Explore the vast range of titles available.

High rates of childhood vaccination defaulting remain a significant barrier to achieving full vaccination coverage in sub-Saharan Africa, contributing to preventable morbidity and mortality. This study evaluated the utility of machine learning algorithms for predicting childhood vaccination defaulters in Ghana, addressing the limitations of traditional statistical methods when handling complex, high-dimensional health data. Using a merged dataset from two malaria vaccine pilot surveys, we engineered novel temporal features, including vaccination timing windows and birth seasonality. Six algorithms, namely logistic regression, support vector machine, random forest, gradient boosting machine, extreme gradient boosting, and artificial neural networks, were compared. Models were trained and validated on both original and synthetically balanced and augmented data. The results showed higher performance across the ensemble tree classifiers. The random forest and extreme gradient boosting models reported the highest F1 scores (0.92) and AUCs (0.95) on augmented unseen data. The key predictors identified include timely receipt of birth and week six vaccines, the child’s age, household wealth index, and maternal education. The findings demonstrate that robust machine learning frameworks, combined with temporal and contextual feature engineering, can improve defaulter risk prediction accuracy. Integrating such models into routine immunization programs could enable data-driven targeting of high-risk groups, supporting policymakers in strategies to close vaccination coverage gaps.

Globally, low and middle-income countries (LMICs) account for the majority of the adverse pregnancy outcomes. Risk prediction models (RPMs) can guide physicians in making clinical decisions to improve maternal and newborn health. However, there is scanty data on RPMs in determining adverse maternal and newborn outcomes in LMICs. Hence, this scoping review aims to describe the RPMs and the risk factors which have been used to determine both maternal and newborn adverse outcomes of pregnancy in LMICs. This scoping review will be guided by the Preferred Reporting Items for Systematic Reviews and Meta-analysis extension for Scoping Reviews (PRISMA-ScR) and the JBI methodology for scoping reviews. The review would employ the Population, Concept, Context (PCC) framework to include studies that reported RPMs to determine either adverse maternal or newborn outcomes of pregnancy or both in LMICs. A literature search will be conducted in four databases for both published and unpublished articles on RPMs for adverse maternal or newborn outcomes from January 1, 2000, to June 26, 2024. We will use the JBI approach for study selection, data extraction, and presentation. The screening and data extraction will be conducted by two independent reviewers. This scoping review will provide a comprehensive assessment of RPMs for adverse maternal and newborn outcomes in LMICs. This study will help gain knowledge on the up-to-date literature on risk prediction models for adverse pregnancy outcomes which can be useful for researchers and clinicians in making clinical decisions. Review registration: Open Science Framework https://doi.org/10.17605/OSF.IO/B9CKJ.

Background Malaria remains a major public health problem, especially among children in sub-Saharan Africa. Knowledge of malaria parasite prevalence informs targeted interventions and helps to monitor the effectiveness of those interventions. This study aimed to determine prevalence and factors associated with malaria in children aged 6 months to 10 years in the Greater Accra Region of Ghana. Methods A community-based cross-sectional study was conducted among 8,741 children aged 6–59 months and 8,292 children aged 5–10 years in all 29 districts of the Greater Accra Region of Ghana in October 2020. Systematic random sampling was used to select communities, households and study participants. A structured questionnaire was used to collect data from caregivers. Rapid diagnostic test kits were used to determine the presence of malaria parasites in blood samples collected by fingerprick. Factors associated with malaria RDT-positivity were determined using multivariate logistic regression analysis. Results Of 8727 children aged 6–59 months and 8279 aged 5–10 years from whom blood samples were obtained, positive results were obtained for 289 (3.3%; 95% CI 3.0–3.7) and 406 (4.9%; 95% CI 4.5–5.4) respectively. Malaria parasite prevalence in the districts ranged from 0.9 to 10.7% and 1.4–15.0% in children aged 6–59 months and 5–10 years respectively. Factors associated with increased odds of malaria included higher age (AOR = 1.43; 95% CI 1.14–1.71), and living in households without nets on the windows (AOR 1.64; 95% CI 1.10–2.45). On the other hand, living in households located in urban communities was associated with a lower risk of malaria (AOR 0.56; 95% CI 0.40–0.78). Conclusion The average prevalence of malaria in the Greater Accra Region is low compared with other regions. However, there are potential hotspots that need to be targeted with appropriate interventions to accelerate the drive towards malaria elimination.

