Effect of women’s empowerment on their utilization of health services: A case of Nepal

Abstract

Over half a million women die annually due to maternity complications. Studies support that utilization of health services reduces maternal mortality. Using a nationally representative sample of 7878 previously pregnant women from Nepal we examined if gender empowerment is associated with health service use. Findings showed that three of the five indicators of empowerment – women’s age at birth of their first child, their education and knowledge about sexually transmitted diseases – significantly increased utilization of health services, especially antenatal and delivery services. Women’s opinion about wife beating and their participation in intra-household financial decisions were insignificant. Implications for social work practice are discussed.

Keywords

antenatal care use gender empowerment hospital delivery maternal health Nepal

Many schools of social work offer courses on international social development, but very few social work educators in North America are focusing their research on maternal mortality in developing countries. Approximately 536,000 women (or one woman every minute) die annually due to maternity complications and nearly all these deaths (99 percent) occur in developing countries (United Nations, 2000a, 2009). Sub-Saharan Africa and Southern Asia together account for 85 percent of all maternal deaths in the world (United Nations, 2009). Most of these deaths could be avoided if pregnant women used antenatal care, delivered at a health facility, and followed up with postnatal care (Jahan, 2007; Khan et al., 2006; Koblinsky et al., 2008; United Nations, 2009). Antenatal visits are universal in developed countries; however, 35 percent (45 million) of the pregnant women in developing countries, 50 percent in South Asia, 85 percent in Nepal, and 92 percent in Afghanistan do not receive any prenatal care (United Nations, 2000a). The two most common causes of maternal deaths – hemorrhage and infection due to delivery complications – claim about 44 percent of maternal deaths in Africa, 42 percent of maternal deaths in Asia, and 29 percent of maternal deaths in Latin America and the Caribbean countries (Khan et al., 2006). In Bangladesh 85 percent of all maternal deaths are due to obstetric causes, including antepartum and postpartum hemorrhage (Anwar et al., 2008; Koblinsky et al., 2008; Koenig et al., 2007). A hospital delivery will reduce maternal deaths from postpartum hemorrhage and sepsis dramatically; however, many women, including 77 percent of the pregnant women in Bangladesh and 58 percent of the pregnant women in India, deliver at home (Anwar et al., 2008; Pagel et al., 2009; Pandey and Lin, 2011). Often, countries with high rates of maternal mortality are also the societies that discriminate against women and girls in their nutrition, food allocation, medical attention and education (Seipel, 1992; Simkhada et al., 2006). Maternal mortality is symptomatic of gender-based social, cultural, economic and educational inequalities that undermine women’s access to health services. Generally, the front-line maternal health workers in developing countries are health professionals. Social workers can work with health professionals side-by-side and play a critical role in reducing maternal mortality by empowering women and girls, their families, and the community in which they reside through counseling, raising awareness and challenging practices (e.g. early marriage, home delivery) that undermine the status of women and girls. In this study we examine the effect of women’s empowerment on their use of antenatal, delivery and postnatal services, controlling for well-known correlates using data from Nepal.

Background

In 1987, international organizations sponsored a global conference and adopted the Safe Motherhood Initiative to reduce the high rate of women dying during pregnancy and childbirth. The Initiative recommended that all countries provide three types of maternity care services for all pregnant women: prenatal care, delivery care, and postnatal care (United Nations, 2000a). Prenatal care services included encouraging a woman with a normal pregnancy to make at least four visits to a skilled health attendant during her pregnancy (with more visits by women with pregnancy complications), and promoting information about maternal nutrition and iron supplements to reduce anemia, underweight and under-nutrition among pregnant women and new mothers. To provide delivery care during childbirth, all member countries were recommended to promote deliveries in health facilities and to promote the attendance of skilled health personnel including a doctor and/or person(s) with midwifery skills who can diagnose and manage obstetrical complications as well as normal delivery. Moreover, the fifth goal of the United Nations Millennium Development Goals (MDGs) is to improve maternal health by reducing maternal mortality by three-quarters between 1990 and 2015 and by achieving universal access to reproductive health by 2015 (United Nations, 2000b, 2010).

In response, many developing countries have expanded their health services. Nepal adopted the Safe Motherhood Initiative in 1997 and increased its focus on maternal and child health through media and local health workers. Health posts staffed by female maternal and child health workers, auxiliary field workers and village health workers were added to offer outreach services (Acharya and Cleland, 2000; Hotchkiss, 2001). Yet the majority of Nepali women continue to deliver at home. The decision to use health services, especially maternal health services, involves a complex set of intra-household, community and structural factors that are not well understood. Many authors have explained health service use behavior employing a broad range of indicators including the patients’ demographics, socio-economic status, cultural factors, health risk behavior, and access to health services. For example, a study from India found that women’s economic status, education, perceived complications during pregnancy, and their greater autonomy (freedom to move) correlated positively with use of antenatal care services (Bloom et al., 2001). In Bangladesh, rural, younger, poorer, and less educated mothers were significantly less likely to seek antenatal, delivery, or postnatal care compared to their urban, older, wealthier, and better educated counterparts (Anwar et al., 2008; Chakroborty et al., 2003; Chowdhury et al., 2007; Mamun et al., 2006). In Nepal, distance to health facility and quality of care correlated with use of maternal health services (Acharya and Cleland, 2000; Hotchkiss, 2001). In this article, we examine the effect of these and other factors on Nepalese women’s use of health services.

