Social factors determining the experience of blindness among pregnant women in developing countries: the case of India.
|Abstract:||Approximately 10 million pregnant women around the world develop night blindness annually. In India, one in 11 pregnant women suffers from night blindness. This study used a nationally representative sample of 35,248 women from India between the ages of 15 and 49 who had given birth in the past five years to understand the effect of women's empowerment on developing blindness during pregnancy. Findings from logistic regression showed that several empowerment-related factors, including women's increased age at the birth of their first child, high school education, and participation in their own health care decisions, significantly reduced the odds of developing blindness, whereas their experience of spousal control, humiliation, and physical abuse significantly increased the odds of developing blindness during pregnancy. Interestingly, antenatal care visits that included nutritional advice and iron supplements during pregnancy had no effect on blindness after controlling for other factors. To reduce blindness during pregnancy in India, more attention should be given to delaying age at marriage and first birth; improving women and girl's education and autonomy; and implementing strategies to reduce spousal control, humiliation, and abuse.|
Developing countries (Health aspects)
Night blindness (Risk factors)
Pregnant women (Social aspects)
Women's issues (Research)
|Publication:||Name: Health and Social Work Publisher: Oxford University Press Audience: Academic; Professional Format: Magazine/Journal Subject: Health; Sociology and social work Copyright: COPYRIGHT 2012 Oxford University Press ISSN: 0360-7283|
|Issue:||Date: August, 2012 Source Volume: 37 Source Issue: 3|
|Topic:||Event Code: 290 Public affairs; 310 Science & research|
|Geographic:||Geographic Scope: India Geographic Code: 0DEVE Developing Countries; 9INDI India|
Globally, pregnant women in 66 countries experience s night
blindness, with moderate to severe public health significance, and
approximately 10 million pregnant women (8 percent of pregnant women)
develop night blindness annually (World Health Organization, 2009).
Night blindness during pregnancy is common in many South Asian
countries. The conditions become more severe in the second and third
trimester and impair pregnant women's ability to be effective
workers during day-time or night-time or both (Christian, Bentley,
Pradhan, & West, 1998). In India, one in 11 pregnant women suffers
from night blindness (Katz et al., 2009). In some parts of India and
Nepal, the prevalence of night blindness during pregnancy is much higher
(16 percent to 24 percent) and is often accepted as a normal symptom of
pregnancy so that treatments are not sought (Christian, Bentley, et al.,
1998; Toteja, Singh, Dhillon, & Saxena, 2002).
Women suffering from night blindness become dependent on their families, and their risk of accidents, personal injury, and death increases (Christian, Bentley, et al., 1998; Christian et al., 2000). Night blindness during pregnancy also has negative health consequences on children. Tielsch et al. (2008) documented that maternal night blindness in India was associated with an increased risk of low birth weight, diarrhea, dysentery, acute respiratory illness, poor growth, underweight, and stunting among children. In Nepal, studies examining the relationship between maternal night blindness and the risk of mortality among infants in the first 6 months of life found that mortality was 3 percent higher among infants of women who had night blindness during pregnancy (Christian et al., 2001).
Most studies have implied poor dietary intake and vitamin A deficiency as key risk factors for night blindness among pregnant women. These studies have documented that pregnant women in South Asia are vulnerable to multiple micronutrient deficiencies, including inadequate vitamin A--rich food and protein malnutrition, resulting in health-related complications such as anemia and night blindness (see, for example, Christian, 2002; Christian, Bentley, et al., 1998; Christian, Schulze, Stoltzfus, & West, 1998; Christian, Thorne-Lyman, et al., 1998; N. Cohen et al., 1987; Haskell et al., 2005; Hussain & Kvale, 1996; Jiang, Christian, Khatry, Wu, & West, 2005; Semba et al., 2010). In addition, such nondietary factors as women's low socioeconomic status--including early marriage, maternal illiteracy, and poverty--increase the risk of night blindness in pregnancy (Christian, West, et al., 1998; Katz et al., 2009; Semba, de Pee, Panagides, Poly, & Bloem, 2003). These same factors reflect that these women have limited control over their own dietary intake.
To prevent night blindness among pregnant women, the World Health Organization (WHO) has recommended three types of micronutrient-related interventions (or a combination of these): to provide nutritional education to encourage consumption of a vitamin A--rich diet, including vegetables (such as carrot, pumpkin, papaya) and dairy products (whole milk, yogurt, cheese); to fortify staple food (such as sugar) with vitamin A; and to supplement pregnant women with vitamin A, especially during the last 12 weeks of pregnancy (WHO, 2009, 2011). In addition, the WHO's Partnership for Maternal, Newborn and Child Health program has recommended that a woman with a normal pregnancy make at least four visits to a skilled health attendant during her pregnancy (and more visits by women with pregnancy complications) and receive maternal nutrition information and iron supplements to reduce under-nutrition, anemia, night blindness, and underweight among pregnant women and new mothers (United Nations, 2000). In response, South Asian countries like India, Nepal, and Bangladesh that have high rates of night blindness among pregnant women have expanded their antenatal services and offered a wide range of pregnancy-related nutrition education and vitamin A supplements free of charge. For example, the government of India has rapidly expanded grass roots organizations, anganwadi (courtyard or a place around the house where people socialize) centers, under the Integrated Child Development Services scheme that began in 1975, with support from the UNICEF. Run by trained, community based anganwadi workers, the main focus of these centers is to provide supplementary nutrition for children and pregnant and lactating mothers.
India has especially targeted the eight Empowered Action Group (EAG) states (Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Orissa, Rajasthan, Uttaranchal, and Uttar Pradesh), which represent nearly half of India's population and report very high maternal and child mortality rates (Arokiasamy & Gautam, 2008; Nandi, 2010).
Existing studies have not systematically examined the extent to which such nondietary factors as women's empowerment are associated with night blindness while controlling for diet-related interventions through antenatal care services including nutrition advice and iron and vitamin A supplements. In this study, we examined the association between women's empowerment and night or day blindness during pregnancy using nationally representative data from India. We controlled for women's barriers to use of health services, their use of antenatal care services and nutrition supplements, and demographic characteristics.
