Assessing the performance of the family folder system for collecting community-based health information in Tigray Region, North Ethiopia: a capture-recapture study.

OBJECTIVES: To assess completeness and accuracy of the family folder in terms of capturing community-level health data. STUDY DESIGN: A capture-recapture method was applied in six randomly selected districts of Tigray Region, Ethiopia. PARTICIPANTS: Child health data, abstracted from randomly selected 24 073 family folders from 99 health posts, were compared with similar data recaptured through household survey and routine health information made by these health posts. PRIMARY AND SECONDARY OUTCOME MEASURES: Completeness and accuracy of the family folder data; and coverage selected child health indicators, respectively. RESULTS: Demographic data captured by the family folders and household survey were highly concordant, concordance correlation for total population, women 15-49 years age and under 5-year child were 0.97 (95% CI 0.94 to 0.99, p<0.001), 0.73 (95% CI 0.67 to 0.88) and 0.91 (95% CI 0.85 to 0.96), respectively. However, the live births, child health service indicators and child health events were more erratically reported in the three data sources. The concordance correlation among the three sources, for live births and neonatal deaths was 0.094 (95% CI -0.232 to 0.420) and 0.092 (95% CI -0.230 to 0.423) respectively, and for the other parameters were close to 0. CONCLUSION: The family folder system comprises a promising development. However, operational issues concerning the seamless capture and recording of events and merging community and facility data at the health centre level need improvement.

Mortality trends and access to care for cardiovascular diseases in Agincourt, rural South Africa: a mixed-methods analysis of verbal autopsy data.

OBJECTIVES: Cardiovascular diseases are the second leading cause of mortality behind HIV/AIDS in South Africa. This study investigates cardiovascular disease mortality trends in rural South Africa over 20+ years and the associated barriers to accessing care, using verbal autopsy data. DESIGN: A mixed-methods approach was used, combining descriptive analysis of mortality rates over time, by condition, sex and age group, quantitative analysis of circumstances of mortality (CoM) indicators and free text narratives of the final illness, and qualitative analysis of free texts. SETTING: This study was done using verbal autopsy data from the Health and Socio-Demographic Surveillance System site in Agincourt, rural South Africa. PARTICIPANTS: Deaths attributable to cardiovascular diseases (acute cardiac disease, stroke, renal failure and other unspecified cardiac disease) from 1993 to 2015 were extracted from verbal autopsy data. RESULTS: Between 1993 and 2015, of 15 305 registered deaths over 1 851 449 person-years of follow-up, 1434 (9.4%) were attributable to cardiovascular disease, corresponding to a crude mortality rate of 0.77 per 1000 person-years. Cardiovascular disease mortality rate increased from 0.34 to 1.12 between 1993 and 2015. Stroke was the dominant cause of death, responsible for 41.0% (588/1434) of all cardiovascular deaths across all years. Cardiovascular disease mortality rate was significantly higher in women and increased with age. The main delays in access to care during the final illness were in seeking and receiving care. Qualitative free-text analysis highlighted delays not captured in the CoM, principally communication between the clinician and patient or family. Half of cases initially sought care outside a hospital setting (50.9%, 199/391). CONCLUSIONS: The temporal increase in deaths due to cardiovascular disease highlights the need for greater prevention and management strategies for these conditions, particularly for the women. Strategies to improve seeking and receiving care during the final illness are needed.

Collective reflections on the first cycle of a collaborative learning platform to strengthen rural primary healthcare in Mpumalanga, South Africa.

