- Email: [email protected]
Antenatal care: provision and inequality in rural north India
ARTICLE IN PRESS
Social Science & Medicine 59 (2004) 1147–1158
Antenatal care: provision and inequality in rural north India Saseendran Pallikadavath, Mary Foss, R William Stones* ‘Opportunities and Choices’ Programme of Reproductive Health Research and School of Nursing and Midwifery, University of Southampton, Southampton SO16 5YA, UK
Abstract The objectives of this paper are to examine factors associated with use of antenatal care in rural areas of north India, to investigate access to specific critical components of care and to study differences in the pattern of services received via health facilities versus home visits. We used the 1998–1999 Indian National Family Health Survey of ever-married women in the reproductive age group and analysed data from the states of Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh (n ¼ 11; 369). Overall, about three-fifths of rural women did not receive any antenatal check-up during their last pregnancy. Services actually received were predominantly provision of tetanus toxoid vaccination and supply of iron and folic acid tablets. Only about 13% of pregnant women had their blood pressure checked and a blood test done at least once. Women visited by health workers received fewer services compared to women who visited a health facility. Home visits were biased towards households with a better standard of living. There was significant under-utilisation of nurse/midwives in the provision of antenatal services and doctors were often the lead providers. The average number of antenatal visits reported in this study was 2.4 and most visits were in the second trimester. Higher social and economic status was associated with increased chances of receiving an antenatal check-up, and of receiving specific components including blood pressure measurement, a blood test and urine testing but not the obstetric physical examination, which was however linked to ever-use of family planning and the education of women and their husbands. Thus, pregnant women from poor and uneducated backgrounds with at least one child were the least likely to receive antenatal check-ups and services in the four large north Indian states. Basic antenatal care components are effective means to prevent a range of pregnancy complications and reduce maternal mortality. The findings indicate substantial limitations of the health services in overcoming socio-economic and cultural barriers to access. r 2004 Elsevier Ltd. All rights reserved. Keywords: Antenatal care; Health services; Socio-economic status; India
Introduction Antenatal care Reducing maternal mortality and morbidity has been a major focus for the developing world since the launch of the Safe Motherhood Initiative in 1987 (WHO, 1996). Over recent years there has been some debate as to the
*Corresponding author. Level F (815), Princess Anne Hospital, Southampton SO16 5YA, UK. Tel.: +44-23-80796033; fax: +44-23-8078-6933. E-mail address: [email protected] (R.W. Stones).
effectiveness of one aspect of maternity provision, that of antenatal care (Carroli, Rooney, & Villar, 2001; Bergsj^, 2001). The main purposes of antenatal care are to prevent certain complications, such as anaemia, and identify women with established pregnancy complications for treatment or transfer. Tetanus toxoid vaccination of the mother prevents neonatal tetanus as protective antibodies are passed across the placenta. It was hoped that women ‘at risk’ of pregnancy complications could be identified so as to avert problems, and scoring systems were developed for this purpose. However, the use of risk assessment has turned out to be unproductive as women may develop complications in pregnancy or childbirth at any time.
0277-9536/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.socscimed.2003.11.045
ARTICLE IN PRESS 1148
S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
Despite the limitations of antenatal risk assessment the basic functions of detection of pre-eclampsia, anaemia and other incipient complications remain essential. In addition, other less tangible benefits may be realised which are not easily evaluated in isolation. Antenatal consultations provide opportunities for health education, health promotion and social support at both the individual and community level. Especially in the rural setting, accessing antenatal care is an important step in bringing women into contact with the health care system. This contact has facilitated women’s access to medical care for future health needs, including postnatal care (Sugathan, Mishra, & Retherford, 2001). Within a context of limited resources there is a need to ensure value for money: overall, antenatal care is considered to represent a cost-effective component of maternity services as part of ‘safe motherhood’ interventions to reduce mortality and morbidity (Jowett, 2000). The Indian context Overall, uptake of antenatal care in India has been low. The 1992–1993 Indian National Family Health Survey showed that only 64% mothers received antenatal check-ups and this increased marginally to 65% in 1998–1999 (International Institute for Population Sciences and ORC Macro, 2000). This lack of improvement occurred despite governmental and non-governmental efforts to strengthen service delivery, and is likely to contribute to the continuing high maternal mortality in the country, especially in the northern states. Maternal mortality in the four states Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh ranged between 450 and 700 per 100,000 live births (International Institute for Population Sciences and ORC Macro, 2000). These states have some of the highest fertility (total fertility rate) in the country: ranging between 3–4 children per woman. Similarly, infant and child mortality rate in these states were reported to be among the highest in India with a range 73–87 infant deaths per 1000 live births. In 2001 female literacy in these states was low. According to the 2001 Census, these states account for about 40% of India’s total population (Registrar General and Census Commissioner, 2001). Achieving significant improvements in maternal health in these states is vital to improving maternal health in India as a whole. In India, antenatal care initiatives began in 1951 with the implementation of the First Five Year Plan (1951–1956). However, a lack of rural health infrastructure, and a focus on family planning using a clinicbased approach limited the benefits to rural women. In the Third Five Year Plan (1961–1966) an extension approach to family planning was adopted with recruitment of auxiliary nurse midwives (ANMs) and Health Assistants. This provided rural women access to some
elements of antenatal care. However, it may be noted that these programmes were primarily geared to family planning service provision. During the Fifth Five Year Plan (1974–1979) maternal and child health services (MCH) were integrated with family planning services and a new programme entitled ‘Family Welfare’ was introduced. This gave impetus to provision of maternal health services in rural areas. In the Seventh Five Year Plan (1985–1990) a Universal Immunisation Programme (UIP) was implemented, greatly increasing pregnant women’s access to tetanus toxoid vaccination. In 1991–1992 a major initiative called ‘Child Survival and Safe Motherhood’ (CSSM) was undertaken by integrating the UIP with expanded MCH activities in states with high mortality rates. Following the recommendations of the International Conference on Population and Development in 1994, in 1997 the Government of India launched the Reproductive and Child Health (RCH) programme for implementation in the Ninth Five Year Plan (1997–2002). This was done by integrating CSSM interventions and adding interventions for reproductive tract infections and sexually transmitted diseases. The idea behind the RCH programme is to provide need based, client centred, demand driven, high quality integrated services to beneficiaries. All districts in the country were covered by the programme as of 1999–2000 (Ministry of Health and Family Welfare, 2001). It is important to note here that Integrated Child Development (ICDS) programmes introduced in 1975 in selected districts also provided women with access to antenatal care. In rural areas, ANMs posted at the sub-Primary Health Centres (sub-PHCs) are responsible for providing antenatal care either in homes or clinics. Sub-PHCs are the grass-roots level government health service points covering a population of 5000 in non-hilly areas. Sub-PHCs refer to Primary Health Centres (PHCs) manned by medical and other paramedical staff. Antenatal care services to pregnant women are also provided by the ‘Anganwadi’ centres working under the ICDS Programmes. Determinants of antenatal care In developing countries uptake of antenatal care is determined by socio-economic factors, demographic factors, and availability and access to health facilities. In many parts of the developing world women’s education was found to be an important predictor of antenatal care utilisation. In Jordan (Obermeyer & Potter, 1991), Nepal (Matsumura & Gubhaju, 2001), and Surabaya in Indonesia (Taguchi, Kawabata, Maruo, & dewata, 2003) higher levels of education of women were associated with greater use of antenatal care. However, in Kenya maternal education was not significantly related to utilisation of antenatal care once
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
other variables were considered (Magadi, Madise, & Rodrigues, 2000). However, the potential impact of husbands’ or partners’ educational status has not been explored. In India, a relationship between maternal education and utilisation of antenatal care is evident in many studies. In Karnataka state in India, Bhatia and Cleland (1995) reported that higher levels of maternal education increased the uptake of antenatal care. Similar finding was reported in a study of four Southern Indian states: Kerala, Tamil Nadu, Karnataka, and Andhra Pradesh (Navaneetham & Dharmalingam, 2002). Indeed a national study conduced by the Indian Medical Council Research (ICMR) in 90 districts of the country in 1999 concluded that literacy of women is the key factor to achieving improvements in antenatal care in India (Singh & Yadav, no date). Demographic factors such as order of pregnancy (or number of children), marital status, maternal age, and marital duration are reported having influence on the utilisation of antenatal care. Less use of antenatal care was found among women having larger numbers of children in Jordan (Obermeyer & Potter, 1991) and Kenya (Magadi et al., 2000). In India utilisation of antenatal care was low among women with higher order pregnancies: Chandrashekar et al. (1998) working in rural Karnataka noted that multiparous and unskilled mothers aged over 30 were less likely to have used antenatal services. Similar findings were reported in Tamil Nadu (Neilsen, Liljestrand, Thilsted, Joseph, & Hedegaard, 2001). Low antenatal care utilisation was reported among teenagers in Jamiaca (McCAW-Binns, Grenade, & Ashley, 1995) and Kenya (Magadi et al., 2000). Other relevant variables were unmarried status, unions of very short duration, desire for large families, never-use of family planning, unwanted pregnancies. In India, Bhatia and Cleland (1995) noted reduced antenatal care use among women aged under 18. In Karnataka, consulting for pregnancy confirmation and a prior history of neonatal or fetal loss were important determinants of subsequent antenatal care utilisation. The interplay of socio-economic factors and accessibility of health care facilities was highlighted in Kenya (Magadi et al., 2000): even if health care services are easily accessible socio-cultural factors at community level determine whether to use the facility. In India, using the 1992–1993 National Family and Health Survey, Misra, Roy, and Rajan Irudaya (1998) revealed that lack of knowledge about antenatal services was an important reason for non-utilisation in the northern states of India. A study in four southern Indian states found that variations in service use were primarily related to availability and access (Navaneetham & Dharmalingam, 2002). In India, religion and caste are two important social factors that influence social practices, including belief
1149
systems surrounding pregnancy. However, these do not follow a stereotypical pattern. For example, while Muslims were more likely to go for medical check-ups n Karnataka they were less likely to do so in Kerala (Bhatia & Cleland, 1995, Navaneetham & Dharmalingam, 2002). Recognising that socio-cultural factors are likely to be critical determinants of care seeking and service utilisation in the context of pregnancy, the objective of this paper is to examine factors associated with use of antenatal care facilities in the rural areas of Bihar, Madhya Pradesh, Rajasthan and Uttar Pradesh (the latter including the newly created state of Uttaranchal). The paper also examines factors associated with access to specific critical components of care. A final objective is to study the differences, if any, in the pattern of antenatal check-ups and services received through health facilities versus home visits.
