Exposure to intimate partner violence and malnutrition among young adult Bangladeshi women : cross-sectional study of a nationally representative sample

This study explores the relationship between malnutrition and intimate partner violence (IPV) among 1,086 young adult Bangladeshi women aged 15-24 years using a cross-sectional data from the 2007 Bangladesh Demographic Health Survey (BDHS). About one-third (33.4%) young adult women experienced physical and/or sexual IPV, 14.5% experienced only sexual IPV and 29% experienced only physical IPV by husbands. About 32.6% young adult women were reported as being underweight (BMI < 18.5) and 6.2% were overweight (BMI ≥ 25). Underweight women experienced more physical IPV (OR = 1.39; 95%CI: 1.09-2.71) and physical and/or sexual IPV (OR = 1.48; 95%CI: 1.12-2.75) than normal range women. Results also indicate a positive association between being overweight/obese and all the forms of IPV. The study findings indicate that the IPV experience plays a significant role in underweight and overweight/obese young adult women and support that younger women’s health and nutrition program and policies need to address IPV. Intimate Partner Violence; Malnutrition; Body Weight; Body Mass Index Correspondence Md. M. Rahman Department of Population Science and Human Resource Development, University of Rajshahi. Rajshahi / Northern – CEP 6205 – Bangladesh. mosfeque@gmail.com 1 University of Rajshahi, Rajshahi, Bangladesh. doi: 10.1590/0102-311X00113916 Cad. Saúde Pública 2018; 34(7):e00113916 ARTIGO ARTICLE This article is published in Open Access under the Creative Commons Attribution license, which allows use, distribution, and reproduction in any medium, without restrictions, as long as the original work is correctly cited. Ferdos J, Rahman MM 2 Cad. Saúde Pública 2018; 34(7):e00113916 Introduction Malnutrition, both undernutrition/underweight and overnutrition/overweight, remains a pervasive and damaging condition in under developed and developing countries. Undernutrition is the leading risk factor for mortality in the world, contributing to nearly 3.5 million deaths 1 and 16% of all disability-adjusted life years (DALYs) lost globally 2. It is also an important determinant of maternal and child death 3. In Bangladesh, 24% of ever married women aged 15-49 years are undernourished (body mass index – BMI < 18.5kg/m2) and women in the age group 15-19 years are more likely to be thin or undernourished than other age cohorts 4. Even among the wealthiest quintile of society, 8.4% of women are underweight 4. Undernutrition in Bangladesh continues to be a serious public health problem. Overnutrition or obesity have significant health and economic consequences and are associated with an increased risk of developing various non-communicable diseases (NCDs), including hypertension, coronary heart disease, diabetes, stroke and some types of cancer 5. Although overnutrition in Bangladesh is still not a major problem, the prevalence of overweight women is increasing. The proportion of women who are overweight or obese (BMI ≥ 25kg/m2) has increased from 2.7% in 1996-1997 to 10.1% in 2007, and 16.5% in 2011 6. Therefore, it is evidently important to determine the prevalence of overnutrition as well as its consequences on woman’s health along with measuring the consequences of undernutrition, particularly in developing countries. The literature on the relationship between being underweight or overweight/obese and experience of intimate partner violence (IPV) is limited. IPV is considered one of the main psychosocial factors that might influence women’s nutritional status 7. IPV increases psychological stress among women, which, in turn, increases oxidative stress 8 and metabolic level 9, risk factors for undernutrition. IPV against women can provoke an eating disorder, unhealthy weight control practices and poorer treatment outcomes for the disorders 10, as well as chronic fatigue, physical inactivity, and disabilities preventing work and daily activities 11. Some recent studies in Bangladesh 12 and India 13 indicate the association between IPV and undernutrition among reproductive-aged women. However, a study in Egypt found a significant relationship between the IPV experience and obesity in adulthood 14. These studies have examined only the relationship between the experiences of IPV and undernutrition/obesity in different settings. A research conducted in Bangladesh indicates that young women experience more IPV than their older counterparts 15. This study, therefore, assessed the association between being underweight or overweight/obese and the experience of IPV among young adult women aged 15-24 years in Bangladesh.


