Dataset Information

A novel screening tool to predict severe acute malnutrition through automated monitoring of weight-for-age growth curves.

ABSTRACT: Weight-for-age (WFA) growth faltering often precedes severe acute malnutrition (SAM) in children, yet it is often missed during routine growth monitoring. Automated interpretation of WFA growth within electronic health records could expedite the identification of children at risk of SAM. This study aimed to develop an automated screening tool to predict SAM risk from WFA growth, and to determine its predictive ability compared with simple changes in weight or WFA z-score. To develop the screening tool, South African child growth experts (n = 30) rated SAM risk on 100 WFA growth curves, which were then used to train an artificial neural network (ANN) to assess SAM risk from consecutive WFA z-scores. The ANN was validated in 185 children under five (63 SAM cases; 122 controls) using diagnostic accuracy methodology. The ANN's performance was compared with that of changes in weight or WFA z-score. Even though experts' SAM risk ratings of the WFA growth curves differed considerably, the ANN achieved a sensitivity of 73.0% (95% confidence interval [CI]: 60.3; 83.4), specificity of 86.1% (95% CI: 78.6; 91.7) and receiver-operating characteristic curve area of 0.795 (95% CI: 0.732; 0.859) during validation with real cases, outperforming changes in weight or WFA z-scores. The ANN, as an automated screening tool, could markedly improve the identification of children at risk of SAM using routinely collected WFA growth information.

SUBMITTER: Nel S

PROVIDER: S-EPMC9218329 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundFor ages 5-19 years, the World Health Organization (WHO) publishes reference charts based on 'core data' from the US National Center for Health Statistics (NCHS), collected from 1963-75 on 22,917 US children. To promote the use of body mass index in older children, weight-for-age was omitted after age 10. Health providers have subsequently expressed concerns about this omission and the selection of centiles. We therefore sought to extend weight-for-age reference curves from 10 to 19 years by applying WHO exclusion criteria and curve fitting methods to the core NCHS data and to revise the choice of displayed centiles.MethodsWHO analysts first excluded ~ 3% of their reference population in order to achieve a "non-obese sample with equal height". Based on these exclusion criteria, 314 girls and 304 boys were first omitted for 'unhealthy' weights-for-height. By applying WHO global deviance and information criteria, optimal Box-Cox power exponential models were used to fit smoothed weight-for-age centiles. Bootstrap resampling was used to assess the precision of centile estimates. For all charts, additional centiles were included in the healthy range (3 to 97%), and the more extreme WHO centiles 0.1 and 99.9% were dropped.ResultsIn addition to weight-for-age beyond 10 years, our charts provide more granularity in the centiles in the healthy range -2 to +2 SD (3-97%). For both weight and BMI, the bootstrap confidence intervals for the 99.9th centile were at least an order of magnitude wider than the corresponding 50th centile values.ConclusionsThese charts complement existing WHO charts by allowing weight-for-age to be plotted concurrently with height in older children. All modifications followed strict WHO methodology and utilized the same core data from the US NCHS. The additional centiles permit a more precise assessment of normal growth and earlier detection of aberrant growth as it crosses centiles. Elimination of extreme centiles reduces the risk of misclassification. A complete set of charts is available at the CPEG web site (http://cpeg-gcep.net).

Project description:BackgroundIn northern Nigeria, UNICEF has supported introduction of a short message service (SMS) system for data transmission in the Community-based Management of Acute Malnutrition (CMAM) programme. The SMS system operates in parallel to the traditional paper-based system, and weekly data are transmitted directly from the health facilities to the federal level. For the paper system, monthly data summaries are passed through all levels of government. We assessed the data quality and performance of both CMAM information systems.MethodsWe undertook a contextualised study in one state in north-west Nigeria, with additional analysis of secondary data from five states. Fieldwork methods included: observation of the data system in nine selected facilities in three local government areas; recounting of data for admissions, exits, and ready-to-use therapeutic food (RUTF) utilisation; and interviews with health workers and government officials.ResultsWhile the small number of facilities does not enable robust generalisation of the quantitative findings, the strengths and weaknesses detected pertain to the whole programme, as they relate to how the system was designed and is operated. We found that the accuracy and reliability of CMAM data were deficient to a similar extent in the paper-based and SMS systems. For the audited month, we found large discrepancies between recounted data and paper records in regard to admissions, exits and RUTF cartons consumed in the majority of facilities visited. There was also a large discrepancy in the reported percentage of "deaths or defaulters" (6.5%) compared to 22% based on a recount of outpatient cards. Errors are mainly introduced during data collection and when completing tallies at facility level.ConclusionOur findings indicate the need for improvements in the design of the monitoring system, training and supervision of data management, and communication of results; as well as clear evidence on how measures to improve data quality may affect performance of individual CMAM clinics. The CMAM default and death rates currently reported in Nigeria are likely to be under-estimates, and therefore provide a misleadingly good impression of CMAM programme performance.

Project description:BACKGROUND:Virtual or in silico ligand screening combined with other computational methods is one of the most promising methods to search for new lead compounds, thereby greatly assisting the drug discovery process. Despite considerable progresses made in virtual screening methodologies, available computer programs do not easily address problems such as: structural optimization of compounds in a screening library, receptor flexibility/induced-fit, and accurate prediction of protein-ligand interactions. It has been shown that structural optimization of chemical compounds and that post-docking optimization in multi-step structure-based virtual screening approaches help to further improve the overall efficiency of the methods. To address some of these points, we developed the program AMMOS for refining both, the 3D structures of the small molecules present in chemical libraries and the predicted receptor-ligand complexes through allowing partial to full atom flexibility through molecular mechanics optimization. RESULTS:The program AMMOS carries out an automatic procedure that allows for the structural refinement of compound collections and energy minimization of protein-ligand complexes using the open source program AMMP. The performance of our package was evaluated by comparing the structures of small chemical entities minimized by AMMOS with those minimized with the Tripos and MMFF94s force fields. Next, AMMOS was used for full flexible minimization of protein-ligands complexes obtained from a mutli-step virtual screening. Enrichment studies of the selected pre-docked complexes containing 60% of the initially added inhibitors were carried out with or without final AMMOS minimization on two protein targets having different binding pocket properties. AMMOS was able to improve the enrichment after the pre-docking stage with 40 to 60% of the initially added active compounds found in the top 3% to 5% of the entire compound collection. CONCLUSION:The open source AMMOS program can be helpful in a broad range of in silico drug design studies such as optimization of small molecules or energy minimization of pre-docked protein-ligand complexes. Our enrichment study suggests that AMMOS, designed to minimize a large number of ligands pre-docked in a protein target, can successfully be applied in a final post-processing step and that it can take into account some receptor flexibility within the binding site area.

Dataset Information

A novel screening tool to predict severe acute malnutrition through automated monitoring of weight-for-age growth curves.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets