Optimizing spatio-temporal correlation structures for modeling food security in Africa: a simulation-based investigation.

This study investigates the impact of spatio- temporal correlation using four spatio-temporal models: Spatio-Temporal Poisson Linear Trend Model (SPLTM), Poisson Temporal Model (TMS), Spatio-Temporal Poisson Anova Model (SPAM), and Spatio-Temporal Poisson Separable Model (STSM) concerning food security and nutrition in Africa. Evaluating model goodness of fit using the Watanabe Akaike Information Criterion (WAIC) and assessing bias through root mean square error and mean absolute error values revealed a consistent monotonic pattern. SPLTM consistently demonstrates a propensity for overestimating food security, while TMS exhibits a diverse bias profile, shifting between overestimation and underestimation based on varying correlation settings. SPAM emerges as a beacon of reliability, showcasing minimal bias and WAIC across diverse scenarios, while STSM consistently underestimates food security, particularly in regions marked by low to moderate spatio-temporal correlation. SPAM consistently outperforms other models, making it a top choice for modeling food security and nutrition dynamics in Africa. This research highlights the impact of spatial and temporal correlations on food security and nutrition patterns and provides guidance for model selection and refinement. Researchers are encouraged to meticulously evaluate the biases and goodness of fit characteristics of models, ensuring their alignment with the specific attributes of their data and research goals. This knowledge empowers researchers to select models that offer reliability and consistency, enhancing the applicability of their findings.

Key predictors of food security and nutrition in Africa: a spatio-temporal model-based study.

There is voluminous literature on Food Security in Africa. This study explicitly considers the spatio-temporal factors in addition to the usual FAO-based metrics in modeling and understanding the dynamics of food security and nutrition across the African continent. To better understand the complex trajectory and burden of food insecurity and nutrition in Africa, it is crucial to consider space-time factors when modeling and interpreting food security. The spatio-temporal anova model was found to be superior(employing statistical criteria) to the other three models from the spatio-temporal interaction domain models. The results of the study suggest that dietary supply adequacy, food stability, and consumption status are positively associated with severe food security, while average food supply and environmental factors have negative effects on Food Security and Nutrition. The findings also indicate that severe food insecurity and malnutrition are spatially and temporally correlated across the African continent. Spatio-temporal modeling and spatial mapping are essential components of a comprehensive practice to reduce the burden of severe food insecurity. likewise, any planning and intervention to improve the average food supply and environment to promote sustainable development should be regional instead of one size fit all.

Application of machine learning methods for predicting under-five mortality: analysis of Nigerian demographic health survey 2018 dataset.

BACKGROUND: Under-five mortality remains a significant public health issue in developing countries. This study aimed to assess the effectiveness of various machine learning algorithms in predicting under-five mortality in Nigeria and identify the most relevant predictors. METHODS: The study used nationally representative data from the 2018 Nigeria Demographic and Health Survey. The study evaluated the performance of the machine learning models such as the artificial neural network, k-nearest neighbourhood, Support Vector Machine, Naïve Bayes, Random Forest, and Logistic Regression using the true positive rate, false positive rate, accuracy, precision, F-measure, Matthew's correlation coefficient, and the Area Under the Receiver Operating Characteristics. RESULTS: The study found that machine learning models can accurately predict under-five mortality, with the Random Forest and Artificial Neural Network algorithms emerging as the best models, both achieving an accuracy of 89.47% and an AUROC of 96%. The results show that under-five mortality rates vary significantly across different characteristics, with wealth index, maternal education, antenatal visits, place of delivery, employment status of the woman, number of children ever born, and region found to be the top determinants of under-five mortality in Nigeria. CONCLUSIONS: The findings suggest that machine learning models can be useful in predicting U5M in Nigeria with high accuracy. The study emphasizes the importance of addressing social, economic, and demographic disparities among the population in Nigeria. The study's findings can inform policymakers and health workers about developing targeted interventions to reduce under-five mortality in Nigeria.

Impact of ambient air pollution exposure during pregnancy on adverse birth outcomes: generalized structural equation modeling approach.

