Unraveling birth weight determinants: Integrating machine learning, spatial analysis, and district-level mapping

Despite a decrease in the prevalence of low birth weight (LBW) over time, its ongoing significance as a public health concern in Bangladesh remains evident. Low birth weight is believed to be a contributing factor to infant mortality, prolonged health complications, and vulnerability to non-communicable diseases. This study utilizes nationally representative data from the Multiple Indicator Cluster Surveys (MICS) conducted in 2012-2013 and 2019 to explore factors associated with birth weight. Modeling birth weight data considers interactions among factors, clustering in data, and spatial correlation. District-level maps are generated to identify high-risk areas for LBW. The average birth weight has shown a modest increase, rising from 2.93 kg in 2012-2013 to 2.96 kg in 2019. The study employs a regression tree, a popular machine learning algorithm, to discern essential interactions among potential determinants of birth weight. Findings from various models, including fixed effect, mixed effect, and spatial dependence models, highlight the significance of factors such as maternal age, household head's education, antenatal care, and few data-driven interactions influencing birth weight. District-specific maps reveal lower average birth weights in the southwestern region and selected northern districts, persisting across the two survey periods. Accounting for hierarchical structure and spatial autocorrelation improves model performance, particularly when fitting the most recent round of survey data. The study aims to inform policy formulation and targeted interventions at the district level by utilizing a machine learning technique and regression models to identify vulnerable groups of children requiring heightened attention.

Geospatial Analyses of Recent Household Surveys to Assess Changes in the Distribution of Zero-Dose Children and Their Associated Factors before and during the COVID-19 Pandemic in Nigeria

The persistence of geographic inequities in vaccination coverage often evidences the presence of zero-dose and missed communities and their vulnerabilities to vaccine-preventable diseases. These inequities were exacerbated in many places during the coronavirus disease 2019 (COVID-19) pandemic, due to severe disruptions to vaccination services. Understanding changes in zero-dose prevalence and its associated risk factors in the context of the COVID-19 pandemic is, therefore, critical to designing effective strategies to reach vulnerable populations. Using data from nationally representative household surveys conducted before the COVID-19 pandemic, in 2018, and during the pandemic, in 2021, in Nigeria, we fitted Bayesian geostatistical models to map the distribution of three vaccination coverage indicators: receipt of the first dose of diphtheria-tetanus-pertussis-containing vaccine (DTP1), the first dose of measles-containing vaccine (MCV1), and any of the four basic vaccines (bacilli Calmette-Guerin (BCG), oral polio vaccine (OPV0), DTP1, and MCV1), and the corresponding zero-dose estimates independently at a 1 × 1 km resolution and the district level during both time periods. We also explored changes in the factors associated with non-vaccination at the national and regional levels using multilevel logistic regression models. Our results revealed no increases in zero-dose prevalence due to the pandemic at the national level, although considerable increases were observed in a few districts. We found substantial subnational heterogeneities in vaccination coverage and zero-dose prevalence both before and during the pandemic, showing broadly similar patterns in both time periods. Areas with relatively higher zero-dose prevalence occurred mostly in the north and a few places in the south in both time periods. We also found consistent areas of low coverage and high zero-dose prevalence using all three zero-dose indicators, revealing the areas in greatest need. At the national level, risk factors related to socioeconomic/demographic status (e.g., maternal education), maternal access to and utilization of health services, and remoteness were strongly associated with the odds of being zero dose in both time periods, while those related to communication were mostly relevant before the pandemic. These associations were also supported at the regional level, but we additionally identified risk factors specific to zero-dose children in each region; for example, communication and cross-border migration in the northwest. Our findings can help guide tailored strategies to reduce zero-dose prevalence and boost coverage levels in Nigeria.

Trends and factors associated with HIV testing among women aged 15-49 years in Belize: an analysis using the Multiple Indicator Cluster Survey 2006, 2011, and 2015-2016

BACKGROUND

Belize has one of the highest human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome prevalence rates in Central America, with women of reproductive age being particularly vulnerable to HIV. Therefore, this study examined the factors associated with HIV testing among women of reproductive age in Belize and trends in HIV testing in 2006, 2011, and 2015-2016.