To better understand empowerment as a protective factor for blindness among pregnant women, we rely on the social relations framework (SRF) and an ecological framework of domestic violence (EFDV). According to the SRF, gendered relations place women in subordinate social positions and limit their capabilities to exercise their full potentials (Kabeer, 1994; Kabeer & Subrahmanian, 1999; Nussbaum, 2000; Sen, 1999). Women's poor nutrition, low education, early marriage, economic dependence on their husbands, and restricted physical mobility are some of the consequences of gendered social relations. Gender gaps in education and paid employment, along with overrepresentation of women in informal employment settings that often lack benefits and security, are widely reported (United Nations, 2010). The effects of women's subordinate role are shown in a study in North India, which found that a young bride in her husband's family had very little voice and autonomy; her mother-in-law and her husband controlled her mobility, including her ability to visit her parents and friends (Gupta, 1995).
In addition, women's capacity to take care of themselves during pregnancy is further compromised if they experience control, humiliation, and physical abuse (Jewkes, 2002). The EFDV underscores the role of women's lack of power in relationships and the influence of socially constructed messages about women's proper place in society in women's experiences of gender-based violence. Heise (1998) used this ecological approach to conceptualize gender-based violence as "a multifaceted phenomenon grounded in an interplay among personal, situational, and sociocultural factors" (pp. 263-264), including male control of family wealth, women's isolation and lower economic status, and enforcement of rigid gender roles. Studies from India have documented that between 16 percent to 56 percent of Indian women suffer some form of domestic violence, including physical, psychological, sexual or other forms of violence (Babu & Kar, 2009; Koenig, Stephenson, Ahmed, Jejeebhoy, & Campball, 2006).
Some of the widely used indicators of empowerment include age at marriage, education, income, employment, and the ability to exercise financial and mobility choices (Kabeer, 1999; Malhotra, Schuler, & Boender, 2002; United Nations Development Programme, 1995; Williams, 2005). In addition, acts of male dominance through the perpetration of control, humiliation, and physical abuse reduce a woman's self-worth (see, for example, Faramarzi, Esmailzadeh, & Mosavi, 2005; Heise, 1998; Rennison & Welchans, 2000) and increase her risk for experiencing infant mortality (Koenig et al., 2010). Using the SRF and EFDV frameworks, we proposed the following three hypotheses:
1. Women's better education, employment, and capacity to make intrahousehold decisions reduce their risk for developing blindness during pregnancy.
2. Women's lower age at the birth of their first child (a proxy for early marriage) increases blindness during pregnancy.
3. Women's experience of domestic abuse (control, humiliation, and physical violence) from their spouses increases the risk for blindness during pregnancy.
The data come from the 2005 through 2006 National Family Health Survey of India, conducted by the International Institute for Population Sciences, with technical assistance from Macro International. Macro International has been conducting detailed surveys on population health in more than 90 countries every five years since the 1990s, with a domestic violence module implemented in some countries (Kishor & Johnson, 2004). In India, the survey was conducted under the auspices of the Ministry of Health and Family Welfare and with funding from multiple sources. The study used a standardized questionnaire to interview a nationally representative sample of 124,385 women between the ages of 15 and 49 from households across India. A multistage sample design was used to interview urban and rural populations in all 29 states with the help of trained interviewers who conducted fieldwork to collect the data (for details, see International Institute for Population Sciences & Macro International, 2007). The questionnaire was translated into local language and pretested in every state. Interviewers were trained to use the domestic violence module and were instructed to proceed with this section of the interview "only if privacy was obtained" (International Institute for Population Sciences & Macro International, 2007, p. 496). For this study, we selected all women who had given birth in the past five years. This resulted in a sample of 36,850 mothers. Respondents who answered "don't know" to any of the survey questions used were excluded, resulting in a sample size of 35,248.
Variables and Measures
Dependent Variables. We used a dichotomous outcome variable, blindness, that included experience of night or daytime blindness during pregnancy. If a woman had experienced blindness during the day or night during the most recent pregnancy, she was coded as 1; all others were coded as 0.
Independent Variables. A set of variables associated with women's empowerment in the literature served as independent variables in the logistic regression. These variables included women's formal education; employment status; age at birth of first child (also a proxy for age at first marriage); role in intrahousehold decisions; and experience of control, humiliation, and physical abuse by their husbands. Education included four dummy variables: (1) no education, (2) primary education (one to five standards of schooling), (3) junior/ middle school education (six to eight standards), and (4) high school and beyond (beyond eighth standard). If women were working or employed at the time of interview, they were coded as 1; otherwise, they were coded as 0. Age at birth of first child was a continuous variable measured in years. Four variables representing women's participation in intrahousehold decisions and three domestic abuse variables that signified husband's control, humiliation, and physical abuse were constructed (see Table 1 for details).
Control Variables. We controlled for use of antenatal care and supplementary nutrition services, barriers to use health services, and individual and household-level demographic variables. Antenatal care visit included three dummy variables: Women who did not use any antenatal care (reference group), those who used one to three antenatal care visits, and those who used four or more antenatal care visits. Women who received supplementary nutrition from local anganwadi center were coded as 1; those who did not were coded as 0.
Women who were advised on pregnancy nutrition at antenatal visits were coded as 1; otherwise, they were coded as 0. To capture women's consumption of micronutrients including vitamin A, we explored their food habits. To this end, if women generally consumed milk, beans, dark green vegetable, fruit, eggs, fish, and meat on daily or weekly basis at the time of interview, they were coded as 1; otherwise, they were coded as 0. To understand if barriers to use of health services led women to live with night blindness, we examined whether women considered distance to health facility, cost of care, and quality of health facility as barriers to use health services (see Table 1 for details).
