Predictors of Contemporary under-5 Child Mortality in Low- and Middle-Income Countries: A Machine Learning Approach.
Ontology highlight
ABSTRACT:
SUBMITTER: Bizzego A
PROVIDER: S-EPMC7908191 | biostudies-literature | 2021 Feb
REPOSITORIES: biostudies-literature
Publications
Predictors of Contemporary under-5 Child Mortality in Low- and Middle-Income Countries: A Machine Learning Approach.
Bizzego Andrea A Gabrieli Giulio G Bornstein Marc H MH Deater-Deckard Kirby K Lansford Jennifer E JE Bradley Robert H RH Costa Megan M Esposito Gianluca G
International journal of environmental research and public health 20210201 3
Child Mortality (CM) is a worldwide concern, annually affecting as many as 6.81% children in low- and middle-income countries (LMIC). We used data of the Multiple Indicators Cluster Survey (MICS) (N = 275,160) from 27 LMIC and a machine-learning approach to rank 37 distal causes of CM and identify the top 10 causes in terms of predictive potency. Based on the top 10 causes, we identified households with improved conditions. We retrospectively validated the results by investigating the associatio ...[more]
Similar Datasets
Project description:ObjectivesTo explore the efficacy of machine learning (ML) techniques in predicting under-five mortality (U5M) in low-income and middle-income countries (LMICs) and to identify significant predictors of U5M.DesignThis is a cross-sectional, proof-of-concept study.Settings and participantsWe analysed data from the Demographic and Health Survey. The data were drawn from 34 LMICs, comprising a total of n=1 520 018 children drawn from 956 995 unique households.Primary and secondary outcome measuresThe primary outcome measure was U5M; secondary outcome was comparing the efficacy of deep learning algorithms: deep neural network (DNN); convolution neural network (CNN); hybrid CNN-DNN with logistic regression (LR) for the prediction of child's survival.ResultsWe found that duration of breast feeding, number of antenatal visits, household wealth index, postnatal care and the level of maternal education are some of the most important predictors of U5M. We found that deep learning techniques are superior to LR for the classification of child survival: LR sensitivity=0.47, specificity=0.53; DNN sensitivity=0.69, specificity=0.83; CNN sensitivity=0.68, specificity=0.83; CNN-DNN sensitivity=0.71, specificity=0.83.ConclusionOur findings provide an understanding of determinants of U5M in LMICs. It also demonstrates that deep learning models are more efficacious than traditional analytical approach.
| S-EPMC7430449 | biostudies-literature
Project description:INTRODUCTION:Machine learning (ML) has been used in bio-medical research, and recently in clinical and public health research. However, much of the available evidence comes from high-income countries, where different health profiles challenge the application of this research to low/middle-income countries (LMICs). It is largely unknown what ML applications are available for LMICs that can support and advance clinical medicine and public health. We aim to address this gap by conducting a scoping review of health-related ML applications in LMICs. METHODS AND ANALYSIS:This scoping review will follow the methodology proposed by Levac et al. The search strategy is informed by recent systematic reviews of ML health-related applications. We will search Embase, Medline and Global Health (through Ovid), Cochrane and Google Scholar; we will present the date of our searches in the final review. Titles and abstracts will be screened by two reviewers independently; selected reports will be studied by two reviewers independently. Reports will be included if they are primary research where data have been analysed, ML techniques have been used on data from LMICs and they aimed to improve health-related outcomes. We will synthesise the information following evidence mapping recommendations. ETHICS AND DISSEMINATION:The review will provide a comprehensive list of health-related ML applications in LMICs. The results will be disseminated through scientific publications. We also plan to launch a website where ML models can be hosted so that researchers, policymakers and the general public can readily access them.
