Causes and age of neonatal death and associations with maternal and newborn care characteristics in Nepal: a verbal autopsy study

Background In Nepal, neonatal mortality fell substantially between 2000 and 2018, decreasing 50% from 40 to 20 deaths per 1,000 live births. Nepal’s success has been attributed to a decreasing total fertility rate, improvements in female education, increases in coverage of skilled care at birth, and community-based child survival interventions. Methods A verbal autopsy study, led by the Integrated Rural Health Development Training Centre (IRHDTC), conducted interviews for 338 neonatal deaths across six districts in Nepal between April 2012 and April 2013. We conducted a secondary analysis of verbal autopsy data to understand how cause and age of neonatal death are related to health behaviors, care seeking practices, and coverage of essential services in Nepal. Results Sepsis was the leading cause of neonatal death (n=159/338, 47.0%), followed by birth asphyxia (n=56/338, 16.6%), preterm birth (n=45/338, 13.3%), and low birth weight (n=17/338, 5.0%). Neonatal deaths occurred primarily on the first day of life (27.2%) and between days 1 and 6 (64.8%) of life. Risk of death due birth asphyxia relative to sepsis was higher among mothers who were nulligravida, had <4 antenatal care visits, and had a multiple birth; risk of death due to prematurity relative to sepsis was lower for women who made ≥1 delivery preparation and higher for women with a multiple birth. Conclusions Our findings suggest cause and age of death distributions typically associated with high mortality settings. Increased coverage of preventive antenatal care interventions and counseling are critically needed. Delays in care seeking for newborn illness and quality of care around the time of delivery and for sick newborns are important points of intervention with potential to reduce deaths, particularly for birth asphyxia and sepsis, which remain common in this population.

Tree-informed Bayesian multi-source domain adaptation: cross-population probabilistic cause-of-death assignment using verbal autopsy

Determining causes of deaths (COD) occurred outside of civil registration and vital statistics systems is challenging. A technique called verbal autopsy (VA) is widely adopted to gather information on deaths in practice. A VA consists of interviewing relatives of a deceased person about symptoms of the deceased in the period leading to the death, often resulting in multivariate binary responses. While statistical methods have been devised for estimating the cause-specific mortality fractions (CSMFs) for a study population, continued expansion of VA to new populations (or "domains") necessitates approaches that recognize between-domain differences while capitalizing on potential similarities. In this paper, we propose such a domain-adaptive method that integrates external between-domain similarity information encoded by a pre-specified rooted weighted tree. Given a cause, we use latent class models to characterize the conditional distributions of the responses that may vary by domain. We specify a logistic stick-breaking Gaussian diffusion process prior along the tree for class mixing weights with node-specific spike-and-slab priors to pool information between the domains in a data-driven way. Posterior inference is conducted via a scalable variational Bayes algorithm. Simulation studies show that the domain adaptation enabled by the proposed method improves CSMF estimation and individual COD assignment. We also illustrate and evaluate the method using a validation data set. The paper concludes with a discussion on limitations and future directions.

Automatic Verbal Autopsy Classification Using Multinomial Logistic Regression Classifier by Using Recursive Feature Elimination

Verbal autopsy is one of the finest medical process to identify automatically the cause of a death afore medical ascendant entities will certify it. Identifying the exact cause is intricate and fuzzy in nature. The dataset with an exact cause of death is a paramount implement for every country to make the presage about the life style and medical facilities available to the people. Multinomial logistic regression was utilized in our study to relegate the exact cause of death. We used standard datasets like PHMRC and Matlab which were potentially accepted in medical field. The reason to utilize the Multinomial logistic Regression is that most of the dataset is consisting of 0 and 1 values which betoken the presence and absence of value in the attribute. We used three standard metrics like the sensitivity, Chance Corrected Concordance (CCC) and Cause-specific mortality fraction (CSMF) for a comparison of our model with precedent models like Insilico VA, Tariff and InterVA-4. Computed results show that proposed model is better than the precedent models.

Improving methods to measure comparable mortality cause (IMMCMC) gold standard verbal autopsy dataset

Objectives Gold standard cause of death data is critically important to improve verbal autopsy (VA) methods in diagnosing cause of death where civil and vital registration systems are inadequate or poor. As part of a three-country research study—Improving Methods to Measure Comparable Mortality by Cause (IMMCMC) study—data were collected on clinicopathological criteria-based gold standard cause of death from hospital record reviews with matched VAs. The purpose of this data note is to make accessible a de-identified format of these gold standard VAs for interested researchers to improve the diagnostic accuracy of VA methods. Data description The study was conducted between 2011 and 2014 in the Philippines, Bangladesh, and Papua New Guinea. Gold standard diagnoses of underlying causes of death for deaths occurring in hospital were matched to VAs conducted using a standardized VA questionnaire developed by the Population Health Metrics Consortium. 3512 deaths were collected in total, comprised of 2491 adults (12 years and older), 320 children (28 days to 12 years), and 702 neonates (0–27 days).