Demographic variables included caste, religion, household wealth index, rural/urban residence, and residence in EAG states. Caste and ethnicity was coded as 1 if women belonged to a scheduled caste, scheduled tribe, or other underprivileged class; otherwise, it was coded as 0. Religion was coded as 1 if the respondents were Hindu; otherwise, it was coded as 0. For wealth index, the National Family Health Survey data classified households into five wealth indices: poorest, poorer, middle, richer, and richest. We recoded this variable into three dummy variables: poor, middle, and rich class. Poor class included poorest and poorer, the middle class remained unchanged, and the rich class included richer and the richest. Residence was coded as 1 if the respondents lived in an urban area, including mega city, large city, small city, large town, and small town; otherwise, it was coded as 0. For EAG states, women who lived in Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Orissa, Rajasthan, Uttaranchal, and Uttar Pradesh were coded as 1; otherwise, they were coded as 0.
Data Analysis Techniques and Diagnostic Tests
Our dependent variable was binomial. Thus, we used descriptive statistics and logistic regression analyses with SAS 9.2 version to assess the relationships between blindness during pregnancy and the independent variables, controlling for other factors. As recommended by Rutstein and Rojas (2006), we weighted all our descriptive results but not the regression results, to avoid biased estimates.
Two variables were excluded from logistic regression--vegetable consumption and nutrition advice. There was not much variance in vegetable consumption as more than 93 percent of the respondents consumed them. In addition, antenatal care and nutrition advice were perfectly correlated, which implied that nutrition advice was given during women's antenatal care visits. We diagnosed any serious evidence of multicollinearity among the remaining independent variables using the tolerance values and the variance inflation factors (VIFs). A VIF of 10 or more or tolerance values of .10 or less indicate serious problems of multicollinearity (Cohen, Cohen, West, & Aiken, 2003; Fox, 1991). In our model, the highest VIF was 3.43, and the lowest tolerance value was .29, which were well within the acceptable range. Large sample size enabled us to retain all theoretically relevant variables available to us and to build a theory-driven model (Hosmer & Lemeshow, 2000).
Logistic regression used 19,902 respondents out of 35,248 as a result of missing values (44 percent). Most of the missing values were generated by domestic violence variables. Given the large sample size and high rate of missing values, we did not use imputation. We examined whether respondents with missing values were significantly different from respondents without missing values in experiencing blindness during pregnancy by regressing the variable blindness on a new dichotomous variable with values of 1 and 0, representing missing and not missing, respectively. The bivariate logistic regression result showed that the respondents with missing values were not significantly different from those without missing values in their experience of blindness during pregnancy (b = .04, Wald [chi square] = 1.61, p = .20).
To assess the fitted model's overall departure from the observed data, we used Hosmer and Lemeshow's goodness-of-fit test. If the difference between the observed and predicted frequencies is high then the obtained chi-square value would be high and the null hypothesis of a close fit between observed and predicted probabilities would be rejected (Hosmer & Lemeshow, 2000). A small chi-square value implies that the model fits the data. In addition, we conducted tests to assess predictive ability of the models.
Approximately 12 percent of women in India experienced blindness during their last pregnancy. Compared with women who did not experience blindness during pregnancy, women who experienced blindness were a year younger at their first marriage (16 versus 17) and at the age of first birth (19 versus 20). Also, compared with women without blindness, a higher proportion of women with blindness had no formal education (63 percent versus 45 percent) and experienced control (58 percent versus 43 percent), humiliation (23 percent versus 14 percent), and physical abuse (51 percent versus 37 percent) from their husbands (see Table 2). A higher percentage of women with blindness were poor, expressed concern over distance, cost and quality of their closest health facilities, and resided in EAG states. Compared with women without blindness, a lower proportion of women with blindness used antenatal care visits and consumed milk, fruits, eggs, meat, or fish.
Effects of Empowerment on Developing Blindness During Pregnancy
Multiple logistic regression was used to predict the effects of women's empowerment on blindness during pregnancy, controlling for use of antenatal care and nutrition supplements, barriers to health service use, and individual and household-level characteristics. Because there was no compelling reason to enter variables in certain order, both the independent and control variables were entered simultaneously, and the results are presented in Table 3. The model was statistically significant (Wald [chi square] = 867.57, p < .000).
The results of logistic regression showed that the independent variables--mothers' education beyond the eighth standard; ability to exercise intrahousehold decisions; age at birth of first child; and spousal control, humiliation, and physical abuse--significantly affected blindness during pregnancy, as hypothesized. The odds of women with more than eighth standard of education developing blindness during pregnancy were 23 percent lower than those without any formal education. In a similar manner, women who had a say in their own health care decisions were 18 percent less likely to develop blindness during pregnancy compared with their counterparts who did not have a say in such decisions. In addition, a one-year increase in age at birth of first child reduced the odds of developing blindness during pregnancy by 3 percent. Finally, women whose husbands controlled, humiliated, and abused them had 49 percent, 31 percent, and 16 percent higher odds, respectively, of developing blindness during pregnancy as compared to women who did not experience such violence from their partners.
With regard to the control variables, wealthier, urban, and Hindu women were less likely to develop blindness during pregnancy as compared with their poor, rural, and non-Hindu counterparts. Women from EAG states had 19 percent higher odds of developing blindness during pregnancy compared with women residing in non-EAG states. Women who were concerned about the distance to a health facility, cost of health care, and quality of care were significantly more likely to develop blindness during pregnancy. Interestingly, antenatal care visits and iron supplement did not have any influence on blindness during pregnancy after controlling for other factors. Women who received supplemental nutrition during pregnancy had 14 percent higher odds of experiencing blindness. With regard to diet, women who drank milk and consumed meat or fish products at least once a week were significantly less likely to experience blindness than their counterparts who did not drink milk or consume meat or fish products that often.