Introduction
Malnutrition, both undernutrition/underweight and overnutrition/overweight, remains a pervasive and damaging condition in under developed and developing countries.Undernutrition is the leading risk factor for mortality in the world, contributing to nearly 3.5 million deaths 1 and 16% of all disability-adjusted life years (DALYs) lost globally 2 .It is also an important determinant of maternal and child death 3 .In Bangladesh, 24% of ever married women aged 15-49 years are undernourished (body mass index -BMI < 18.5kg/m 2 ) and women in the age group 15-19 years are more likely to be thin or undernourished than other age cohorts 4 .Even among the wealthiest quintile of society, 8.4% of women are underweight 4 .Undernutrition in Bangladesh continues to be a serious public health problem.
Overnutrition or obesity have significant health and economic consequences and are associated with an increased risk of developing various non-communicable diseases (NCDs), including hypertension, coronary heart disease, diabetes, stroke and some types of cancer 5 .Although overnutrition in Bangladesh is still not a major problem, the prevalence of overweight women is increasing.The proportion of women who are overweight or obese (BMI ≥ 25kg/m 2 ) has increased from 2.7% in 1996-1997 to 10.1% in 2007, and 16.5% in 2011 6 .Therefore, it is evidently important to determine the prevalence of overnutrition as well as its consequences on woman's health along with measuring the consequences of undernutrition, particularly in developing countries.
The literature on the relationship between being underweight or overweight/obese and experience of intimate partner violence (IPV) is limited.IPV is considered one of the main psychosocial factors that might influence women's nutritional status 7 .IPV increases psychological stress among women, which, in turn, increases oxidative stress 8 and metabolic level 9 , risk factors for undernutrition.IPV against women can provoke an eating disorder, unhealthy weight control practices and poorer treatment outcomes for the disorders 10 , as well as chronic fatigue, physical inactivity, and disabilities preventing work and daily activities 11 .Some recent studies in Bangladesh 12 and India 13 indicate the association between IPV and undernutrition among reproductive-aged women.However, a study in Egypt found a significant relationship between the IPV experience and obesity in adulthood 14 .These studies have examined only the relationship between the experiences of IPV and undernutrition/obesity in different settings.A research conducted in Bangladesh indicates that young women experience more IPV than their older counterparts 15 .This study, therefore, assessed the association between being underweight or overweight/obese and the experience of IPV among young adult women aged 15-24 years in Bangladesh.

Sample
This cross-sectional study used data from the 2007 Bangladesh Demographic Health Survey (BDHS), conducted by the National Institute of Population Research and Training of the Ministry of Health and Family Welfare of Bangladesh, from March to August 2007 16 .The survey used a sample drawn from the total population of Bangladesh residing in private dwellings.A stratified, multistage cluster sample of 361 primary sampling units was conducted, 227 in rural areas and 134 in urban areas.A total of 11,178 eligible women aged 15-49 years were identified to participate in the survey; 10,996 were interviewed, a response rate of 98.4%.Data collection procedures for the BDHS were approved by the ORC Macro Institutional Review Board.Details of data collection and management procedures are described elsewhere 16 .In our analyses, we included only currently married young adult women aged 15-24 years (N = 1,086), who were not pregnant at the time of the survey.

Measures
The BMI, calculated as weight in kilograms divided by height in meters squared, was used to assess undernutrition and overnutrition in this study.BMI is chosen to assess underweight and overweight Cad.Saúde Pública 2018; 34 (7):e00113916 mothers because it does not require a reference table from a well-nourished population.To fulfill the requirement of this study, variable was created to define the categories underweight, normal and overweight/obese for BMI < 18.5kg/m 2 , between 18.5 to 24.99kg/m 2 , and ≥ 25kg/m 2 , respectively 17 .
Women's experience of IPV was the main exposure of interest in this study.The survey measured IPV with a shortened and modified Conflict Tactics Scale (CTS) 18 .Each married woman was asked "(Has) your (last) husband ever done any of these things to you: (i) pushed you, shaken you, or throwed something at you; (ii) slapped you; (iii) twisted your arm or pulled your hair; (iv) punched you with his fist or with something that could hurt you; (v) kicked you, dragged you, or beaten you up; (vi) tried to choke you or burnt you on purpose; (vii) threatened or attacked you with a knife, gun, or any other weapon; (viii) physically forced you to have sexual intercourse with him even when you did not want to?"A positive answer to any of these questions [(i)-(vii)] indicated physical perpetration and a positive answer to question (viii) indicated sexual IPV perpetration.Again, questions [(i)-(iv)] indicate experience of minor physical violence and questions [(v)-(vii)] indicate experience of severe physical violence.
Several socioeconomic and demographic variables were included in the analysis as potential confounding factors, which were associated with the experience of IPV in previous studies 12,13,19,20 .Age at marriage was categorized as being married before 18 years vs. 18 years or older.The women's educational level was classified as: no education (0 year), primary (1-5 years), secondary or higher (6 years or more).Place of residence was categorized as rural vs. urban.The wealth variable categorized respondents into quintiles according to the household's score on the DHS wealth index, which is based on the household's amenities, assets and living conditions 21 .The sample was then divided into quintiles with 1 = poorest and 5 = wealthiest, 20% of household.Total numbers of household members were classified as (2-4, 5-6 or ≥ 7).Maternal working status was created as classifying women according to whether they were working at the time of the survey or not.Maternal height was categorized into two categories: < 145 or ≥ 145cm.

