnancy and childbirth related causes is more among the very young and old age groups [WHO 1991].
An effort has been made here to understand whether the number of children born has any relation to maternal deaths. Fourteen per cent of the deaths have occurred to women who do not have a single child. Majority of the deaths have been reported for women who have one child or have three or more. It does not show any specific pattern. Maternal deaths are actually seen more in cases of women with no children or for higher order births. Looking at the month of pregnancy at the time of death, most of them (62 per cent) died when they were in the third trimester. About a quarter died during the first trimester whereas 13 per cent died in the second trimester. It shows that in most of the cases complications occurred at the end of the term, i e, in spite of carrying a pregnancy till almost full term they could not sail through safely.
Economic and Political Weekly August 4, 2007
As has been mentioned above, maternal the end of the pregnancy or within six deaths can be classified into four cate-weeks of childbirth (34 per cent). The gories such as, (i) woman was pregnant deaths associated with abortion are 10 per when she died, (ii) died within six weeks cent and deaths during childbirth are 15 of abortion or at the time of abortion, per cent. The data for India shows that
(iii) died during childbirth, and (iv) died maternal mortality due to abortions is 8 per within six weeks of childbirth or at the end cent [SRS 2006]. Those who died within of pregnancy. six weeks of childbirth or at the end of
Table 4 shows that most maternal deaths pregnancy might have sustained injury at occurred when the woman was pregnant the time of childbirth or could have suc(40 per cent) or deaths have occurred at cumbed to a fatal infection. The share of
Table 5: Percentage Distribution of Maternal Deaths by Timing of Death and SelectedBackground/Demographic Characteristics
Background/Demographic Was Pregnant Died within Six Died during Died within Six Total Characteristics When Weeks of Abortion Childbirth Weeks of She Died or at the Time Childbirth or at the of Abortion End of Pregnancy
Place of residence Rural 39.1 8.4 17.4 35.1 478 Urban 45.1 15.8 8.3 8.3 133
Caste SC, ST, backward castes 42.6 8.4 15.6 33.4 479 Other than Backward castes 32.8 16 13.4 37.8 119 Don’t know 30.8 15.4 23.1 30.8 13
Standard of living index Low 44.3 7.9 14.2 33.5 379 Medium 35.8 11.2 17.2 35.8 151 High 30.9 17.3 17.3 34.6 81
Age at death 15-19 42.3 16.2 16.2 25.2 111 20-24 35.9 10.2 16.4 37.4 195 25-29 44.7 6.8 12.4 36 161 30-34 44 8.3 15.5 32.1 84 35-39 31.4 8.6 25.7 34.3 35 40-44 41.7 8.3 8.3 41.7 24
Not reported ---100 1 Parity
Total 40.4 10 15.4 34.2 611
the postnatal deaths is high in India, which needs to be restricted to reduce maternal mortality.
It has been seen that overall the differentials have been observed by place of residence, caste and standard of living. Thus to explore further, distribution of deaths has also been studied by timing of death.
Everywhere except in case of abortions, maternal deaths have been seen more for rural and backward caste women than for urban and advanced caste women. Once again it is reflected that abortion is an urban phenomenon, where access to medical facilities is easier. Urban maternal mortality associated with abortions is double compared to that of rural. The same is true in the case of backward and other than backward castes. As regards standard of living, when it increases, deaths due to abortion also increase.
No trend has been observed in the timing of the death by age when the woman died while pregnant and when the woman died during childbirth. In case of women who died within six weeks of childbirth or at the end of pregnancy, chances of death increase with age. Similarly, a steadily increasing trend has been observed in case of deaths associated with abortions. Deaths due to abortions are sizeably more in cases of young women. Similarly, no trend has been seen by parity. Thus one can conclude that biological characteristics of women on which the data is available do not indicate any particular patterns of maternal deaths, though they are known to contribute to associated morbidity.