Data and methods Data from the 1998–1999 National Family Health Survey (NFHS-2) were utilized in the present study. This survey was undertaken by the International Institute for Population Sciences, Mumbai on behalf of the Government of India, Ministry of Health and Family Welfare. This was a nationally representative population based sample survey of 90,303 ever-married women aged 15–49 from 26 states comprising 99% of India’s population. The survey methods are described in detail together with univariate tabulations in an overview report (International Institute for Population Sciences and ORC Macro, 2000). Briefly, the sample size of the survey was set for each state considering the size of the state, resources available for the survey, and the aggregate level at which separate estimates were needed. The planned sample size was 4000 completed interviews for states with populations of more than 25 million, 3000 for states with populations of between 2 and 25 million and 1500 for states with populations of less than 2 million. Because of a need to provide region-wise estimates for the populous northern states larger samples of 10,000 for Uttar Pradesh and 7000 each for Madhya Pradesh, Bihar and Rajasthan were planned. Rural and urban samples within each state were drawn separately and in proportion to the size of the urban and rural populations as far as possible. The survey adopted a uniform sampling strategy in all the states. In rural areas selections of villages identified from the 1991 Census list were designated as primary sampling units with probability proportional to population size (PPS); within these units households were randomly selected. In urban areas, wards were selected using the PPS method followed by random selection of one census enumeration block (CEB). In the final stage,
ARTICLE IN PRESS 1150
S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
households were randomly selected within each sample CEB. All households selected were contacted during the survey, and no replacement was made if eligible members of a selected household could not be interviewed. The survey data were provided in anonymised form for academic use by ORC MACRO (Claverton, MD, USA), for which ethical approval was not required. The states of Bihar, Madhya Pradesh (MP), Rajasthan and Uttar Pradesh (UP) were selected for analysis. Uttar Pradesh has since been split into Uttarachal and Uttar Pradesh but these new states are considered together. The NFHS-2 collected information about antenatal care, components of antenatal care, women’s autonomy, fertility, family planning and women’s socio-economic background. These details are available for the last births of all women interviewed in the survey. This paper considers antenatal check-ups for women pertaining to their last birth only. In this paper, ‘antenatal check-up’ means that a pregnant woman either visited a health facility or was visited at home by a health worker. An ‘antenatal visit’ does not necessarily mean that a woman received any antenatal care services. Therefore, ‘antenatal check-up’ and ‘antenatal care services’ are different: the former pertains only to the visit and the later relates to services actually received, for example, blood pressure measurement. Logistic regression analysis was applied to study the association between occurrence of antenatal check-ups and a set of independent variables. The dependent variable is whether or not women received an antenatal check-up with a Yes/No response. The set of independent variables hypothesised to have associations with antenatal care and included in the analysis are (1) women’s education (2) women’s socio-economic characteristics (3) demographic characteristics and (4) exposure to media. Women’s socio-economic characteristics include education, an index of autonomy, working status, standard of living, caste, and religion. The autonomy index was constructed using seven questions about various dimensions of women’s autonomy. These were: decision about items to cook; who makes decisions on women’s health care; who makes decisions on costly purchases; who makes decisions on a woman’s stay with her parents; whether permission is required to go to market; whether permission required to visit friends; whether permission required to keep money aside. Each item was assigned a score 1 or 2; 1 to a situation where women is not a decision maker and 2 where the women is the decision maker. Thus the score for each person ranged between 7 and 14. A score 7–10 was taken to indicate low autonomy and a score 11–14 to indicate high autonomy. Demographic characteristics included age at marriage, number of children ever born, and family planning use. In order to identify multicollinearity among the variables ‘tolerance values’ were generated, and in none of the cases those were less than
0.2. Similarly none of the variance inflation factors (VIF) was greater than 10. Two models were constructed for each state. The first model included only antenatal check-ups received through visits to a health facility and the second model included cases from both visits to a health facility and visits to the women at her residence by a health visitor. In the survey, those women who had received antenatal check-ups (through both means) were further asked about the services they received at least once during their antenatal check-ups in the last pregnancy. Critical antenatal services potentially influencing pregnancy outcomes were further analysed using logistic regression. The services included in this analysis were urine testing, measurement of blood pressure, blood testing, abdomen examination, internal examination, tetanus toxoid vaccination, and obtaining iron and folic acid tablets. The dependent variables were Yes/No answers to each service and the independent variables were the same used in the logistic regression for uptake of antenatal check-ups.
Results Antenatal check-ups Table 1 provides information about antenatal checkups in the four northern states in India. In Bihar and Uttar Pradesh nearly 70% of women in their last pregnancy did not receive any antenatal check-up. In Rajasthan, antenatal check-ups were slightly more common, and in Madhya Pradesh more than half of women received antenatal care during their last pregnancy. The proportion of women who received any antenatal check-up through visits to a health facility was greatest in MP (38%) and lowest in Bihar (21%). The proportion of women who received antenatal check-ups through home visits by health workers was highest in MP (12%), it was lowest in UP (4%). Few women received antenatal check-ups through both modes in all states. During visits, 62% of women were seen by a nurse/ midwife and 55% by a doctor. In Bihar and Uttar Pradesh more women were seen by doctors than by nurses or midwives. Among those women who attended at least one antenatal check-up, the average number of antenatal visits was 2.4; the lowest was in UP and the highest in Rajasthan, indicating that many are not achieving the RCH programme guideline of at least 3 antenatal visits during pregnancy. Regarding the timing of antenatal visits, only 38% women made their first antenatal visit during their first trimester of pregnancy, 44% in the second trimester, and remaining 18% in the last trimester. Although there were some differences across the states, the pattern was similar. This contrasts
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
1151
Table 1 Percentage of rural women who received at least one antenatal check-up during their last pregnancy, northern states, India, 1998–1999 Antenatal care uptake status Received by visit to a facility Received through visit of a health worker Both Not received Total
Bihar
Madhya Pradesh
21.0 6.3 5.7 67.0 100.0 N ¼ 3064
37.6 12.4 6.1
26.8 11.0 5.4
43.9 100.0 N ¼ 1950
with the recommendation of a commonly cited Indian reference text of one visit each during the third, sixth, eighth, and ninth months of pregnancy (Park & Park, 1989).
Components of antenatal care Those women who had an antenatal check-up during their last pregnancy were further asked whether they received any of the specific services listed in Table 2. Among women who visited a health facility for at least one antenatal check-up more than 50% received tetanus toxoid vaccination, abdominal examination, iron and folic acid tablets, and advice on diet. The service with the highest uptake was tetanus toxoid vaccination. Among women visited by a health worker the only service received by more than 50% of women was tetanus toxoid vaccination. In general, women visited by health visitors for antenatal check-ups received substantially fewer antenatal services compared to women who visited a health facility. In particular, blood pressure, blood tests and urine tests were provided to less than 10% of women visited by health workers compared to 40% of those who visited a health facility. Those who both visited a health facility and were visited by a health worker had a higher rate of internal examination than those who consulted in either setting alone. Women obtaining check-ups from both sources were more likely to receive tetanus toxoid vaccination, abdominal examination, internal examination, explanation of danger signs in pregnancy and family planning advice compared to women who received antenatal check-ups through either of the sources alone. The analysis identified some women who reported receiving components of antenatal care despite having not attended an antenatal consultation. In particular, many had received tetanus toxoid. This reflects the wide availability of the vaccination through other routes such as village pharmacies. However, the likelihood of having been vaccinated was still greater among those who had attended antenatal checkups (60% versus 89%).
Rajasthan
56.8 100.0 N ¼ 1471
Uttar Pradesh 24.6 3.7 2.1 69.6 100.0 N ¼ 4884
All four states 26.1 6.9 4.2 62.8 100.0 N ¼ 11369
Antenatal check-ups by background characteristics Table 3 shows antenatal check-ups by background characteristics of women. Uptake of antenatal check-ups through visits to a health facility increased with husband’s education in all the four states. However, antenatal care received through home visits did not increase with husband’s education. In particular, where husband’s education was the highest a lower proportion of women received antenatal care compared with wives of husbands with lower educational levels. Similarly, women’s education consistently showed a positive association with antenatal check-ups through visits to a health facility in all the states. As in the case of husband’s education, a lower proportion of women with higher levels of education received antenatal care through home visits. Except in Rajasthan, the likelihood of antenatal check-ups from either source did not vary significantly with women’s autonomy. Interestingly, in all states uptake though visits to a health facility was higher among non-working women compared to working women. Although not substantial, the proportion of women receiving antenatal care through home visits was higher among working women. A higher standard of living was associated with a greater likelihood of receiving antenatal check-ups though visits to a health facility in all the states. The proportion of women who received antenatal care through home visits was higher in ‘high standard of living’ households in Bihar and Madhya Pradesh, but not significantly different in the other two states. Religion and caste showed unique state specific features. For example, whilst among all the religious groups Muslims in Madhya Pradesh were the most likely to access antenatal care through visits to a health facility, Muslims in the other states had the lowest uptake. However, in Madhya Pradesh, Muslims had a low uptake of antenatal care through home visits. With regard to caste, members of ‘other backward classes’ had a higher uptake of antenatal care compared to ‘scheduled castes’ and ‘scheduled tribes’. In all states antenatal check-ups through visits to a health facility were more likely among women who married at the age
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
1152
Table 2 Percentage of rural women receiving specific components of antenatal care, northern states, India, 1998–1999 Antenatal care component
Only among women who visited a health facility
Only among women visited by a health worker
Both visited a health facility and visited by a health worker
Among all women who received antenatal care
Among all women eligible for antenatal care
Tetanus toxoid Abdomen examined Iron folic acid Diet Blood test Blood pressure Urine test Weight Internal examination Danger signs Newborn care Family planning Height Ultrasound X-ray Amniocentesis
88.3 65.9
90.1 27.8
94.4 70.9
89.3 59.4
59.5 21.9
63.6 54.5 41.7 40.6 38.4 28.1 23.3
46.4 42.2 9.1 6.1 5.7 5.1 4.9
61.6 50.0 40.2 34.6 37.1 25.0 36.9
60.2 51.7 35.5 33.6 32.2 23.5 21.5
30.8 19.1 13.1 12.4 11.9 8.6 7.9
24.0 23.8 14.6 10.3 8.4 3.8 0.8
9.1 10.3 12.9 3.0 0.4 0.1 0.3
25.7 23.5 18.7 7.0 3.8 2.9 0.4
21.4 21.3 14.8 8.6 6.4 3.0 0.7
7.9 7.8 5.5 0.1 2.3 1.1 0.3
Note: N ¼ 3454 for the first four columns; N ¼ 11369 for the last column.