BACKGROUND: Air pollution and several prenatal factors, such as socio-demographic, behavioural, physical activity and clinical factors influence adverse birth outcomes. The study aimed to investigate the impact of ambient air pollution exposure during pregnancy adjusting prenatal risk factors on adverse birth outcomes among pregnant women in MACE birth cohort. METHODS: Data for the study was obtained from the Mother and Child in the Environment (MACE) birth cohort study in Durban, South Africa from 2013 to 2017. Land use regression models were used to determine household level prenatal exposure to PM2.5, SO2 and NOx. Six hundred and fifty-six births of pregnant females were selected from public sector antenatal clinics in low socio-economic neighbourhoods. We employed a Generalised Structural Equation Model with a complementary log-log-link specification. RESULTS: After adjustment for potential prenatal factors, the results indicated that exposure to PM2.5 was found to have both significant direct and indirect effects on the risk of all adverse birth outcomes. Similarly, an increased level of maternal exposure to SO2 during pregnancy was associated with an increased probability of being small for gestational age. Moreover, preterm birth act a mediating role in the relationship of exposure to PM2.5, and SO2 with low birthweight and SGA. CONCLUSIONS: Prenatal exposure to PM2.5 and SO2 pollution adversely affected birth outcomes after controlling for other prenatal risk factors. This suggests that local government officials have a responsibility for better control of air pollution and health care providers need to advise pregnant females about the risks of air pollution during pregnancy.

Rural-urban differentials in child body mass index over time.

BACKGROUND: The body mass index is a simple index based on weight and height that can be used to screen children and adults for potential weight problems. The objective of this study was to investigate urban-rural variations in child BMI and its distribution from 2006 to 2016 in four low and middle-income countries. METHODS: This study used data from the Young Lives prospective cohort study conducted in Ethiopia, India, Peru, and Vietnam to assess the BMI change for children aged 5 to 15 between 2006 and 2016. We adopted a mixed-effect model to analyze the data. RESULTS: The study revealed substantial changes and rises in BMI in Vietnam, Peru, India, and Ethiopia between 2006 and 2016. Peru had the highest BMI changes in both urban-rural areas. A low BMI was observed in Ethiopia and India. Urban-rural differences had a significant role in determining BMI variation. In urban Ethiopia, the mean BMI increased from 14.56 kg/m2 to 17.52 kg/m2, and in rural areas, it increased from 14.57 kg/m2 to 16.67 kg/m2. Similarly, in urban Vietnam, the BMI increased from 16 kg/m2 to 20.3 kg/m2, and in rural areas, it increased from 14.69 kg/m2 to 18.93 kg/m2. CONCLUSIONS: The findings showed an increase in BMI changes in Ethiopia, India, Peru, and Vietnam from 2006 to 2016. Urban-rural differences have a significant contribution to determining BMI variation.

Classification of imbalanced data using machine learning algorithms to predict the risk of renal graft failures in Ethiopia.

INTRODUCTION: The prevalence of end-stage renal disease has raised the need for renal replacement therapy over recent decades. Even though a kidney transplant offers an improved quality of life and lower cost of care than dialysis, graft failure is possible after transplantation. Hence, this study aimed to predict the risk of graft failure among post-transplant recipients in Ethiopia using the selected machine learning prediction models. METHODOLOGY: The data was extracted from the retrospective cohort of kidney transplant recipients at the Ethiopian National Kidney Transplantation Center from September 2015 to February 2022. In response to the imbalanced nature of the data, we performed hyperparameter tuning, probability threshold moving, tree-based ensemble learning, stacking ensemble learning, and probability calibrations to improve the prediction results. Merit-based selected probabilistic (logistic regression, naive Bayes, and artificial neural network) and tree-based ensemble (random forest, bagged tree, and stochastic gradient boosting) models were applied. Model comparison was performed in terms of discrimination and calibration performance. The best-performing model was then used to predict the risk of graft failure. RESULTS: A total of 278 completed cases were analyzed, with 21 graft failures and 3 events per predictor. Of these, 74.8% are male, and 25.2% are female, with a median age of 37. From the comparison of models at the individual level, the bagged tree and random forest have top and equal discrimination performance (AUC-ROC = 0.84). In contrast, the random forest has the best calibration performance (brier score = 0.045). Under testing the individual model as a meta-learner for stacking ensemble learning, the result of stochastic gradient boosting as a meta-learner has the top discrimination (AUC-ROC = 0.88) and calibration (brier score = 0.048) performance. Regarding feature importance, chronic rejection, blood urea nitrogen, number of post-transplant admissions, phosphorus level, acute rejection, and urological complications are the top predictors of graft failure. CONCLUSIONS: Bagging, boosting, and stacking, with probability calibration, are good choices for clinical risk predictions working on imbalanced data. The data-driven probability threshold is more beneficial than the natural threshold of 0.5 to improve the prediction result from imbalanced data. Integrating various techniques in a systematic framework is a smart strategy to improve prediction results from imbalanced data. It is recommended for clinical experts in kidney transplantation to use the final calibrated model as a decision support system to predict the risk of graft failure for individual patients.