Symptoms
Table 6: Percentage Distribution of Symptoms by the Timing of Maternal Death
Over 80 per cent of maternal deaths in
Symptoms Observed Was Pregnant Died within Died during Died within when Dead Six Weeks of Childbirth Six Weeks of India, as elsewhere in the world are due Abortion or Time Childbirth or at the to six medical causes, hemorrhage,of Abortion End of Pregnancy
eclampsia, obstructed labour, sepsis, com-Swelling of hands and feet 71 (28.7) 18 (29.5) 26 (27.6) 46 (22.0) plications arising due to unsafe abortion Paleness 77 (31.2) 16 (26.2) 29 (30.8) 42 (20.1) Visual disturbances 44 (17.8) 7 (11.5) 12 (12.8) 23 (11.0)
Table 7: Per Cent Distribution of Causes
Bleeding 43 (17.4) 11 (18.0) 24 (25.5) 26 (12.4)
of Maternal Deaths by Three
Convulsions 38 (15.4) 8 (13.1) 20 (21.3) 24 (11.5)
Different Sources
Weak or no movement of foetus 36 (14.6) 7 (11.5) 11 (11.7) 16 (7.6) Abnormal position of foetus 27 (10.9) 5 (8.2) 11 (11.7) 12 (5.7)
Symptoms SRS, Arora, SRS, Premature labour --20 (21.3) 18 (8.6)
2006 2005 1998
Prolonged labour --24 (25.5) 23 (11.0) Breech presentation --5 (5.3) 12 (5.7) Haemorrhage 38 25.6 29.6 High fever -15 (24.6) -41 (19.6) Sepsis 11 13 16.1 Lower abdominal pain -17 (27.9) -43 (20.6) Hypertensive disorders 5 --Foul smelling discharge -4 (6.6) -17 (8.1) Obstructed labour 5 6.2 9.5 Excessive bleeding -12 (19.7) -49 (23.4) Abortion 8 8 8.9 Dizziness -17 (27.9) -40 (19.1) Toxemia -11.9 -Severe headache -14 (23.0) -42 (20.1) Eclampsia --8.3 Total 247 61 94 209 Anaemia --19.0
Other conditions 34 35.3 -Note: * Percentage may not add to 100 because of multiple symptoms reported.
Economic and Political Weekly August 4, 2007 and pre-existing conditions such as anaemia or malaria. Information in RCH-2 has also been collected about the symptoms at the time of maternal death. The distribution of the symptoms has been presented by the timing of maternal death in Table 6. It gives the details as reported in the survey. As expected, multiple symptoms have been reported for one deceased woman.
An attempt has been made in Table 6 to present the percentage distribution by causes of maternal mortality, as in other studies. But considering the fact that symptoms have been reported by nonmedical personnel, there is a possibility that some symptoms might not have been reported.
From the list of symptoms available one can broadly classify the problems as anaemia (114); swelling of hands and feet and paleness, toxemia (10); swelling on hands and feet, visual disturbances, convulsions, dizziness and severe headache, haemorrhage as bleeding, excessive bleeding, sepsis (15); foul smelling discharge and high fever and obstructed labour (2); abnormal position of the foetus, prolonged labour, breech presentation and lower abdominal pain. These numbers have been calculated by clubbing the listed symptoms together just to get crude estimates and are on the lower side, considering the respondent’s knowledge.
Just to get the comparative picture, the estimates from other studies conducted in India have been compiled in Table 7.
There seems to be a large-scale variation in the distribution of causes of maternal death. Still, haemorrhage is the most reported cause. Already anaemic women experience heavy bleeding causing death. The other causes relate to sepsis, toxemia and abortion. The share of obstructed labour has gone down sizeably over the period of eight years, from 1998 to 2006, indicating possible improvement in service delivery.
Table 8 shows that place of delivery or type of service provider does not matter when it comes to maternal death. The death is because of a problem which may not be in the hands of medical personnel. This data does not support the view that place of delivery, public or private, matters. Maternal deaths have been reported to be more if deliveries are conducted at home. Arora (2005) also reports that MMR would be high if the deliveries are conducted at home by untrained persons. A large proportion of home deliveries is conducted by untrained personnel, so it is difficult to address this mortality. However, at the institutional level, in some cases, the associated factors would be poor care provided or poorly developed infrastructure, under-staffed public health facility that is unable to provide right type of care [Rawal 2003].
Discussion
The overall maternal mortality ratio for India was about 400 in 1997-98, which has gone down to 301 in 2001-03, registering a decline of 24 per cent during this period. However, the time bound target of 200 maternal deaths by 2007 and 109 deaths by 2015 is a challenge.