METHODS

Cross-sectional data were analyzed using three Belize Multiple Indicator Cluster Surveys. The number of participants were 1,675, 4,096, and 4,699 women aged 15-49 years in 2006, 2011, and 2015-2016, respectively. We used variance-weighted least-squares regression to estimate annual changes. Multivariate logistic regression analysis was performed to evaluate the associated factors. Analyses were conducted using Stata version 15, and weights were applied for generalization to the population.

RESULTS

HIV testing rates increased from 47.7% in 2006 to 66.5% in 2015, with an average annual change of 0.082 (95% confidence interval: 0.07-0.09). Logistic regression models showed that women aged 15-24 years were less likely to have been tested for HIV compared to women aged 25-34 years. Women from the Mayan ethnic group were less likely to have been tested than those from other ethnic groups. Compared to women who spoke Spanish, those who spoke English/Creole were more likely to have been tested for HIV; additionally, those who spoke minority languages were less likely to have been tested. Being married and having given birth were associated with increased odds of HIV testing. Living in rural areas and households with the poorest wealth indices were associated with decreased odds of being tested for HIV. Women with good HIV knowledge and accepting attitudes towards people living with HIV were more likely to be tested.

CONCLUSIONS

From 2006 to 2015, HIV testing in women of reproductive age showed an increasing trend in Belize. We recommend interventions to expand HIV testing for women of reproductive age in Belize, particularly those aged 15-24 years, speaking minority languages, living in rural areas, and having a low socioeconomic status.

Child marriage, marital disruption, and marriage thereafter: evidence from a national survey

BACKGROUND

This study examines the relationship between women's ages at their first marriages and the marital disruption among those who experienced child marriages and those who did not as well as identifies some compromises that women make in their remarriages after previous marital disruptions.

METHODS

The data of 57,476 women from the 2019 Bangladesh Multiple Indicator Cluster Survey were analysed using multivariable logistic and linear regressions. Women's compromises in their remarriages were examined by determining the age differences with their current husbands, whether the current husband has another wife and their attitudes toward the justification of intimate partner violence by husbands.

RESULTS

Almost 65% of women experienced child marriage, and its prevalence is higher in rural (66.5%) than in urban areas (59.2%). The probability of marital disruptions decreases as the ages at the first marriages rise among women who experienced child marriages and increase among women who did not. Women living in rural areas are less likely than those living in urban areas to report marital dissolution (AOR 0.81, 95% CI 0.73-0.90). Also, women who completed relatively more years of education or have greater wealth are less likely to report marital disruptions and those who never gave birth are more likely to report these (AOR 3.54, 95% CI 3.14-3.99). Women who remarried after previous disruptions are more likely to report that their new husbands are, on average, almost 12 years older than they are, and have another wife. Also, those who experienced marital disruptions are more likely than others to believe that husbands are justified in beating their wives in certain circumstances.

CONCLUSION

The odd of marital disruption decreases with the ages at first marriage among women who experienced child marriage and increase among women who did not. There is a curvilinear relationship between women's ages at their first marriages and the probability of marital disruptions. Making compromises in remarriages after disruptions is common. Because marital disruption is increasing, appropriate policies are needed to address the adverse outcomes of divorces that ensue.

Parents' Attitudes Toward Domestic Violence as a Risk Factor for Early Childhood Development: Testing an Actor-Partner Interdependence Model Using UNICEF MICS

Purpose: While the importance of domestic violence has been examined in relation to parenting behaviors and child development, less is known about the link between justifying attitudes toward wife beating and parenting, especially in low- and middle-income countries (LMIC). This study employs an actor-partner interdependence mediation model to examine how parents' justifying attitudes toward violence against women relate to their own (actor effects) and their partners' (partner effects) level of parental involvement, which then influence their preschool children's early development. Method: Using data from mothers, fathers, and children in 16,010 families residing in LMIC that participated in UNICEF's Multiple Indicator Cluster Survey Waves 4-5-6 and a dyadic mediation modeling approach, we have examined the associations between maternal and paternal justification of violence against women, parental involvement, and children's early development. Results: Results revealed that mothers' greater justification of violence against themselves were associated with decreased level of maternal (actor effect) and paternal (partner effect) involvement, while fathers' greater justification of violence against their wives was related to decreased paternal involvement (actor effect). Furthermore, mediation tests indicated that paternal justification of violence was negatively and indirectly associated with early childhood development through paternal involvement. Moreover, maternal justification of violence was negatively and indirectly associated with early childhood development through both maternal and paternal involvement. Conclusion: Mothers' and fathers' justifying attitudes toward domestic violence emerge as a possible risk factor for child development that can be addressed by preventive interventions.