| S-EPMC7223147 | biostudies-literature
Project description:BackgroundIncome inequality and national wealth are strong determinants for health, but few studies have systematically investigated their influence on mortality across the early life-course, particularly outside the high-income world.MethodsWe performed cross-sectional regression analyses of the relationship between income inequality (national Gini coefficient) and national wealth (Gross Domestic Product (GDP) averaged over previous decade), and all-cause and grouped cause national mortality rate amongst infants, 1-4, 5-9, 10-14, 15-19 and 20-24 year olds in low and middle-income countries (LMIC) in 2012. Gini models were adjusted for GDP.ResultsData were available for 103 (79%) countries. Gini was positively associated with increased all-cause and communicable disease mortality in both sexes across all age groups, after adjusting for national wealth. Gini was only positively associated with increased injury mortality amongst infants and 20-24 year olds, and increased non-communicable disease mortality amongst 20-24 year old females. The strength of these associations tended to increase during adolescence. Increasing GDP was negatively associated with all-cause, communicable and non-communicable disease mortality in males and females across all age groups. GDP was also associated with decreased injury mortality in all age groups except 15-19 year old females, and 15-24 year old males. GDP became a weaker predictor of mortality during adolescence.ConclusionPolicies to reduce income inequality, rather than prioritising economic growth at all costs, may be needed to improve adolescent mortality in low and middle-income countries, a key development priority.
| S-EPMC5425964 | biostudies-literature
Project description:Poverty is an important social determinant of health that is associated with increased risk of death1-5. Cash transfer programmes provide non-contributory monetary transfers to individuals or households, with or without behavioural conditions such as children's school attendance6,7. Over recent decades, cash transfer programmes have emerged as central components of poverty reduction strategies of many governments in low- and middle-income countries6,7. The effects of these programmes on adult and child mortality rates remains an important gap in the literature, however, with existing evidence limited to a few specific conditional cash transfer programmes, primarily in Latin America8-14. Here we evaluated the effects of large-scale, government-led cash transfer programmes on all-cause adult and child mortality using individual-level longitudinal mortality datasets from many low- and middle-income countries. We found that cash transfer programmes were associated with significant reductions in mortality among children under five years of age and women. Secondary heterogeneity analyses suggested similar effects for conditional and unconditional programmes, and larger effects for programmes that covered a larger share of the population and provided larger transfer amounts, and in countries with lower health expenditures, lower baseline life expectancy, and higher perceived regulatory quality. Our findings support the use of anti-poverty programmes such as cash transfers, which many countries have introduced or expanded during the COVID-19 pandemic, to improve population health.
| S-EPMC10231298 | biostudies-literature
Project description:Childhood malnutrition is associated with high morbidity and mortality globally1. Undernourished children are more likely to experience cognitive, physical, and metabolic developmental impairments that can lead to later cardiovascular disease, reduced intellectual ability and school attainment, and reduced economic productivity in adulthood2. Child growth failure (CGF), expressed as stunting, wasting, and underweight in children under five years of age (0-59 months), is a specific subset of undernutrition characterized by insufficient height or weight against age-specific growth reference standards3-5. The prevalence of stunting, wasting, or underweight in children under five is the proportion of children with a height-for-age, weight-for-height, or weight-for-age z-score, respectively, that is more than two standard deviations below the World Health Organization's median growth reference standards for a healthy population6. Subnational estimates of CGF report substantial heterogeneity within countries, but are available primarily at the first administrative level (for example, states or provinces)7; the uneven geographical distribution of CGF has motivated further calls for assessments that can match the local scale of many public health programmes8. Building from our previous work mapping CGF in Africa9, here we provide the first, to our knowledge, mapped high-spatial-resolution estimates of CGF indicators from 2000 to 2017 across 105 low- and middle-income countries (LMICs), where 99% of affected children live1, aggregated to policy-relevant first and second (for example, districts or counties) administrative-level units and national levels. Despite remarkable declines over the study period, many LMICs remain far from the ambitious World Health Organization Global Nutrition Targets to reduce stunting by 40% and wasting to less than 5% by 2025. Large disparities in prevalence and progress exist across and within countries; our maps identify high-prevalence areas even within nations otherwise succeeding in reducing overall CGF prevalence. By highlighting where the highest-need populations reside, these geospatial estimates can support policy-makers in planning interventions that are adapted locally and in efficiently directing resources towards reducing CGF and its health implications.