The ratio of maternal deaths to live births measures the risk of dying as a result of pregnancy. In case of India, with an MMR of 301, a woman’s chance of dying each time she becomes pregnant is about 1 in
330. If she becomes pregnant three times, her lifetime risk is 1 in 110.
Looking at a risk, it is necessary to understand the factors responsible for maternal mortality so as to plan efficient interventions that would help reducing mortality [Anandlakshmy et al 1993]. Maternal deaths assume importance, not just for health reasons but also when the woman dies there are significant social and economic losses. Children who lose their mothers suffer the most. The risk of death for children under five years increases if the mother dies.
It has become clear that women residing in rural areas, belonging to backward castes and having a low standard of living are the most vulnerable in terms of maternal mortality. A lesser number of maternal deaths in the urban areas reflects the easier access of the city-dwellers to medical services [WHO 1991]. A large number of women need to fetch water from outside and need to go out to the toilet, excessive physical activity is a precipitating factor for maternal mortality; especially when the women’s nutritional status is poor. This data set does not support the view that biological characteristics of the deceased woman at large have a role, though it is known that the risk of dying is not spread evenly for all pregnancies. The age of the woman and the number of pregnancies have a bearing on the outcome. Women who become pregnant in their teens are at greater risk of complications such as obstructed labour and death. These are young women who have also married early. Physically they are not fully capable of bearing a child. Similarly, older women also face an increased risk of death than those in prime childbearing ages [WHO 1991]. It is seen here that 18 per cent of the total maternal deaths belong to the age
REVIEW OF AGRICULTURE June 30, 2007
Growth Crisis in Agriculture: Severity and Options at National and State Levels – Ramesh Chand, S S Raju, L M Pandey
Modern Agriculture and the Ecologically Handicapped:
| Fading Glory of Boro Paddy Cultivation in West Bengal | – Sreerupa Ray, Dhrubajyoti Ghosh |
|---|---|
| Implications of Alternative Institutional Arrangements in Groundwater Sharing: Evidence from West Bengal | – Aditi Mukherji |
| Agricultural Credit in the Post-Reform Era: A Target of Systematic Policy Coarctation | – P Satish |
| Turnaround in Financial Recovery in Maharashtra’s |
Irrigation Sector – A Narayanamoorthy
Impact of Declining Groundwater Levels on Acreage Allocation in Haryana – Priya Bhalla
For copies write to Circulation Manager
Economic and Political Weekly
Hitkari House, 284, Shahid Bhagatsingh Road, Mumbai 400 001 email: circulation@epw.org.in
Economic and Political Weekly August 4, 2007
group 15 to 19 years. As far as parity is concerned, studies show that whatever is the age of the mother, the second and third births are the most trouble free and the risk of serious complications and deaths increases steadily thereafter [WHO 1991]. Therefore child bearing and child spacing remains a major gray area in our society [Arora 2005].
Anandlakshmy et al (1993) identified the following risk factors in order of degree of risk. Severe anaemia assessed by haemoglobin levels below 8.5 g per cent, presence of haemorrhage and toxemia as a complication of pregnancy, a short birth interval of less than two years, mother’s age below 20 and above 35 years and parity 0 or 5 or more. Thus once the associated factors are identified, to combat high maternal mortality, there is a strong need for effective interventions.
A lesser proportion of women belonging to the lower socio-economic strata go through abortion and thus the risk of dying due to abortion also is less. If there are clandestine, unsafe abortions, information on maternal deaths also would be concealed. Whatever the case, abortions are known to pose major risks in terms of maternal health.
Haemorrhages – both during the pregnancy and after delivery – are the commonest cause of maternal death. The risk is aggravated with anaemia which is widely prevalent in India. Infection or sepsis denotes the entry of germs into the genital tract through the use of unwashed hands and unsterilised instruments during delivery. It is a frequent complication in case of illegal abortions or in case of women who remain undelivered many hours after their membranes have ruptured. It is the first cause of death to recede when maternal mortality falls. Reduction of deaths from sepsis generally reflects an increase in standards of delivery care and lower case fatality resulting from wider use of antibiotics [WHO 1991]. Apart from hypertensive disorders, often seen among young women pregnant for first time, obstructed labour needs attention.