Goodness-of-Fit Test and Sensitivity Analysis
The chi-square for the Hosmer and Lemeshow goodness-of-fit test and the corresponding p value was insignificant [[chi square](8, N = 19,902) = 13.35, p = .10], indicating that the observed and model predicted frequencies were not statistically different and that the model had a good fit. In addition, to assess misclassification of binary outcome variables, we conducted sensitivity analyses. The results for sensitivity, specificity, false positive, and false negative at three different probability cutoff points, chosen to reflect prevalence of blindness in India, are reported in Table 4. Correct classification rates ranged from 62 percent to 89 percent. Further, as sensitivity and specificity rely on a single cutoff point to classify a response, we also evaluated the receiver operating characteristic curve derived by plotting true response (sensitivity) and false response (1-specificity), which provides a more complete assessment of accuracy (Hosmer & Lemeshow, 2000). Theoretically, the value of concordance index (c statistic) or the area under the curve ranges from .50 to 1, where .50 suggests that the model randomly predicts the response, and 1 indicates that the model perfectly predicts the response. In other words, the closer the value of c to 1, the higher the level of correct classification. In our model, the c statistic equals .70, which indicates that the model has acceptable levels of discrimination (Hosmer & Lemeshow, 2000).
Using the SRF and EFDV frameworks, we examined whether women's empowerment-related variables--including age at the birth of their first child, formal education, role in intrahousehold decisions, and domestic abuse--were associated with experiencing blindness during pregnancy while controlling for other variables. The results provide evidence that women's empowerment serves as a protective factor against developing blindness during pregnancy. In this section, we discuss some of the key findings.
First, consistent with previous literature (Katz et al., 2009), the present study showed that low maternal education was a risk factor for experiencing blindness during pregnancy. The data we analyzed were cross-sectional. In a woman's life course in India, educational attainment, especially primary and secondary education, invariably precedes marriage and child birth. Nearly one in two women in India had no formal education, and only one in four women had education beyond the eighth standard (or year). If girls have secondary education, their risk of experiencing blindness during pregnancy as adults will be substantially reduced. Yet, globally, more than 69 million school-age children (18 million in southern Asia alone, many of whom were girls) were not in school in 2008; of those who were enrolled in school, girls tended to drop out more quickly than boys, especially in rural and poor households (United Nations, 2010). It will be difficult to eliminate blindness among pregnant women as long as gender-based inequalities in education persist. Our goal as social workers should be to empower women early on in their life course to eventually reduce blindness during pregnancy and other health disparities.
Second, women who were able to make decisions in personal health care were significantly less likely to develop blindness during pregnancy. Given this information, boosting women's status within the household is critical to attain improved health. A number of evidence-based strategies exist that may empower women within the household. Some of these include improving women's rights to property and access to financial institutions.
Third, experience of domestic abuse is a significant risk factor for developing blindness during pregnancy, controlling for other factors. Given that a high proportion of women in India are living in abusive relationships--with 45 percent experiencing control, 15 percent experiencing humiliation, and 38 percent experiencing physical abuse--strategies to reduce domestic violence must be introduced. This will have to include changing the social acceptance of gendered violence by both men and women. As long as social norms encourage "men's sense of entitlement and ownership of women, support the use of violence in conflict resolution, and condone the abuse of women" (Koenig et al., 2006, p. 137), it will be difficult to reduce blindness during pregnancy. Improving women's social status through education, property rights, and access to public health services must be a primary focus to prevent domestic violence and blindness during pregnancy.
Some studies have shown that if wives' occupational status surpasses that of their husbands, especially among women engaged in small businesses and fanning, this increase in gender empowerment may exacerbate the risk of domestic violence as men respond to this societal shift (Koenig, Ahmed, Hossain, & Khorshed Alam Mozumder, 2003; Majumdar, 2004-2005; Simister & Mehta, 2010). Other studies have clearly indicated that higher levels of education (of both husbands and wives) and household wealth serve as protective measures against physical violence (Babu & Kar, 2010; Christy-McMullin & Shobe, 2007; Koenig et al., 2006). To improve women's health, men's absolute control over property, household economics, and women's ability to generate income must be challenged (Rajan, 2004), and we must foster an increased level of awareness for both men and women about the negative effects of domestic violence (Majumdar, 2004-2005).
Fourth, younger age at the birth of a woman's first child increased the risk for blindness during pregnancy. This finding is consistent with the literature that indicates that younger women in India have a higher risk of suffering from vitamin A deficiency (Pathak et al., 2008), which could be associated with the fact that Indian women's status generally falls immediately after marriage, although it may level off and even improve with age over the life course (Gupta, 1995). Moreover, early marriage increases the risk for domestic violence against women in India (Babu & Kar, 2010; Speizer & Pearson, 2011). These findings further underscore the importance of delaying marriage and increasing access to reproductive health services among young brides. Given that the average age at first marriage was 17 (only 16 for women who experienced blindness during pregnancy) and the average age at birth of first child was 19, many of these women's marriages are taking place in violation of India's Prohibition of Child Marriage Act of 2006 that bans marriage of girls and boys before age 18 and 21, respectively (Ministry of Law and Justice of India, 2007).
Fifth, several control variables were significant. As noted in the previous literature, increased household wealth reduced the risk for blindness during pregnancy (Katz et al., 2009). However, nearly 45 percent of all pregnant women and 66 percent of all women who developed blindness during pregnancy were identified as being from a poor economic background. These were the most vulnerable women, for whom such poverty-eradication programs as access to financial institutions, property rights, and income generation activities might indirectly reduce the risk of blindness during pregnancy. Women in EAG states were more likely to experience blindness during pregnancy even after controlling for a wide range of factors. India has identified these as priority states; they have very low levels of female literacy rates, and most women give birth at home (Pandey & Lin, 2012). More effort must be placed to reach out to pregnant women in these states.
In addition, to combat blindness during pregnancy, the WHO has recommended promoting nutrition education, access to vitamin A--rich diets, and dietary diversification including consumption of vegetables and dairy products (WHO, 2011). Surprisingly, prenatal visits, also a proxy for nutritional advice and supplementary nutrition during pregnancy, did not reduce night blindness after controlling for other variables. Approximately 77 percent of pregnant women in India reported at least one prenatal visit, although only 37 percent reported four (or more) visits, the minimum number of visits recommended by the WHO in a normal pregnancy. More research is needed to understand the nature of services and advice women receive during their prenatal visits and to examine the extent to which women are able to execute such advice (nutritional or otherwise). It is possible that antenatal care and nutrition advice has helped reduce blindness during pregnancy among wealthier and empowered women but that the problem continues to be prevalent among disempowered and poor women who cannot execute the advice they receive. In addition, women who received supplemental nutrition during pregnancy had 14 percent higher odds of experiencing blindness during pregnancy. One possibility is that the nutrition program may have specifically targeted pregnant women with symptoms of blindness. Another possibility is that women who receive nutritional supplements may be giving away their meal to others (family or nonfamily) for various reasons. More research is needed to understand the nature, content, and users of supplemental nutrition.