Identifying Potential Factors Associated with High HIV viral load in KwaZulu-Natal, South Africa using Multiple Correspondence Analysis and Random Forest Analysis.

BACKGROUND: Sustainable Human Immunodeficiency Virus (HIV) virological suppression is crucial to achieving the Joint United Nations Programme of HIV/AIDS (UNAIDS) 95-95-95 treatment targets to reduce the risk of onward HIV transmission. Exploratory data analysis is an integral part of statistical analysis which aids variable selection from complex survey data for further confirmatory analysis. METHODS: In this study, we divulge participants' epidemiological and biological factors with high HIV RNA viral load (HHVL) from an HIV Incidence Provincial Surveillance System (HIPSS) sequential cross-sectional survey between 2014 and 2015 KwaZulu-Natal, South Africa. Using multiple correspondence analysis (MCA) and random forest analysis (RFA), we analyzed the linkage between socio-demographic, behavioral, psycho-social, and biological factors associated with HHVL, defined as ≥400 copies per m/L. RESULTS: Out of 3956 in 2014 and 3868 in 2015, 50.1% and 41% of participants, respectively, had HHVL. MCA and RFA revealed that knowledge of HIV status, ART use, ARV dosage, current CD4 cell count, perceived risk of contracting HIV, number of lifetime HIV tests, number of lifetime sex partners, and ever diagnosed with TB were consistent potential factors identified to be associated with high HIV viral load in the 2014 and 2015 surveys. Based on MCA findings, diverse categories of variables identified with HHVL were, did not know HIV status, not on ART, on multiple dosages of ARV, with less likely perceived risk of contracting HIV and having two or more lifetime sexual partners. CONCLUSION: The high proportion of individuals with HHVL suggests that the UNAIDS 95-95-95 goal of HIV viral suppression is less likely to be achieved. Based on performance and visualization evaluation, MCA was selected as the best and essential exploration tool for identifying and understanding categorical variables' significant associations and interactions to enhance individual epidemiological understanding of high HIV viral load. When faced with complex survey data and challenges of variables selection in research, exploratory data analysis with robust graphical visualization and reliability that can reveal divers' structures should be considered.

Space-time dynamics regression models to assess variations of composite index for anthropometric failure across the administrative zones in Ethiopia.

BACKGROUND: A single anthropometric index such as stunting, wasting, or underweight does not show the holistic picture of under-five children's undernutrition status. To alleviate this problem, we adopted a multifaceted single index known as the composite index for anthropometric failure (CIAF). Using this undernutrition index, we investigated the disparities of Ethiopian under-five children's undernutrition status in space and time. METHODS: Data for analysis were extracted from the Ethiopian Demographic and Health Surveys (EDHSs). The space-time dynamics models were formulated to explore the effects of different covariates on undernutrition among children under five in 72 administrative zones in Ethiopia. RESULTS: The general nested spatial-temporal dynamic model with spatial and temporal lags autoregressive components was found to be the most adequate (AIC = -409.33, R2 = 96.01) model. According to the model results, the increase in the percentage of breastfeeding mothers in the zone decreases the CIAF rates of children in the zone. Similarly, the increase in the percentages of parental education, and mothers' nutritional status in the zones decreases the CIAF rate in the zone. On the hand, increased percentages of households with unimproved water access, unimproved sanitation facilities, deprivation of women's autonomy, unemployment of women, and lower wealth index contributed to the increased CIAF rate in the zone. CONCLUSION: The CIAF risk factors are spatially and temporally correlated across 72 administrative zones in Ethiopia. There exist geographical differences in CIAF among the zones, which are influenced by spatial neighborhoods of the zone and temporal lags within the zone. Hence these findings emphasize the need to take the spatial neighborhood and historical/temporal contexts into account when planning CIAF prevention.