BACKGROUND: Frontline managers and health service providers are constrained in many contexts from responding to community priorities due to organizational cultures focused on centrally defined outputs and targets. This paper presents an evaluation of the Verbal Autopsy with Participatory Action Research (VAPAR) programme-a collaborative learning platform embedded in the local health system in Mpumalanga, South Africa-for strengthening of rural primary healthcare (PHC) systems. The programme aims to address exclusion from access to health services by generating and acting on research evidence of practical, local relevance. METHODS: Drawing on existing links in the provincial and national health systems and applying rapid, participatory evaluation techniques, we evaluated the first action-learning cycle of the VAPAR programme (2017-19). We collected data in three phases: (1) 10 individual interviews with programme stakeholders, including from government departments and parastatals, nongovernmental organizations and local communities; (2) an evaluative/exploratory workshop with provincial and district Department of Health managers; and (3) feedback and discussion of findings during an interactive workshop with national child health experts. RESULTS: Individual programme stakeholders described early outcomes relating to effective research and stakeholder engagement, and organization and delivery of services, with potential further contributions to the establishment of an evidence base for local policy and planning, and improved health outcomes. These outcomes were verified with provincial managers. Provincial and national stakeholders identified the potential for VAPAR to support engagement between communities and health authorities for collective planning and implementation of services. Provincial stakeholders proposed that this could be achieved through a two-way integration, with VAPAR stakeholders participating in routine health planning and review activities and frontline health officials being involved in the VAPAR process. Findings were collated into a revised theory of change. CONCLUSIONS: The VAPAR learning platform was regarded as a feasible, acceptable and relevant approach to facilitate cooperative learning and community participation in health systems. The evaluation provides support for a collaborative learning platform within routine health system processes and contributes to the limited evaluative evidence base on embedded health systems research.

Understanding non-communicable diseases: combining health surveillance with local knowledge to improve rural primary health care in South Africa.

Background: NCDs are non-infectious, long-term conditions that account for 40 million deaths per annum. 87% of premature NCD mortality occurs in low- and middle-income countries. Objective: The aims were:develop methods to provide integrated biosocial accounts of NCD mortality; and explore the practical utility of extended mortality data for the primary health care system. Methods: We drew on data from research programmes in the study area. Data were analysed in three steps: [a]analysis of levels, causes and circumstances of NCD mortality [n = 4,166] from routine census updates including Verbal Autopsy and of qualitative data on lived experiences of NCDs in rural villages from participatory research; [b] identifying areas of convergence and divergence between the analyses; and [c]exploration of the practical relevance of the data drawing on engagements with health systems stakeholders. Results: NCDs constituted a significant proportion of mortality in this setting [36%]. VA data revealed multiple barriers to access in end-of-life care. Many deaths were attributed to problems with resources and health systems [21%;19% respectively]. The qualitative research provided rich complementary detail on the processes through which risk originates, accumulates and is expressed in access to end-of-life care, related to chronic poverty and perceptions of poor quality care in clinics. The exploration of practical relevance revealed chronic under-funding for NCD services, and an acute need for robust, timely data on the NCD burden. Conclusions: VA data allowed a significant burden of NCD mortality to be quantified and revealed barriers to access at and around the time of death. Qualitative research contextualised these barriers, providing explanations of how and why they exist and persist. Health systems analysis revealed shortages of resources allocated to NCDs and a need for robust research to provide locally relevant evidence to organise and deliver care. Pragmatic interdisciplinary and mixed method analysis provides relevant renditions of complex problems to inform more effective responses.

The 2020 report of The Lancet Countdown on health and climate change: responding to converging crises.

TRANSLATIONS: For the Chinese, French, German, and Spanish translations of the abstract see Supplementary Materials section.

Eco-epidemiological assessment of the COVID-19 epidemic in China, January-February 2020.

Background: The outbreak of COVID-19 in China in early 2020 provides a rich data source for exploring the ecological determinants of this new infection, which may be of relevance as the pandemic develops.Objectives: Assessing the spread of the COVID-19 across China, in relation to associations between cases and ecological factors including population density, temperature, solar radiation and precipitation.Methods: Open-access COVID-19 case data include 18,069 geo-located cases in China during January and February 2020, which were mapped onto a 0.25° latitude/longitude grid together with population and weather data (temperature, solar radiation and precipitation). Of 15,539 grid cells, 559 (3.6%) contained at least one case, and these were used to construct a Poisson regression model of cell-weeks. Weather parameters were taken for the preceding week given the established 5-7 day incubation period for COVID-19. The dependent variable in the Poisson model was incident cases per cell-week and exposure was cell population, allowing for clustering of cells over weeks, to give incidence rate ratios.Results: The overall COVID-19 incidence rate in cells with confirmed cases was 0.12 per 1,000. There was a single confirmed case in 113/559 (20.2%) of cells, while two grid cells recorded over 1,000 confirmed cases. Weekly means of maximum daily temperature varied from -28.0°C to 30.1°C, minimum daily temperature from -42.4°C to 23.0°C, maximum solar radiation from 0.04 to 2.74 MJm-2 and total precipitation from 0 to 72.6 mm. Adjusted incidence rate ratios suggested brighter, warmer and drier conditions were associated with lower incidence.Conclusion: Though not demonstrating cause and effect, there were appreciable associations between weather and COVID-19 incidence during the epidemic in China. This does not mean the pandemic will go away with summer weather but demonstrates the importance of using weather conditions in understanding and forecasting the spread of COVID-19.

