Adequate antenatal care visits and higher wealth index are associated with lower odds of low birthweight in Ghana: cross-sectional study

¹Trauma and Specialist Hospital, Winneba, Ghana, ²Department of Medical Laboratory Science, University of Ghana, Legon, Accra, Ghana, ³Department of Biostatistics and Epidemiology, College of Public Health, East Tennessee State University, Johnson City, Tennessee, United States of America, ⁴Vanke School of Public Health, Tsinghua University, Beijing, China

^&Corresponding author
Emmanuel Angmorteh Mensah, Department of Biostatistics and Epidemiology, College of Public Health, East Tennessee State University, Johnson City, Tennessee, United States of America

Introduction    

Low birthweight (LBW), defined by the World Health Organization (WHO) as a birth weight of less than 2,500 grams [1], remains a significant public health concern globally, particularly in low- and middle-income countries. It serves as a key indicator of neonatal health and a strong predictor of infant mortality, morbidity, and long-term developmental challenges [2,3]. Globally, an estimated 15% to 20% of all births, corresponding to over 20 million annual births, are classified as LBW [4]. However, the burden of LBW is unevenly distributed, with the majority of cases concentrated in low- and middle-income countries, often among the most vulnerable populations [5,6]. Regional estimates reveal striking disparities, with LBW rates of 28% occurring in South Asia; 9% in sub-Saharan Africa and 9% in Latin America [7]. In Ghana, approximately 10% of newborns are classified as having LBW. While this is lower than the rates observed in South Asia, it is still a significant public health concern, due to profound implications for neonatal survival and long-term outcomes such as undernutrition, stunting, wasting and poor school performance [8,9]. Notably, LBWs are associated with substantially increased mortality risk during the neonatal period and infancy compared to normal-weight in Ghana [10,11]. Birth weight is not just a clinical metric but a mirror reflecting the quality of healthcare services and the broader socioeconomic environment [12]. It is closely tied to neonatal survival, with lower birth weights strongly associated with increased neonatal and perinatal mortality rates [2].

The United Nations' Sustainable Development Goal 3.2, which Ghana subscribes to, aims to lower neonatal mortality to 12 or fewer deaths per 1,000 live births by 2030 [13,14]. However, Ghana remains far from achieving these neonatal mortality rate targets, with a current rate of 22.8 per 1,000 live births [15]. Given the well-established link between LBW and neonatal mortality [16], addressing LBW is a vital step in moving closer to these international health objectives. To make substantial progress, it is crucial to gain a comprehensive understanding of the factors that contribute to LBW in Ghana. Findings from this study will contribute to formulating evidence-based policies and targeted interventions aimed at improving birth outcomes and helping the country meet its global health goals. In Ghana, although some studies have explored the prevalence and determinants of LBW, most have been limited to specific hospitals, districts, or regions of the country [11,17-20]. The context-specific evidence generated by the existing studies leaves significant gaps and makes it difficult to fully understand LBW determinants at the national landscape. Therefore, this study addresses this critical research gap by conducting a comprehensive, nationwide analysis of LBW determinants. Using data from a large, nationally representative sample. This aims to identify the risk factors, particularly sociodemographic and maternal reproductive factors, associated with LBW among infants in Ghana.

Methods     

Study design and setting: a cross-sectional study on the prevalence of low birthweight and its associated factors in Ghana. A nationwide study conducted in Ghana among women of fertility age. Respondents for the study were drawn across all administrative regions in the country. Ghana is a lower-middle-income coastal country in West Africa with sixteen (16) administrative regions. It shares borders with Togo, Burkina Faso, and Ivory Coast. The 2021 population and housing census preliminary report projected the country's population at approximately 30.8 million (50.69% females) and an average household size of 3.6 members [21].

Study population: women aged 15-49 who had given birth within the five years preceding the survey were included in this study. The most recent birth of these women was used for this study. The most recent birth in the past five years with recorded birth weights was 4,037; this number of observations served as the sample size for the study. Deliveries without birth weight records

Data collection: this study utilized secondary data from the 2022 Ghana Demographic and Health Survey (GDHS), the seventh survey in the DHS series conducted by the Ghana Statistical Service (GSS). Ghana Demographic and Health Survey was designed to gather data on a wide array of health-related indicators, including fertility, family planning, maternal and child health, childhood mortality, nutrition, HIV prevention, violence against women, women's empowerment, health insurance, water and sanitation, and malaria prevalence. Data collection was conducted using Computer-Assisted Personal Interviewing (CAPI), with the application pre-tested and refined based on initial results. The GDHS consists of Household, Women's, Men's, and Biomarker questionnaires. This study analyzed data from the Women's Questionnaire, focusing on birth history and other relevant health indicators for women aged 15-49 and their children in selected households. The GDHS used a stratified cluster sampling method across Ghana's 16 administrative regions to select respondents. The data were obtained upon justifiable request from the Demographic and Health Survey programme.