| S-EPMC7015855 | biostudies-literature
Project description:OBJECTIVE:Using estimates from the Global Burden of Disease (GBD) study, we examined differences in unintentional occupational injury mortality rates from 1990 to 2016 between high-income countries (HICs) and low- and middle-income countries (LMICs). METHODS:Unintentional occupational injury mortality rates were obtained through the GBD online visualization tool. We quantified mortality changes over time for common external causes of injury for ages 15?49 years and 50?69 years separately in HICs and LMICs using negative binomial regression models. RESULTS:In 2016, there were 24,396 and 303,999 unintentional occupational injury deaths among individuals aged 15 to 69 years in HICs and LMICs, respectively, corresponding to 3.1 and 7.0 per 100,000 people. Between 1990 and 2016, unintentional occupational injury mortality for people aged 15?69 years dropped 46% (from 5.7 to 3.1 per 100,000 people) in HICs and 42% in LMICs (from 13.2 to 7.0 per 100,000 people). Sustained and large disparities existed between HICs and LMICs for both sexes and both age groups during 1990?2016 (mortality rate ratio: 2.2?2.4). All unintentional occupational injury causes of death displayed significant reduction with one exception (ages 15?49 years in HICs). Country-specific analysis revealed large variations in unintentional occupational injury mortality and changes in occupational injury mortality between 1990 and 2016. CONCLUSIONS:Despite substantial decreases in mortality between 1990 and 2016 for both HICs and LMICs, a large disparity continues to exist between HICs and LMICs. Multifaceted efforts are needed to reduce the disparity.
| S-EPMC6210857 | biostudies-other
Project description:ImportanceThe burden of cancer falls disproportionally on low-middle-income countries (LMICs). It is not well known how novel therapies are tested in current clinical trials and the extent to which they match global disease burden.ObjectivesTo describe the design, results, and publication of oncology randomized clinical trials (RCTs) and examine the extent to which trials match global disease burden and how trial methods and results differ across economic settings.Design, setting, and participantsIn this retrospective cohort study, a literature search identified all phase 3 RCTs evaluating anticancer therapies published from 2014 to 2017. Randomized clinical trials were classified based on World Bank economic classification. Descriptive statistics were used to compare RCT design and results from high-income countries (HICs) and low/middle-income countries (LMICs). Statistical analysis was conducted in January 2020.Main outcomes and measuresDifferences in the design, results, and output of RCTs between HICs and LMICs.ResultsThe study cohort included 694 RCTs: 636 (92%) led by HICs and 58 (8%) led by LMICs. A total of 601 RCTs (87%) tested systemic therapy and 88 RCTs (13%) tested radiotherapy or surgery. The proportion of RCTs relative to global deaths was higher for breast cancer (121 RCTs [17%] and 7% of deaths) but lower for gastroesophageal cancer (38 RCTs [6%] and 14% of deaths), liver cancer (14 RCTs [2%] and 8% of deaths), pancreas cancer (14 RCTs [2%] and 5% of deaths), and cervical cancer (9 RCTs [1%] and 3% of deaths). Randomized clinical trials in HICs were more likely than those in LMICs to be funded by industry (464 [73%] vs 24 [41%]; P < .001). Studies in LMICs were smaller than those in HICs (median, 219 [interquartile range, 137-363] vs 474 [interquartile range, 262-743] participants; P < .001) and more likely to meet their primary end points (39 of 58 [67%] vs 286 of 636 [45%]; P = .001). The observed median effect size among superiority trials was larger in LMICs compared with HICs (hazard ratio, 0.62 [interquartile range, 0.54-0.76] vs 0.84 [interquartile range, 0.67-0.97]; P < .001). Studies from LMICs were published in journals with lower median impact factors than studies from HICs (7 [interquartile range, 4-21] vs 21 [interquartile range, 7-34]; P < .001). Publication bias persisted when adjusted for whether a trial was positive or negative (median impact factor: LMIC negative trial, 5 [interquartile range, 4-6] vs HIC negative trial, 18 [interquartile range, 6-26]; LMIC positive trial, 9 [interquartile range, 5-25] vs HIC positive trial, 25 [interquartile range, 10-48]; P < .001).Conclusions and relevanceThis study suggests that oncology RCTs are conducted predominantly by HICs and do not match the global burden of cancer. Randomized clinical trials from LMICs are more likely to identify effective therapies and have a larger effect size than RCTs from HICs. This study suggests that there is a funding and publication bias against RCTs led by LMICs. Policy makers, research funders, and journals need to address this issue with a range of measures including building capacity and capability in RCTs.