Background There is growing interest in the use of reliable evidence for health decision-making among low-and middle-income countries. Ghana has deployed DHIMS2 to replace the previously existing manual data harmonization processes. Methods This cross-sectional study was conducted in 12 districts comprising 12 district directorates, 10 district hospitals, 29 sub-district health centers, and 38 community health facilities in the Brong-Ahafo Region. Data collection tools were developed based on the Measure Evaluate assessment tools designed for evaluating the performance of routine information systems management tools. Utilization was assessed based on documented evidence and data was analyzed using STATA version 14. Results Although 93% of the health facilities studied submitted data unto the DHIMS2 platform, evidence suggested low use of this data in decision-making, particularly at the community level facilities where only 26% of the facilities used data from DHIMS2 to inform annual action plans and even less than 20% examined findings and issued directives for action. At the district level, 58% issued directives based on DHIMS2 information, 50% used DHIMS2 information for Advocacy purposes and 58% gave feedback reports based on DHIMS2 data for action. Functional computers were lacking across all facilities. Conclusions Activities relating to the use of DHIMS2 information skew towards data quality checking with less focus on examining findings, making comparisons, and taking action-based decisions from findings and comparisons. Improving factors like internet access, availability of functional ICTs, frequency of supervisory visits, staff training and the provision of training manuals may facilitate the use of DHIMS2 in decision-making at all levels of the district health system.

Background In 2019, the RTS,S/AS01 E malaria vaccine (RTS,S) was introduced into Ghana’s routine health system as part of the Malaria Vaccine Implementation Programme (MVIP). Household surveys were conducted prior to vaccine introduction and approximately 18 and 30 months post-introduction. We present a description of the area in Ghana based on the baseline household survey including malaria prevalence, malnutrition, wealth, insecticide-treated net (ITN) coverage, other health interventions (deworming, Vitamin A supplementation (VAS)), coverage of Expanded Programme on Immunization (EPI) vaccines, and health-seeking behaviour for febrile children. Methods The baseline household survey was conducted between 25 February and 18 March 2019 in a representative sample of 6778 households across 66 districts (33 in each of the implementing and comparator areas) in Ghana. Caregivers of children aged 5–48 months were interviewed. For each child, vaccination details were transcribed from the maternal and child health record book, and we measured the mid-upper arm circumference and obtained a malaria Rapid Diagnostic Test (RDT). Survey-weighted coverage estimates were obtained using standard survey methods. Survey Poisson regression was used to estimate prevalence ratios. Results Overall, 7768 children were included in the study, and 21% (95% CI 18–23) tested positive for malaria parasitemia by RDT. About 87%, 95%CI (85–89) of all households owned at least one ITN, and 62%, 95%CI (59–64) of children aged 5–48 months slept under an insecticide-treated net (ITN) the night before the survey. Additionally, 22%, 95%CI (21–24) of children reported having fever in the two weeks preceding the survey; among those with reported fever, 72%, 95%CI (69–74) sought advice or treatment, 40%, 95%CI (37–44) were tested for malaria, and 42%, 95%CI (39–46) of those with fever took an antimalarial drug. Additionally, 17%, 95%CI (16–19) had a mid-upper arm circumference (MUAC) ≤ 13.5 cm, and 1%, 95%CI (0–1) had a (MUAC) ≤ 11.5 cm. The uptake of vitamin A VAS in the 6 months prior to the survey was 36%, based on routine delivery through EPI, and deworming coverage was 29%. Coverage of EPI vaccines was > 90%. Indicators in comparison and implementation areas were comparable. Conclusions The pilot implementation and evaluation of the RTS,S malaria vaccine in Ghana was conducted in an area with substantial malaria transmission and illness, modest health-seeking behaviour and ITN use, and good EPI vaccine coverage. This study has established the baseline comparability between implementation and comparator areas, which serves as the foundation for future feasibility assessments.