Definitions: only quantitative variables were used in this study. The outcome variable LBW was defined as a birthweight below 2.5 kilograms, according to the WHO criteria. Birthweight was therefore recoded for these analyses, with LBW cases documented in the study ranging from 0 grams to 2.49 kilograms. Birthweights at or above 2.5 kg were classified as non-low birthweight (NLBW). The independent variables included both socio-demographic factors and maternal reproductive factors. Key among the potential determinants were maternal age, which was categorised (15-19, 20-34, and 35+ years), child's sex (male, female), and educational level of mother (no education, primary, junior high, senior high, and higher education). The remaining variables were recoded as well. Marital status was recoded into never married, married/partners, widowed/divorced and employment status was recoded as employed and unemployed. Other recoding including wealth index (low, middle, high), ethnicity (Akan, Mole-Dagbani, Ewe, others), residence (urban and rural), and region of residence (southern belt, middle belt, northern belt). The southern belt of Ghana included administrative regions such as Western, Central, Greater Accra, and Volta. The middle belt includes Eastern, Ashanti, Western North, Ahafo, Bono and Bono East regions. The northern zone regions were Oti, Northern, Savannah, North East, Upper East and Upper West administrative regions. In addition, antenatal care (ANC) visits (1-3 visits or more than 4 visits), first parity (yes, no) and gestational age (preterm; born before 9 months, full-term) were factored into the study.

Statistical analysis: statistical analysis was conducted using Stata version 18. Descriptive statistics were performed for birthweight and all independent variables. From the frequency distribution analyses, unweighted frequencies, weighted percentages, and associated confidence intervals were reported. Bivariate analyses using Chi-square tests were conducted to assess associations between sociodemographic and reproductive factors and LBW. Variables that showed significant associations at 95% confidence intervals based on the designed corrected chi-square statistics and their corresponding p-values were included in a further analysis: a bivariate and multivariate complex survey logistic regression. Crude Odds Ratios (CORs) and Adjusted Odds Ratios (AORs) were reported to establish the relationship between low birthweight and the potential determinant.

Ethical considerations: the GSS and the administrators of the DHS survey received approval from the Ethical Review Committee of the Ghana Health Service, ensuring that Ghanaian ethical research standards were followed. No ethical approval number was found in the survey documents. Respondents were informed of the advantages and disadvantages of taking part in the study. Participation in the survey was optional, and prior to administering the Household or Woman's Questionnaire, eligible respondents were asked directly for their informed consent.

Results     

Background characteristics of participants and low birth weight: the study involved 4,037 respondents. They represent the most recent births of women between 19 and 49 years who had children from 2014 to 2019. From the weighted frequency distribution shown in Table 1, the majority of respondents were aged between 20 and 34 years. This age group accounted for 68.61% of the study population. In terms of gestational age, the vast majority of infants (96.85%) were born at full term, with only 3.15% being preterm births. While 17.48% of the participants had no formal education, 14.16% and 57.10% had primary and secondary education, respectively, and only 11.25% had tertiary education. The participants resided mostly in urban areas (53.05%). The wealth index distribution indicated that 41.21% of respondents were in the high wealth category, 38.61% in the low category, and 20.18% in the medium category. The majority of respondents were married or in partnerships (82.48%), while 12.54% had never married. Employment status revealed that 81.95% of respondents were employed, while 18.05% were unemployed. Ethnicity data showed that the Akan ethnic group constituted the largest proportion (41.06%), followed by Mole-Dagbani (23.59%), Ewe (10.85%), and other ethnic groups (24.50%). The sex distribution of infants was nearly equal, with 52.28% male and 47.72% female. Antenatal care utilization was high, with 92.48% of respondents attending 4 or more visits, while 7.52% attended 1-3 visits. Parity data indicated that 28.50% of respondents were first-time mothers, while 71.50% had given birth previously. Finally, birthweight data revealed that 90.09% of infants had a normal birth weight, while 9.91% were classified as low birth weight with a 95% confidence interval of 8.76-11.20.