of 19 or above compared to women married at a younger age. However, the pattern of women receiving antenatal care though home visits varied across the states. In all the states the likelihood of an antenatal check-up though visits to a health facility was lower among women with two or more children compared to women with one child. In Rajasthan, the uptake of antenatal care through home visits was higher among women who had two or more children. In all states women who had ever used a modern family planning method had a higher uptake of antenatal care though visits to a health facility compared to those who had never used family planning. However, in Madhya Pradesh antenatal care uptake was high among women who had never used a family planning method. In all the states, women who reported exposure to media were more likely to have had a check-up though visits to a health facility than those who did not. Multivariate analysis Results from logistic regression analysis are given in Table 4, in which two models for each state are presented. Women’s socio-economic characteristics As expected, women’s educational attainment is positively associated with antenatal care in the four states and for the two models. The importance of
women’s education in predicting their access to antenatal care was most evident in Rajasthan but in all the states the influence of education became apparent from the secondary level. Some variation in the importance of women’s educational status is evident between models for antenatal care with and without home visits. The striking finding is that the marked difference between the ‘no education’ group and others was not reduced when modelling care including home visits. A possible interpretation is that health workers tend to visit more educated women for antenatal care. In general, women’s autonomy measured through seven indicators was positively related to use of antenatal care in all the states, but this reached significance only in MP and UP. In the two states, the odds of ‘high autonomy’ women receiving antenatal care were 1.4 and 1.3 relative to those with ‘low autonomy’. However, autonomy was no longer predictive in models including antenatal care provided via home visits. Thus, there may be an underlying tendency for health workers to visit ‘low autonomy’ women at home in these two states, perhaps reflecting a recognition of the constraints on such women to leave the home. In contrast to the univariate findings, possibly indicating an adverse effect of work on access to care, in multivariate models women’s work status was either non-significant (three states) or showed a positive relationship with access (UP) with the odds of receiving antenatal care of about 1.2 for working women
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
1153
Table 3 Percentage of rural women who received at least one antenatal check-up by background characteristics, northern states, India, 1998–1999 Characteristics
Bihar Visit to health facility
Madhya Pradesh
Uttar Pradesh
Visit to health facility
Visit to home
Visit to health facility
Visit to home
Visit to health facility
Visit to home
5 6 8 6
29 36 44 54
13 11 16 8
18 25 32 43
12 14 10 7
15 19 28 41
3 5 4 4
Women’s socio-economic status Education No education 16 Primary 24 Secondary 42 High 71
6 11 8 5
31 45 62 75
12 16 11 0
22 41 48 78
12 9 8 0
19 32 43 71
3 5 6 0
Women’s autonomy Low High
21 15
6 12
37 44
13 5
26 7
11 6
25 27
4 1
Women’s work status Not working 23 Working 14
6 7
45 33
10 14
28 26
11 12
26 21
4 4
Standard of living index Low standard 15 High standard 30
5 8
31 42
11 14
19 30
12 11
17 29
4 4
Religion Hindu Muslim Others
22 17 24
7 5 8
37 57 35
12 9 18
27 28 40
11 9 7
25 21 51
4 4 6
15 20 20
7 4 7
38 26 42
12 13 12
24 24 29
14 12 12
22 19 22
3 3 4
32
5
50
12
27
8
29
4
7 11
36 52
13 11
25 35
12 7
23 37
4 3
Husband’s education No education Primary Secondary High
Caste Scheduled caste Scheduled tribe Other backward classes Others
12 17 25 46
Demographic factors marriage 18 years or less 19 19 years or more 22
Visit to home
Rajasthan
Children ever born 1 child 2 or more children
28 19
9 6
46 35
13 12
33 25
9 13
34 22
4 4
Family planning use Never used Ever used
19 31
6 9
35 44
13 9
25 32
11 12
22 32
4 3
Exposure to media No Yes
18 47
6 8
32 52
13 11
23 48
11 8
20 45
4 3
N ¼ 3064
N ¼ 1950
N ¼ 1471
N ¼ 4884
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
1154
Table 4 Logistic regression showing variables influencing the odds of receiving antenatal care, rural residents of northern states, India, 1998–1999 Characteristics
Husband’s education No educationa Primary Secondary High
Bihar
Madhya Pradesh
Rajasthan Excluding Home visit
Excluding Home visit
Including home visits
Excluding Home visit
Including home visits
1.4 1.6 2.5
1.3 1.6 2.2
1.2 1.2 1.7
1.1 1.3 1.5
1.5 1.6 1.9
Uttar Pradesh Including home visits
Excluding Home visit
Including home visits
1.5 1.4 1.5
1.0 1.5 1.6
1.1 1.5 1.6
Women’s socio-economic No educationa Primary Secondary High
status 1.0 1.7 2.9
1.1 1.7 2.8
1.2 2.6 8.0
1.5 3.2 5.1
2.0 2.2 20.0
2.0 2.2 13.0
1.6 2.3 4.8
1.6 2.6 4.3
Autonomy (high) Work status (working) Standard of living (high)
1.3 0.8 1.2
1.2 0.9 1.3
1.4 0.9 1.1
1.0 1.1 1.3
1.4 1.1 1.2
1.4 1.1 1.1
1.3 1.2 1.3
1.1 1.2 1.2
Caste Scheduled castea Scheduled tribe Other backward classes Others
1.3 1.2
0.9 1.3
0.6 0.9
0.7 0.9
1.1 1.0
1.1 1.0
0.6 0.9
0.7 0.9
1.5
1.2
0.9
0.9
0.8
0.7
1.0
1.1
1.0 1.1
1.0 1.2
3.7 0.5
4.9 0.8
1.3 0.9
1.2 0.9
0.9 3.2
0.9 3.5
1.0 0.7 1.5 2.6 N ¼ 3064
1.0 0.6 1.8 2.6
1.4 0.7 1.2 1.8 N ¼ 1950
1.4 0.7 1.1 1.8
1.2 0.8 1.4 2.2
1.0 0.6 1.3 2.0 N ¼ 4884
0.9 0.6 1.3 1.9
Religion Hindua Muslim Others Demographic factors Age at marriage (19+) No of children (2+) FP use (Ever use) Exposure to media (Yes)
1.5 0.8 1.4 2.3 N ¼ 1471
po0:010; o0:05: a Reference category.
compared to others. Inclusion of antenatal care delivered through visits of health workers did not change the relationship in either magnitude or direction in any of the states. Except in Rajasthan, the standard of living was significantly associated with women’s use of antenatal care, but in the case of Bihar and MP this became significant only when home visits were included. This may indicate a preference for health workers to visit households with a higher standard of living. State-wise variation was evident in relation to the impact of caste and religion. In MP the odds of receiving antenatal care were 30–40% lower for members of ‘scheduled tribes’ relative to ‘scheduled castes’. Interestingly, relative to those from ‘scheduled castes’, Rajasthani women belonging to the ‘forward’ caste groups
had 20–30% lower odds for receiving antenatal care, especially when home visits were included, a finding that may indicate specific features of these communities with regard to care seeking behaviour. In Bihar and Uttar Pradesh caste membership did not influence women’s use of antenatal care. Muslim women in MP had 3.7–4.9 times greater odds of using antenatal care compared to Hindus. In UP, on the other hand, Muslims and Hindus were similar while members of ‘other’ religions, including Sikhs, Buddhists and Christians were more likely to access antenatal care with odds of 2.3–3.2 relative to Hindus. Demographic characteristics Age at marriage was positively associated with access to antenatal care, but this reached significance only in
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
MP and Rajasthan, with odds of 1.4 and 1.5 among women married at 19 or more compared to those married at less than 19 years of age. This association was no longer evident for Rajasthan when antenatal care via home visits was included, suggesting that health workers might preferentially visit this group. Parity was negatively associated with use of antenatal care in all the states except Rajasthan. The odds of women with two or more children (i.e., one or more children at the time of receiving antenatal care) receiving antenatal care were 20–40% lower compared to women with one child (i.e., no children at the time of receiving antenatal care). The influence of parity did not change after the inclusion of home visits in the model. Use of family planning was positively associated with antenatal care in all the states except MP. Women who had ever used a modern family planning method had 1.3–1.8-fold greater odds of obtaining antenatal care than women who had not, for Bihar, Rajasthan and UP. In Bihar, the model including antenatal home visits increased the odds ratio slightly, possibly suggesting that health workers preferred to visit women who were family planning users. While it is difficult to draw conclusions as to whether antenatal care increases family planning acceptance or vice versa, it appears that these two aspects of health service utilisation are inter-related. Depending on the stage in formation of the family the direction of the relationship could change. For example, it is likely that women drawn into antenatal care might subsequently use modern family planning methods for either spacing or limiting family size (Misra et al., 1998). Similarly, parous women who use family planning methods for spacing are perhaps more likely to use antenatal care in subsequent pregnancies. However, given the massive dominance of female sterilisation in the Indian family planning programme, the main impact is likely to be an effect of antenatal care in facilitating future family planning use. Exposure to media Watching television every week substantially increased the odds of women seeking antenatal care in all the states. This was the second strongest factor after education that positively influenced antenatal care in the four states. The odds of women obtaining antenatal care were 1.9–2.8 times greater for those who watched television every week compared to those who did not. The greatest impact of weekly television watching was seen in Bihar and the least in Rajasthan. Inclusion of home visits did not alter the findings. Factors associated with receipt of specific components of antenatal care Table 5 shows factors associated with access to specific components of antenatal care. The previous
1155
section demonstrated the importance of women’s education in predicting access to antenatal care in the four states. This influence varied in relation to specific components of antenatal care, indicating differential influences on the actual composition of care received. Autonomy did not predict access to most components, but was actually associated with reduced odds of receiving tetanus toxoid vaccination. In contrast, urine testing was more likely among women with ‘high autonomy’. Women’s work status was significantly and negatively associated with access to most of the antenatal care components, with the exception of provision of iron supplementation. Thus while employment is generally associated with increased access to antenatal care, some specific constraints seem to apply to consultations by working women such that they are less likely to receive certain components of care. Higher age at marriage was associated with a greater likelihood of receiving urine and blood tests in all the four states. Access to other components of care were unaffected by age at marriage. Parity significantly and negatively related to the uptake of all critical antenatal services except provision of iron supplementation, with odds 20–40% lower. Thus parity influenced both the likelihood of accessing care and its content in a similar manner. Use of modern family planning methods predicted access to all components of antenatal care, including internal examination. This suggests either an element of women’s familiarity with aspects of clinical care derived from education about family planning, or greater ease for providers in dealing with clients already known. While weekly television watching was associated with greater access to antenatal care and non-intimate components of care it did not predict physical examination during antenatal consultations, suggesting that social customs are a barrier to quality of care. Similarly, women from households with a higher standard of living were more likely to receive antenatal components involving tests or medicines but not physical examination. While members of ‘scheduled castes’ and ‘scheduled tribes’ had reduced odds of obtaining most of the components of antenatal care, the data show that Muslim women were more likely than others to receive components of care. This suggests that the stereotype of Muslim women being secluded from care and unable to access care of adequate quality does not hold in this context.