| S-EPMC7844695 | biostudies-literature
Project description:BackgroundNon-biological childhood mortality sex ratios may reflect community sex preferences and gender discrimination in health care.ObjectiveWe assessed the association between contextual factors and gender bias in under-five mortality rates (U5MR) in low- and middle-income countries.MethodsFull birth histories available from Demographic and Health Surveys and Multiple Indicator Cluster Surveys (2010-2018) in 80 countries were used to estimate U5MR male-to-female sex ratios. Expected sex ratios and their residuals (difference of observed and expected) were derived from a linear regression model, adjusted for overall mortality. Negative residuals indicate more likelihood of discrimination against girls, and we refer to this as a measure of potential gender bias. Associations between residuals and national development and gender inequality indices and with survey-derived child health care indicators were tested using Spearman's correlation.ResultsMortality residuals for under-five mortality were not associated with national development, education, religion, or gender inequality indices. Negative residuals were more common in countries where boys were more likely to be taken to health services than girls (rho -0.24, 95% confidence interval -0.45, -0.01).ConclusionsCountries where girls were more likely to die than boys, accounting for overall mortality levels, were also countries where boys were more likely to receive health care than girls. Further research is needed to understand which national characteristics explain the presence of gender bias, given that the analyses of development levels and gender equality did not discriminate between countries with or without excess mortality of girls. Reporting on child mortality separately by sex is required to enable such advances.
| S-EPMC8453971 | biostudies-literature
Project description:Sustainable Development Goal 2.2-to end malnutrition by 2030-includes the elimination of child wasting, defined as a weight-for-length z-score that is more than two standard deviations below the median of the World Health Organization standards for child growth1. Prevailing methods to measure wasting rely on cross-sectional surveys that cannot measure onset, recovery and persistence-key features that inform preventive interventions and estimates of disease burden. Here we analyse 21 longitudinal cohorts and show that wasting is a highly dynamic process of onset and recovery, with incidence peaking between birth and 3 months. Many more children experience an episode of wasting at some point during their first 24 months than prevalent cases at a single point in time suggest. For example, at the age of 24 months, 5.6% of children were wasted, but by the same age (24 months), 29.2% of children had experienced at least one wasting episode and 10.0% had experienced two or more episodes. Children who were wasted before the age of 6 months had a faster recovery and shorter episodes than did children who were wasted at older ages; however, early wasting increased the risk of later growth faltering, including concurrent wasting and stunting (low length-for-age z-score), and thus increased the risk of mortality. In diverse populations with high seasonal rainfall, the population average weight-for-length z-score varied substantially (more than 0.5 z in some cohorts), with the lowest mean z-scores occurring during the rainiest months; this indicates that seasonally targeted interventions could be considered. Our results show the importance of establishing interventions to prevent wasting from birth to the age of 6 months, probably through improved maternal nutrition, to complement current programmes that focus on children aged 6-59 months.
| S-EPMC10511327 | biostudies-literature
Project description:The 21st Century has been marked by increased volatility in food prices, with global price spikes in 2007-08, 2010-11, and again in 2021-22. The impact of food inflation on the risk of child undernutrition is not well understood, however. This study explores the potential impacts of food inflation on wasting and stunting among 1.27 million pre-school children from 44 developing countries. On average, a 5 percent increase in the real price of food increases the risk of wasting by 9 percent and severe wasting by 14 percent. These risks apply to young infants, suggesting a prenatal pathway, as well as to older children who typically experience a deterioration in diet quality in the wake of food inflation. Male children and children from poor and rural landless households are more severely impacted. Food inflation during pregnancy and the first year after birth also increases the risk of stunting for children 2-5 years of age. This evidence provides a strong rationale for interventions to prevent food inflation and mitigate its impacts on vulnerable children and their mothers.
| S-EPMC10505228 | biostudies-literature