Statistical analysis
Descriptive statistics were used to describe the distribution variables of interest.Descriptive data were presented as standard deviations means for continuous variables, and frequencies and percentages for categorical variables.Chi-square tests were used to investigate associations between IPV and sociodemographic variables.Multivariate logistic regression was performed to examine the relationship between undernutrition and overnutrition/obesity and maternal experience of IPV.The multicollinearity of the variables was verified by examining the variance inflation factors; in all cases, the values were lower than 2.0, indicating that multicollinearity was low.In all the analyses, the level of significance was p-value < 0.05.All statistical analyses were conducted using IBM SPSS 20.0 for Windows (IBM Corp., Armonk, USA) to accommodate the complex sampling design of BDHS.

Ethical considerations
The 2007 BDHS follows a standard protocol and was given complete approval by the ORC Macro Institutional Review Board.The protocol of the survey was also reviewed and approved by the National Ethics Committee of the Bangladesh Ministry of Health and Family Welfare.Informed consent was obtained from each respondent for this survey.This study was exempt from full review as it was based on secondary use of anonymous data set with no identifiable information on the survey participants.
The bivariate analyses revealed several significant differences in the experience of IPV across various sociodemographic groups (Table 2).Age at marriage was significantly associated with all forms of IPV.Higher percentages of women who were married before 18 years of age experienced significantly more physical IPV, physical and/or sexual IPV, minor physical IPV, and severe physical IPV.Women whose height fell below the cutoff of 145cm were also more likely to report physical IPV, sexual IPV, physical and/or sexual IPV and severe physical IPV.Physical IPV, physical and/or sexual IPV, minor physical IPV and severe physical IPV were significantly associated with the BMI of women.A higher percentage of underweight women reported experiencing IPV than the normal and overweight/obese category (Table 2).