This study has limitations. Some of the critical risk variables we controlled were poorly measured. For example, the questions about blindness during pregnancy were asked to women who had given birth in the past five years. The micronutrient consumption questions, however, asked in general how often the women consumed such items as milk, beans, dark green leafy vegetables, fruits, eggs, fish, and chicken during the week prior to the interview. The food that was available and consumed during pregnancy may have been different. The information about micronutrient consumption they provided may not have captured their nutritional intake during their pregnancy.
Future studies should measure the nature of micronutrient consumption during pregnancy. Also, the employment variable captured women's work status at the time of interview. It is not clear whether women had similar work status during their last pregnancy. Moreover, we are not sure how reliable the domestic violence variables were; they contributed the highest number of missing values, which could be because interviews were discontinued or that women did not feel comfortable answering these questions.
In addition, the WHO does not recommend vitamin A supplementation during pregnancy as part of routine antenatal care; however, in areas where there is a severe public health problem related to vitamin A deficiency (prevalence of night blindness is 5 percent or higher), vitamin A supplementation during pregnancy is recommended at doses of up to 10,000 international units (IU) daily or 25,000 IU weekly (oral doses) (WHO, 2011). Given that southern Asian countries (for example, India, Nepal, Bangladesh) have a high concentration of women experiencing night blindness, it is possible that some of the women in the sample may have received vitamin A doses. The survey we used did not ask if any of the women in the sample received vitamin A supplements during their pregnancy; thus, we were not able to account for use of vitamin A supplementation. In spite of these limitations, this study clearly shows that blindness among women during pregnancy in India is an outcome of gendered social relations that place women in subordinate positions.
The findings of this study have important implications. Given that women's status in India rises and falls during the life course (Gupta, 1995), to prevent night blindness during pregnancy, empowerment of women must start early. For instance, if we can enforce India's compulsory education of girls and boys and ensure their primary and secondary education, their risk of developing blindness when they get married and become pregnant is low. Moreover, delaying first childbirth is critical to prevent experience of night blindness during pregnancy. Social workers and community organizers can work with family and communities to enforce India's Prohibition of Child Marriage Act of 2006 (Ministry of Law and Justice of India, 2007) more aggressively. Once married, young brides in India benefit from being able to make critical intrahousehold decisions. Such evidence-based strategies as women's rights to property and access to financial institutions could empower married women within the household and may also reduce violence against them.
This study may also have implications in the United States. Globally, approximately 15 percent of pregnant women are vitamin A deficient (biochemically) and 8 percent are night blind (WHO, 2009). The WHO considers the United States and other countries with gross domestic product (GDP) per capita income higher than $15,000 to be free from vitamin A deficiency as a public health problem and calculates prevalence of night blindness for preschool children and pregnant women in the remaining 156 member states with GDP per capita of $15,000 or less (WHO, 2009); thus, it is not clear how prevalent is night blindness among pregnant women in the United States. There is evidence that anemia, iron deficiency, and vitamin A deficiency are common among refugee populations (woodruff et al., 2006). It is possible that immigrant and refugee women in the United States suffer from blindness during pregnancy, especially those arriving from countries that are known to have a high prevalence of this condition. Therefore, it makes sense to collect data to see if immigrant and refugee women in the United States suffer from blindness during pregnancy.
Amkiasamy, P., & Gautam, A. (2008). Neonatal mortality in the empowered action group states of India: Trends and determinants. Journal of Biosocial Science, 40, 183-201.
Bahu, B. V., & Kar, S. K. (2009). Domestic violence against women in eastern India: A population-based study on prevalence and related issues. BMC Public Health, 9, 129.
Babu, B. V., & Kar, S. K. (2010). Domestic violence in Eastern India: Factors associated with victimization and perpetration. Public Health, 124, 136-148.
Christian, P. (2002). Maternal nutrition, health, and survival. Nutrition Reviews, 60, S59-S63.
Christian, P., Bentley, M. E., Pradhan, R., & West, K. P. Jr. (1998). An ethnographic study of night blindness "ratauni" among women in the Terai of Nepal. Social Science & Medicine, 46, 879-889.
Christian, P., Schulze, K., Stoltzfus, R. J., & West, K. P. Jr. (1998). Hyporetinolemia, illness symptoms, and acute phase protein response in pregnant women with and without night blindness. American Journal of Clinical Nutrition, 67, 1237-1243.
Christian, P., Thorne-Lyman, A. L., West, K. P. Jr., Bentley, M. E., Khatry, S. K., Pradhan, E. K., et al. (1998). Working after the sun goes down: Exploring how night blindness impairs women's work activities in rural Nepal. European Journal of Clinical Nutrition, 52, 519-524.
Christian, P., West, K. P. Jr., Khatry, S. K., Katz, J., Shrestha, S. R., Pradhan, E. K., et al. (1998). Night blindness of pregnancy in rural Nepal--Nutritional and health risks. International Journal of Epidemiology, 27, 231-237.
Christian, P., West, K. P., Jr., Khatry, S. K., Kimbrough-Pradhan, E., LeClerq, S. C., Katz, J., et al. (2000). Night blindness during pregnancy and subsequent mortality among women in Nepal: effects of vitamin A and beta-carotene supplementation. American Journal of Epidemiology, 152, 542-547.
Christian, P., West, K. P., Jr., Khatry, S. K., LeClerq, S. C., Kimbrough-Pradhan, E., Katz, J., et al. (2001). Maternal night blindness increases risk of mortality in the first 6 months of life among infants in Nepal. Journal of Nutrition, 131, 1510-1512.
