Prevalence of undernutrition among pregnant women and its differences across relevant subgroups in rural Ethiopia: a community-based cross-sectional study

{"background": "Maternal undernutrition is a major public health problem in low- and middle-income countries, with global prevalence around 10% and pooled African prevalence of 20%. In Ethiopia, prevalence estimates among pregnant women range from 14% to 45%, with rural areas disproportionately affected (approximately 1.5 times higher than urban). Maternal undernutrition is associated with low birth weight, preterm birth, intrauterine growth restriction, stillbirth, and accounts for roughly 20% of maternal deaths. Multiple factors have been implicated, including maternal age, dietary practices, food taboos, education, income, and reproductive history. Local traditions in the Sidama region, including avoidance of nutritious foods during pregnancy, prompted this investigation to identify high-risk subgroups for targeted intervention.", "methods": "A community-based cross-sectional study was conducted from April 30 to May 30, 2019 across six randomly selected districts of Sidama Region, Southern Ethiopia. Sample size was calculated using a single population proportion formula assuming 31.8% prevalence, yielding 550 participants after adjustment for design effect and non-response. Multistage sampling was used, with 18 kebeles selected and pregnant women identified through Health Extension Worker registers. Trained BSc nurses collected data via structured questionnaires (translated to Sidaamu Afoo) and measured MUAC at the mid-point of the left arm. Undernutrition was defined as MUAC \u226423 cm. Data were entered in Epi-data 3.1, analyzed in SPSS v24 and Stata v14.0 using multilevel mixed-effect logistic regression to account for hierarchical clustering. Variables with p<0.20 in bivariate analysis, plus age, education, and income as confounders, were entered into the multivariable model. Statistical significance was set at p<0.05.", "key_results": "Of 550 sampled, 530 women participated (96% response rate). Mean age was 27\u00b14.9 years (range 18\u201342), 52% had no formal education, 50% were housewives, and 99% earned \u22641000 ETB/month. All were married, with 10% in polygamous unions. The prevalence of undernutrition (MUAC \u226423 cm) was 38% (95% CI: 34.2\u201342.3). Food taboos were practiced by 32% of women, most commonly avoiding eggs, butter, meat, and avocado; 44% cited fear of delivering a big baby as the reason. 58% received nutritional counseling during pregnancy. In multivariable analysis, significant independent risk factors were: history of miscarriage (AOR: 3.18; 95% CI: 1.77\u20135.70), absence of nutritional counseling (AOR: 2.97; 95% CI: 1.79\u20134.95), practice of food taboos (AOR: 2.23; 95% CI: 1.47\u20133.39), and multigravidity (AOR: 1.66; 95% CI: 1.02\u20132.71). The intra-class correlation coefficient was 4.1% (95% CI: 2.2\u20135.9%), and the median odds ratio was 1.2 (95% CI: 0.8\u20131.5). Among women with all four risk factors, prevalence reached 95% (OR: 6.95; 95% CI: 0.80\u201360.41). Undernutrition was approximately 70% among women with repeated pregnancies and miscarriage history, and nearly 60% among those practicing food taboos. The difference across risk-factor strata was statistically significant (p<0.001).", "clinical_implications": "The 38% prevalence of maternal undernutrition in rural Sidama substantially exceeds global and African pooled estimates, indicating a severe public health burden. The identification of modifiable risk factors\u2014particularly food taboos and absence of nutritional counseling\u2014suggests that integrating structured nutrition counseling into routine antenatal care could substantially reduce maternal undernutrition. Family planning interventions to increase inter-pregnancy intervals are warranted for multigravid women. The exceptionally high prevalence (95%) among women with combined risk factors supports targeted, subgroup-specific interventions rather than universal approaches. Health extension workers and ANC providers should screen for nutritional status, address food taboos through culturally appropriate health education, and provide counseling to all pregnant women. The cross-sectional design precludes causal inference, and the over-fitted subgroup model warrants validation in longitudinal studies."}