Shared component modelling of early childhood anaemia and malaria in Kenya, Malawi, Tanzania and Uganda.

BACKGROUND: Malaria and anaemia contribute substantially to child morbidity and mortality. In this study, we sought to jointly model the residual spatial variation in the likelihood of these two correlated diseases, while controlling for individual-level, household-level and environmental characteristics. METHODS: A child-level shared component model was utilised to partition shared and disease-specific district-level spatial effects. RESULTS: The results indicated that the spatial variation in the likelihood of malaria was more prominent compared to that of anaemia, for both the shared and specific spatial components. In addition, approximately 30% of the districts were associated with an increased likelihood of anaemia but a decreased likelihood of malaria. This suggests that there are other drivers of anaemia in children in these districts, which warrants further investigation. CONCLUSIONS: The maps of the shared and disease-specific spatial patterns provide a tool to allow for more targeted action in malaria and anaemia control and prevention, as well as for the targeted allocation of limited district health system resources.

Studying latent change process in height growth of children in Ethiopia, India, Peru and Vietnam.

BACKGROUND: Anthropometric measurements of healthy children differ in different parts of the world due to the diverse ethnicity and cultural backgrounds of families. In longitudinal studies, appropriate modeling of repeated anthropometric measures can improve the understanding of patterns of change, determinants of patterns, and variations in patterns of change over time. The objective of this study was to examine the latent change in physical height of children in Ethiopia, India, Peru, and Vietnam. METHOD: Longitudinal data of 6601 children aged 1 to 15 years were obtained from the Young Lives cohort study. The data were analyzed using a latent basis growth curve model. RESULTS: The findings of the study revealed that the rates of growth did not remain constant across the time intervals, which indicates the nonlinearity of the growth trajectory over time. For instance, children had the highest rate of growth between age 1 and 5 years, then between age 8 and 12 years, and a low rate of growth was observed between age 12 and 15 years. At the first measurement occasion (age 1 year) females were 0.826 cm (p < 0.0001) times shorter than males. The mean height at one year of age ranged from 72.13 cm in Ethiopia to 72.62 cm in India. Children in India and Vietnam had higher mean height at age one year. However, no significant difference in mean height at age one year was found between Ethiopian and Peruvian children, ([Formula: see text]). Peruvian and Vietnamese children grew at a faster rate, while Indian children grew at a slower rate than Ethiopian children. CONCLUSION: We found substantial latent growth variations among children in four low- and middle-income countries. The latent trajectories differed by gender and country. The outcomes of the study could aid in detecting inequalities in children's height growth.

Small Area Estimation of Undernutrition under Age Five Children Based on Spatial-Temporal Models.

The mean flow of direct survey estimates is mainly concerning the sample adequacy fulfillment unless it has been produced large variance estimates, and therefore, the small area estimations are developed to manage this flaw of the path. Small area estimation improved the direct survey estimates by borrowing strength from the census data and at the same time by using historical data from consecutive surveys. In this paper, we applied the spatiotemporal Fay-Herriot (STFH) model for producing fairly reliable disaggregate-level estimates of undernutrition indicators across all zones. The STFH model is an appropriately fitted model to the undernutrition data since it has the lowest information criteria (IC) value. The spatiotemporal estimates improved both the direct and spatial estimates of undernutrition under the FH model and have brought efficiency gain in the percent coefficient of variation (CV). These results may provide useful information to the government's planners, policymakers, and legislative organs for effective policy formulation and budget allocation in all zones.

