Modelling of Nutritional Status of Ever-Married Women in Bangladesh

Abstract

The change in nutritional status plays an important role in the course of a person’s health. Body mass index (BMI) is measure of body fat based on height and weight of a person and it is used to screen for weight categories that may lead to health problems. The first main objective of this thesis was to observe the presence of secular trends in nutritional status of Bangladeshi married women over time. A total of 45,572 Bangladeshi non-pregnant married women in reproductive age were used in this study with average age 30.11±9.04 years. The secondary cross-sectional data used in this study was extracted from Bangladesh Demographic and Health Survey (BDHS) and they collected data from Bangladeshi ever-married women using multistage stratified cluster sampling. The last five rounds (1996-97, 1999-2000, 2004, 2007 and 2011) BDHS measured height and weight from their selected sample, the data was derived from the measurement years 1996-97 to 2011. Descriptive statistics, Analysis of variance (ANOVA) and linear regression analysis were used in this study. The mean value of BMI of Bangladeshi non-pregnant married women was 20.65±3.67 kg/m2 with height and weight were 150.60±5.44 cm and 46.93±9.24 kg, respectively. Using ANOVA, height, weight and BMI showed significant (p<0.001) differences among the measurement year cohorts from 1996-97 to 2011. The increasing tendency in height, weight and BMI of Bangladeshi married women in reproductive age were showed during the measurement years from 1996-97 to 2011. The sample was divided into 40 groups according their birth year cohorts from 1955 to 1995. ANOVA also demonstrated that the variation of height, weight and BMI were significant (p<0.001) among the birth year cohorts from 1955 to 1995. The slope of linear regression demonstrated that an increasing tendency in BMI was found among birth year cohorts from 1955 to 1971, but a decreasing tendency was observed during the birth year cohorts from 1972 to 1995. In this study found that more than 50% women were normal weight and a remarkable number of married women (31.4%) were underweight. More than 10% women were overweight and only 2.1% were obese among non-pregnant Bangladeshi married women. It was found that the proportion of underweight women has been increasing and the opposite direction also observed for the number of obese in those born during the last 40 years of the study period (1955 to 1995). Underweight can be considered as the major health problems of Bangladeshi married women and requires attention. The second objective of the thesis was to investigate the influencing factors of body mass index among Bangladeshi married women. Hierarchy data was used in this study that was collected by Bangladesh Demographic and Health Survey-2011 (BDHS-2011) using multistage stratified cluster sampling. Usually, large-scale survey for public health, demography and sociology follow a hierarchical data structure as the surveys are based on multistage stratified cluster sampling. The appropriate approach to analyzing such survey data is therefore based on nested sources of variability which come from different levels of the hierarchy. The single level statistical model is not appropriate for analyzing such kind of data set. Multilevel (two level) linear regression was utilized in the present study to remove the cluster effect for each outcome. The mean BMI of Bangladeshi married women in reproductive age was 21.60±3.86 kg/m2 with height and weight was 150.92 ±5.36 cm and 49.28 ± 9.69 kg, respectively. The prevalence of underweight, normal weight, overweight and obese were 22.8%, 59.1%, 14.9% and 3.2%, respectively of Bangladeshi non-pregnant married women in reproductive age for BDHS-2011 data set. Analysis of variance (ANOVA) demonstrated that the BMI of Bangladeshi non-pregnant married women in reproductive age was significantly different among respondents’ and her husbands’ education level (p<0.001) and wealth index (p<0.001). The t-test showed that the differences of BMI was significant between (i) urban and rural (p<0.001), (ii) watching television, yes and no ((p<0.001), (iii) currently breastfeeding, yes and no (p<0.001), (iv) currently use contraceptive, yes and no (p<0.05), (v) living with husband, yes and no (p<0.01) and (v) currently working status (p<0.05). Person’s correlation coefficients exhibited that respondent’s age, age at first marriage and age at first birth were significantly (p<0.001) positively related with BMI of married women in reproductive age, and there was a negative association between number of ever born children and BMI (p<0.001). Two level regression model demonstrated that age and age at first marriage of women were positively (p<0.01) related with their BMI. Parity was negatively related with women’s BMI. BMI was especially less pronounced among non-pregnant married women who came from rural, uneducated, uneducated husband, came from poor family, house wife, currently not use contraceptive and currently not breastfeeding. Under-nutrition is considered as a major health problems in Bangladesh especially of married women. Government and non-government organization of Bangladesh should take step to increase people education level and improve economic condition for removing under-nutrition among married women. The third and final main objective of this thesis was to quantify the prevalence of anemia and assessed how various nutritional, socio-economic and demographic factors associated with anemia among non-pregnant women in Bangladesh. Anemia is one of the most common and global public health problems. Highly prevalence of anemia was found among women of reproductive age in the world especially in developing countries. Study on anemia among Bangladeshi non-pregnant married women is poorly documented. Cross-sectional data were used in this study. Data was extracted from Bangladesh Demographic and Health Survey (BDHS) 2011. The nationally representative sample (5,293 married women) was selected by multistage stratified cluster sampling and data was collected from July 8, 2011 to December 27, 2011. Multilevel (two level) logistic regression analysis was done in this study. The prevalence of anemia among Bangladeshi married women was more than 41% and among anemic women, 35.5% was mild, 5.6% and only 0.2% were moderate and severe anemic respectively. Chi-square (χ2-test) test was utilized in this study for selecting independent factors for multilevel logistic regression analysis. χ2-test demonstrated that residence (p<0.001), respondents’ and her husbands’ education levels (p<0.001), currently breastfeeding (p<0.001), currently amenorrhea (p<0.001), currently use contraceptive (p<0.001), toilet facility (p<0.01), religion (p<0.001), wealth index (p<0.001), BMI (p<0.001), age group (p<0.001), number of ever born children (p<0.001) were significantly associated with anemia. The two level logistic regression model demonstrated that women who were currently breastfeeding and amenorrhea had more likely (p<0.01) to get anemia than their counterparts. Underweight women had a higher chance (p<0.01) to get anemia than normal weight, overweight and obese. Uneducated women were more likely to get anemia (p<0.01) than secondary and higher educated. Anemia was especially less pronounced among non-pregnant women who are currently use contraceptive (p<0.05), Muslim (p<0.01) and came from rich family (p<0.01). Moreover, women who were 30-49 years old had more likely to get anemia than younger (p<0.01). Under-nutrition is most important predictor for anemia among Bangladeshi married women and undernourished women are living under poor condition. Government of Bangladesh should take appropriate initiative to improve economic condition for removing anemia among married women.