Discussion
This article presents the association between being both under-and overweight and experience of IPV among young adult Bangladeshi women aged 15-24 years.Our study findings indicate that one-third of young adult women are suffering from chronic undernutrition and 6.2% of women are overweight or obese.Young Bangladeshi women are more likely to be underweight than older women but less likely to be overweight/obese 4,6,16 .These findings suggest that both underweight and overweight/ obese women coexist and accounted for two-fifths among young adult Bangladeshi women.The fact that this population is often at the bottom of the food chain, with little or no decision-making power Cad.Saúde Pública 2018; 34(7):e00113916  about food distribution in the household, could lead to food security issues and may contribute to their poor nutritional status.Moreover, women aged 15-24 years need adequate nutrients to support fast physical, mental and emotional growth.The unawareness of their own health and nutritional status could be another reason associated with their poor nutritional status.A study from South-Asia showed that the prevalence of overweight-obesity among women of reproductive age has risen between 1996 and 2006 and overweight-obesity was positively related to age 22 .This study also found that 29% and 14.5% of young Cad. Saúde Pública 2018; 34(7):e00113916 adult women experienced physical and sexual violence, respectively.These findings are similar to the earlier findings of experiencing IPV among women between 15-49 years in Bangladesh 16 , including young adults 15 .Comparing with the earlier study 15 , the likelihood of experiencing IPV decreases with age, which suggests that as women grow older, their roles as mothers become more prominent than their roles as wives, and they achieve a certain status at the household and community levels 23 .
The results from nationally representative data set provide evidence for a relation between malnutrition and IPV experience among young adult Bangladeshi women.Underweight woman experienced more physical and physical and/or sexual IPV than a normal weight range woman.In this South-Asian region, the withholding of food is a documented form of abuse that is likely correlated with the perpetration of physical violence 24 .This withholding of food through psychological or emotional abuse resulted in insufficient intake of diet among these populations and could mediate the relation between physical IPV and nutrient deficiencies that may cause undernutrition.Furthermore, IPV is strongly associated with a woman's inability to make decisions for herself and her family, including the choice of types and quantities of food that a woman prepares as she cares for herself and her children 25 .This might also explain how IPV reflects the nutritional status among these women.
Another explanation underpinning the association between IPV and undernutrition is that the relation between IPV and nutritional deficiencies may also involve a mediating effect of psychological stress.A study in South-Asia documented that women who experience IPV tend to have higher levels of psychological stress 26 , which could also be related to underweight status.Psychological stress increases oxidative stress 8 , a term used to describe a number of chemical reactions that produce free radicals and other organic molecules capable of damaging living tissue, risk factors for undernutrition.However, chronic stress also increases metabolic rate and energy expenditure 9,27 , a process that could cause weight loss in a person with a fixed and limited caloric intake.That psychological stress may be associated with undernutrition in a low-income country is consistent with the result of a longitudinal study from the United Kingdom, which found that, under stressful conditions, lean subjects tend to lose weight 28 .
Our findings also indicated that overweight/obese women experienced significantly more IPV (all the forms of IPV) than normal range women.Experience of sexual IPV is significantly associated with overweight/obese women, which corroborates the findings of the study of Egypt 14 .Thus, Bangladeshi young adult women who experience sexual IPV may experience adverse psychological after-effects, which leads to excess energy intake or a more sedentary lifestyle, and ultimately to a higher chance of being obese.Most research in food-rich environments in industrialized nations has documented a link between chronic stress and obesity mediated by metabolic changes 29 or behavioral adjustments 30 that promote abdominal adipose tissue deposition.Findings from a longitudinal study in the United Kingdom indicated that, under stressful conditions, overweight subjects tend to gain weight 28 .Thus, young adult Bangladeshi women, who are on the verge of the normal nutritional range (BMI ≥ 23.0 and < 25.0), may gain weight due to stressful conditions in the family after experiencing sexual IPV and enter the overweight nutritional status.
Although our findings indicate important insights into the nutritional status and its association with IPV in Bangladesh, the results should be interpreted in the light of several limitations.First, the findings rely on self-reported data and retrospective accounts, which are vulnerable to recall bias and desirability bias.The second limitation is that a single item of sexual IPV precludes a more thorough documentation of marital sexual abuse; use of multiple behaviorally specific questions about the type or types of sexual abuse inflicted could have elicited a greater number of reports of sexual IPV 31 .Third, the BDHS is a cross-sectional observational survey, which prevents us from inferring that domestic violence causes malnutrition among these women.Indeed, the causal pathway may be reversed, with violence being a husband's reaction against his malnourished wife.Longitudinal designs that can address causality or cross-sectional path or mediational models using exposure to IPV and outcome variables will be more relevant than mere correlational designs.Fourth, as there is a strong psychological element used in this study, it will be more informative to shine some light on these women's feelings.However, BDHS survey did not evaluate the participating subjects for mood symptoms or negative moods, such as depression or stress.Finally, the sample of this study is limited to young adult Bangladeshi women and should not be generalized to women of other age groups and other countries, though the association found in this study is consistent with those of the few studies of this type.
Cad. Saúde Pública 2018; 34( 7):e00113916 Despite these potential limitations, our findings reveal a significant relationship between IPV and malnutrition among young adult Bangladeshi women.The findings strongly suggest that malnutrition in such settings may partly have its roots in the subordination of women.Future longitudinal studies are needed to investigate the influence of potential mechanisms mediating the relationship between IPV and undernutrition and overnutrition among these populations.

Table 1 (continued)
.2% were overweight/obese (BMI ≥ 25).Approximately one-third (34.3%) of the young adult women reported they had experienced physical and/or sexual violence from their husbands in the 12 months preceding the survey: 29% reported physical violence only in the absence of sexual violence, 14.4% reported sexual violence only in the absence of physical violence (Table 6

Table 3
Adjusted odds ratios (OR) from multivariate logistic regression analyses examining the association between maternal body mass index (BMI) and maternal experience of intimate partner violence (IPV).