Conceptual framework

We start with the assumption that women’s empowerment is central to understanding the use of maternity health services. The word empowerment is widely used in the development literature. Nearly all definitions concur that empowerment is power attainment among those individuals, families, and communities who were powerless before. The Oxford English Dictionary defines the word empower as giving an individual ‘the authority or power to do something or to make someone stronger and more confident, especially in controlling their life and claiming their rights’. Amartya Sen, economist and Nobel laureate, introduced a capabilities approach to development and argued that nations should be judged based on their citizens’ capabilities to exercise their full potentials so that each person may realize his/her maximum functioning (Sen, 1999). Philosopher Martha Nussbaum (2000) saw gender inequality as the most important barrier to development; with examples of women from India, she lays out how women’s full potentials are undermined due to unequal power structures between genders. In the discussion of gender empowerment, other authors have also compared the power discrepancy of men versus women within a household, in a community, in a nation, and at the global level in their social, political, and economic opportunities (Kabeer, 1994, 1997, 1999, 2005; Malhotra et al., 2002; United Nations Development Programme [UNDP], 1995). Kabeer’s definition of empowerment, ‘as a process of change’ or ‘the processes by which those who have been denied the ability to make choices acquire such an ability’ (2005: 437), implies that the process of empowerment must improve women’s social, political, and economic choices. The process may require men to relinquish some of their power over women; eventually, however, both men and women and the entire society benefits when full potentials of all members of a society are utilized (Ehrhardt et al., 2009). Several authors have attempted to operationalize the concept of women’s empowerment (Malhotra et al., 2002; Pradhan, 2003; Williams, 2005). Some of the frequently used indicators include women’s education, income, employment, and participation in politics (UNDP, 1995). Reproductive health indicators include ability to decide when to get married, when to have children, and being able to control the number of pregnancies and birth. Williams (2005) used a series of variables as measures of women’s relative ability to exercise their power within a social system as measures of empowerment, including women’s ability to make general decisions at the community level, financial and purchasing decisions for small and large items at the household level, and women’s mobility choices. We hypothesized that increased empowerment of women will increase their use of antenatal, delivery, and postnatal care services.

In addition, the commonly used model, the behavioral model of health service use, implies that service utilization is a function of an individual’s predisposing, enabling, and perceived and evaluated need-related characteristics (Aday and Andersen, 1974; Andersen, 1995; Andersen and Newman, 1973; Davidson et al., 2004; Gelberg et al., 2000; Phillips et al., 1998). Using the two conceptual frameworks – gender empowerment and the behavioral model of health service use – we examined the relationship between women’s empowerment and their use of health services, controlling for other variables.

Methodology

This study utilized secondary data from the 2006 Nepal Demographic and Health Survey (NDHS) collected by New ERA with technical support from Macro International, under the auspices of the Ministry of Health and Population, and was funded by the United States Agency for International Development (Ministry of Health and Population, New ERA and Macro International, 2007). The study interviewed a nationally representative sample of 10,793 women between the ages of 15 and 49 from 8707 households across Nepal. We selected all previously pregnant women, which resulted in 7878 women (unweighted).

Measures

Dependent variable

We examined four outcome variables that reflected use of health services. The first variable covered women’s overall use of health services. The question asked: ‘In the last 12 months, have you visited a health facility for care for yourself or your children?’. Women who visited a health service were coded as 1 and those who did not visit received a score of 0. The next three dichotomous variables, antenatal care, delivery care, and postnatal care, captured utilization of pregnancy and birth-related services by a sub-sample of 4182 women (unweighted) who had given birth in the past five years. Women who did not use any antenatal care during pregnancy were coded as 0 and those who made at least one visit were coded as 1. Women who delivered at a health facility were coded as 1, and those who gave birth at home were coded as 0. Women who had their health checked from a health care provider, including traditional birth attendant after the delivery, were coded as 1 and those who did not utilize any postnatal visit(s) were coded as 0.