Discussion The present study shows continuing inequity in access to antenatal care in northern India. As the two successive NFHS used similar methods to sample a
ARTICLE IN PRESS 1156
S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
Table 5 Logistic regression showing variables influencing the odds of receiving specific components of antenatal care, rural residents of northern states, India, 1998–1999 Characteristics
Abdominal examination
Internal examination
Urine Test
Blood Pressure
Blood test
Tetanus toxoid
Iron/folic acid
Husband’s education No educationa Primary Secondary High
1.2 1.1 1.3
0.8 1.1 1.5
0.9 1.0 1.4
1.2 1.1 1.6
0.9 1.0 1.3
1.4 1.4 2.2
1.4 1.3 1.5
Women’s socio-economic status Education No educationa Primary Secondary High
1.1 1.4 2.6
0.9 1.4 2.1
1.0 1.6 2.9
1.2 1.3 3.1
1.1 1.4 3.1
1.2 2.3 5.3
1.2 1.4 1.5
Autonomy (High) Work status (Work) Standard of living (High)
1.1 0.7 1.1
1.2 0.6 1.1
1.6 0.8 1.3
1.3 0.8 1.1
1.2 1.0 1.2
0.6 0.9 0.9
0.8 1.4 1.3
Caste Scheduled castea Scheduled tribe Other backward classes Others
0.9 1.0 1.1
0.5 1.2 0.9
0.8 0.9 1.0
1.3 1.3 1.4
1.1 1.0 1.3
0.7 0.8 0.7
1.3 1.0 1.0
Religion Hindua Muslim Others
0.9 1.5
1.4 0.5
1.3 1.2
1.3 1.6
1.0 1.4
0.8 3.6
0.9 1.9
Demographic factors Age at marriage (19+) Children ever born (2+) Family planning use (Ever use) Exposure to media (Yes)
0.9 0.7 1.2 1.1
1.1 0.8 1.6 0.9
1.3 0.6 1.1 1.3
1.2 0.7 1.4 1.4
1.3 0.7 1.1 1.3
0.8 0.8 1.5 1.5
1.1 0.9 1.3 1.3
po0:010; o0:05: a Reference category.
population of ever-married women it is likely that the present findings truly reflect a lack of substantial progress towards increasing uptake in antenatal care. The impact of nationally driven service initiatives implemented since the survey, especially the RCH programme, cannot of course be assessed in the present study. The present findings show population sub-groups less likely to receive care, and also indicate wide variations in the content of antenatal care delivered either through clinics or home visits. Variables positively associated with receiving antenatal care include women’s educational attainment, use of family planning services, older age at marriage, low parity, and access to television. It is possible that health workers are more reluctant to visit the homes of poor women in some settings; conversely, some better off
women may prefer not to invite health workers into their homes perhaps for reasons of caste or status. The impact of women’s employment and of caste and religion varied in the different states. In the present analysis the construct of autonomy did not prove a powerful explanator of variation in access: for example, it was non-predictive of access to care in Rajasthan and Bihar and only moderately predictive for Madhya Pradesh and Uttar Pradesh. However, we combined questionnaire items to create a single autonomy variable that may have masked the influence of some of the individual items or underlying factors. The concept of women’s autonomy is complex; we used seven items generally viewed as reflecting women’s autonomy but the results should be interpreted bearing in mind the limitations of this approach. In an urban north Indian context ‘freedom
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
of movement’ was an important predictor of antenatal care seeking whereas ‘control over finances’ and ‘decision making power’ were non-significant in multivariate models (Bloom, Wypij, & Das Gupta, 2001). The differences in social conditions between the different states and between communities in each state are reflected in the diverse associations between caste and religious affiliation seen in the present analyses, highlighting the need for state level policy formulation that takes such conditions into account. Higher parity was associated with reduced use of antenatal care, as also seen in the study cited above. In south India in a context of very high rural uptake of antenatal care, women were less likely to obtain antenatal care commencing in the first trimester in their second or third pregnancies compared to their first, but in fourth or subsequent pregnancies early care seeking was again the norm (Matthews, Mahendra, Kilaru, & Ganapathy, 2001). Reduced uptake of antenatal care for second or third pregnancies may reflect both women’s perceptions of risk and service provision: women who have experienced a previous pregnancy without complications may feel little need to seek care, and as preeclampsia in particular is less common in subsequent compared to first pregnancies health workers may consider parous women to need less attention. Practical issues such as the difficulty in attending a health facility when caring for small children also need consideration. If, as suggested by the present study and others, both social dynamics and biological factors are influential in individual antenatal care seeking, community level perceptions of the value of antenatal care assume considerable importance. Rural antenatal care has historically been provided through the state, with more recent emphasis on the involvement of private practitioners in the RCH programme. Rural women also have access to practitioners of Indian traditional medical systems; care seeking in this sector was not explored in the NFHS-2. Care seeking will reflect both perceptions of illness and wellness in relation to a normal physiological event, and perceptions of the value and quality of health providers. Similar considerations apply in other contexts where normal human behaviour is medicalised, especially family planning, where the tools of social marketing have been used to link cost, value and quality in design of service provision. Stephenson and Tsui (2002) applied a multilevel modelling approach to north Indian data to demonstrate the inter-relationships between individual variables such as parity and education, facility and community level variables such as health infrastructure, the operation of health campaigns and the size of the village population. Individual-level variation was evident in contraceptive use, STI treatment, antenatal care and place of birth after controlling for demographic and socioeconomic factors. Facility level and community level influences were different for
1157
the four types of service. Analysis of unexplained heterogeneity in the resulting models showed that community level variation in the use of antenatal care was greater for women with secondary or higher educational attainment. Such findings indicate the complex interplay of drivers of service access and the need for careful planning of interventions that take due account of individual and community factors. The variable content of antenatal care provision is demonstrated in the present study. Many components of care shown to be lacking do not require expensive equipment or extensive training, but indicate deficiencies in training and supervision of health workers. Whereas a reluctance to undertake internal examination antenatally is not likely directly to impact negatively on birth outcomes, failure to examine the abdomen represents a missed opportunity to detect obstetric abnormalities and refer for delivery under medical supervision. Failure to measure the blood pressure or test the urine means that pre-eclampsia, usually asymptomatic, will not be detected with a consequent risk of eclampsia. Provision of iron and folic acid supplementation, with or without blood testing for anaemia, represents one of the simplest and cheapest ‘safe motherhood’ interventions in this context. Antenatal anaemia results directly in maternal death due to the increased circulatory demands of pregnancy, and indirectly through reducing the maternal reserve should antepartum or postpartum haemorrhage occur. For example, in a hospital series of maternal deaths and ‘near miss’ episodes in Haryana, severe anaemia was responsible for 2/31 deaths and 35/224 cases of major morbidity (Khosla, Dahiya, & Sangwana, 2000). The data indicate that skill mix in service provision may not be optimal: around 62% of those who had received antenatal care reported consultations with a nurse/midwife, and 55% with a doctor. Reorientation of the current service model should emphasise the role of non-medical staff undertaking those components of antenatal care most likely to prevent maternal complications, and free up medical attention for those with established complications. Further health economic analysis is need to assess the impact of such reorientation on the total service costs, on the costs of providing adequate care to an increased proportion of women, and to what extent these might be offset by health gains. Critical assessment is also required of the advice components within antenatal care as currently provided, such as birth preparedness and awareness of danger signs. There is potential to link health education components with understanding of the community level dynamics: when women attend consultations with family members or are seen at home, those family members also receive the advice and influence women’s responses to it. There are some indications from the present study of inequities in provision of components of care at clinics
ARTICLE IN PRESS 1158
S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
to certain groups, and reluctance either by staff to undertake home visits or by women to be visited at home by health workers. Further research on consulting dynamics including observation of consultations, exit interviews at facilities and interviews at home is required. Provision of iron and folic acid supplementation for pregnant women could be a suitable topic for community mobilisation and/or community-based distribution, thus reducing dependence on the health system for access to this simple but effective intervention. In conclusion, pregnant women from poor and uneducated backgrounds, with at least one child were the most unlikely to receive antenatal check-ups and services in the four large north Indian states. Basic antenatal care components are effective means to prevent a range of pregnancy complications and reduce maternal mortality. The findings indicate substantial limitations of the health services in overcoming socioeconomic and cultural barriers to access.
Acknowledgements This study was funded by the UK Government Department for International Development.
References Bergsj^, P. (2001). What is the evidence for the role of antenatal care strategies in the reduction of maternal mortality and morbidity? In Vincent De Brouwere, Wim Van Lerberghe (Eds.), Safe Motherhood Strategies: A review of the evidence. Studies in Health Services Organisation and Policy no 17. Antwerp: ITG Press. Bhatia, J. C., & Cleland, J. (1995). Determinants of maternal care in a region of South India. Health Transition Review, 5, 127–142. Bloom, S., Wypij, D., & Das Gupta, M. (2001). Dimensions of women’s autonomy and the influence on maternal health care utilisation in a north Indian city. Demography, 38, 67–78. Chandrashekar, S., Phaneendran, R. S., & Nair, N. S. (1998). Socio-demographic determinants of antenatal care. Tropical Doctor, 28, 206–209. Carroli, G., Rooney, C., & Villar, J. (2001). How effective is antenatal care in preventing maternal mortality and serious morbidity? An overview of the evidence. Paediatric and Perinatal Epidemiology, 15(suppl. 1), 1–42. International Institute for Population Sciences and ORC Macro (2000). National Family Health Survey (NFHS-2), 1998–99: India. Mumbai: IIPS. Jowett, M. (2000). Safe motherhood interventions in lowincome countries: An economic justification and evidence of cost effectiveness. Health Policy, 53, 201–228.
Khosla, A. H., Dahiya, K., & Sangwana, K. (2000). Maternal mortality and ‘near-miss’ in rural north India. International Journal of Gynecology & Obstetrics, 68, 163–164. Matsumura, M., & Gubhaju, B. (2001). Women’s status, household structure and the utilisation of maternal health services in Nepal. Asia-Pacific Population Journal, 16(1), 23–44. Matthews, Z., Mahendra, S., Kilaru, A., & Ganapathy, S. (2001). Antenatal care, care seeking and morbidity in rural Karnataka: Results of a prospective study. Asia-Pacific Population Journal, 16, 11–28. Magadi, M. A., Madise, N. J., & Rodrigues, R. N. (2000). Frequency and timing of antenatal care in Kenya: Explaining the variations between women of different communities. Social Science & Medicine, 51, 551–561. McCAW-Binns, A., Grenade, J. L., & Ashley, D. (1995). Under-users of antenatal care: A comparison of nonattenders and late attenders for antenatal care, with early attenders, Social Science & Medicine, 40, 1003–1012. Ministry of Health and Family Welfare. (2001). Annual Report 2000–2001. Misra, U. S., Roy, T. K., & Rajan Irudaya, S. (1998). Antenatal care and contraceptive behaviour in India: Some evidence from the National Family Health Survey. The Journal of Family Welfare, 44, 1–14. Neilsen, B. B., Liljestrand, J., Thilsted, S. H., Joseph, A., & Hedegaard, M. (2001). Characteristics of antenatal care attenders in a rural population in Tamil Nadu, South India: A community-based cross-sectional study. Health and Social Care in the Community, 9, 327–333. Obermeyer, C. M., & Potter, J. (1991). Maternal health care utilisation in Jordan: A study of patterns and determinants. Studies in Family Planning, 22, 177–187. Park, J. E., & Park, K. (1989). Textbook of preventive and social medicine (12th Edition.). Jabalpur: Banarsidas Bhanot. Registrar General and Census Commissioner. (2001). Census of India, Provisional Population tables, Paper 1 of 2001. Singh, P., & Yadav, R. H. (no date). Antenatal care of pregnant women in India, Institute for Research in Medical Statistics, Indian Council of Medical Research, New Delhi. http:// themedicaljobs.com/at/a-com2.htm, date accessed, 25, November 2003. Stephenson, R., & Tsui, A. O. (2002). Contextual influences on reproductive health service use in Uttar Pradesh, India. Studies in Family Planning, 33, 309–320. Sugathan, K. S., Mishra, V., & Retherford, R. D. (2001). Promoting institutional deliveries in rural India: The role of antenatal-care services. National Family Health Survey Subject Reports no. 20, International Institute for Population Sciences and East-West Centre. Taguchi, N., Kawabata, M., Maruo, T., Aditiawarman, & Dewata, L. (2003). Influence of socio-economic background and maternal care programmes on maternal mortality in Surabaya, Indonesia. Tropical Medicine and International Health, 8, 847–852. World Health Organsiation. (1996). Care in Normal Birth: A Practical Guide (Geneva, WHO/FRH/MSM 96-24).
