Objectives
Efforts to shrink the U.S. Government have decimated the existing American data infrastructure. By tracking the impact of executive orders and administration policies on data integrity, supply, and demand, we show outcomes for state and local governments, non-profit and social sector organizations, and the private sector.
Methods
Soon after the inauguration, government agencies had to remove data and findings about topics including diversity, equity and inclusion; sexual orientation and gender identity; and climate change. This involved dataset delisting and funding restrictions in affected programs. Data specialists began data rescue efforts to preserve copies of existing data sources. Additional data experts started data monitoring efforts to spot changes and deletions. At time same time, cuts to federal staffing and budgets were monitored to identify disruptions to linked and modeled data measures essential for the functioning of the economy and society.
Results
A dedicated community emerged in February 2025 including academic and private sector data users, archivists and librarians, and former data producers. They track what is gone, what is changing, how critical data series are affected (e.g., economic indicators and population estimates), and source alternative options. Others assess the data inputs for AI models, and propose protections to protect privacy and to reduce risks of AI perpetuating error or bias. Policy proposals on how to restore transparency involve physical and information technology security, as well as policies and protocols. These proposals aim to restore trust that data will be used appropriately and that questions to government officials will be answered.
Conclusion
The previous governance structures and norms about use of data are gone. The new normal must be monitored and questioned, specifically whether increased use of AI achieves efficiencies and harms. We outline plans to build on the current infrastructure to ensure reliable and responsible use of personal data.
Objectives
Linking large-scale datasets is challenging due to the computational power required. This research explores using Locality-Sensitive-Hashing (LSH) as a novel blocking method to facilitate linking large administrative data. LSH uses an efficient technique of hashing data while preserving similarity, reducing the search space and processing power required to find links.
Approach
Samples were taken from a gold-standard dataset and blocked using LSH. Various parameters were tested to establish LSH’s optimal performance at retrieving linked pairs and reducing the search space size.
Results
Testing the method on small datasets gave promising results, with the LSH method creating ~9,000 candidate pairs. This is a improvement over our traditional blocking method or cartesian product, which created ~70,000 and 23.4 million candidate pairs, respectively. We have therefore shown that LSH can significantly reduce the search-space size. A further advantage of the method is its capability in handling alternative variables such as those that may be present in longitudinal or composite data, without needing to manually anticipate the different combinations of variables which may occur. Variable formats to simulate agreement weighting were also considered, with encouraging results.
Conclusion and Implications
Our research shows that LSH can be used to drastically reduce the search space when blocking for data linkage This suggests developing LSH as a blocking method could result in more effective blocking, quicker linkage, and no loss of quality compared to traditional methods. However, further research on increasing the method’s scale is necessary.
When record linkage efforts involve complex characteristics there is ample potential for general purpose machine learning (ML) techniques to succeed where traditional probabilistic approaches might fall short. However, there can still be pre-processing (e.g. geocoding) and hand-picked comparators that can further improve linkage outcomes using standard ML models. In this project we present a fusion of these sides we are calling an Augmented Twin Neural Network. This approach leverages the inherent flexibility of Twin Neural Networks in a record linkage context while adding additional layers to allow for hand curated comparators that may be difficult for ML optimizers to implicitly identify without sufficiently large, labeled data sets. The framework is used to match establishments from the BLS Survey of Occupational Injuries and Illnesses to establishments in the OSHA Injury Tracking Application data. The difficulties inherent in matching company names and addresses and the existence of multi-establishment firms make this a prime application for testing. Linkage outcome metrics of this augmented algorithm are compared both with results from probabilistic methods (e.g. Fellegi-Sunter) and standard machine learning methods to illustrate the added benefits.
Theodora Kokosi, Bianca De Stavola, Robin Mitra
et al.
Use of administrative data for research and for planning services has increased over recent decades due to the value of the large, rich information available. However, concerns about the release of sensitive or personal data and the associated disclosure risk can lead to lengthy approval processes and restricted data access. This can delay or prevent the production of timely evidence. A promising solution to facilitate more efficient data access is to create synthetic versions of the original datasets which do not hold any confidential information and can minimise disclosure risk. Such data may be used as an interim solution, allowing researchers to develop their analysis plans on non-disclosive data, whilst waiting for access to the real data. We aim to provide an overview of the background and uses of synthetic data, describe common methods used to generate synthetic data in the context of UK administrative research, propose a simplified terminology for categories of synthetic data, and illustrate challenges and future directions for research.
Daniela K Schlüter, Rowena Griffiths, Ashley Akbari
et al.
Introduction
As people with cystic fibrosis (CF) lead longer, healthier lives, educational qualifications and employment prospects are increasingly important. However, little is known about the social consequences of CF, in particular, any impact on educational achievements and the support children with CF receive in schools.
Objectives
To assess the educational achievements of children with CF in Wales compared to the general Welsh population, and the additional learning support children with CF receive in schools.
Methods
We conducted a population-scale data linkage study of all children born in Wales using the Secure Anonymised Information Linkage (SAIL) Databank. We used anonymised individual-level population-scale health and administrative data sources to identify children with CF born between 2000 -- 2015, linked to educational attainment records. We calculated the percentage of children that reached expected levels in statutory assessment at age 10-11, Key Stage
2 (KS2), and compared this to educational outcomes in the general population. We also assessed the percentage of children with CF that received extra learning support.