Spatial data analysis of malnutrition among children under-five years in Ethiopia.

BACKGROUND: Childhood malnutrition is a major cause of child mortality under the age of 5 in the sub-Saharan Africa region. This study sought to identify the risk factors and spatial distribution of the composite index of anthropometric failure (CIAF). METHODS: Secondary data from 2000, 2005, 2011, and 2016 Ethiopian Health and Demographic Survey (EDHS) were used. The generalized geo-additive mixed model was adopted via the Integrated Nested Laplace Approximation (INLA) with a binomial family and logit link function. RESULTS: The CIAF status of children was found to be positively associated with the male gender, the potency of contracting a disease, and multiple births. However, it was negatively associated with family wealth quartiles, parental level of education, place of residence, unemployment status of mothers, improved sanitation, media exposure, and survey years. Moreover, the study revealed significant spatial variations on the level of CIAF among administrative zones. CONCLUSIONS: The generalized geo-additive mixed-effects model results identified gender of the child, presence of comorbidity, size of child at birth, dietary diversity, birth type, place of residence, age of the child, parental level of education, wealth index, sanitation facilities, and media exposure as main drivers of CIAF. The results would help decision-makers to develop and carry out target-oriented programs.

Mortality-related risk factors of COVID-19: a systematic review and meta-analysis of 42 studies and 423,117 patients.

BACKGROUND: Mortality rates of coronavirus disease-2019 (COVID-19) continue to rise across the world. The impact of several risk factors on coronavirus mortality has been previously reported in several meta-analyses limited by small sample sizes. In this systematic review, we aimed to summarize available findings on the association between comorbidities, complications, smoking status, obesity, gender, age and D-dimer, and risk of mortality from COVID-19 using a large dataset from a number of studies. METHOD: Electronic databases including Google Scholar, Cochrane Library, Web of Sciences (WOS), EMBASE, Medline/PubMed, COVID-19 Research Database, and Scopus, were systematically searched till 31 August 2020. We included all human studies regardless of language, publication date or region. Forty-two studies with a total of 423,117 patients met the inclusion criteria. To pool the estimate, a mixed-effect model was used. Moreover, publication bias and sensitivity analysis were evaluated. RESULTS: The findings of the included studies were consistent in stating the contribution of comorbidities, gender, age, smoking status, obesity, acute kidney injury, and D-dimer as a risk factor to increase the requirement for advanced medical care. The analysis results showed that the pooled prevalence of mortality among hospitalized patients with COVID-19 was 17.62% (95% CI 14.26-21.57%, 42 studies and 423,117 patients). Older age has shown increased risk of mortality due to coronavirus and the pooled odds ratio (pOR) and hazard ratio (pHR) were 2.61 (95% CI 1.75-3.47) and 1.31 (95% CI 1.11-1.51), respectively. A significant association were found between COVID-19 mortality and male (pOR = 1.45; 95% CI 1.41-1.51; pHR = 1.24; 95% CI 1.07-1.41), and current smoker (pOR = 1.42; 95% CI 1.01-1.83). Furthermore, risk of mortality among hospitalized COVID-19 patients is highly influenced by patients with Chronic Obstructive Pulmonary Disease (COPD), Cardiovascular Disease (CVD), diabetes, hypertension, obese, cancer, acute kidney injury and increase D-dimer. CONCLUSION: Chronic comorbidities, complications, and demographic variables including acute kidney injury, COPD, diabetes, hypertension, CVD, cancer, increased D-dimer, male gender, older age, current smoker, and obesity are clinical risk factors for a fatal outcome associated with coronavirus. The findings could be used for disease's future research, control and prevention.

The impact of antenatal care on under-five mortality in Ethiopia: a difference-in-differences analysis.