Social Science & Medicine 59 (2004) 1147–1158
Antenatal care: provision and inequality in rural north India Saseendran Pallikadavath, Mary Foss, R William Stones* ‘Opportunities and Choices’ Programme of Reproductive Health Research and School of Nursing and Midwifery, University of Southampton, Southampton SO16 5YA, UK
Abstract The objectives of this paper are to examine factors associated with use of antenatal care in rural areas of north India, to investigate access to specific critical components of care and to study differences in the pattern of services received via health facilities versus home visits. We used the 1998–1999 Indian National Family Health Survey of ever-married women in the reproductive age group and analysed data from the states of Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh (n ¼ 11; 369). Overall, about three-fifths of rural women did not receive any antenatal check-up during their last pregnancy. Services actually received were predominantly provision of tetanus toxoid vaccination and supply of iron and folic acid tablets. Only about 13% of pregnant women had their blood pressure checked and a blood test done at least once. Women visited by health workers received fewer services compared to women who visited a health facility. Home visits were biased towards households with a better standard of living. There was significant under-utilisation of nurse/midwives in the provision of antenatal services and doctors were often the lead providers. The average number of antenatal visits reported in this study was 2.4 and most visits were in the second trimester. Higher social and economic status was associated with increased chances of receiving an antenatal check-up, and of receiving specific components including blood pressure measurement, a blood test and urine testing but not the obstetric physical examination, which was however linked to ever-use of family planning and the education of women and their husbands. Thus, pregnant women from poor and uneducated backgrounds with at least one child were the least likely to receive antenatal check-ups and services in the four large north Indian states. Basic antenatal care components are effective means to prevent a range of pregnancy complications and reduce maternal mortality. The findings indicate substantial limitations of the health services in overcoming socio-economic and cultural barriers to access. r 2004 Elsevier Ltd. All rights reserved. Keywords: Antenatal care; Health services; Socio-economic status; India
Introduction Antenatal care Reducing maternal mortality and morbidity has been a major focus for the developing world since the launch of the Safe Motherhood Initiative in 1987 (WHO, 1996). Over recent years there has been some debate as to the
*Corresponding author. Level F (815), Princess Anne Hospital, Southampton SO16 5YA, UK. Tel.: +44-23-80796033; fax: +44-23-8078-6933. E-mail address: [email protected] (R.W. Stones).
effectiveness of one aspect of maternity provision, that of antenatal care (Carroli, Rooney, & Villar, 2001; Bergsj^, 2001). The main purposes of antenatal care are to prevent certain complications, such as anaemia, and identify women with established pregnancy complications for treatment or transfer. Tetanus toxoid vaccination of the mother prevents neonatal tetanus as protective antibodies are passed across the placenta. It was hoped that women ‘at risk’ of pregnancy complications could be identified so as to avert problems, and scoring systems were developed for this purpose. However, the use of risk assessment has turned out to be unproductive as women may develop complications in pregnancy or childbirth at any time.
0277-9536/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.socscimed.2003.11.045
ARTICLE IN PRESS 1148
S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
Despite the limitations of antenatal risk assessment the basic functions of detection of pre-eclampsia, anaemia and other incipient complications remain essential. In addition, other less tangible benefits may be realised which are not easily evaluated in isolation. Antenatal consultations provide opportunities for health education, health promotion and social support at both the individual and community level. Especially in the rural setting, accessing antenatal care is an important step in bringing women into contact with the health care system. This contact has facilitated women’s access to medical care for future health needs, including postnatal care (Sugathan, Mishra, & Retherford, 2001). Within a context of limited resources there is a need to ensure value for money: overall, antenatal care is considered to represent a cost-effective component of maternity services as part of ‘safe motherhood’ interventions to reduce mortality and morbidity (Jowett, 2000). The Indian context Overall, uptake of antenatal care in India has been low. The 1992–1993 Indian National Family Health Survey showed that only 64% mothers received antenatal check-ups and this increased marginally to 65% in 1998–1999 (International Institute for Population Sciences and ORC Macro, 2000). This lack of improvement occurred despite governmental and non-governmental efforts to strengthen service delivery, and is likely to contribute to the continuing high maternal mortality in the country, especially in the northern states. Maternal mortality in the four states Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh ranged between 450 and 700 per 100,000 live births (International Institute for Population Sciences and ORC Macro, 2000). These states have some of the highest fertility (total fertility rate) in the country: ranging between 3–4 children per woman. Similarly, infant and child mortality rate in these states were reported to be among the highest in India with a range 73–87 infant deaths per 1000 live births. In 2001 female literacy in these states was low. According to the 2001 Census, these states account for about 40% of India’s total population (Registrar General and Census Commissioner, 2001). Achieving significant improvements in maternal health in these states is vital to improving maternal health in India as a whole. In India, antenatal care initiatives began in 1951 with the implementation of the First Five Year Plan (1951–1956). However, a lack of rural health infrastructure, and a focus on family planning using a clinicbased approach limited the benefits to rural women. In the Third Five Year Plan (1961–1966) an extension approach to family planning was adopted with recruitment of auxiliary nurse midwives (ANMs) and Health Assistants. This provided rural women access to some
elements of antenatal care. However, it may be noted that these programmes were primarily geared to family planning service provision. During the Fifth Five Year Plan (1974–1979) maternal and child health services (MCH) were integrated with family planning services and a new programme entitled ‘Family Welfare’ was introduced. This gave impetus to provision of maternal health services in rural areas. In the Seventh Five Year Plan (1985–1990) a Universal Immunisation Programme (UIP) was implemented, greatly increasing pregnant women’s access to tetanus toxoid vaccination. In 1991–1992 a major initiative called ‘Child Survival and Safe Motherhood’ (CSSM) was undertaken by integrating the UIP with expanded MCH activities in states with high mortality rates. Following the recommendations of the International Conference on Population and Development in 1994, in 1997 the Government of India launched the Reproductive and Child Health (RCH) programme for implementation in the Ninth Five Year Plan (1997–2002). This was done by integrating CSSM interventions and adding interventions for reproductive tract infections and sexually transmitted diseases. The idea behind the RCH programme is to provide need based, client centred, demand driven, high quality integrated services to beneficiaries. All districts in the country were covered by the programme as of 1999–2000 (Ministry of Health and Family Welfare, 2001). It is important to note here that Integrated Child Development (ICDS) programmes introduced in 1975 in selected districts also provided women with access to antenatal care. In rural areas, ANMs posted at the sub-Primary Health Centres (sub-PHCs) are responsible for providing antenatal care either in homes or clinics. Sub-PHCs are the grass-roots level government health service points covering a population of 5000 in non-hilly areas. Sub-PHCs refer to Primary Health Centres (PHCs) manned by medical and other paramedical staff. Antenatal care services to pregnant women are also provided by the ‘Anganwadi’ centres working under the ICDS Programmes. Determinants of antenatal care In developing countries uptake of antenatal care is determined by socio-economic factors, demographic factors, and availability and access to health facilities. In many parts of the developing world women’s education was found to be an important predictor of antenatal care utilisation. In Jordan (Obermeyer & Potter, 1991), Nepal (Matsumura & Gubhaju, 2001), and Surabaya in Indonesia (Taguchi, Kawabata, Maruo, & dewata, 2003) higher levels of education of women were associated with greater use of antenatal care. However, in Kenya maternal education was not significantly related to utilisation of antenatal care once
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
other variables were considered (Magadi, Madise, & Rodrigues, 2000). However, the potential impact of husbands’ or partners’ educational status has not been explored. In India, a relationship between maternal education and utilisation of antenatal care is evident in many studies. In Karnataka state in India, Bhatia and Cleland (1995) reported that higher levels of maternal education increased the uptake of antenatal care. Similar finding was reported in a study of four Southern Indian states: Kerala, Tamil Nadu, Karnataka, and Andhra Pradesh (Navaneetham & Dharmalingam, 2002). Indeed a national study conduced by the Indian Medical Council Research (ICMR) in 90 districts of the country in 1999 concluded that literacy of women is the key factor to achieving improvements in antenatal care in India (Singh & Yadav, no date). Demographic factors such as order of pregnancy (or number of children), marital status, maternal age, and marital duration are reported having influence on the utilisation of antenatal care. Less use of antenatal care was found among women having larger numbers of children in Jordan (Obermeyer & Potter, 1991) and Kenya (Magadi et al., 2000). In India utilisation of antenatal care was low among women with higher order pregnancies: Chandrashekar et al. (1998) working in rural Karnataka noted that multiparous and unskilled mothers aged over 30 were less likely to have used antenatal services. Similar findings were reported in Tamil Nadu (Neilsen, Liljestrand, Thilsted, Joseph, & Hedegaard, 2001). Low antenatal care utilisation was reported among teenagers in Jamiaca (McCAW-Binns, Grenade, & Ashley, 1995) and Kenya (Magadi et al., 2000). Other relevant variables were unmarried status, unions of very short duration, desire for large families, never-use of family planning, unwanted pregnancies. In India, Bhatia and Cleland (1995) noted reduced antenatal care use among women aged under 18. In Karnataka, consulting for pregnancy confirmation and a prior history of neonatal or fetal loss were important determinants of subsequent antenatal care utilisation. The interplay of socio-economic factors and accessibility of health care facilities was highlighted in Kenya (Magadi et al., 2000): even if health care services are easily accessible socio-cultural factors at community level determine whether to use the facility. In India, using the 1992–1993 National Family and Health Survey, Misra, Roy, and Rajan Irudaya (1998) revealed that lack of knowledge about antenatal services was an important reason for non-utilisation in the northern states of India. A study in four southern Indian states found that variations in service use were primarily related to availability and access (Navaneetham & Dharmalingam, 2002). In India, religion and caste are two important social factors that influence social practices, including belief
1149
systems surrounding pregnancy. However, these do not follow a stereotypical pattern. For example, while Muslims were more likely to go for medical check-ups n Karnataka they were less likely to do so in Kerala (Bhatia & Cleland, 1995, Navaneetham & Dharmalingam, 2002). Recognising that socio-cultural factors are likely to be critical determinants of care seeking and service utilisation in the context of pregnancy, the objective of this paper is to examine factors associated with use of antenatal care facilities in the rural areas of Bihar, Madhya Pradesh, Rajasthan and Uttar Pradesh (the latter including the newly created state of Uttaranchal). The paper also examines factors associated with access to specific critical components of care. A final objective is to study the differences, if any, in the pattern of antenatal check-ups and services received through health facilities versus home visits.
