Scholarlink

Background <p>Child malnutrition is a critical public health issue in Bangladesh, significantly affecting child development and health. This study analyzes the prevalence and determinants of malnutrition among children under five years old using data from the Bangladesh Demographic and Health Survey (BDHS) 2017–18.</p> Methods <p>The study employed Generalized Linear Mixed Models (GLMM) and Generalized Estimating Equations (GEE) to examine socio-economic, demographic, and health-related factors associated with stunting, wasting, and underweight conditions among children.</p> Results <p>The study found that 31% of children were stunted, 22% were underweight, and 8% were wasted in Bangladesh in 2017–18, with overlaps likely among these forms of malnutrition. Higher parental education levels and wealthier household status were significantly (<i>p&#xa0;</i>&lt; 0.05<i>)</i> associated with a&#xa0;lower prevalence of malnutrition. Children from Sylhet had 1.3 times higher odds of being stunted (AOR = 1.3, 95% C.I. = 1.05–1.65) and 1.46 times higher odds of being underweight (AOR = 1.46, 95% C.I. = 1.14–1.88) compared to children from Barisal. Mothers with normal BMI were significantly less likely to have stunted (AOR = 0.67, 95% C.I. = 0.56–0.79), wasted (AOR = 0.49, 95% C.I. = 0.37–0.66), and underweight (AOR = 0.63, 95% C.I. = 0.53–0.75) children compared to mothers with underweight BMI. Both GLMM and GEE models identified the same associated factors for stunting, wasting, and underweight, with close estimates. However, GLMM was found to have better predictive power for all three models, as indicated by higher area under the curve (AUC) values.</p> Conclusion <p>The findings emphasize the association of poor parental education, economic conditions, and maternal health with child malnutrition in Bangladesh. The GLMM demonstrated better predictive power based on AUC values across all three outcomes, making it a more reliable choice for this type of analysis. Policymakers should prioritize enhancing maternal education, household economic status, and access to healthcare services.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prevalence and determinants of child malnutrition in Bangladesh: a comparative analysis of multilevel modeling

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月31日

查看全文
  • Afroza Jannat Suchana,
  • Syed Toukir Ahmed Noor,
  • Md. Rafil Tazir Shah,
  • Ahmad Kabir,
  • Mohammad Romel Bhuia

摘要

Background

Child malnutrition is a critical public health issue in Bangladesh, significantly affecting child development and health. This study analyzes the prevalence and determinants of malnutrition among children under five years old using data from the Bangladesh Demographic and Health Survey (BDHS) 2017–18.

Methods

The study employed Generalized Linear Mixed Models (GLMM) and Generalized Estimating Equations (GEE) to examine socio-economic, demographic, and health-related factors associated with stunting, wasting, and underweight conditions among children.

Results

The study found that 31% of children were stunted, 22% were underweight, and 8% were wasted in Bangladesh in 2017–18, with overlaps likely among these forms of malnutrition. Higher parental education levels and wealthier household status were significantly (p < 0.05) associated with a lower prevalence of malnutrition. Children from Sylhet had 1.3 times higher odds of being stunted (AOR = 1.3, 95% C.I. = 1.05–1.65) and 1.46 times higher odds of being underweight (AOR = 1.46, 95% C.I. = 1.14–1.88) compared to children from Barisal. Mothers with normal BMI were significantly less likely to have stunted (AOR = 0.67, 95% C.I. = 0.56–0.79), wasted (AOR = 0.49, 95% C.I. = 0.37–0.66), and underweight (AOR = 0.63, 95% C.I. = 0.53–0.75) children compared to mothers with underweight BMI. Both GLMM and GEE models identified the same associated factors for stunting, wasting, and underweight, with close estimates. However, GLMM was found to have better predictive power for all three models, as indicated by higher area under the curve (AUC) values.

Conclusion

The findings emphasize the association of poor parental education, economic conditions, and maternal health with child malnutrition in Bangladesh. The GLMM demonstrated better predictive power based on AUC values across all three outcomes, making it a more reliable choice for this type of analysis. Policymakers should prioritize enhancing maternal education, household economic status, and access to healthcare services.