BACKGROUND: Sub-Saharan Africa, as opposed to other regions, has the highest under-five mortality rates yet makes the least improvement in reducing under-five mortality. Despite the decline, Ethiopia is among the top ten countries contributing the most to global under-five mortalities. This article examines the impact of the number of antenatal care and the timing of first antenatal care on child health outcomes. We specifically investigated if the utilization of antenatal care services positively affects the reduction of under-five mortality. METHODS: We employ a difference-in-differences design with propensity score matching to identify direct causal effects of antenatal care on under-five mortality based on the Ethiopian Demographic Health Survey data of 2011 and 2016. Our sample includes 22 295 women between the ages of 14-49 who had antenatal care visits at different times before delivery. RESULTS: The study revealed 1 481 cases of reported under-five mortality. 99.0% of that under-five mortality cases are women who had less than eight antenatal care visits, while only 1% of that is by women who had eight or more antenatal care visits. Antenatal care visit decreases the likelihood of under-five mortality in Ethiopia by 45.2% (CI = 19.2-71.3%, P-value < 0.001) while the timing of first antenatal care within the first trimester decreases the likelihood of under-five mortality by 10% (CI = 5.7-15.6%, P-value < 0.001). CONCLUSIONS: To achieve a significant reduction in the under-five mortality rate, Intervention programs that encourages more antenatal care visits should be considered. This will improve child survival and help in attaining Sustainable Development Goal targets.

Association of medical male circumcision and sexually transmitted infections in a population-based study using targeted maximum likelihood estimation.

BACKGROUND: Epidemiological theory and many empirical studies support the hypothesis that there is a protective effect of male circumcision against some sexually transmitted infections (STIs). However, there is a paucity of randomized control trials (RCTs) to test this hypothesis in the South African population. Due to the infeasibility of conducting RCTs, estimating marginal or average treatment effects with observational data increases interest. Using targeted maximum likelihood estimation (TMLE), a doubly robust estimation technique, we aim to provide evidence of an association between medical male circumcision (MMC) and two STI outcomes. METHODS: HIV and HSV-2 status were the two primary outcomes for this study. We investigated the associations between MMC and these STI outcomes, using cross-sectional data from the HIV Incidence Provincial Surveillance System (HIPSS) study in KwaZulu-Natal, South Africa. HIV antibodies were tested from the blood samples collected in the study. For HSV-2, serum samples were tested for HSV-2 antibodies via an ELISA-based anti-HSV-2 IgG. We estimated marginal prevalence ratios (PR) using TMLE and compared estimates with those from propensity score full matching (PSFM) and inverse probability of treatment weighting (IPTW). RESULTS: From a total 2850 male participants included in the analytic sample, the overall weighted prevalence of HIV was 32.4% (n = 941) and HSV-2 was 53.2% (n = 1529). TMLE estimates suggest that MMC was associated with 31% lower HIV prevalence (PR: 0.690; 95% CI: 0.614, 0.777) and 21.1% lower HSV-2 prevalence (PR: 0.789; 95% CI: 0.734, 0.848). The propensity score analyses also provided evidence of association of MMC with lower prevalence of HIV and HSV-2. For PSFM: HIV (PR: 0.689; 95% CI: 0.537, 0.885), and HSV-2 (PR: 0.832; 95% CI: 0.709, 0.975). For IPTW: HIV (PR: 0.708; 95% CI: 0.572, 0.875), and HSV-2 (PR: 0.837; 95% CI: 0.738, 0.949). CONCLUSION: Using a TMLE approach, we present further evidence of a protective association of MMC against HIV and HSV-2 in this hyper-endemic South African setting. TMLE has the potential to enhance the evidence base for recommendations that embrace the effect of public health interventions on health or disease outcomes.

Decomposing the urban-rural inequalities in the utilisation of maternal health care services: evidence from 27 selected countries in Sub-Saharan Africa.

BACKGROUND: There has been a substantial improvement in reducing maternal mortality in the Sub-Saharan African region. The vast rural-urban gap in maternal health outcomes, however, is obscured by this average achievement. This study attempts to measure the contribution of identified risk factors to describe the average rural-urban difference in the use of antenatal care, health facilities for delivery, and health professional assistance at delivery. METHOD: To achieve this objective, we used descriptive analysis and Fairlie non-linear decomposition method to quantify covariates' contribution in explaining the urban-rural difference in maternal healthcare services utilisation. RESULT: The study's finding shows much difference between urban and rural areas in the use of maternal healthcare services. Socio-economic factors such as household wealth index, exposure to media, and educational level of women and their husbands/partners contributed the most in explaining the gap between urban and rural areas in healthcare services utilisation. CONCLUSIONS: Interventions to bridge the gap between urban and rural areas in maternal healthcare services utilisation in Sub-Saharan Africa should be centred towards socio-economic empowerment. Government can enforce targeted awareness campaigns to encourage women in rural communities in Sub-Sharan Africa to take the opportunity and use the available maternal health care services to be at par with their counterparts in urban areas.