Data and methods Data from the 1998–1999 National Family Health Survey (NFHS-2) were utilized in the present study. This survey was undertaken by the International Institute for Population Sciences, Mumbai on behalf of the Government of India, Ministry of Health and Family Welfare. This was a nationally representative population based sample survey of 90,303 ever-married women aged 15–49 from 26 states comprising 99% of India’s population. The survey methods are described in detail together with univariate tabulations in an overview report (International Institute for Population Sciences and ORC Macro, 2000). Briefly, the sample size of the survey was set for each state considering the size of the state, resources available for the survey, and the aggregate level at which separate estimates were needed. The planned sample size was 4000 completed interviews for states with populations of more than 25 million, 3000 for states with populations of between 2 and 25 million and 1500 for states with populations of less than 2 million. Because of a need to provide region-wise estimates for the populous northern states larger samples of 10,000 for Uttar Pradesh and 7000 each for Madhya Pradesh, Bihar and Rajasthan were planned. Rural and urban samples within each state were drawn separately and in proportion to the size of the urban and rural populations as far as possible. The survey adopted a uniform sampling strategy in all the states. In rural areas selections of villages identified from the 1991 Census list were designated as primary sampling units with probability proportional to population size (PPS); within these units households were randomly selected. In urban areas, wards were selected using the PPS method followed by random selection of one census enumeration block (CEB). In the final stage,
ARTICLE IN PRESS 1150
S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
households were randomly selected within each sample CEB. All households selected were contacted during the survey, and no replacement was made if eligible members of a selected household could not be interviewed. The survey data were provided in anonymised form for academic use by ORC MACRO (Claverton, MD, USA), for which ethical approval was not required. The states of Bihar, Madhya Pradesh (MP), Rajasthan and Uttar Pradesh (UP) were selected for analysis. Uttar Pradesh has since been split into Uttarachal and Uttar Pradesh but these new states are considered together. The NFHS-2 collected information about antenatal care, components of antenatal care, women’s autonomy, fertility, family planning and women’s socio-economic background. These details are available for the last births of all women interviewed in the survey. This paper considers antenatal check-ups for women pertaining to their last birth only. In this paper, ‘antenatal check-up’ means that a pregnant woman either visited a health facility or was visited at home by a health worker. An ‘antenatal visit’ does not necessarily mean that a woman received any antenatal care services. Therefore, ‘antenatal check-up’ and ‘antenatal care services’ are different: the former pertains only to the visit and the later relates to services actually received, for example, blood pressure measurement. Logistic regression analysis was applied to study the association between occurrence of antenatal check-ups and a set of independent variables. The dependent variable is whether or not women received an antenatal check-up with a Yes/No response. The set of independent variables hypothesised to have associations with antenatal care and included in the analysis are (1) women’s education (2) women’s socio-economic characteristics (3) demographic characteristics and (4) exposure to media. Women’s socio-economic characteristics include education, an index of autonomy, working status, standard of living, caste, and religion. The autonomy index was constructed using seven questions about various dimensions of women’s autonomy. These were: decision about items to cook; who makes decisions on women’s health care; who makes decisions on costly purchases; who makes decisions on a woman’s stay with her parents; whether permission is required to go to market; whether permission required to visit friends; whether permission required to keep money aside. Each item was assigned a score 1 or 2; 1 to a situation where women is not a decision maker and 2 where the women is the decision maker. Thus the score for each person ranged between 7 and 14. A score 7–10 was taken to indicate low autonomy and a score 11–14 to indicate high autonomy. Demographic characteristics included age at marriage, number of children ever born, and family planning use. In order to identify multicollinearity among the variables ‘tolerance values’ were generated, and in none of the cases those were less than
0.2. Similarly none of the variance inflation factors (VIF) was greater than 10. Two models were constructed for each state. The first model included only antenatal check-ups received through visits to a health facility and the second model included cases from both visits to a health facility and visits to the women at her residence by a health visitor. In the survey, those women who had received antenatal check-ups (through both means) were further asked about the services they received at least once during their antenatal check-ups in the last pregnancy. Critical antenatal services potentially influencing pregnancy outcomes were further analysed using logistic regression. The services included in this analysis were urine testing, measurement of blood pressure, blood testing, abdomen examination, internal examination, tetanus toxoid vaccination, and obtaining iron and folic acid tablets. The dependent variables were Yes/No answers to each service and the independent variables were the same used in the logistic regression for uptake of antenatal check-ups.
Results Antenatal check-ups Table 1 provides information about antenatal checkups in the four northern states in India. In Bihar and Uttar Pradesh nearly 70% of women in their last pregnancy did not receive any antenatal check-up. In Rajasthan, antenatal check-ups were slightly more common, and in Madhya Pradesh more than half of women received antenatal care during their last pregnancy. The proportion of women who received any antenatal check-up through visits to a health facility was greatest in MP (38%) and lowest in Bihar (21%). The proportion of women who received antenatal check-ups through home visits by health workers was highest in MP (12%), it was lowest in UP (4%). Few women received antenatal check-ups through both modes in all states. During visits, 62% of women were seen by a nurse/ midwife and 55% by a doctor. In Bihar and Uttar Pradesh more women were seen by doctors than by nurses or midwives. Among those women who attended at least one antenatal check-up, the average number of antenatal visits was 2.4; the lowest was in UP and the highest in Rajasthan, indicating that many are not achieving the RCH programme guideline of at least 3 antenatal visits during pregnancy. Regarding the timing of antenatal visits, only 38% women made their first antenatal visit during their first trimester of pregnancy, 44% in the second trimester, and remaining 18% in the last trimester. Although there were some differences across the states, the pattern was similar. This contrasts
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
1151
Table 1 Percentage of rural women who received at least one antenatal check-up during their last pregnancy, northern states, India, 1998–1999 Antenatal care uptake status Received by visit to a facility Received through visit of a health worker Both Not received Total
Bihar
Madhya Pradesh
21.0 6.3 5.7 67.0 100.0 N ¼ 3064
37.6 12.4 6.1
26.8 11.0 5.4
43.9 100.0 N ¼ 1950
with the recommendation of a commonly cited Indian reference text of one visit each during the third, sixth, eighth, and ninth months of pregnancy (Park & Park, 1989).
Components of antenatal care Those women who had an antenatal check-up during their last pregnancy were further asked whether they received any of the specific services listed in Table 2. Among women who visited a health facility for at least one antenatal check-up more than 50% received tetanus toxoid vaccination, abdominal examination, iron and folic acid tablets, and advice on diet. The service with the highest uptake was tetanus toxoid vaccination. Among women visited by a health worker the only service received by more than 50% of women was tetanus toxoid vaccination. In general, women visited by health visitors for antenatal check-ups received substantially fewer antenatal services compared to women who visited a health facility. In particular, blood pressure, blood tests and urine tests were provided to less than 10% of women visited by health workers compared to 40% of those who visited a health facility. Those who both visited a health facility and were visited by a health worker had a higher rate of internal examination than those who consulted in either setting alone. Women obtaining check-ups from both sources were more likely to receive tetanus toxoid vaccination, abdominal examination, internal examination, explanation of danger signs in pregnancy and family planning advice compared to women who received antenatal check-ups through either of the sources alone. The analysis identified some women who reported receiving components of antenatal care despite having not attended an antenatal consultation. In particular, many had received tetanus toxoid. This reflects the wide availability of the vaccination through other routes such as village pharmacies. However, the likelihood of having been vaccinated was still greater among those who had attended antenatal checkups (60% versus 89%).
Rajasthan
56.8 100.0 N ¼ 1471
Uttar Pradesh 24.6 3.7 2.1 69.6 100.0 N ¼ 4884
All four states 26.1 6.9 4.2 62.8 100.0 N ¼ 11369
Antenatal check-ups by background characteristics Table 3 shows antenatal check-ups by background characteristics of women. Uptake of antenatal check-ups through visits to a health facility increased with husband’s education in all the four states. However, antenatal care received through home visits did not increase with husband’s education. In particular, where husband’s education was the highest a lower proportion of women received antenatal care compared with wives of husbands with lower educational levels. Similarly, women’s education consistently showed a positive association with antenatal check-ups through visits to a health facility in all the states. As in the case of husband’s education, a lower proportion of women with higher levels of education received antenatal care through home visits. Except in Rajasthan, the likelihood of antenatal check-ups from either source did not vary significantly with women’s autonomy. Interestingly, in all states uptake though visits to a health facility was higher among non-working women compared to working women. Although not substantial, the proportion of women receiving antenatal care through home visits was higher among working women. A higher standard of living was associated with a greater likelihood of receiving antenatal check-ups though visits to a health facility in all the states. The proportion of women who received antenatal care through home visits was higher in ‘high standard of living’ households in Bihar and Madhya Pradesh, but not significantly different in the other two states. Religion and caste showed unique state specific features. For example, whilst among all the religious groups Muslims in Madhya Pradesh were the most likely to access antenatal care through visits to a health facility, Muslims in the other states had the lowest uptake. However, in Madhya Pradesh, Muslims had a low uptake of antenatal care through home visits. With regard to caste, members of ‘other backward classes’ had a higher uptake of antenatal care compared to ‘scheduled castes’ and ‘scheduled tribes’. In all states antenatal check-ups through visits to a health facility were more likely among women who married at the age
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
1152
Table 2 Percentage of rural women receiving specific components of antenatal care, northern states, India, 1998–1999 Antenatal care component
Only among women who visited a health facility
Only among women visited by a health worker
Both visited a health facility and visited by a health worker
Among all women who received antenatal care
Among all women eligible for antenatal care
Tetanus toxoid Abdomen examined Iron folic acid Diet Blood test Blood pressure Urine test Weight Internal examination Danger signs Newborn care Family planning Height Ultrasound X-ray Amniocentesis
88.3 65.9
90.1 27.8
94.4 70.9
89.3 59.4
59.5 21.9
63.6 54.5 41.7 40.6 38.4 28.1 23.3
46.4 42.2 9.1 6.1 5.7 5.1 4.9
61.6 50.0 40.2 34.6 37.1 25.0 36.9
60.2 51.7 35.5 33.6 32.2 23.5 21.5
30.8 19.1 13.1 12.4 11.9 8.6 7.9
24.0 23.8 14.6 10.3 8.4 3.8 0.8
9.1 10.3 12.9 3.0 0.4 0.1 0.3
25.7 23.5 18.7 7.0 3.8 2.9 0.4
21.4 21.3 14.8 8.6 6.4 3.0 0.7
7.9 7.8 5.5 0.1 2.3 1.1 0.3
Note: N ¼ 3454 for the first four columns; N ¼ 11369 for the last column.