Christy-McMullin, K., & Shobe, M. A. (2007). The role of economic resources and human capital with woman abuse. Journal of Policy Practice, 6, 3-26.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences. Mahwah, NJ: Lawrence Erlbaum Associates.
Cohen, N., Rahman, H., Mitra, M., Sprague, J., Islam, S., Leemhuis de Regt, E., et al. (1987). Impact of massive doses of vitamin A on nutritional blindness in Bangladesh. American Journal of Clinical Nutrition, 45, 970-976.
Faramarzi, M., Esmailzadeh, S., & Mosavi, S. (2005). A comparison of abused and non-abused women's definitions of domestic violence and attitudes to acceptance of male dominance. European Journal of Obstetrics & Gynecology & Reproductive Biology, 122, 225-231.
Fox, J. (1991). Regression diagnostics. Newbury Park, CA: Sage Publications.
Gupta, M. D. (1995). Life course perspectives on women's autonomy and health outcomes. American Anthropologist, 97, 481-491.
Haskell, M. J., Pandey, P., Graham, J. M., Peerson, J. M., Shrestha, R. K., & Brown, K. H. (2005). Recovery from impaired dark adaptation in nightblind pregnant Nepali women who receive small daily doses of vitamin A as amaranth leaves, carrots, goat fiver, vitamin A-fortified rice, or retinyl palmitate. American Journal of Clinical Nutrition, 81, 461-471.
Heise, L. L. (1998). Violence against women: An integrated, ecological framework. Violence Against Women, 4, 262-290.
Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic regression. New York: John Wiley & Sons.
Hussain, A., & Kvale, G. (1996). Sustainability of a nutrition education programme to prevent night-blindness in Bangladesh. Tropical Medicine and International Health, 1, 43-51.
International Institute for Population Sciences & Macro International. (2007). National Family Health Survey (NFHS-3), 2005-06: India: Volume I. Retrieved from http://www.nflasindia.org/NFHS-3%20Data/ VOL-1/India_volume_I_corrected_17oct08.pdf
Jewkes, R. (2002). Intimate partner violence: Causes and prevention. Lancet, 359, 1423-1429.
Jiang, T., Christian, P., Khatry, S. K., Wu, L., & West, K. P., Jr. (2005). Micronutrient deficiencies in early pregnancy are common, concurrent, and vary by season among rural Nepali pregnant women. Journal of Nutrition, 135, 1106-1112.
Kabeer, N. (1994). Reversed realities: Gender hierarchies in development thought. London: Verso.
Kabeer, N. (1999). Resources, agency, achievements: Reflections on the measurement of women's empowerment. Development and Change, 30, 435-464.
Kabeer, N., & Subrahmanian, R. (Eds.). (1999). Institutions, relations and outcomes: A framework and case studies for gender-aware planning. New Delhi, India: Raj Press.
Katz, J., Tielsch, J. M., Thulasiraj, R. D., Coles, C., Sheeladevi, S., Yanik, E. L., & Rahmathulla, L. (2009). Risk factors for maternal night blindness in rural South India. Ophthalmic Epidemiology, 16, 193-197.
Kishor, S., & Johnson, K. (2004). Profiling domestic violence: A multi-country study. Retrieved from http://www. measuredhs.com/pubs/pdf/OD31/OD31.pdf
Koenig, M. A., Ahmed, S., Hossain, M. B., & Khorshed Alam Mozumder, A. B. (2003). Women's status and domestic violence in rural Bangladesh: Individual- and community-level effects. Demography, 40, 269-288.
Koenig, M. A., Stephenson, R., Acharya, R., Barrick, L., Ahmed, S., & Hindin, M. (2010). Domestic violence and early childhood mortality in rural India: Evidence from prospective data. International Journal of Epidemiology, 39, 825-833.
Koenig, M. A., Stephenson, R., Ahmed, S., Jejeebhoy, S. J., & Campball, J. (2006). Individual and contextual determinants of domestic violence in North India. American Journal of Public Health, 96, 132-137.
Majumdar, S. (2004-2005). In India, domestic violence rises with education. Women Lawyers Journal, 90, 24-26. Retrieved from http://heinonline.org/HOL/ Page?handle=hein.journals/wolj90&div=28&g_sent=1&collection=journals
Malhotra, A., Schuler, S. R., & Boender, C. (2002). Measuring women's empowerment as a variable in international development. Retrieved from http://siteresources. worldbank.org/INTEMPOWERAVIENT/Resources/ 486312-1095970750368/529763-1095970803335/ malhotra.pdf
Ministry of Law and Justice of India. (2007). The Prohibition of Child Marriage Act, 2006. Retrieved from http:// wcd.nic.in/cma2006.pdf
Nandi, A. (2010). If you build it, will they come? Demand-side incentives for public healthcare delivery in India. Retrieved from http://www.cddep.org/blog/posts/ if_you_build_it_will they_come_demand side incentives_public_healthcare_delivery_india
Nussbaum, M. C. (2000). Women and human development: The capabilities approach. Cambridge, England: Cambridge University Press.
Pandey, S., & Lin, Y. (2012, January). Mothers who are more likely to experience child mortality in India: The role of gender empowerment. Paper presented at the 16th Annual Conference of the Society for Social Work and Research, Washington, DC.
Pathak, P., Kapil, U., Raghuvanshi, R. S., Singh, V., Dwivedi, S. N., & Singh, R. (2008). Night blindness among pregnant women: A study in a rural block of Uttaranchal State, India. Sight and Life Magazine, 1, 22-24.
Rajan, R. S. (2004). Rethinking law and violence: The Domestic Violence (Prevention) Bill in India, 2002. Gender & History, 16, 769-793.
Rennison, C. M., & Welchans, S. (2000). Intimate partner violence (BJS Special Report No. NCJ 178247). Washington, DC: U.S. Department of Justice.