Independent (empowerment) variables

A set of variables that was associated with women’s power in the literature was considered independent in the regression models. They included women’s age at birth of their first child, their formal education, their knowledge about Sexually Transmitted Diseases (STD) and Acquired Immune Deficiency Syndrome (AIDS), their attitude toward wife beating, and their role in intra-household financial and health care decision making. Women’s age at birth of first child was a continuous variable. Education included four dummy variables: no formal education, primary education (one to five years of schooling), secondary education (six to ten years but did not complete the school-leaving certificate [SLC]), and SLC and beyond. Women who had heard about STD and AIDS were coded as 1 and those who had not were coded as 0. Attitude towards wife beating was constructed by combining the responses to five items: ‘Sometimes a husband is annoyed or angered by things that his wife does. In your opinion, is a husband justified in hitting or beating his wife in the following situations: (1) If she goes out without telling him? (2) If she neglects the children? (3) If she argues with him? (4) If she refuses to have sex with him? (5) If she burns the food?’ If the answer was ‘yes’ to any one of these items, wife beating was coded 1; otherwise it was coded 0. Financial or health-related decision variables included the responses to the following items: ‘(1) Who usually makes decisions about health care for you? (2) Who usually makes decisions about making major household purchases? (3) Who usually makes decisions about making purchases for daily household needs? (4) Who usually makes decisions about visits to your family or relatives?’ If women made decisions or if their opinion was considered in making decisions in any one of these items, they were coded as 1; otherwise they were coded as 0.

Control variables

A number of predisposing, enabling, and needs variables based on Andersen’s behavior model were controlled before examining the effect of women’s empowerment on the outcome variable. Predisposing factors included age, gender of household head, caste, and religion. Age was a continuous variable measured in years. Household head was coded as 1 for female and 0 for male. Caste and ethnicity was dummy coded into three variables: (1) Chhetri and Brahmin (generally economically privileged); (2) occupational caste; and (3) indigenous non-caste (Newar, Gurung, Magar, Tamang/Sherpa, Rai/Limbu, Muslim/Churaute, Tharu/Rajbanshi, Yadav/Ahir, other hill origin, and other terai origin). Religion was coded as 1 for Hindu and 0 for other (Buddhist, Muslim, Christian, and other).

Enabling factors included ecological region, development region, household’s wealth index, cost of care, quality of care, and distance to health facility. Ecological region included three dummy variables: Mountain, Hill, and Terai. Development region included five dummy coded administrative regions: Eastern, Central, Western, Midwestern, and Far-western development regions. For household wealth index, the NDHS 2006 data classified women into five wealth categories: poorest, poorer, middle, richer, and richest, using a wide range of household asset data including ownership of a number of consumer items ranging from a television to a bicycle or car, as well as dwelling characteristics, such as source of drinking water, sanitation facilities and type of material used for flooring (Ministry of Health and Population, New ERA and Macro International, 2007). We recoded this variable into three dummy variables – poor, middle, and upper class. The poor class combined the poorest and the poorer, the middle class remained unchanged, and the upper class included the richer and the richest.

Given the geographic and socio-economic conditions of Nepal various factors may act as a barrier to health-seeking behavior. In particular, we were interested in learning if respondents thought that cost of care, quality of health facility, and distance to health facility posed barriers to their health service use. To this end, we utilized the following questions: ‘Many different factors can prevent women from getting medical advice or treatment for themselves. When you are sick and want to get medical advice or treatment, is each of the following a big problem or not?’ The sub-questions included getting money needed for treatment; the distance to the health facility; concern that there may not be any health provider; concern that there may not be a female health provider; and concern that there may be no drugs available. For the variable money or cost a big problem, women who answered affirmatively were coded as 1 and those who answered otherwise were coded as 0. If distance was a big problem they were coded as 1; otherwise they were coded as 0. Finally, quality of health facility combined the three concerns and created one variable: women who were concerned about any of the three situations – concern that there may not be any health provider or a female health provider or prescription drugs – were coded as 1; all others were coded as 0.

Needs variables included women’s anemia status, body mass index (BMI), whether they ever had pregnancies terminated, whether they ever had a child or neonatal death, and whether they had given birth in the last five years. Anemia was measured at four levels: 1 = ‘severe’ to 4 = ‘not anemic’. We recoded this variable (anemic = 1; not anemic = 0). Using the internationally accepted cut off points, we dummy coded BMI into three variables, under weight (BMI = <18.59), normal (BMI = 18.60 to 24.99), and overweight (BMI > 24.99). Women who had ever terminated pregnancy were coded as 1; otherwise they were coded as 0. Women who ever had a child death or neonatal death were coded as 1; otherwise they were coded as 0. Finally, women who had given birth in the last five years were coded as 1; otherwise they were coded as 0.

Health-risk behavior captured women’s cigarette smoking habit; if a woman smoked, the variable was coded as 1; otherwise it was coded as 0.

We conducted descriptive and logistic regression analyses using the STATA software. All the results were weighted using probability weight (pweight).

Results

On average, women were 32 years of age and their age at marriage was 17, with first birth at age 19. Forty-four percent did not visit a health facility in the past 12 months for care for themselves or for their children (see Table 1). One in every four women was underweight and 36 percent were anemic. About 21 percent lived in women-headed households, 65 percent had no education, but 70 percent had heard about sexually transmitted diseases. About 72 percent made financial or health care related decisions. One in four women believed that wife beating was justified if the wives did not follow their husbands’ directions. About 52 percent had given birth in the past five years; a closer look at these 52 percent of the women showed that 74 percent had used antenatal care but only 22 percent delivered at a health facility and only 11 percent had used postnatal care.