Factors associated with low birthweight: from Table 2, the sociodemographic, maternal, and infant factors selected as potential determinant of low birthweight based on the Chi-square analyses included gestational age (χ² = 171.57, p < 0.001), sex at birth (χ² = 14.91, p < 0.001), antenatal care (ANC) visits (χ² = 10.56, p= 0.001), wealth index (χ² = 3.65, p= 0.027), ethnicity (χ² = 3.47, p= 0.016), and region (χ²= 4.08, p= 0.018). From the final adjusted model as presented in Table 3, gestational age was the strongest predictor of low birthweight, with preterm infants having 19.63 times (aOR: 19.63, 95% CI: 11.22-34.34; P <0.001) higher odds of low birthweight compared to full-term infants. Sex at birth was also significantly associated with low birthweight, with female infants (aOR: 1.73; 95% CI:1.31-2.29; P =0.001) having 1.73 times higher odds of low birthweight compared to male infants. Wealth index showed that infants from high-wealth households had 46% (aOR: 0.54, 95% CI 0.36-0.81; P =0.003) lower odds of low birthweight compared to those from low-wealth households. Ethnicity was another significant factor, with Ewe infants having 47% (aOR: 0.53, 95% CI 0.28-0.98; P =0.045) lower odds of low birthweight compared to Akan infants. Antenatal care (ANC) visits were associated with reduced odds of low birthweight, as mothers who attended 4 or more ANC visits had 41% (aOR: 0.59, 95% CI 0.39-0.89; P =0.012) lower odds of having a low birthweight infant compared to those who attended 1-3 visits.

Discussion     

This study was performed to determine the prevalence and the factors associated with LBW among infants in Ghana. This nationwide analysis of low birthweight (LBW) in Ghana highlights a prevalence of 9.91%. Gestational age, wealth index, ethnicity, sex at birth, and antenatal care (ANC) visits were identified as key predictors of LBW. In comparison to other African nations, Ghana's LBW prevalence is similar to countries like Uganda (10%); Burkina Faso (13.4%), and Malawi (12.1%), while being lower than rates observed in Ethiopia (18%) and Senegal (15.7%) [22,23]. This highlights LBW as a pervasive issue across the African continent. Notably, this study presents a lower LBW rate than previous investigations conducted within Ghana, which may imply improvements in the quality of ANC services, particularly through advanced clinical interventions for screening, diagnosis, and community-based health interventions [24]. One of the key findings from the surveys was the strong association between preterm delivery and low birth weight (LBW) among mothers, compared to those with term deliveries. This observation aligns with findings from similar studies conducted in various settings, including Rwanda, Tanzania, Kenya, and Ethiopia [25,26]. The consistency of this association across diverse geographical and socio-demographic contexts stresses the universality of the relationship between preterm birth and LBW. This suggests that preterm delivery is a significant risk factor for LBW, regardless of regional or population-specific differences. These can be attributed to the fact that preterm infants often have underdeveloped organs, including the lungs, liver, and brain, which can further complicate their health and growth, thereby affecting the birth weight of the child. At 32 weeks' gestation, foetuses who will be delivered preterm are already significantly smaller than those who will be delivered at term [27].

Wealth index also played a significant role, with mothers from lower socioeconomic backgrounds more likely to deliver LBW babies. This finding reflects broader global patterns where several studies have indicated a relationship between the wealth index and LBW [28]. Socioeconomic inequalities in access to healthcare, nutrition, and other social determinants of health likely contribute to disparities in LBW prevalence [29]. The low economic status of pregnant women might hinder their accessibility to healthcare, medicine, and quality nutritional needs before and during pregnancy [30]. An explanation for this phenomenon could be that with improving economic conditions, there may be enhanced opportunities for early detection of fetal growth retardation among expectant mothers. Ethnicity emerged as another influential factor, and this is consistent with reports by related studies [31,32]. The Akan ethnic group exhibited a remarkably higher prevalence of LBW compared to Ewes, highlighting the influence of ethnicity on birth weight outcomes. The observed variations in LBW risk among different ethnic groups indicate a complex interplay of socio-cultural factors in shaping birth weight outcomes. Ethnicity is a proxy for various cultural, social, and environmental influences that can impact maternal and child health. Understanding these influences is essential for developing culturally sensitive approaches to maternal and child health promotion and addressing disparities in birth weight outcomes among vulnerable ethnic groups. This study also found a significantly higher incidence of LBW among female infants compared to males; a trend observed in several studies [33]. Although the reasons for this gender disparity are not fully understood, potential explanations may include biological factors influencing fetal growth and development, as well as differential exposure to maternal risk factors during pregnancy [34].