Time-critical conditions: assessment of burden and access to care using verbal autopsy in Agincourt, South Africa.

Background: Time-critical conditions (TCC) are estimated to cause substantial mortality in low and middle-income countries. However, quantification of deaths and identification of contributing factors to those deaths are challenging in settings with poor health records. Aim: To use verbal autopsy (VA) data from the Agincourt health and sociodemographic surveillance system in rural South Africa to quantify the burden of deaths from TCC and to evaluate the barriers in seeking, reaching and receiving quality care for TCC leading to death. Methodology: Deaths from 1993 to 2015 were analysed to identify causality from TCC. Deaths due to TCC were categorised as communicable, non-communicable, maternal, neonatal or injury-related. Proportion of deaths from TCC by age, sex, condition type and temporal trends was described. Deaths due to TCC from 2012 to 2015 were further examined by circumstances of mortality (CoM) indicators embedded in VA. Healthcare access, at illness onset and during the final day of life, as well as place of death, was extracted from free text summaries. Summaries were also analysed qualitatively using a Three Delays framework to identify barriers to healthcare. Results: Of 15 305 deaths, 5885 (38.45%) were due to TCC. Non-communicable diseases were the most prevalent cause of death from TCC (2961/5885 cases, 50.31%). CoM indicators highlighted delays in a quarter of deaths due to TCC, most frequently in seeking care. The most common pattern of healthcare access was to die outwith a facility, having sought no healthcare (409/1324 cases, 30.89%). Issues in receipt of quality care were identified by qualitative analysis. Conclusion: TCCs are responsible for a substantial burden of deaths in this rural South African population. Delays in seeking and receiving quality care were more prominent than those in reaching care, and thus further research and solution development should focus on healthcare-seeking behaviour and quality care provision.

Building cooperative learning to address alcohol and other drug abuse in Mpumalanga, South Africa: a participatory action research process.

Background: Alcohol and other drug (AOD) abuse is a major public health challenge disproportionately affecting marginalised communities. Involving communities in the development of responses can contribute to acceptable solutions.Objectives: To: (1) document forms, processes, and contexts of engaging communities to nominate health concerns and generate new knowledge for action; (2) further build participation in the local health system by reflecting on and adapting the process.Methods: PAR was progressed with 48 community stakeholders across three rural villages in the MRC/Wits Agincourt Health and Socio Demographic Surveillance System (HDSS) in Mpumalanga, South Africa. A series of workshops explored community-nominated topics, systematised lived experience into shared accounts and considered actions to address problems identified. Photovoice was also used to generate visual evidence. Narrative and visual data were thematically analysed, situated within practice frameworks, and learning and adaption elicited.Results: AOD abuse was identified as a topic of high priority. It was understood as an entrenched social problem with destructive effects. Biopsychosocial impacts were mapped and related to unemployment, poverty, stress, peer pressure, criminal activity, corruption, and a proliferating number of taverns. Integrated action agendas were developed focussed on demand, supply, and harm reduction underpinned by shared responsibility among community, state, and non-state actors. Community stakeholders appreciated systematising and sharing knowledge, taking active roles, developing new skills in planning and public speaking, and progressing shared accountability processes. Expectations required sensitive management, however.Conclusion: There is significant willingness and capacity among community stakeholders to work in partnership with authorities to address priority health concerns. As a process, participation can help to raise and frame issues, which may help to better inform action and encourage shared responsibility. Broader understandings of participation require reference to, and ultimately transfer of power towards, those most directly affected, developing community voice as continuous processes within social and political environments.