Childhood vaccination saves millions of lives yearly, yet over a million children in low-and middle-income countries die from vaccine-preventable diseases each year. Predicting childhood vaccination defaulter risk with analytical models requires understanding how to represent different individual demographics, community structures, and environmental factors that feed input data. This review explores features for analysing childhood vaccination defaulter risk in low-resource settings with a focus on feature encoding, engineering and representation. Articles published from 2018 to January 2025 were searched using PubMed, Google Scholar, ACM Digital Library, and references from the searched articles. Search was limited to low- and middle-income countries, focusing on African countries. We included studies that utilised either statistics or machine learning for analysis. Of the 4,174 articles retrieved, 55 were eligible, 41 were then excluded after full-text review, and 4 were added from references. Cross-cutting features included maternal education and health service utilisation. Novel features included community rates of poverty, maternal education and maternal unemployment. Variations in encoding strategies, engineering techniques and feature representation were marginal. Categorical data were mainly encoded as binary inputs, while features with high dimensionality like socio-economic status were condensed by using principal component analysis. A review of existing feature representations can serve as a feature construction reference to improve the exploitation of machine learning techniques within the context of childhood vaccination defaulter risk prediction. Future studies can exploit other representations different from binary encoding, like frequency encoding, to introduce elements of weighting into multi-categorical features.

The reliability of counts for estimating population dynamics and disease burdens in communities depends on the availability of a common unique identifier for matching general population data with health facility data. Biometric data has been explored as a feasible common identifier between the health data and sociocultural data of resident members in rural communities within the Kintampo Health and Demographic Surveillance System located in the central part of Ghana. Our goal was to assess the feasibility of using fingerprint identification to link community data and hospital data in a rural African setting. A combination of biometrics and other personal identification techniques were used to identify individual's resident within a surveillance population seeking care in two district hospitals. Visits from resident individuals were successfully recorded and categorized by the success of the techniques applied during identification. The successes of visits that involved identification by fingerprint were further examined by age. A total of 27,662 hospital visits were linked to resident individuals. Over 85% of those visits were successfully identified using at least one identification method. Over 65% were successfully identified and linked using their fingerprints. Supervisory support from the hospital administration was critical in integrating this identification system into its routine activities. No concerns were expressed by community members about the fingerprint registration and identification processes. Fingerprint identification should be combined with other methods to be feasible in identifying community members in African rural settings. This can be enhanced in communities with some basic Demographic Surveillance System or census information.

Background The reliability of counts for estimating population dynamics and disease burdens in communities depends on the availability of a common unique identifier for matching general population data with health facility data. Biometric data has been explored as a feasible common identifier between the health data and sociocultural data of resident members in rural communities within the Kintampo Health and Demographic Surveillance System located in the central part of Ghana. Objective Our goal was to assess the feasibility of using fingerprint identification to link community data and hospital data in a rural African setting. Design A combination of biometrics and other personal identification techniques were used to identify individual's resident within a surveillance population seeking care in two district hospitals. Visits from resident individuals were successfully recorded and categorized by the success of the techniques applied during identification. The successes of visits that involved identification by fingerprint were further examined by age. Results A total of 27,662 hospital visits were linked to resident individuals. Over 85% of those visits were successfully identified using at least one identification method. Over 65% were successfully identified and linked using their fingerprints. Supervisory support from the hospital administration was critical in integrating this identification system into its routine activities. No concerns were expressed by community members about the fingerprint registration and identification processes. Conclusions Fingerprint identification should be combined with other methods to be feasible in identifying community members in African rural settings. This can be enhanced in communities with some basic Demographic Surveillance System or census information.