Maternal health refers to the health of women throughout pregnancy, delivery, and the postnatal period. Each step should be a good experience that ensures mothers, and their infants realize their maximum health and well-being potential. In this study, we used individual, demographic, and socio-economic characteristics to measure the urban­rural discrepancies in maternal health care services in Sub-Saharan Africa. We used Information of 220 164 women of child-bearing age (15­49) gathered from National Demographic Health Surveys from 27 countries in the Sub-Sahara African region. We found 46.1% of women in rural areas had no education, 39.7% of the women in rural areas have husbands/partners with no education, and 60.1% of the women in rural areas are from households with poor wealth indexes. The use of maternal health care services found to be predominant in the urban areas than rural areas, and the measure of this difference can inform policymakers on the level of effort that needed to be put in place to balance the discrepancies and improve maternal health in general.

A machine learning classifier approach for identifying the determinants of under-five child undernutrition in Ethiopian administrative zones.

BACKGROUND: Undernutrition is the main cause of child death in developing countries. This paper aimed to explore the efficacy of machine learning (ML) approaches in predicting under-five undernutrition in Ethiopian administrative zones and to identify the most important predictors. METHOD: The study employed ML techniques using retrospective cross-sectional survey data from Ethiopia, a national-representative data collected in the year (2000, 2005, 2011, and 2016). We explored six commonly used ML algorithms; Logistic regression, Least Absolute Shrinkage and Selection Operator (L-1 regularization logistic regression), L-2 regularization (Ridge), Elastic net, neural network, and random forest (RF). Sensitivity, specificity, accuracy, and area under the curve were used to evaluate the performance of those models. RESULTS: Based on different performance evaluations, the RF algorithm was selected as the best ML model. In the order of importance; urban-rural settlement, literacy rate of parents, and place of residence were the major determinants of disparities of nutritional status for under-five children among Ethiopian administrative zones. CONCLUSION: Our results showed that the considered machine learning classification algorithms can effectively predict the under-five undernutrition status in Ethiopian administrative zones. Persistent under-five undernutrition status was found in the northern part of Ethiopia. The identification of such high-risk zones could provide useful information to decision-makers trying to reduce child undernutrition.

Modelling HIV disease process and progression in seroconversion among South Africa women: using transition-specific parametric multi-state model.

BACKGROUND: HIV infected patients may experience many intermediate events including between-event transition throughout their follow up. Through modelling these transitions, we can gain a deeper understanding of HIV disease process and progression and of factors that influence the disease process and progression pathway. In this work, we present transition-specific parametric multi-state models to describe HIV disease process and progression. METHODS: The data is from an ongoing prospective cohort study conducted amongst adult women who were HIV-infected in KwaZulu-Natal, South Africa. Participants were enrolled during the acute HIV infection phase and then followed up during chronic infection, up to ART initiation. RESULTS: Transition specific distributions for multi-state models, including a variety of accelerated failure time (AFT) models and proportional hazards (PH) models, were presented and compared in this study. The analysis revealed that women enrolling with a CD4 count less than 350 cells/mm3 (severe and advanced disease stages) had a far lower chance of immune recovery, and a considerably higher chance of immune deterioration, compared to women enrolling with a CD4 count of 350 cells/mm3 or more (normal and mild disease stages). Our analyses also showed that older age, higher educational levels, higher scores for red blood cell counts, higher mononuclear scores, higher granulocytes scores, and higher physical health scores, all had a significant effect on a shortened time to immunological recovery, while women with many sex partners, higher viral load and larger family size had a significant effect on accelerating time to immune deterioration. CONCLUSION: Multi-state modelling of transition-specific distributions offers a flexible tool for the study of demographic and clinical characteristics' effects on the entire disease progression pathway. It is hoped that the article will help applied researchers to familiarize themselves with the models, including interpretation of results.