Global burden of respiratory infections associated with seasonal influenza in children under 5 years in 2018: a systematic review and modelling study.

BACKGROUND: Seasonal influenza virus is a common cause of acute lower respiratory infection (ALRI) in young children. In 2008, we estimated that 20 million influenza-virus-associated ALRI and 1 million influenza-virus-associated severe ALRI occurred in children under 5 years globally. Despite this substantial burden, only a few low-income and middle-income countries have adopted routine influenza vaccination policies for children and, where present, these have achieved only low or unknown levels of vaccine uptake. Moreover, the influenza burden might have changed due to the emergence and circulation of influenza A/H1N1pdm09. We aimed to incorporate new data to update estimates of the global number of cases, hospital admissions, and mortality from influenza-virus-associated respiratory infections in children under 5 years in 2018. METHODS: We estimated the regional and global burden of influenza-associated respiratory infections in children under 5 years from a systematic review of 100 studies published between Jan 1, 1995, and Dec 31, 2018, and a further 57 high-quality unpublished studies. We adapted the Newcastle-Ottawa Scale to assess the risk of bias. We estimated incidence and hospitalisation rates of influenza-virus-associated respiratory infections by severity, case ascertainment, region, and age. We estimated in-hospital deaths from influenza virus ALRI by combining hospital admissions and in-hospital case-fatality ratios of influenza virus ALRI. We estimated the upper bound of influenza virus-associated ALRI deaths based on the number of in-hospital deaths, US paediatric influenza-associated death data, and population-based childhood all-cause pneumonia mortality data in six sites in low-income and lower-middle-income countries. FINDINGS: In 2018, among children under 5 years globally, there were an estimated 109·5 million influenza virus episodes (uncertainty range [UR] 63·1-190·6), 10·1 million influenza-virus-associated ALRI cases (6·8-15·1); 870 000 influenza-virus-associated ALRI hospital admissions (543 000-1 415 000), 15 300 in-hospital deaths (5800-43 800), and up to 34 800 (13 200-97 200) overall influenza-virus-associated ALRI deaths. Influenza virus accounted for 7% of ALRI cases, 5% of ALRI hospital admissions, and 4% of ALRI deaths in children under 5 years. About 23% of the hospital admissions and 36% of the in-hospital deaths were in infants under 6 months. About 82% of the in-hospital deaths occurred in low-income and lower-middle-income countries. INTERPRETATION: A large proportion of the influenza-associated burden occurs among young infants and in low-income and lower middle-income countries. Our findings provide new and important evidence for maternal and paediatric influenza immunisation, and should inform future immunisation policy particularly in low-income and middle-income countries. FUNDING: WHO; Bill & Melinda Gates Foundation.

Enhancing the value of mortality data for health systems: adding Circumstances Of Mortality CATegories (COMCATs) to deaths investigated by verbal autopsy.

Half of the world's deaths and their causes pass unrecorded by routine registration systems, particularly in low- and middle-income countries. Verbal autopsy (VA) collects information on medical signs, symptoms and circumstances from witnesses of a death that is used to assign likely medical causes. To further contextualise information on mortality, understanding underlying determinants, such as logistics, barriers to service utilisation and health systems responses, is important for health planning. Adding systematic methods for categorising circumstantial determinants of death to conventional VA tools is therefore important. In this context, the World Health Organization (WHO) leads the development of international standards for VA, and added questions on the social and health systems circumstances of death in 2012. This paper introduces a pragmatic and scalable approach for assigning relevant Circumstances Of Mortality CATegories (COMCATs) within VA tools, and examines their consistency, reproducibility and plausibility for health policy making, as well as assessing additional effort and cost to the routine VA process. This innovative COMCAT model is integrated with InterVA-5 software (which processes WHO-2016 VA data), for assigning numeric likelihoods to six circumstantial categories for each death. VA data from 4,116 deaths in the Agincourt Health and Socio-Demographic Surveillance System in South Africa from 2012 to 2016 were used to demonstrate proof of principle for COMCATs. Lack of resources to access health care, poor recognition of diseases and inadequate health systems responses ranked highest among COMCATs in the demonstration dataset. COMCATs correlated plausibly with age, sex, causes of death and local knowledge of the demonstration population. The COMCAT approach appears to be plausible, feasible and enhances the functionality of routine VA to account for critical limiting circumstances at and around the time of death. It is a promising tool for evaluating progress towards the Sustainable Development Goals and the roll-out of Universal Health Coverage.

Correlation between noncommunicable disease mortality in people aged 30-69 years and those aged 70-89 years.

OBJECTIVE: To investigate whether the key metric for monitoring progress towards sustainable development goal target 3.4 that is measuring premature noncommunicable disease mortality (deaths among people aged 30-69 years), is ageist. METHODS: To examine the relationship between premature noncommunicable disease mortality and noncommunicable disease mortality in older people, a database of mortality rates for cardiovascular disease, cancer, chronic obstructive pulmonary disease and diabetes in people aged 30 to 69 years and 70 to 89 years was compiled using estimates from the Global Burden of Disease Study 2017. The data covered 195 countries, six time-points and both sexes, giving 2340 instances. The World Health Organization's (WHO's) life-table method for the premature noncommunicable disease mortality metric was applied to the data. FINDINGS: There was a strong correlation between noncommunicable disease mortality patterns in the premature and older age groups, which suggests that measuring premature noncommunicable disease mortality is informative about such mortality in later life. Neither time nor geographical location had a substantial effect on this correlation. However, there were female-to-male differences in age-specific probabilities of death due to noncommunicable disease, implying that noncommunicable disease mortality should be assessed using a sex-disaggregated approach. CONCLUSION: As the established WHO metric for premature noncommunicable disease mortality was predictive of noncommunicable disease mortality in older people, the metric should not be construed as ageist. Focusing resources on measuring premature noncommunicable disease mortality will be appropriate, particularly in settings without universal civil death registration. This approach should not prejudice the provision of health services throughout the life-course.

'Water is life': developing community participation for clean water in rural South Africa.

BACKGROUND: South Africa is a semiarid country where 5 million people, mainly in rural areas, lack access to water. Despite legislative and policy commitments to the right to water, cooperative governance and public participation, many authorities lack the means to engage with and respond to community needs. The objectives were to develop local knowledge on health priorities in a rural province as part of a programme developing community evidence for policy and planning. METHODS: We engaged 24 participants across three villages in the Agincourt Health and Socio-Demographic Surveillance System and codesigned the study. This paper reports on lack of clean, safe water, which was nominated in one village (n=8 participants) and in which women of reproductive age were nominated as a group whose voices are excluded from attention to the issue. On this basis, additional participants were recruited (n=8). We then held a series of consensus-building workshops to develop accounts of the problem and actions to address it using Photovoice to document lived realities. Thematic analysis of narrative and visual data was performed. RESULTS: Repeated and prolonged periods when piped water is unavailable were reported, as was unreliable infrastructure, inadequate service delivery, empty reservoirs and poor supply exacerbated by droughts. Interconnected social, behavioural and health impacts were documented combined with lack of understanding, cooperation and trust between communities and authorities. There was unanimity among participants for taps in houses as an overarching goal and strategies to build an evidence base for planning and advocacy were developed. CONCLUSION: In this setting, there is willingness among community stakeholders to improve water security and there are existing community assemblies to support this. Health and Socio-Demographic Surveillance Systems provide important opportunities to routinely connect communities to resource management and service delivery. Developing learning platforms with government and non-government organisations may offer a means to enable more effective public participation in decentralised water governance.

Social determinants of under-5 child health: A qualitative study in Wolkayit Woreda, Tigray Region, Ethiopia.

Despite the significant reductions seen in under-5 child mortality in Ethiopia over the last two decades, more than 10,000 children still die each year in Tigray Region alone, of whom 75% die from preventable diseases. Using an equity lens, this study aimed to investigate the social determinants of child health in one particularly vulnerable district as a means of informing the health policy decision-making process. An exploratory qualitative study design was adopted, combining focus group discussions and qualitative interviews. Seven Focus Group Discussions with mothers of young children, and 21 qualitative interviews with health workers were conducted in Wolkayit district in May-June 2015. Data were subjected to thematic analysis. Mothers' knowledge regarding the major causes of child mortality appeared to be good, and they also knew about and trusted the available child health interventions. However, utilization and practice of these interventions was limited by a range of issues, including cultural factors, financial shortages, limited female autonomy on financial resources, seasonal mobility, and inaccessible or unaffordable health services. Our findings pointed to the importance of a multi-sectoral strategy to improve child health equity and reduce under-5 mortality in Wolkayit. Recommendations include further decentralizing child health services to local-level Health Posts, and increasing the number of Health Facilities based on local topography and living conditions.

Rethinking collaboration: developing a learning platform to address under-five mortality in Mpumalanga province, South Africa.

Following 50 years of apartheid, South Africa introduced visionary health policy committing to the right to health as part of a primary health care (PHC) approach. Implementation is seriously challenged, however, in an often-dysfunctional health system with scarce resources and a complex burden of avoidable mortality persists. Our aim was to develop a process generating evidence of practical relevance on implementation processes among people excluded from access to health systems. Informed by health policy and systems research, we developed a collaborative learning platform in which we worked as co-researchers with health authorities in a rural province. This article reports on the process and insights brought by health systems stakeholders. Evidence gaps on under-five mortality were identified with a provincial Directorate after which we collected quantitative and qualitative data. We applied verbal autopsy to quantify levels, causes and circumstances of deaths and participatory action research to gain community perspectives on the problem and priorities for action. We then re-convened health systems stakeholders to analyse and interpret these data through which several systems issues were identified as contributory to under-five deaths: staff availability and performance; service organization and infrastructure; multiple parallel initiatives; and capacity to address social determinants. Recommendations were developed ranging from immediate low- and no-cost re-organization of services to those where responses from higher levels of the system or outside were required. The process was viewed as acceptable and relevant for an overburdened system operating 'in the dark' in the absence of local data. Institutional infrastructure for evidence-based decision-making does not exist in many health systems. We developed a process connecting research evidence on rural health priorities with the means for action and enabled new partnerships between communities, authorities and researchers. Further development is planned to understand potential in deliberative processes for rural PHC.

External injuries, trauma and avoidable deaths in Agincourt, South Africa: a retrospective observational and qualitative study.

OBJECTIVE: Injury burden is highest in low-income and middle-income countries. To reduce avoidable deaths, it is necessary to identify health system deficiencies preventing timely, quality care. We developed criteria to use verbal autopsy (VA) data to identify avoidable deaths and associated health system deficiencies. SETTING: Agincourt, a rural Bushbuckridge municipality, Mpumalanga Province, South Africa. PARTICIPANTS: Agincourt Health and Socio-Demographic Surveillance System and healthcare providers (HCPs) from local hospitals. METHODS: A literature review to explore definitions of avoidable deaths after trauma and barriers to access to care using the 'three delays framework' (seeking, reaching and receiving care) was performed. Based on these definitions, this study developed criteria, applicable for use with VA data, for identifying avoidable death and which of the three delays contributed to avoidable deaths. These criteria were then applied retrospectively to the VA-defined category external injury deaths (EIDs-a subset of which are trauma deaths) from 2012 to 2015. The findings were validated by external expert review. Key informant interviews (KIIs) with HCPs were performed to further explore delays to care. RESULTS: Using VA data, avoidable death was defined with a focus on survivability, using level of consciousness at the scene and ability to seek care as indicators. Of 260 EIDs (189 trauma deaths), there were 104 (40%) avoidable EIDs and 78 (30%) avoidable trauma deaths (41% of trauma deaths). Delay in receiving care was the largest contributor to avoidable EIDs (61%) and trauma deaths (59%), followed by delay in seeking care (24% and 23%) and in reaching care (15% and 18%). KIIs revealed context-specific factors contributing to the third delay, including difficult referral systems. CONCLUSIONS: A substantial proportion of EIDs and trauma deaths were avoidable, mainly occurring due to facility-based delays in care. Interventions, including strengthening referral networks, may substantially reduce trauma deaths.

An integrated approach to processing WHO-2016 verbal autopsy data: the InterVA-5 model.

BACKGROUND: Verbal autopsy is an increasingly important methodology for assigning causes to otherwise uncertified deaths, which amount to around 50% of global mortality and cause much uncertainty for health planning. The World Health Organization sets international standards for the structure of verbal autopsy interviews and for cause categories that can reasonably be derived from verbal autopsy data. In addition, computer models are needed to efficiently process large quantities of verbal autopsy interviews to assign causes of death in a standardised manner. Here, we present the InterVA-5 model, developed to align with the WHO-2016 verbal autopsy standard. This is a harmonising model that can process input data from WHO-2016, as well as earlier WHO-2012 and Tariff-2 formats, to generate standardised cause-specific mortality profiles for diverse contexts. The software development involved building on the earlier InterVA-4 model, and the expanded knowledge base required for InterVA-5 was informed by analyses from a training dataset drawn from the Population Health Metrics Research Collaboration verbal autopsy reference dataset, as well as expert input. RESULTS: The new model was evaluated against a test dataset of 6130 cases from the Population Health Metrics Research Collaboration and 4009 cases from the Afghanistan National Mortality Survey dataset. Both of these sources contained around three quarters of the input items from the WHO-2016, WHO-2012 and Tariff-2 formats. Cause-specific mortality fractions across all applicable WHO cause categories were compared between causes assigned in participating tertiary hospitals and InterVA-5 in the test dataset, with concordance correlation coefficients of 0.92 for children and 0.86 for adults. The InterVA-5 model's capacity to handle different input formats was evaluated in the Afghanistan dataset, with concordance correlation coefficients of 0.97 and 0.96 between the WHO-2016 and the WHO-2012 format for children and adults respectively, and 0.92 and 0.87 between the WHO-2016 and the Tariff-2 format respectively. CONCLUSIONS: Despite the inherent difficulties of determining "truth" in assigning cause of death, these findings suggest that the InterVA-5 model performs well and succeeds in harmonising across a range of input formats. As more primary data collected under WHO-2016 become available, it is likely that InterVA-5 will undergo minor re-versioning in the light of practical experience. The model is an important resource for measuring and evaluating cause-specific mortality globally.

A comparison of all-cause and cause-specific mortality by household socioeconomic status across seven INDEPTH network health and demographic surveillance systems in sub-Saharan Africa.

BACKGROUND: Understanding socioeconomic disparities in all-cause and cause-specific mortality can help inform prevention and treatment strategies. OBJECTIVES: To quantify cause-specific mortality rates by socioeconomic status across seven health and demographic surveillance systems (HDSS) in five countries (Ethiopia, Kenya, Malawi, Mozambique, and Nigeria) in the INDEPTH Network in sub-Saharan Africa. METHODS: We linked demographic residence data with household survey data containing living standards and education information we used to create a poverty index. Person-years lived and deaths between 2003 and 2016 (periods varied by HDSS) were stratified in each HDSS by age, sex, year, and number of deprivations on the poverty index (0-8). Causes of death were assigned to each death using the InterVA-4 model based on responses to verbal autopsy questionnaires. We estimated rate ratios between socioeconomic groups (2-4 and 5-8 deprivations on our poverty index compared to 0-2 deprivations) for specific causes of death and calculated life expectancy for the deprivation groups. RESULTS: Our pooled data contained almost 3.5 million person-years of observation and 25,038 deaths. All-cause mortality rates were higher among people in households with 5-8 deprivations on our poverty index compared to 0-2 deprivations, controlling for age, sex, and year (rate ratios ranged 1.42 to 2.06 across HDSS sites). The poorest group had consistently higher death rates in communicable, maternal, neonatal, and nutritional conditions (rate ratios ranged 1.34-4.05) and for non-communicable diseases in several sites (1.14-1.93). The disparities in mortality between 5-8 deprivation groups and 0-2 deprivation groups led to lower life expectancy in the higher-deprivation groups by six years in all sites and more than 10 years in five sites. CONCLUSIONS: We show large disparities in mortality on the basis of socioeconomic status across seven HDSS in sub-Saharan Africa due to disparities in communicable disease mortality and from non-communicable diseases in some sites. Life expectancy gaps between socioeconomic groups within sites were similar to the gaps between high-income and lower-middle-income countries. Prevention and treatment efforts can benefit from understanding subpopulations facing higher mortality from specific conditions.