Table 1.

Weighted descriptive analysis (unweighted N = 7878)

Categorical variables (frequency distribution)		(%)
Dependent variables
Used health services in past 12 months
No		44.05
Yes		55.95
Received antenatal care¹
No		26.21
Yes		73.79
Delivered at health facility¹
No (delivered at home)		78.45
Yes (delivered at health facility)		21.55
Received postnatal care¹
No		88.82
Yes		11.18
Independent variables (empowerment-related measures)
Education
No education		65.04
Primary: 1-5 grade		16.02
Secondary or higher : 6–10 but not SLC		12.93
SLC and above		6.01
Health knowledge (heard about STD and AIDS)
Never heard about STD and AIDS		29.76
Ever heard about STD or AIDS		70.24
Wife beating
Not OK to beat		76.76
OK to beat		23.24
Makes decisions related to financial and health care use
No		27.55
Yes		72.45
Control variables
Predisposing variables
Household head
Male		78.86
Female		21.14
Caste and ethnicity
Privileged caste (Brahmin or Chhetri)		30.56
Occupational caste		12.63
Indigenous non-caste		56.81
Religion
Hindu		85.60
Other		14.40
Enabling variables
Ecological region
Mountain		6.97
Hill		41.44
Terai		51.62
Development region
Eastern		21.48
Central		33.51
Western		19.17
Mid-western		11.72
Far western		14.12
Distance to health facility is a big problem?
No		58.01
Yes		41.99
Getting money needed for treatment is a big problem?
No		59.73
Yes		40.27
Quality of health facility is a big problem?
No		33.44
Yes		66.56
Wealth index
Lower class		38.43
Middle class		20.95
Upper class		40.61
Need variables
Anemic
No		63.84
Yes		36.16
Body Mass Index (BMI)
Underweight		24.01
Normal		65.48
Overweight		10.51
Ever had a terminated pregnancy
No		78.56
Yes		21.44
Ever had a child or neonatal death
No		95.17
Yes		4.83
Smokes
No		79.89
Yes		20.11
Has given birth in last five years
No		48.43
Yes		51.57
Continuous variables	Mean	(SD)

Age (predisposing)	32.33	8.69
Age at birth of first child (empowerment)	19.37	3.06

Included only the women who had given birth in the past five years (unweighted n = 4182)

To test the effect of women’s empowerment on health service use, we regressed women’s use of health services in the past 12 months on the independent and control variables using the total sample in Model 1. We then selected only the women who had given birth in the past five years (unweighted n = 4,182) and regressed their use of antenatal (Model II), delivery (Model III), and postnatal care (Model IV) on the independent and control variables. Results are presented in Table 2. Adjusted odds ratio of 1 or close to 1 indicated absence of relationship between the independent variable and the outcome variable; that is, there was no difference between the users and nonusers of health services. An odds ratio greater than 1 implied that the odds of a woman using health services improved when the value on the associated independent variable increased, and an odds ratio of less than 1 indicated that the odds of a woman using health services dropped when the associated independent variable increased.

Table 2.

Weighted predicted odds of health service use by women who had given birth in past five years

Parameters	Odds ratio
	Health service utilization (Model I)	Received antenatal care (Model II)	Delivered at health facility (Model III)	Received postnatal care (Model IV)
Empowerment related measures
Education (no education = 0)
Primary (1-5th grade)	1.25 *	1.79 ***	1.65 **	1.46*
Secondary: 6-10 but not SLC	1.58 ***	2.35 ***	2.73 ***	2.17***
SLC and above	1.76 ***	18.62 ***	5.10 ***	2.21*
Heard about STD and AIDS (yes = 1; no = 0)	1.55 ***	1.38 **	1.45 *	1.43
Wife beating (OK = 1; Not OK = 0)	0.99	1.04	1.04	1.72**
Makes financial and health decisions (yes = 1; no = 0)	1.01	1.15	1.12	1.39 *
Age at first birth	1.01	1.07***	1.16***	1.01
Predisposing variables
Age	0.97 ***	0.95***	0.95***	0.98
Gender of HH head (female = 1; male = 0)	0.95	0.98	1.05	0.80
Privilege caste (Brahmin/Chhetri = 0)
Occupational caste	0.75 **	0.65**	0.89	0.65
Indigenous non-caste	0.88	0.52***	1.06	0.64*
Religion (Hindu = 1; other = 0)	1.09	1.09	1.23	0.80
Enabling variables
Ecological region (Hill = 0)
Mountain	0.83	0.71 *	0.56**	0.82
Terai	0.87	2.40 ***	1.08	1.64**
Development region (Central = 0)
Eastern	0.96	1.24	0.51***	0.91
Western	0.96	0.51 ***	0.41***	1.02
Mid-western	1.00	0.59 ***	0.72	0.50**
Far western	0.74 **	1.11	0.45***	0.96
Distance a big problem (yes = 1; no = 0)	0.87 *	0.80*	0.70**	1.03
Money/cost a big problem (yes = 1; no = 0)	1.00	0.83	0.94	1.00
Quality care a big problem (yes = 1; no = 0)	0.91	1.04	0.95	0.74
Wealth index (Poor class = 0)
Middle class	0.94	1.55***	1.15	1.29
Upper class	1.08	2.38 ***	2.75***	2.02***
Need variables
Anemic (yes = 1; no = 0)	0.94	0.93	0.90	0.97
Body Mass Index, BMI (Normal = 0)
Underweight	1.22 **	1.21	0.88	1.03
Overweight	1.22	1.39	3.36 ***	0.88
Ever terminated pregnancy (yes = 1; no = 0)	1.07	1.08	1.03	0.82
Ever had child or neonatal death (yes = 1; no = 0)	0.79	0.73*	1.22	1.00
Smokes (yes = 1; no = 0)	1.01	0.69**	0.59*	0.89
Has given birth in last five years (yes = 1; no = 0)	2.26 ***
Wald χ²	586.83***	473.51***	566.72***	156.97***
Pseudo R²	0.10	0.19	0.26	0.09
Observations (unweighted)	7686	4135	4136	3256