Married Women's Attitude toward Intimate Partner Violence Is Influenced by Exposure to Media: A Population-Based Spatial Study in Bangladesh

This study estimated the attitudes of women toward accepting IPV at district level in Bangladesh and examined its relationship with sociodemographic predictors including exposure to media (e.g., newspaper, radio and television) using the Multiple Indicator Cluster Survey-2019 with a sample of 63,689 women. Around 25.6% women accepted IPV that geographically varied from 1.78% (Pirojpur) to 57.14% (Kurigram). Women regularly exposed to media were 17% less likely to accept IPV. Attitude toward accepting IPV was found to be higher among the illiterate women in disadvantaged circumstances, patriotically from poorer households living in remote areas, which suggest that planned interventions are needed for this vulnerable group of women to improve their living status by providing access to education and media. Further research is necessary to assess the impact of women's empowerment on their attitude toward acceptance of IPV.

Trends in Socioeconomic Inequalities in the Use of Antenatal Care Services by Women Aged 15 to 49 Years in Vietnam

Despite considerable variations in the use of antenatal care (ANC) services in Vietnam, limited information is available on socioeconomic inequalities concerning the use of ANC services. This study aimed to assess the trends and changes in socioeconomic inequalities in the use of ANC services by women aged 15 to 49 years in Vietnam from 2006 to 2014. We used data from the Multiple Indicator Cluster Survey conducted in 2006, 2011, and 2014. The percentage of women who received ANC services increased significantly from 26.5% in 2006 to 42.7% in 2011 and reached 56.6% in 2014. We found a decreasing trend in the concentration indices of the use of ANC services from 0.36 in 2006 to 0.19 in 2014. The common factors significantly associated with the higher percentage of the use of ANC services in 2006, 2011, and 2014 were women belonging to the Kinh and Hoa ethnic groups and belonging to wealthier groups. Our study showed a reduction in socioeconomic inequality in the use of ANC services between 2004 and 2014. However, significant inequalities still exist in the use of ANC services based on women's education, ethnicity, and economic status.

Water-sanitation-hygiene mapping: an improved approach for data collection at local level

Strategic planning and appropriate development and management of water and sanitation services are strongly supported by accurate and accessible data. If adequately exploited, these data might assist water managers with performance monitoring, benchmarking comparisons, policy progress evaluation, resources allocation, and decision making. A variety of tools and techniques are in place to collect such information. However, some methodological weaknesses arise when developing an instrument for routine data collection, particularly at local level: i) comparability problems due to heterogeneity of indicators, ii) poor reliability of collected data, iii) inadequate combination of different information sources, and iv) statistical validity of produced estimates when disaggregated into small geographic subareas. This study proposes an improved approach for water, sanitation and hygiene (WASH) data collection at decentralised level in low income settings, as an attempt to overcome previous shortcomings. The ultimate aim is to provide local policymakers with strong evidences to inform their planning decisions. The survey design takes the Water Point Mapping (WPM) as a starting point to record all available water sources at a particular location. This information is then linked to data produced by a household survey. Different survey instruments are implemented to collect reliable data by employing a variety of techniques, such as structured questionnaires, direct observation and water quality testing. The collected data is finally validated through simple statistical analysis, which in turn produces valuable outputs that might feed into the decision-making process. In order to demonstrate the applicability of the method, outcomes produced from three different case studies (Homa Bay District-Kenya-; Kibondo District-Tanzania-; and Municipality of Manhiça-Mozambique-) are presented.