Rutstein, S. O., & Rojas, G. (2006). Guide to DHS statistics. Retrieved from http://www.measuredhs.com/pubs/ pdf/DHSG1/Guide_DHS_Statistics.pdf
Semba, R. D., de Pee, S., Panagides, D., Poly, O., & Bloem, M. W. (2003). Risk factors for nightblindness among women of childbearing age in Cambodia. European Journal of Clinical Nutrition, 57, 1627-1632.
Semba, R. D., de Pee, S., Sun, K., Akhter, N., Bloem, M. W., & Raju, V. K. (2010). Coverage of vitamin A capsule programme in Bangladesh and risk factors associated with non-receipt of vitamin A. Journal of Health Population and Nutrition, 28, 143-148.
Sen, A. (1999). Development as freedom. Oxford, England: Oxford University Press.
Simister, J., & Mehta, P. S. (2010). Gender-based violence in India: Long-term trends. Journal of Interpersonal Violence, 25, 1594-1611.
Speizer, I. S., & Pearson, E. (2011). Association between early marriage and intimate partner violence in India: A focus on youth from Bihar and Rajasthan. Journal of Interpersonal Violence, 26, 1963-1981.
Tielsch, J. M., Rahmathullah, L., Katz, J., Thulasiraj, R. D., Coles, C., Sheeladevi, S., & Prakash, K. (2008). Maternal night blindness during pregnancy is associated with low birthweight, morbidity, and poor growth in South India. Journal of Nutrition, 138, 787-792.
Toteja, G. S., Singh, P., Dhillon, B. S., & Saxena, B. N. (2002). Vitamin A deficiency disorders in 16 districts of India. Indian Journal of Pediatrics, 69, 603-605.
United Nations. (2000). World's women: Trends and statistics. New York, NY: United Nations.
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
United Nations Development Programme. (1995). Human development report. New York, NY: Oxford University Press.
Williams, J. (2005). Measuring gender and women's empowerment using confirmatory factor analysis [Working paper]. Boulder: University of Colorado at Boulder, Institute of Behavioral Science.
Woodruff, B. A., Blanck, H. M., Slutsker, L., Cookson, S. T., Larson, M. K., Duffield, A., & Bhatia, R. (2006). Anaemia, iron status and vitamin A deficiency among adolescent refugees in Kenya and Nepal. Public Health Nutrition, 9, 26-34.
World Health Organization. (2009). Global prevalence of vitamin A deficiency in populations at risk 1995-2005: WHO global database on vitamin A deficiency. Retrieved from http://whqlibdoc.who.int/publications/2009/ 9789241598019_eng.pdf
World Health Organization. (2011). Guideline: Vitamin A supplementation in pregnant women. Retrieved from http:// whqlibdoc.who.int/publications/2011/9789241501781_eng.pdf
Shanta Pandey, PhD, is associate professor, George Warren Brown School of Social Work, Washington University in St. Louis, One Brookings Drive, Box 1196, St. Louis, MO 63130; e-mail: email@example.com. Yuan Lin, MSW, is research and program evaluation specialist, Assisted Recovery Centers of America, LLC, St. Louis. Shannon Collier-Tenison, PhD, MSW, is associate professor and coordinator of the BSW program, School of Social Work, University of Arkansas at Little Rock. Jamie Bodden is a graduate student, George Warren Brown School of Social Work, Washington University in St. Louis.
Original manuscript received February 1, 2012
Accepted February 6, 2012
Advance Access Publication October 9, 2012
Table 1: Indicators of Participation in Intrahousehold Decisions, Domestic Abuse, and Barriers to Use of Health Services Indicator Question Participates in Who usually males the decisions about: (1) Health intrahousehold care for yourself. (2) Major household Purchases? decisions (3) Purchases for daily household needs. (4) Visits to your family or relatives? If women usually made decisions or if their opinion was considered in making decisions in each of these areas, they were coded as 1; otherwise, they were coded as 0. Domestic abuse Husband (1) He is jealous or angry if you talk to other controls men. (2) He frequently accuses you of being unfaithful. (3) He does not permit you to meet your female friends. (4) He tries to limit your contact with your family. (5) He insists on knowing where you are at all times. (6) He does not trust you with any money. If answer was yes" to any one of these items, husband controls was coded as 1; otherwise, it was coded as 0. Husband Does your husband: (1) Say or do something to humiliates humiliate you in front of others? (2) Threaten to hurt or harm you or someone close to you? (3) Insult you or make you feel bad about yourself? Husband humiliates was coded as 1 if women answered 'yes" to any one of these questions; otherwise, it was coded as 0. Husband Does your husband: (1) Slap you? (2) Twist your physically arm or pull your hair? (3) Push you, shake you, or abuses throw something at you? (4) Punch you with his fist or with something that could hurt you? (5) Kick you, drag you, or beat you up? (G) Try to choke you or burn you on purpose? (7) Threaten or attack you with a knife, gun, or any other weapon? (8) Physically force you to have sexual intercourse with him even when you did not want to? (9) Force you to perform any sexual acts you did not want to? If the answer was "yes "to any one of these items, husband physically abuses was coded as 1; otherwise, it was coded as 0. Barriers to use Many different factors can prevent women from health services 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 problem or not? Distance If distance to the health facility was a problem, women were coded as 1; otherwise, they were coded as 0. Cost If getting money needed for treatment was a problem, women were coded as l; otherwise, they were coded as 0. Quality Do you have: (1) Concern that there may not be any