Modelling of viral load dynamics and CD4 cell count progression in an antiretroviral naive cohort: using a joint linear mixed and multistate Markov model.

BACKGROUND: Patients infected with HIV may experience a succession of clinical stages before the disease diagnosis and their health status may be followed-up by tracking disease biomarkers. In this study, we present a joint multistate model for predicting the clinical progression of HIV infection which takes into account the viral load and CD4 count biomarkers. METHODS: The data is from an ongoing prospective cohort study conducted among antiretroviral treatment (ART) naïve HIV-infected women in the province of KwaZulu-Natal, South Africa. We presented a joint model that consists of two related submodels: a Markov multistate model for CD4 cell count transitions and a linear mixed effect model for longitudinal viral load dynamics. RESULTS: Viral load dynamics significantly affect the transition intensities of HIV/AIDS disease progression. The analysis also showed that patients with relatively high educational levels (ß = - 0.004; 95% confidence interval [CI]:-0.207, - 0.064), high RBC indices scores (ß = - 0.01; 95%CI:-0.017, - 0.002) and high physical health scores (ß = - 0.001; 95%CI:-0.026, - 0.003) were significantly were associated with a lower rate of viral load increase over time. Patients with TB co-infection (ß = 0.002; 95%CI:0.001, 0.004), having many sex partners (ß = 0.007; 95%CI:0.003, 0.011), being younger age (ß = 0.008; 95%CI:0.003, 0.012) and high liver abnormality scores (ß = 0.004; 95%CI:0.001, 0.01) were associated with a higher rate of viral load increase over time. Moreover, patients with many sex partners (ß = - 0.61; 95%CI:-0.94, - 0.28) and with a high liver abnormality score (ß = - 0.17; 95%CI:-0.30, - 0.05) showed significantly reduced intensities of immunological recovery transitions. Furthermore, a high weight, high education levels, high QoL scores, high RBC parameters and being of middle age significantly increased the intensities of immunological recovery transitions. CONCLUSION: Overall, from a clinical perspective, QoL measurement items, being of a younger age, clinical attributes, marital status, and educational status are associated with the current state of the patient, and are an important contributing factor to extend survival of the patients and guide clinical interventions. From a methodological perspective, it can be concluded that a joint multistate model approach provides wide-ranging information about the progression and assists to provide specific dynamic predictions and increasingly precise knowledge of diseases.

Multilevel ordinal model for CD4 count trends in seroconversion among South Africa women.

BACKGROUND: Ordinal health longitudinal response variables have distributions that make them unsuitable for many popular statistical models that assume normality. We present a multilevel growth model that may be more suitable for medical ordinal longitudinal outcomes than are statistical models that assume normality and continuous measurements. METHODS: The data is from an ongoing prospective cohort study conducted amongst adult women who are HIV-infected patients in Kwazulu-Natal, South Africa. Participants were enrolled into the acute infection, then into early infection subsequently into established infection and afterward on cART. Generalized linear multilevel models were applied. RESULTS: Multilevel ordinal non-proportional and proportional-odds growth models were presented and compared. We observed that the effects of covariates can't be assumed identical across the three cumulative logits. Our analyses also revealed that the rate of change of immune recovery of patients increased as the follow-up time increases. Patients with stable sexual partners, middle-aged, cART initiation, and higher educational levels were more likely to have better immunological stages with time. Similarly, patients having high electrolytes component scores, higher red blood cell indices scores, higher physical health scores, higher psychological well-being scores, a higher level of independence scores, and lower viral load more likely to have better immunological stages through the follow-up time. CONCLUSION: It can be concluded that the multilevel non-proportional-odds method provides a flexible modeling alternative when the proportional-odds assumption of equal effects of the predictor variables at every stage of the response variable is violated. Having higher clinical parameter scores, higher QoL scores, higher educational levels, and stable sexual partners were found to be the significant factors for trends of CD4 count recovery.