of 19 or above compared to women married at a younger age. However, the pattern of women receiving antenatal care though home visits varied across the states. In all the states the likelihood of an antenatal check-up though visits to a health facility was lower among women with two or more children compared to women with one child. In Rajasthan, the uptake of antenatal care through home visits was higher among women who had two or more children. In all states women who had ever used a modern family planning method had a higher uptake of antenatal care though visits to a health facility compared to those who had never used family planning. However, in Madhya Pradesh antenatal care uptake was high among women who had never used a family planning method. In all the states, women who reported exposure to media were more likely to have had a check-up though visits to a health facility than those who did not. Multivariate analysis Results from logistic regression analysis are given in Table 4, in which two models for each state are presented. Women’s socio-economic characteristics As expected, women’s educational attainment is positively associated with antenatal care in the four states and for the two models. The importance of
women’s education in predicting their access to antenatal care was most evident in Rajasthan but in all the states the influence of education became apparent from the secondary level. Some variation in the importance of women’s educational status is evident between models for antenatal care with and without home visits. The striking finding is that the marked difference between the ‘no education’ group and others was not reduced when modelling care including home visits. A possible interpretation is that health workers tend to visit more educated women for antenatal care. In general, women’s autonomy measured through seven indicators was positively related to use of antenatal care in all the states, but this reached significance only in MP and UP. In the two states, the odds of ‘high autonomy’ women receiving antenatal care were 1.4 and 1.3 relative to those with ‘low autonomy’. However, autonomy was no longer predictive in models including antenatal care provided via home visits. Thus, there may be an underlying tendency for health workers to visit ‘low autonomy’ women at home in these two states, perhaps reflecting a recognition of the constraints on such women to leave the home. In contrast to the univariate findings, possibly indicating an adverse effect of work on access to care, in multivariate models women’s work status was either non-significant (three states) or showed a positive relationship with access (UP) with the odds of receiving antenatal care of about 1.2 for working women
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
1153
Table 3 Percentage of rural women who received at least one antenatal check-up by background characteristics, northern states, India, 1998–1999 Characteristics
Bihar Visit to health facility
Madhya Pradesh
Uttar Pradesh
Visit to health facility
Visit to home
Visit to health facility
Visit to home
Visit to health facility
Visit to home
5 6 8 6
29 36 44 54
13 11 16 8
18 25 32 43
12 14 10 7
15 19 28 41
3 5 4 4
Women’s socio-economic status Education No education 16 Primary 24 Secondary 42 High 71
6 11 8 5
31 45 62 75
12 16 11 0
22 41 48 78
12 9 8 0
19 32 43 71
3 5 6 0
Women’s autonomy Low High
21 15
6 12
37 44
13 5
26 7
11 6
25 27
4 1
Women’s work status Not working 23 Working 14
6 7
45 33
10 14
28 26
11 12
26 21
4 4
Standard of living index Low standard 15 High standard 30
5 8
31 42
11 14
19 30
12 11
17 29
4 4
Religion Hindu Muslim Others
22 17 24
7 5 8
37 57 35
12 9 18
27 28 40
11 9 7
25 21 51
4 4 6
15 20 20
7 4 7
38 26 42
12 13 12
24 24 29
14 12 12
22 19 22
3 3 4
32
5
50
12
27
8
29
4
7 11
36 52
13 11
25 35
12 7
23 37
4 3
Husband’s education No education Primary Secondary High
Caste Scheduled caste Scheduled tribe Other backward classes Others
12 17 25 46
Demographic factors marriage 18 years or less 19 19 years or more 22
Visit to home
Rajasthan
Children ever born 1 child 2 or more children
28 19
9 6
46 35
13 12
33 25
9 13
34 22
4 4
Family planning use Never used Ever used
19 31
6 9
35 44
13 9
25 32
11 12
22 32
4 3
Exposure to media No Yes
18 47
6 8
32 52
13 11
23 48
11 8
20 45
4 3
N ¼ 3064
N ¼ 1950
N ¼ 1471
N ¼ 4884
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
1154
Table 4 Logistic regression showing variables influencing the odds of receiving antenatal care, rural residents of northern states, India, 1998–1999 Characteristics
Husband’s education No educationa Primary Secondary High
Bihar
Madhya Pradesh
Rajasthan Excluding Home visit
Excluding Home visit
Including home visits
Excluding Home visit
Including home visits
1.4 1.6 2.5
1.3 1.6 2.2
1.2 1.2 1.7
1.1 1.3 1.5
1.5 1.6 1.9
Uttar Pradesh Including home visits
Excluding Home visit
Including home visits
1.5 1.4 1.5
1.0 1.5 1.6
1.1 1.5 1.6
Women’s socio-economic No educationa Primary Secondary High
status 1.0 1.7 2.9
1.1 1.7 2.8
1.2 2.6 8.0
1.5 3.2 5.1
2.0 2.2 20.0
2.0 2.2 13.0
1.6 2.3 4.8
1.6 2.6 4.3
Autonomy (high) Work status (working) Standard of living (high)
1.3 0.8 1.2
1.2 0.9 1.3
1.4 0.9 1.1
1.0 1.1 1.3
1.4 1.1 1.2
1.4 1.1 1.1
1.3 1.2 1.3
1.1 1.2 1.2
Caste Scheduled castea Scheduled tribe Other backward classes Others
1.3 1.2
0.9 1.3
0.6 0.9
0.7 0.9
1.1 1.0
1.1 1.0
0.6 0.9
0.7 0.9
1.5
1.2
0.9
0.9
0.8
0.7
1.0
1.1
1.0 1.1
1.0 1.2
3.7 0.5
4.9 0.8
1.3 0.9
1.2 0.9
0.9 3.2
0.9 3.5
1.0 0.7 1.5 2.6 N ¼ 3064
1.0 0.6 1.8 2.6
1.4 0.7 1.2 1.8 N ¼ 1950
1.4 0.7 1.1 1.8
1.2 0.8 1.4 2.2
1.0 0.6 1.3 2.0 N ¼ 4884
0.9 0.6 1.3 1.9
Religion Hindua Muslim Others Demographic factors Age at marriage (19+) No of children (2+) FP use (Ever use) Exposure to media (Yes)
1.5 0.8 1.4 2.3 N ¼ 1471
po0:010; o0:05: a Reference category.
compared to others. Inclusion of antenatal care delivered through visits of health workers did not change the relationship in either magnitude or direction in any of the states. Except in Rajasthan, the standard of living was significantly associated with women’s use of antenatal care, but in the case of Bihar and MP this became significant only when home visits were included. This may indicate a preference for health workers to visit households with a higher standard of living. State-wise variation was evident in relation to the impact of caste and religion. In MP the odds of receiving antenatal care were 30–40% lower for members of ‘scheduled tribes’ relative to ‘scheduled castes’. Interestingly, relative to those from ‘scheduled castes’, Rajasthani women belonging to the ‘forward’ caste groups
had 20–30% lower odds for receiving antenatal care, especially when home visits were included, a finding that may indicate specific features of these communities with regard to care seeking behaviour. In Bihar and Uttar Pradesh caste membership did not influence women’s use of antenatal care. Muslim women in MP had 3.7–4.9 times greater odds of using antenatal care compared to Hindus. In UP, on the other hand, Muslims and Hindus were similar while members of ‘other’ religions, including Sikhs, Buddhists and Christians were more likely to access antenatal care with odds of 2.3–3.2 relative to Hindus. Demographic characteristics Age at marriage was positively associated with access to antenatal care, but this reached significance only in
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
MP and Rajasthan, with odds of 1.4 and 1.5 among women married at 19 or more compared to those married at less than 19 years of age. This association was no longer evident for Rajasthan when antenatal care via home visits was included, suggesting that health workers might preferentially visit this group. Parity was negatively associated with use of antenatal care in all the states except Rajasthan. The odds of women with two or more children (i.e., one or more children at the time of receiving antenatal care) receiving antenatal care were 20–40% lower compared to women with one child (i.e., no children at the time of receiving antenatal care). The influence of parity did not change after the inclusion of home visits in the model. Use of family planning was positively associated with antenatal care in all the states except MP. Women who had ever used a modern family planning method had 1.3–1.8-fold greater odds of obtaining antenatal care than women who had not, for Bihar, Rajasthan and UP. In Bihar, the model including antenatal home visits increased the odds ratio slightly, possibly suggesting that health workers preferred to visit women who were family planning users. While it is difficult to draw conclusions as to whether antenatal care increases family planning acceptance or vice versa, it appears that these two aspects of health service utilisation are inter-related. Depending on the stage in formation of the family the direction of the relationship could change. For example, it is likely that women drawn into antenatal care might subsequently use modern family planning methods for either spacing or limiting family size (Misra et al., 1998). Similarly, parous women who use family planning methods for spacing are perhaps more likely to use antenatal care in subsequent pregnancies. However, given the massive dominance of female sterilisation in the Indian family planning programme, the main impact is likely to be an effect of antenatal care in facilitating future family planning use. Exposure to media Watching television every week substantially increased the odds of women seeking antenatal care in all the states. This was the second strongest factor after education that positively influenced antenatal care in the four states. The odds of women obtaining antenatal care were 1.9–2.8 times greater for those who watched television every week compared to those who did not. The greatest impact of weekly television watching was seen in Bihar and the least in Rajasthan. Inclusion of home visits did not alter the findings. Factors associated with receipt of specific components of antenatal care Table 5 shows factors associated with access to specific components of antenatal care. The previous
1155
section demonstrated the importance of women’s education in predicting access to antenatal care in the four states. This influence varied in relation to specific components of antenatal care, indicating differential influences on the actual composition of care received. Autonomy did not predict access to most components, but was actually associated with reduced odds of receiving tetanus toxoid vaccination. In contrast, urine testing was more likely among women with ‘high autonomy’. Women’s work status was significantly and negatively associated with access to most of the antenatal care components, with the exception of provision of iron supplementation. Thus while employment is generally associated with increased access to antenatal care, some specific constraints seem to apply to consultations by working women such that they are less likely to receive certain components of care. Higher age at marriage was associated with a greater likelihood of receiving urine and blood tests in all the four states. Access to other components of care were unaffected by age at marriage. Parity significantly and negatively related to the uptake of all critical antenatal services except provision of iron supplementation, with odds 20–40% lower. Thus parity influenced both the likelihood of accessing care and its content in a similar manner. Use of modern family planning methods predicted access to all components of antenatal care, including internal examination. This suggests either an element of women’s familiarity with aspects of clinical care derived from education about family planning, or greater ease for providers in dealing with clients already known. While weekly television watching was associated with greater access to antenatal care and non-intimate components of care it did not predict physical examination during antenatal consultations, suggesting that social customs are a barrier to quality of care. Similarly, women from households with a higher standard of living were more likely to receive antenatal components involving tests or medicines but not physical examination. While members of ‘scheduled castes’ and ‘scheduled tribes’ had reduced odds of obtaining most of the components of antenatal care, the data show that Muslim women were more likely than others to receive components of care. This suggests that the stereotype of Muslim women being secluded from care and unable to access care of adequate quality does not hold in this context.