Note: p* < .05, p** < .01, p*** < .001.

Among the indicators of empowerment, women’s education clearly improved service use even after controlling for a wide range of predisposing, enabling, needs and health risk behavior variables. The odds for women with primary education utilizing a health facility in the past 12 months were 25 percent higher than for women without any education. Among women who had given birth in the past five years, as compared to those without any education, those with primary education were 1.79, 1.65, and 1.46 times as likely to have used antenatal, delivery, and postnatal care respectively. The role of education was stronger with secondary education. Women with a school leaving certificate (SLC) and higher were 1.76 times as likely to have utilized a health facility in the past 12 months as women without any education. Moreover, of the women who had given birth in the past five years, those with an SLC (and higher) were 18.62, 5.10, and 2.21 times as likely to have utilized antenatal, delivery, and postnatal care respectively as their counterparts without any education. Women who had heard about STDs and AIDS were significantly more likely to use overall health services (OR = 1.55), antenatal care (OR = 1.38), and delivery care services (OR = 1.45) compared to women who had not heard about STDs and AIDS. In addition, a one-year increase in women’s age at birth of first child improved the odds of using antenatal and delivery care by 7 percent and 16 percent respectively. Interestingly, women’s attitude towards wife beating when wives failed to meet husband’s expectations and women’s involvement in intra-household decisions had no effect on their use of overall care, antenatal care, and delivery care. These two variables, however, significantly improved women’s use of postnatal care.

Among the control variables, women from privileged castes – Chhetri and Brahmin – were significantly more likely to have visited a health facility in the past 12 months and used antenatal care, but they were as likely to deliver at home and receive postnatal care as the less privileged caste or indigenous non-caste women. Household wealth was positively associated with maternity care use across the board. The odds of receiving antenatal, delivery, and postnatal care for women from the highest wealth index were 2.38, 2.75, and 2.02 times the odds for women from the lowest wealth index respectively. Of the three barriers to health service use, women who indicated that distance to health facility was a big problem were significantly less likely to use health services (OR = 0.87), antenatal (OR = 0.80), and delivery care (OR = 0.70), but their use of postnatal care was similar to those who did not see distance as a problem. Interestingly, the other two barriers – money and quality of care – were insignificant in every model. Among the need variables, overweight women were significantly more likely to have delivered at a health facility (OR = 3.36), but their use of antenatal and postnatal care was similar to their normal weight counterparts. Women who had experienced child or neonatal death were significantly less likely to have used antenatal care, but their use of delivery and postnatal care was similar to women who had not experienced child or neonatal death. Interestingly, mothers who smoked used antenatal and hospital delivery services significantly less than their nonsmoking counterparts.

Discussion

Globally, between 1990 and 2005 the maternal mortality ratio declined by only 5 % in developing countries, the ratio of decline was even lower (.03 percent) (United Nations, 2009). Pregnancy and delivery experiences continue to be ‘a privilege of the rich’ (United Nations, 2009: 26) and life threatening for many women in developing countries (Horton, 2009). The key reason for high maternal mortality is low use of antenatal, delivery, and postnatal care. Prevention of deaths due to hemorrhage and sepsis after delivery alone would drop maternal deaths by one-third (United Nations, 2009). It is, therefore, critical to understand the underlying social reasons for low use of maternal health services. In the discussion below, we highlight the key findings that predicted Nepali women’s utilization of antenatal, delivery, and postnatal care and the roles social workers can play in addressing the social problems that prevent women from their use of health services.