In India about half the women in the reproductive age group and two thirds of pregnant women are anaemic. It is difficult for such women to sustain the haemorrhages, either intra-partum or postpartum. Similarly, because of anaemia overall immunity goes down and these women are more susceptible to infections
– when there are home deliveries conducted in not-so-hygienic manner – and cannot recover speedily. Anaemia among the women in the reproductive age group and issues related to their social status also have to be addressed on the backdrop of maternal mortality.
Maternal mortality is strongly associated with the services provided. Proper antenatal, natal and postnatal care would certainly lower the number of maternal deaths significantly. These services should be more accessible to all those who are poor and residing in remote areas, especially where the services do not percolate or are provided inefficiently. Though place of delivery/abortion has no significant role in case of maternal death, antenatal checkups would definitely identify the problem on time and suggest referrals, if required. This would help save the life of mother and the child. The major problems need to be addressed: an absence of links between communities, subcentre and referral facilities, shortages of equipment and trained staff at referral facilities and a lack of emergency transport [Arora 2005].
The immediate risks of maternal death begin with the stage of conception itself. The major complications that cause maternal death may be present initially in a mild form, progressing to severe
Table 8: Per Cent Distribution of Maternal Deaths by Timing of Death and Place of Delivery/Abortion
| Place of Delivery/ | Died within | Died during | Died within | Total |
| Abortion | Six Weeks of | Childbirth | Six Weeks of | |
| Abortion or at the | Childbirth or | |||
| Time of Abortion | End of Pregnancy | |||
| Government | 18.5 | 22.2 | 59.3 | 54 |
| Private | 22.2 | 18.5 | 59.3 | 54 |
| Government ISM | - | 33.3 | 66.7 | 3 |
| Private ISM | - | - | 100.0 | 1 |
| Home | 8.3 | 31.9 | 59.7 | 144 |
| Other | 17.8 | 28.6 | 53.6 | 28 |
| Not reported | 27.5 | 21.2 | 51.3 | 80 |
| Not applicable | - | - | - | 247 |
| Total | 61 | 94 | 209 | 611 |
complications in some 12-15 per cent of all pregnancies. Interventions can stop this progression at several points:
Historical evidence from developed countries suggests that a significant decline in maternal mortality coincided with the use of antibiotics, blood transfusion and management of hypertensive disorders of pregnancy [WHO 1991]. Many developing countries still lack these interventions during pregnancy and childbirth.
Undoubtedly a major factor which determines the pregnant women’s risk of death is the lack of access to well-equipped healthcare services [WHO 1991]. The interventions to prevent maternal deaths have to occur immediately after the onset of early symptoms such as bleeding during pregnancy, non-progression of labour, fever, etc, because the estimated average interval from onset to death for major obstetric complications is very short, meaning that interventions have to occur with a sense of urgency that is almost rare in our public health system.
EPWEmail: aradkar@unipune.ernet.in sulabhap@rediffmail.com
References
Anandlakshmy, P N, P P Talwar, K Buckshee and V Hingorani (1993): ‘Demographic, Socioeconomic and Medical Factors Affecting Maternal Mortality – An India Experience’, The Journal of Family Welfare, 39(3), September.
Arora, Punita (2005): ‘Maternal Mortality: Indian Scenario’, MJAFI, 61(3).
Motashaw, Nergesh D (1997): ‘Root Causes of Maternal Mortality: Infancy to Motherhood’, Journal of Family Welfare, 43(2), June.
Rawal, Asha (2003): ‘Trends in Maternal Mortality and Some Policy Concerns’, Indian Journal of Community Medicine, 28(1), January-March.
SRS (2006): ‘Maternal Mortality in India: 19972003, Trends, Causes and Risk Factors’, Registrar General of India, New Delhi.
The Hindu (2005): ‘India’s Maternal Mortality Ratio Highest in South Asia’, December 28.
Tsui, Amy O, Judith N Wasserheit and John G Hagga (eds) (1997): Reproductive Health in Developing Countries, Expanding Dimensions, Building Solutions, National Academy Press, Washington DC.
WHO (1991): ‘Maternal Mortality: A Global Factbook’, WHO, Geneva.
Economic and Political Weekly August 4, 2007