health provider? (2) Concern that there may not be a female health provider? (3) Concern that there may be no drugs available? Women who were concerned about any of the three situations were coded as 1; all others were coded as 0. Table 2: Weighted Descriptive Results of All Women Who Had Given Birth in the Past Five Years in India, by Experience of Blindness, 2005-2006 (Unweighted N = 35,248) Women Women with without All women Blindness Blindness Variable (%) (%) (%) Categorical Experienced daytime or nighttime blindness 11.87 Education None 47.02 63.33 44.82 Primary (1-5 standards) 13.87 15.17 13.70 Junior/middle (6-8 standards) 14.25 11.15 14.66 High school and above (>8 standards) 24.87 10.35 26.83 Participates in purchase of major household items 47.43 43.29 47.99 Participates in purchase of items for daily use 54.13 52.64 54.32 Participates in own health care decisions 58.90 52.74 59.73 Participates in own visits to family/ relatives 54.54 51.20 54.98 Mother worked 29.82 34.76 29.15 Said husband controls (for example, jealous, cannot meet friends) 44.62 58.13 42.76 Said husband humiliates (for example, insults inpublic) 15.03 23.00 13.93 Said husband abuses physically (for example, slaps, kicks, punches) 38.44 50.90 36.72 Hindu (religion) 79.11 78.98 79.14 Scheduled caste/scheduled tribe/ other underprivileged class 71.68 78.48 70.77 Wealth index Poor class 45.29 66.24 42.46 Middle class 19.75 16.46 20.18 Rich class 34.97 17.30 37.36 Urban residence 26.89 13.37 28.71 EAG state residence 52.32 67.96 50.23 Said distance to health facility is a problem 56.90 71.66 54.91 Said getting money needed for treatment is a problem 44.42 57.54 42.65 Said quality of health facility is a problem 42.95 59.43 40.79 Number of antenatal care visits during pregnancy 0 22.84 34.21 21.31 1-3 39.56 43.79 39.00 4 or more 37.60 22.00 39.70 Received iron supplement 65.65 56.12 66.93 Received supplementary nutrition during pregnancy 19.19 19.71 19.12 Received advise on pregnancy nutrition 66.15 58.57 67.01 Consumed milk at least once a week 52.25 41.66 53.68 Consumed beans at least once a week 90.06 88.61 90.25 Consumed green vegetables at least once a week 92.97 92.06 93.10 Consumed fruit at least once a week 34.27 22.88 35.80 Consumed eggs at least once a week 30.35 22.37 31.42 Consumed meat or fish at least once a week 33.23 25.37 34.29 Continuous M SD M SD M SD Women's age (in years) 26.40 5.81 26.66 6.71 26.36 5.70 Age at first marriage 17.26 3.45 16.38 3.18 17.38 3.46 Age at birth of first child 19.44 3.60 18.61 3.45 19.55 3.60 Note: EAG = Empowered Action Group (EAG states are Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Orlssa, Rajasthan, Uttaranchal and Uttar Pradesh). Table 3: Logistic Regression Predicting Experience of Blindness among Pregnant Women in India, 2005-2006 (N = 19,902) Parameter OddsRatio 95& CI Independent Variables Mother's empowerment Education (no education = 0) Primary (1-5 standards) 1.04 [0.90, 1.19] Junior/middle (6-8 standards) 1.00 [0.85, 1.16] High school and above (>8 standards) 0.77 ** [0.65, 0.91] Mother employed (yes = 1; no = 0) 1.02 [0.92, 1.13] Participates in purchases of major household items (yes = 1, no = 0) 0.91 [0.78, 1.03] Participates in purchase of daily use items (yes = 1, no = 0) 1.03 [0.90, 1.17] Participates in decisions for own health care (yes = 1, no = 0) 0.82 *** [0.73, 0.92] Participates in own visits to family/ relatives (yes = 1, no = 0) 1.14 [1.00, 1.29] Age at birth of first child 0.97 *** [0.96, 0.99] Domestic abuse Husband controls (yes = 1, no = 0) 1.49 *** [1.34, 1.64] Husband humiliates (yes = 1, no = 0) 1.31 *** [1.43, 1.49] Husband abuses physically (yes = 1, no = 0) 1.16 ** [1.04, 1.30] Control Variables Antenatal care visit (none = 0) 1-3 1.07 [0.93, 1.22] 4 or more 1.03 [0.87, 1.21] Received supplementary nutrition (yes = 1, no = 0) 1.14 * [1.01, 1.28] Received iron supplement (yes = 1, no = 0) 1.00 [0.89, 1.12] Milk (drank milk at least once a week = 1, otherwise = 0) 0.87 * [0.79, 0.97] Beans (consumed beans at least once a week = 1, otherwise = 0) 1.02 [0.89, 1.17] Fruits (consumed fruits at least once a week = 1, otherwise = 0) 1.01 [0.90, 1.14] Eggs (consumed eggs at least once a week = 1, othertwise = 0) 0.93 [0.82, 1.06] Meat/fish (consumed meat/fish at least once a week = 1, otherwise = 0) 0.78 *** [0.68, 0.88] Distance to health facility big problem? (yes = 1, no = 0) 1.15 * [1.02, 1.30] Money/cost a problem? (yes = 1, no = 0) 1.12 * [1.01, 1.25] Quality of care a problem? (yes = 1, no = 0) 1.34 *** [1.20, 1.50] Caste (scheduled caste/scheduled tribe/ other underprivileged class = 1, otherwise = 0) 1.12 [1.00, 1.25] Religion (Hindu = 1, not Hindu = 0) 0.82 *** [0.74, 0.92] EAG states (yes= 1, no = 0) 1.19 ** [1.06, 1.33] Wealth index (poor class = 0) Middle class 0.79 *** [0.69, 0.90] Rich class 0.79 ** [0.68, 0.93] Residence (urban = 1, rural = 0) 0.68 *** [0.60, 0.78] Wald [chi square] 867.57 ** Pseudo [R.sup.2] 0.10 * Note: CI = confidence interval; EAG = Empowered Action Group (EAG states are Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Orissa, Rajasthan, Uttaranchal and Uttar Pradesh). * p < .05. ** p < .01. *** p < .001. Table 4: Classification Table Probability Correct Cutoff Point Classification Sensitivity Specificity .10 61.6 68.6 60.8 .20 84.1 23.9 91.0 .30 89.1 04.3 98.8 Probability Cutoff Point FALSE Positive FALSE Negative .10 83.3 5.6 .20 76.6 8.7 .30 70.8 10.0 Notes: Probability cutoff points reflect prevalence of blindness in India. All other values represent percentages.
|Gale Copyright:||Copyright 2012 Gale, Cengage Learning. All rights reserved.|