Discussion The present study shows continuing inequity in access to antenatal care in northern India. As the two successive NFHS used similar methods to sample a
ARTICLE IN PRESS 1156
S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
Table 5 Logistic regression showing variables influencing the odds of receiving specific components of antenatal care, rural residents of northern states, India, 1998–1999 Characteristics
Abdominal examination
Internal examination
Urine Test
Blood Pressure
Blood test
Tetanus toxoid
Iron/folic acid
Husband’s education No educationa Primary Secondary High
1.2 1.1 1.3
0.8 1.1 1.5
0.9 1.0 1.4
1.2 1.1 1.6
0.9 1.0 1.3
1.4 1.4 2.2
1.4 1.3 1.5
Women’s socio-economic status Education No educationa Primary Secondary High
1.1 1.4 2.6
0.9 1.4 2.1
1.0 1.6 2.9
1.2 1.3 3.1
1.1 1.4 3.1
1.2 2.3 5.3
1.2 1.4 1.5
Autonomy (High) Work status (Work) Standard of living (High)
1.1 0.7 1.1
1.2 0.6 1.1
1.6 0.8 1.3
1.3 0.8 1.1
1.2 1.0 1.2
0.6 0.9 0.9
0.8 1.4 1.3
Caste Scheduled castea Scheduled tribe Other backward classes Others
0.9 1.0 1.1
0.5 1.2 0.9
0.8 0.9 1.0
1.3 1.3 1.4
1.1 1.0 1.3
0.7 0.8 0.7
1.3 1.0 1.0
Religion Hindua Muslim Others
0.9 1.5
1.4 0.5
1.3 1.2
1.3 1.6
1.0 1.4
0.8 3.6
0.9 1.9
Demographic factors Age at marriage (19+) Children ever born (2+) Family planning use (Ever use) Exposure to media (Yes)
0.9 0.7 1.2 1.1
1.1 0.8 1.6 0.9
1.3 0.6 1.1 1.3
1.2 0.7 1.4 1.4
1.3 0.7 1.1 1.3
0.8 0.8 1.5 1.5
1.1 0.9 1.3 1.3
po0:010; o0:05: a Reference category.
population of ever-married women it is likely that the present findings truly reflect a lack of substantial progress towards increasing uptake in antenatal care. The impact of nationally driven service initiatives implemented since the survey, especially the RCH programme, cannot of course be assessed in the present study. The present findings show population sub-groups less likely to receive care, and also indicate wide variations in the content of antenatal care delivered either through clinics or home visits. Variables positively associated with receiving antenatal care include women’s educational attainment, use of family planning services, older age at marriage, low parity, and access to television. It is possible that health workers are more reluctant to visit the homes of poor women in some settings; conversely, some better off
women may prefer not to invite health workers into their homes perhaps for reasons of caste or status. The impact of women’s employment and of caste and religion varied in the different states. In the present analysis the construct of autonomy did not prove a powerful explanator of variation in access: for example, it was non-predictive of access to care in Rajasthan and Bihar and only moderately predictive for Madhya Pradesh and Uttar Pradesh. However, we combined questionnaire items to create a single autonomy variable that may have masked the influence of some of the individual items or underlying factors. The concept of women’s autonomy is complex; we used seven items generally viewed as reflecting women’s autonomy but the results should be interpreted bearing in mind the limitations of this approach. In an urban north Indian context ‘freedom
ARTICLE IN PRESS S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
of movement’ was an important predictor of antenatal care seeking whereas ‘control over finances’ and ‘decision making power’ were non-significant in multivariate models (Bloom, Wypij, & Das Gupta, 2001). The differences in social conditions between the different states and between communities in each state are reflected in the diverse associations between caste and religious affiliation seen in the present analyses, highlighting the need for state level policy formulation that takes such conditions into account. Higher parity was associated with reduced use of antenatal care, as also seen in the study cited above. In south India in a context of very high rural uptake of antenatal care, women were less likely to obtain antenatal care commencing in the first trimester in their second or third pregnancies compared to their first, but in fourth or subsequent pregnancies early care seeking was again the norm (Matthews, Mahendra, Kilaru, & Ganapathy, 2001). Reduced uptake of antenatal care for second or third pregnancies may reflect both women’s perceptions of risk and service provision: women who have experienced a previous pregnancy without complications may feel little need to seek care, and as preeclampsia in particular is less common in subsequent compared to first pregnancies health workers may consider parous women to need less attention. Practical issues such as the difficulty in attending a health facility when caring for small children also need consideration. If, as suggested by the present study and others, both social dynamics and biological factors are influential in individual antenatal care seeking, community level perceptions of the value of antenatal care assume considerable importance. Rural antenatal care has historically been provided through the state, with more recent emphasis on the involvement of private practitioners in the RCH programme. Rural women also have access to practitioners of Indian traditional medical systems; care seeking in this sector was not explored in the NFHS-2. Care seeking will reflect both perceptions of illness and wellness in relation to a normal physiological event, and perceptions of the value and quality of health providers. Similar considerations apply in other contexts where normal human behaviour is medicalised, especially family planning, where the tools of social marketing have been used to link cost, value and quality in design of service provision. Stephenson and Tsui (2002) applied a multilevel modelling approach to north Indian data to demonstrate the inter-relationships between individual variables such as parity and education, facility and community level variables such as health infrastructure, the operation of health campaigns and the size of the village population. Individual-level variation was evident in contraceptive use, STI treatment, antenatal care and place of birth after controlling for demographic and socioeconomic factors. Facility level and community level influences were different for
1157
the four types of service. Analysis of unexplained heterogeneity in the resulting models showed that community level variation in the use of antenatal care was greater for women with secondary or higher educational attainment. Such findings indicate the complex interplay of drivers of service access and the need for careful planning of interventions that take due account of individual and community factors. The variable content of antenatal care provision is demonstrated in the present study. Many components of care shown to be lacking do not require expensive equipment or extensive training, but indicate deficiencies in training and supervision of health workers. Whereas a reluctance to undertake internal examination antenatally is not likely directly to impact negatively on birth outcomes, failure to examine the abdomen represents a missed opportunity to detect obstetric abnormalities and refer for delivery under medical supervision. Failure to measure the blood pressure or test the urine means that pre-eclampsia, usually asymptomatic, will not be detected with a consequent risk of eclampsia. Provision of iron and folic acid supplementation, with or without blood testing for anaemia, represents one of the simplest and cheapest ‘safe motherhood’ interventions in this context. Antenatal anaemia results directly in maternal death due to the increased circulatory demands of pregnancy, and indirectly through reducing the maternal reserve should antepartum or postpartum haemorrhage occur. For example, in a hospital series of maternal deaths and ‘near miss’ episodes in Haryana, severe anaemia was responsible for 2/31 deaths and 35/224 cases of major morbidity (Khosla, Dahiya, & Sangwana, 2000). The data indicate that skill mix in service provision may not be optimal: around 62% of those who had received antenatal care reported consultations with a nurse/midwife, and 55% with a doctor. Reorientation of the current service model should emphasise the role of non-medical staff undertaking those components of antenatal care most likely to prevent maternal complications, and free up medical attention for those with established complications. Further health economic analysis is need to assess the impact of such reorientation on the total service costs, on the costs of providing adequate care to an increased proportion of women, and to what extent these might be offset by health gains. Critical assessment is also required of the advice components within antenatal care as currently provided, such as birth preparedness and awareness of danger signs. There is potential to link health education components with understanding of the community level dynamics: when women attend consultations with family members or are seen at home, those family members also receive the advice and influence women’s responses to it. There are some indications from the present study of inequities in provision of components of care at clinics
ARTICLE IN PRESS 1158
S. Pallikadavath et al. / Social Science & Medicine 59 (2004) 1147–1158
to certain groups, and reluctance either by staff to undertake home visits or by women to be visited at home by health workers. Further research on consulting dynamics including observation of consultations, exit interviews at facilities and interviews at home is required. Provision of iron and folic acid supplementation for pregnant women could be a suitable topic for community mobilisation and/or community-based distribution, thus reducing dependence on the health system for access to this simple but effective intervention. In conclusion, pregnant women from poor and uneducated backgrounds, with at least one child were the most unlikely to receive antenatal check-ups and services in the four large north Indian states. Basic antenatal care components are effective means to prevent a range of pregnancy complications and reduce maternal mortality. The findings indicate substantial limitations of the health services in overcoming socioeconomic and cultural barriers to access.
Acknowledgements This study was funded by the UK Government Department for International Development.
References Bergsj^, P. (2001). What is the evidence for the role of antenatal care strategies in the reduction of maternal mortality and morbidity? In Vincent De Brouwere, Wim Van Lerberghe (Eds.), Safe Motherhood Strategies: A review of the evidence. Studies in Health Services Organisation and Policy no 17. Antwerp: ITG Press. Bhatia, J. C., & Cleland, J. (1995). Determinants of maternal care in a region of South India. Health Transition Review, 5, 127–142. Bloom, S., Wypij, D., & Das Gupta, M. (2001). Dimensions of women’s autonomy and the influence on maternal health care utilisation in a north Indian city. Demography, 38, 67–78. Chandrashekar, S., Phaneendran, R. S., & Nair, N. S. (1998). Socio-demographic determinants of antenatal care. Tropical Doctor, 28, 206–209. Carroli, G., Rooney, C., & Villar, J. (2001). How effective is antenatal care in preventing maternal mortality and serious morbidity? An overview of the evidence. Paediatric and Perinatal Epidemiology, 15(suppl. 1), 1–42. International Institute for Population Sciences and ORC Macro (2000). National Family Health Survey (NFHS-2), 1998–99: India. Mumbai: IIPS. Jowett, M. (2000). Safe motherhood interventions in lowincome countries: An economic justification and evidence of cost effectiveness. Health Policy, 53, 201–228.
Khosla, A. H., Dahiya, K., & Sangwana, K. (2000). Maternal mortality and ‘near-miss’ in rural north India. International Journal of Gynecology & Obstetrics, 68, 163–164. Matsumura, M., & Gubhaju, B. (2001). Women’s status, household structure and the utilisation of maternal health services in Nepal. Asia-Pacific Population Journal, 16(1), 23–44. Matthews, Z., Mahendra, S., Kilaru, A., & Ganapathy, S. (2001). Antenatal care, care seeking and morbidity in rural Karnataka: Results of a prospective study. Asia-Pacific Population Journal, 16, 11–28. Magadi, M. A., Madise, N. J., & Rodrigues, R. N. (2000). Frequency and timing of antenatal care in Kenya: Explaining the variations between women of different communities. Social Science & Medicine, 51, 551–561. McCAW-Binns, A., Grenade, J. L., & Ashley, D. (1995). Under-users of antenatal care: A comparison of nonattenders and late attenders for antenatal care, with early attenders, Social Science & Medicine, 40, 1003–1012. Ministry of Health and Family Welfare. (2001). Annual Report 2000–2001. Misra, U. S., Roy, T. K., & Rajan Irudaya, S. (1998). Antenatal care and contraceptive behaviour in India: Some evidence from the National Family Health Survey. The Journal of Family Welfare, 44, 1–14. Neilsen, B. B., Liljestrand, J., Thilsted, S. H., Joseph, A., & Hedegaard, M. (2001). Characteristics of antenatal care attenders in a rural population in Tamil Nadu, South India: A community-based cross-sectional study. Health and Social Care in the Community, 9, 327–333. Obermeyer, C. M., & Potter, J. (1991). Maternal health care utilisation in Jordan: A study of patterns and determinants. Studies in Family Planning, 22, 177–187. Park, J. E., & Park, K. (1989). Textbook of preventive and social medicine (12th Edition.). Jabalpur: Banarsidas Bhanot. Registrar General and Census Commissioner. (2001). Census of India, Provisional Population tables, Paper 1 of 2001. Singh, P., & Yadav, R. H. (no date). Antenatal care of pregnant women in India, Institute for Research in Medical Statistics, Indian Council of Medical Research, New Delhi. http:// themedicaljobs.com/at/a-com2.htm, date accessed, 25, November 2003. Stephenson, R., & Tsui, A. O. (2002). Contextual influences on reproductive health service use in Uttar Pradesh, India. Studies in Family Planning, 33, 309–320. Sugathan, K. S., Mishra, V., & Retherford, R. D. (2001). Promoting institutional deliveries in rural India: The role of antenatal-care services. National Family Health Survey Subject Reports no. 20, International Institute for Population Sciences and East-West Centre. Taguchi, N., Kawabata, M., Maruo, T., Aditiawarman, & Dewata, L. (2003). Influence of socio-economic background and maternal care programmes on maternal mortality in Surabaya, Indonesia. Tropical Medicine and International Health, 8, 847–852. World Health Organsiation. (1996). Care in Normal Birth: A Practical Guide (Geneva, WHO/FRH/MSM 96-24).