While some research has identified an association between female gender and low birthweight, the specific biological and environmental factors contributing to this disparity require deeper exploration. There is limited literature on the subject, so not much is known about the underlying factors contributing to this issue. Further research is needed to uncover the underlying causes of this gender-specific difference in birth weight outcomes. Mothers with fewer than four ANC visits had a higher likelihood of delivering LBW babies, consistent with Ghana Health Service recommendations that at least four visits are necessary to improve birth outcomes. Antenatal care visits offer essential opportunities for health education, nutritional support, and the early detection of pregnancy complications [35]. However, barriers such as long wait times and inadequate education at ANC centres prevent many Ghanaian women from achieving this recommendation. Addressing these barriers is critical for reducing LBW rates. These findings have important implications for maternal and child health efforts in Ghana. The strong association between preterm birth and low birth weight highlights the need for improved prevention, early detection, and management of preterm pregnancies within antenatal care services. Strengthening access to and the quality of antenatal care is particularly important, as adequate attendance was linked to lower odds of low birth weight. The protective effect observed among women in higher wealth quintiles also suggests that socioeconomic inequalities continue to influence birth outcomes, indicating the need for targeted support for disadvantaged mothers. In addition, differences across ethnic groups suggest that cultural and contextual factors should be considered when designing maternal health interventions aimed at reducing the burden of low birth weight.

The study used a nationally representative sample selected through probability proportional-to-size sampling and applied regional weighting to ensure accurate representation of the population. As a result, the findings can reasonably be generalized to Ghana and may also provide useful insights for other developing countries on LBW. Despite its valuable insights, the study has limitations. The use of observational data and cross-sectional analysis restricts the ability to establish causal relationships between variables. Important risk factors such as medical history, pregnancy complications, and lifestyle factors like smoking and alcohol use were not included in the analysis, potentially affecting birth outcomes. Reporting bias may also exist, particularly among participants who were unable to provide accurate birth weight records, leading to unreliable data. The findings of this study are generalizable to every geographic unit of Ghana because the GDHS had a nationwide representative sample. Furthermore, the results may apply to neighbouring countries with similar developmental characteristics as Ghana in West Africa.

Conclusion     

Low birth weight remains a notable public health concern in Ghana, with a prevalence of 9.9% in this study. The findings indicate that preterm gestational age and female birth were associated with higher odds of low birth weight, while higher household wealth, Ewe ethnicity, and adequate antenatal care attendance were associated with reduced odds. These results demonstrate the importance of strengthening antenatal care services and addressing socioeconomic disparities to improve birth outcomes in Ghana. Continued research is needed to better understand why female births were associated with low birth weight to improve neonatal health outcomes in Ghana.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Conception and study design: George Boamah Appiah and Zhihui LI. Data analysis and interpretation: Emmanuel Angmorteh Mensah, Benjamin Ansah Dortey, Abdul Rashid Adams and Derrick Nyantakyi Owusu. Manuscript drafting: George Boamah Appiah, Zhihui LI and Emmanuel Angmorteh Mensah. Manuscript revision: George Boamah Appiah, Zhihui LI,Emmanuel Angmorteh Mensah, Benjamin Ansah Dortey, Abdul Rashid Adams and Derrick Nyantakyi Owusu. Authors approved final version of the manuscript: George Boamah Appiah, Zhihui LI, Emmanuel Angmorteh Mensah, Benjamin Ansah Dortey, Abdul Rashid Adam and Derrick Nyantakyi Owusu. All the authors have read and agreed to the final manuscript.

Acknowledgement     

The authors acknowledge the Vanke School of Public Health, Tsinghua University, for supporting the first author, GBA, during his MPH studies.

Tables     

Table 1: background characteristics of study respondents and infants, demographic and health survey; 2022

Table 2: sociodemographic, maternal, and infant factors associated with low birthweight in Ghana, demographic and health survey; 2022

Table 3: logistic regression sociodemographic, maternal, and infant factors associated with low birthweight in Ghana, demographic and health survey; 2022

References