Given that there was a robust correlation between three of the five indicators of women’s empowerment and outcome variables, women’s empowerment should be placed within a broader context of public health initiatives to eventually reduce maternal mortality rate. Educated women were significantly more likely to seek treatment at critical times. This finding is consistent with previous studies (Anwar et al., 2008; Chakraborty et al., 2003; Furuta and Salway, 2006; Koblinsky et al., 2008; Mamun et al., 2006; Simkhada et al., 2006; World Health Organization [WHO], 2008). A strong relationship between health service use and women and girls’ education is telling. Yet nearly 50 percent of the women in Nepal have no education. As a part of the Millennium Development Goal 2, all member nations of the United Nations have pledged to achieve universal primary education by 2015. Given the high proportion of illiterate women around the world, it is unlikely that many countries in Asia and Africa will meet this goal by the target date. As we move beyond 2015, social workers from around the world must act with public health professionals to relieve millions of girls from their current obligations of picking rags, fetching water, fetching fuel wood, and attending to domestic chores that undermine their enrollment in schools. Increasing girls’ access to education should be framed not only as a means to reduce gender inequality but also as a means to advance overall health, especially maternal health (World Bank, 2000). In addition to the ongoing expansion of health facilities and training of health professionals, government and non-governmental organizations should draw upon the strengths of social workers in individual counseling, advocacy, and community organization to increase girls’ access to schools and young married women’s access to reproductive health services as part of a public health initiative. One such NGO that comes to our mind is Action Research and Training for Health (ARTH) in Rajasthan, India, which has both health professionals and social workers working jointly to improve maternal and child health. This NGO’s model of help is considered one of the most successful models in South Asia.

Similarly, women who had heard about STDs and AIDs were significantly more likely to use three of the four types of health services. However, nearly one in three previously pregnant women were not aware of STDs and AIDS. Here again, social workers can promote information about sexually transmitted diseases and AIDS, especially to poor, uneducated, low-caste, and marginalized women and improve their rates of health service utilization.

Women’s age at birth of their first child was an important variable; the younger the women were at birth of their first child, the lower the likelihood of their utilization of maternity-related health services. Average age of mothers at first birth was 19; 59 percent were teenagers (under 20 years of age) when their first child arrived. Almost 28 percent of the mothers were 17 and under when they gave birth to their first child. These women have the highest risk of developing maternity-related complications and are least likely to use health services. Social workers can work with health workers as a team to inform and counsel women and young teenage brides about reproductive health and contraceptives.

Women’s attitude toward wife beating and their intra-household decisions had no effect on their use of overall antenatal and delivery care. Future research should explore the nature of intra-household decisions that women are engaged in and the role they play in these decisions.

Several control variables were significant. Women from economically well-off households were significantly more likely to use health services. This finding underscores the importance of programs that strengthen a household’s wealth. Although Nepal has expanded its health infrastructures, this study showed that for many women, distance continues to be a barrier to health service use. This finding is consistent with literature (Anwar et al., 2008; Pagel et al., 2009; United Nations, 2009). Interestingly, cost and quality of care had no effect on use of health services after controlling for other factors. This finding is inconsistent with studies conducted in Nepal and Bangladesh that found cost and quality of care to be significant barriers to utilization of health services (Acharya and Cleland, 2000; Hotchkiss, 2001; Koblinsky et al., 2008; Koenig et al., 2007). Further studies are needed to unravel the effect of cost and quality of care on health services use.

Conclusion

Social work values are in line with gender empowerment. High illiteracy among girls, early marriage (against national laws), and lack of power, capability, and freedom to decide when and how often to reproduce all contribute to high rates of maternal mortality in developing countries. Social workers have the skills to tackle these problems at the individual, family, and community level. They can intervene in social practices that undermine women’s power, including practice of early marriage and differential priorities given to sons and daughters in their nutrition and education. They can also help marginalized women and girls gain access to power through their increased access to education, fertility choices, and health services.

Social workers have much to contribute in increasing use of maternal health services. In many countries in South Asia, a young woman’s decision to use health services, including the selection of place of delivery (home or institution), are not hers alone. In Nepal other family members in power – husband, mothers-in-law, sisters-in-law, and other significant members – influence these decisions (Simkhada et al., 2006). Among the women who delivered at home in India and Bangladesh, over 75 percent did so because they or their family members (including husband, in-laws, and other relatives) did not see it as necessary to go to a hospital for delivery (Koblinsky et al., 2008; Pandey and Lin, 2011). Similarly, young brides are often unaware of different methods of fertility choices; when they become pregnant, they often rely on their family to access information necessary to follow through pregnancy and childbirth. These are the sorts of social practices that health professionals may not be able to solve without input from social workers. Social workers can act as an advocate on behalf of pregnant women, especially marginalized women and women with complicated pregnancies, and work closely with their families and communities to ensure safe deliveries.