Integrating community-based verbal autopsy into civil registration and vital statistics: lessons learnt from five countries

This paper describes the lessons from scaling up a verbal autopsy (VA) intervention to improve data about causes of death according to a nine-domain framework: governance, design, operations, human resources, financing, infrastructure, logistics, information technologies and data quality assurance. We use experiences from China, Myanmar, Papua New Guinea, Philippines and Solomon Islands to explore how VA has been successfully implemented in different contexts, to guide other countries in their VA implementation. The governance structure for VA implementation comprised a multidisciplinary team of technical experts, implementers and staff at different levels within ministries. A staged approach to VA implementation involved scoping and mapping of death registration processes, followed by pretest and pilot phases which allowed for redesign before a phased scale-up. Existing health workforce in countries were trained to conduct the VA interviews as part of their routine role. Costs included training and compensation for the VA interviewers, information technology (IT) infrastructure costs, advocacy and dissemination, which were borne by the funding agency in early stages of implementation. The complexity of the necessary infrastructure, logistics and IT support required for VA increased with scale-up. Quality assurance was built into the different phases of the implementation. VA as a source of cause of death data for community deaths will be needed for some time. With the right technical and political support, countries can scale up this intervention to ensure ongoing collection of quality and timely information on community deaths for use in health planning and better monitoring of national and global health goals.

Trends and causes of adult mortality from 2007 to 2017 using verbal autopsy method, Addis Ababa, Ethiopia

ObjectivesWe aim to analyse the trends and causes of mortality among adults in Addis Ababa.SettingThis analysis was conducted using verbal autopsy data from the Addis Ababa Mortality Surveillance in Addis Ababa, Ethiopia.ParticipantsAll deceased adults aged 15 years and above between 2007–2012 and 2015–2017 were included in the analysis.Outcome measuresWe collected verbal autopsy and conducted physician review to ascertain cause of death.ResultA total of 7911 data were included in this analysis. Non-communicable disease (NCD) accounted for 62.8% of adult mortality. Mortality from communicable diseases, maternal conditions and nutritional deficiencies followed this by accounting for 30.3% of total mortality. Injury accounted for 6.8% of total mortality. We have observed a significant decline in mortality attributed to group one cause of death (43.25% in 2007 to 12.34% in 2017, p<0.001). However, we observed a significant increase in mortality attributed to group II cause of death (from 49.95% in 2007 to 81.17% in 2017, p<0.001). The top five leading cause of death in 2017 were cerebrovascular disease (12.8%), diabetes mellitus (8.1%), chronic liver disease (6.3%), hypertension (5.7%), ischaemic heart disease (5.7%) and other specified neoplasm (5.2%).ConclusionWe documented an epidemiological shift in cause of mortality from communicable diseases to NCD over 10 years. There is a great progress in reducing mortality due to communicable diseases over the past years. However, the burden of NCDs call for actions for improving access to quality health service, improved case detection and community education to increase awareness. Integrating NCD intervention in to a well-established and successful programme targeting communicable diseases in the country might be beneficial for improving provision of comprehensive healthcare.

Comparison of machine learning algorithms applied to symptoms to determine infectious causes of death in children: national survey of 18,000 verbal autopsies in the Million Death Study in India

Background Machine learning (ML) algorithms have been successfully employed for prediction of outcomes in clinical research. In this study, we have explored the application of ML-based algorithms to predict cause of death (CoD) from verbal autopsy records available through the Million Death Study (MDS). Methods From MDS, 18826 unique childhood deaths at ages 1–59 months during the time period 2004–13 were selected for generating the prediction models of which over 70% of deaths were caused by six infectious diseases (pneumonia, diarrhoeal diseases, malaria, fever of unknown origin, meningitis/encephalitis, and measles). Six popular ML-based algorithms such as support vector machine, gradient boosting modeling, C5.0, artificial neural network, k-nearest neighbor, classification and regression tree were used for building the CoD prediction models. Results SVM algorithm was the best performer with a prediction accuracy of over 0.8. The highest accuracy was found for diarrhoeal diseases (accuracy = 0.97) and the lowest was for meningitis/encephalitis (accuracy = 0.80). The top signs/symptoms for classification of these CoDs were also extracted for each of the diseases. A combination of signs/symptoms presented by the deceased individual can effectively lead to the CoD diagnosis. Conclusions Overall, this study affirms that verbal autopsy tools are efficient in CoD diagnosis and that automated classification parameters captured through ML could be added to verbal autopsies to improve classification of causes of death.