Moreover, social workers can play a critical role in the empowerment of the whole community. For example, a low-caste woman may choose to deliver at home in spite of having affordable health facility near her home because she feels discriminated against by high-caste health care providers. As caste-based discrimination is illegal in most South Asian countries, social workers can empower low-caste women to demand quality services and increase health service use among these marginalized populations. Thus, social work educators and researchers interested in women’s health can empower marginalized women, who are dying in tens of thousands and whose deaths could be prevented, sometimes with such simple means as increasing a year of education, delaying marriage and childbirth by one year, or changing the place of delivery from home to a health facility.

Finally, a limitation of this study is that the measures of empowerment reflect individual psychological empowerment variables and ignore collective community-level empowerment measures due to data limitations. Future studies should consider including these variables in the model. Also, the current study did not examine the relationship between women’s political empowerment and maternal mortality. Future studies should keep in mind that women’s political empowerment may play an important role in reducing maternal mortality. The study also did not consider the contributions of international social work communities that are working with local NGOs to reduce maternal mortality. In future, researchers should document some of the key roles social workers have been playing at the community level to reduce maternal mortality.

Footnotes

Author biographies

Shanta Pandey is an Associate Professor in the George Warren Brown School of Social Work, Washington University, USA.

Gyanesh Lama is a Doctoral Student in the George Warren Brown School of Social Work, Washington University, St Louis, USA.

Haenim Lee is an MSW student in the George Warren Brown School of Social Work, Washington University, St Louis, USA.

References

Acharya

L.B.

Cleland

(2000) ‘Maternal and Child Health Services in Rural Nepal: Does Access or Quality Matter More?’, Health Policy and Planning 15(2): 223–9.

Aday

L.A.

Andersen

R.M.

(1974) ‘A Framework for the Study of Access to Medical Care’, Health Services Research 9(3): 208–20.

Andersen

R.M.

(1995) ‘Revisiting the Behavioral Model and Access to Medical Care: Does it Matter?’, Journal of Health and Social Behavior 36: 1–10.

Andersen

R.M.

Newman

(1973) ‘Societal and Individual Determinants of Medical Care Utilization in the United States’, Milbank Memorial Fund Quarterly 51: 95–124.

Anwar

Sami

Akhtar

Chowdhury

M.E.

Salma

Rahman

Koblinsky

(2008) ‘Inequity in Maternal Health-care Services: Evidence from Home-based Skilled-birth-attendant Programmes in Bangladesh’, Bulletin of the World Health Organization 86(4): 252–9.

Bloom

S.S.

Wypij

Gupta

M.D.

(2001) ‘Dimensions of Women’s Autonomy and the Influence on Maternal Health Care Utilization in a North Indian City’, Demography 38(1): 67–78.

Chakraborty

Islam

M.A.

Chowdhury

R.I.

Bari

Akhter

H.H.

(2003) ‘Determinants of the Use of Maternal Health Services in Rural Bangladesh’, Health Promotion International 18(4): 327–37.

Chowdhury

Islam

M.A.

Gulshan

Chakraborty

(2007) ‘Delivery Complications and Healthcare-seeking Behavior: The Bangladesh Demographic Health Survey, 1999–2000’, Health and Social Care in the Community 15(3): 254–64.

Davidson

P.L.

Andersen

R.M.

Wyn

Brown

E.R.

(2004) ‘A Framework for Evaluating Safety-net and Other Community-level Factors on Access for Low-income Populations’, Inquiry 41(1): 21–38.

10.

Ehrhardt

A.A.

Sawires

McGovern

Peacock

Westen

(2009) ‘Gender, Empowerment, and Health: What is it? How Does it Work?’, Journal of Acquired Immune Deficiency Syndromes 51: S96–S105.

11.

Furuta

Salway

(2006) ‘Women’s Position Within the Household as a Determinant of Maternal Health Care Use in Nepal’, International Family Planning Perspectives 32(1): 17–27.

12.

Gelberg

Andersen

R.M.

Leake

B.D.

(2000) ‘The Behavioral Model for Vulnerable Populations: Application to Medical Care Use and Outcomes for Homeless People,’ Health Services Research 34(6): 1273–302.

13.

Horton

(2009) ‘What Will it Take to Stop Maternal Deaths?’, The Lancet 374(9699): 1400–2.

14.

Hotchkiss

D.R.

(2001) ‘Expansion of Rural Health Care and the Use of Maternal Services in Nepal’, Health & Place 7: 39–45.

15.

Jahan

(2007) ‘Securing Maternal Health Through Comprehensive Reproductive Health Services: Lessons from Bangladesh’, American Journal of Public Health 97(7): 1186–190.

16.

Kabeer

(1994) Reversed Realities: Gender Hierarchies in Development Thought. London: Verso.

17.

Kabeer

(1997) ‘Women, Wages and Intra-household Power Relations in Urban Bangladesh’, Development and Change 28: 261–302.

18.

Kabeer

(1999) ‘Resources, Agency, Achievements: Reflections on the Measurement of Women’s Empowerment’, Development and Change 30: 435–64.

19.

Kabeer

(2005) ‘Gender Equality and Women’s Empowerment: A Critical Analysis of the Third Millennium Development Goal 1’, Gender & Development 13(1): 13–24.

20.

Khan

K.S.

Wojdyla

Say

Gülmezoglu

A.M.

Van Look

P.F.A.

(2006) ‘WHO Analysis of Causes of Maternal Death: A Systematic Review’, The Lancet 367: 1066–74.

21.

Koblinsky

Anwar

Mridha

M.K.

Chowdhury

M.E.

Botlero

(2008) ‘Reducing Maternal Mortality and Improving Maternal Health: Bangladesh and MDG 5’, Journal of Health, Population, & Nutrition 26(3): 280–94.

22.

Koenig

M.A.

Jamil

Streatfield

P.K.

Saha

Al-Sabir

Arifeen

S.E.

Hill

Haque

(2007) ‘Maternal Health and Care-seeking Behavior in Bangladesh: Findings from a National Survey’, International Family Planning Perspective 33(2): 75–82.

23.

Malhotra

Schuler

S.R.

Boender

(2002) ‘Measuring Women’s Empowerment as a Variable in International Development’. Available at: http://siteresources.worldbank.org/INTEMPOWERMENT/Resources/486312-1095970750368/529763-1095970803335/malhotra.pdf

24.

Mamun

A.A.

Padmadas

S.S.

Khatun

(2006) ‘Maternal Health During Pregnancy and Perinatal Mortality in Bangladesh: Evidence from a Large-scale Community-based Clinical Trial’, Pediatric and Perinatal Epidemiology 20: 482–90.

25.

Ministry of Health and Population, New ERA and Macro International (2007) Nepal Demographic and Health Survey 2006. Kathmandu, Nepal: Ministry of Health and Population, New ERA and Macro International.

26.

Nussbaum

M.C.

(2000). Women and Human Development: The Capabilities Approach. Cambridge, UK: Cambridge University Press.

27.

Pagel

Lewycka

Colbourn

Mwansambo

Meguid

Chiudzu

Utley

Costello

A.M.

(2009) ‘Estimation of Potential Effects of Improved Community-based Drug Provision, to Augment Health-facility Strengthening, on Maternal Mortality due to Postpartum Hemorrhage and Sepsis in Sub-Saharan Africa: An Equity-effectiveness Model’, The Lancet 374(9699): 1441–8.

28.

Pandey

Lin

(2011) Predictors of Antenatal Care Use and Institutional Delivery in India. Working Paper. George Warren Brown School of Social Work, Washington University: St. Louis, MO.

29.

Phillips

K.A.

Morrison

K.R.

Andersen

Aday

L.A.

(1998) ‘Understanding the Context of Healthcare Utilization: Assessing Environmental and Provider-related Variables in the Behavioral Model of Utilization’, Health Services Research 33(3): 571–96.

30.

Pradhan

(2003) ‘Measuring Empowerment: A Methodological Approach’, Development 46(2): 51–7.

31.

Seipel

M.M.O.

(1992) ‘Promoting Maternal Health in Developing Countries’, Health & Social Work 17(3): 200–6.

32.

Sen

(1999) Development as Freedom. Oxford: Oxford University Press.

33.

Simkhada

Teijlingen

E.R.

Porter

Simkhada

(2006) ‘Major Problems and Key Issues in Maternal Health in Nepal’, Kathmandu University Medical Journal 4(14): 258–63.

34.

United Nations Development Programme (UNDP) (1995) Human Development Report. New York: Oxford University Press.

35.

United Nations (2000a) World’s Women: Trends and Statistics. New York: United Nations.

36.

United Nations (2000b) ‘Resolution Adopted by the General Assembly, United Nations Millennium Declaration’. Retrieved from http://www.un.org/millennium/declaration/ares552e.pdf

37.

United Nations (2009) The Millennium Development Goals Report 2009. Retrieved from http://www.un.org/millenniumgoals/pdf/MDG_Report_2009_ENG.pdf

38.

United Nations (2010) The Millennium Development Goals Report. Retrieved from http://unstats.un.org/unsd/mdg/Resources/Static/Products/Progress2010/MDG_Report_2010_En.pdf

39.

Williams

(2005) Measuring Gender and Women’s Empowerment Using Confirmatory Factor Analysis. Boulder, CO: Population Program, Institute of Behavioral Science, University of Colorado.

40.

World Bank (2000) Nepal: Operational Issues and Prioritization of Resources in the Health Sector. Report No. 19613. Health, Nutrition and Population Unit, South Asia Region: The World Bank. Retrieved from http://www.healthnet.org.np/healthstat/worldbank/nepaloperational.pdf

41.

World Health Organization (2008) The World Health Report 2008: Primary Health Care Now More Than Ever. Geneva, Switzerland: WHO Press.