Results
Out of 150 eligible children, 119 had KS2 results. 77% (95% CI: 69%-84%) of children achieved expected levels in English, 81% (95% CI: 73% -87%) in Mathematics and 82% (95% CI: 75% - 88%) in Science. In the comparable general Welsh population, 83.4% to 91.1% achieved the expected level in English, 84.9% to 91.6% in Maths, and 87.1% to 92.2% in Science across the years of the study. 70% of children with CF received extra learning support.
Conclusions
Children with CF in Wales may have worse educational achievements than the general population. More research is needed to inform policies and interventions to better support children with CF to reach their full educational potential and employment opportunities.
INTRODUCTION Smoking produces significant health problems affecting millions of people globally. It is the largest single cause of preventable deaths. This study aims to describe the characteristics of Filipino adolescent smokers aged 13–15 years and determine the factors associated with the intention to continue to use tobacco in the next 12 months among current smokers. METHODS This was a secondary analysis of the 2015 Global Youth Tobacco Survey (GYTS) in the Philippines to determine the association of the following factors with the intention to continue to use tobacco in the next 12 months among current smokers: 1) exposure to anti-tobacco media messages, 2) exposure to health warnings on cigarette packages, and 3) being taught about the dangers of tobacco in any class. RESULTS The adjusted odds of continuing to use of any form of tobacco in the next 12 months was higher (AOR=1.24; 95% CI: 0.62–2.49) among those who have not seen health warnings on cigarette packages and higher (AOR=1.32; 95% CI: 0.71–2.45) when not taught about tobacco in any of their classes. Finally, those who have not seen or heard antitobacco messages on television, radio, internet, billboards, posters, newspapers, magazines, or movies, were more likely (AOR=2.19; 95% CI: 1.14–4.21) to report their intention to continue to use tobacco. CONCLUSIONS Understanding the relationship of different tobacco control strategies to adolescents’ intention to continue using tobacco can contribute to the improvement of tobacco policies and programs and achieve effective tobacco control among the youth. A targeted anti-tobacco campaign in both traditional and new media should be considered an essential part of a comprehensive tobacco control program. Health warnings on cigarette packs should be supported by a policy banning the sale of single stick cigarettes. INTRODUCTION Tobacco use is one of the biggest public health threats globally and the most significant single cause of preventable deaths. Globally, it kills more than seven million a year, with more than six million of those deaths attributable to direct tobacco use. Around 80% of smokers worldwide live in lowand middle-income countries such as the Philippines1. Smoking, especially during adolescence, produces significant health problems, including respiratory illnesses, cancer, and cardiovascular disease2. To address the growing tobacco epidemic, the Philippines started implementing tobacco control measures as early as 1987. Several tobacco control policies were enacted to: 1) increase tobacco taxes, 2) ban tobacco advertising or promotion, 3) designate smoke-free places, and 4) ban the sale of tobacco products to minors3. Despite these actions, 47.5% of Filipino students aged 13–15 years were not prevented from buying cigarettes despite their age, while only 67.1% were taught in school about the dangers of tobacco use4. Moreover, there were significant decreases in the prevalence of Filipino youth who saw anti-tobacco messages in the media (from 89.4% in 2011 to 71.6% in 2015) and at sports or community events (from 88.8% in 2011 to 53.2% in 2015)4. In 2015, 14.5% of Filipino students were current tobacco smokers (smoked tobacco anytime during the past 30 Research Paper| Population Medicine Popul. Med. 2021;3(November):30 https://doi.org/10.18332/popmed/143526 2 days), while 12% were current cigarette smokers (smoked cigarettes anytime during the past 30 days)4. Current cigarette smoking among Filipino youth showed a declining trend from the year 20004. However, a 34.8% increase in current cigarette smoking was observed between 2011 and 20154. The latest Global Youth Tobacco Survey (GYTS) shows that one in ten students are current smokers and 0.7% have smoked cigarettes on 20 or more days in the past 30 days4. It is crucial to look at adolescents’ intention to continue to use tobacco, which can predict behavior. According to the Theory of Planned Behavior, a behavior is a result of the intention to do the behavior and the perceived control over it. Behavioral intentions are often influenced by the attitude towards the behavior, the subjective evaluation of the results of the behavior, and the perceived control over the behavior. While the relationship between intention and behavior is not always perfect, the intention is often used as a proxy for actual behavior. Given inadequate information, young people often underestimate the risk of becoming addicted to smoking due to nicotine, which relates to underestimating future costs and consequences5. This myopic view on the ills of tobacco use and the external costs imposed on others, which result in premature death and illness, justifies government intervention through strategies including regulation, taxation, and information dissemination. Hence, to reduce the prevalence of current tobacco use among the youth by 2% per year, the Philippines has to devise or reinvent its tobacco control and prevention strategies for the youth4. This study aims to look at three of the most common methods of message delivery and their relationship to adolescent smokers’ intention to continue using tobacco. Specifically, this study aimed to describe the characteristics of Filipino adolescent smokers aged 13–15 years and determine the association of the following factors with the intention to continue to use tobacco in the next 12 months among current smokers: 1) exposure to anti-tobacco media messages, 2) exposure to health warnings on cigarette packages, and 3) being taught about the dangers of tobacco in any class. METHODS Data source This study was a secondary dataset analysis of the 2015 Global Youth Tobacco Survey in the Philippines, which was administered from January to March 20154. The GYTS was developed by the Tobacco Free Initiative (TFI), United States Centers for Disease Control and Prevention (US CDC), and World Health Organization (WHO) in collaboration with representatives from the six WHO regions6. The GYTS is a school-based survey designed to monitor tobacco use among the youth and guide tobacco control program implementation and evaluation6. The GYTS is a school-based survey that utilized a twostage cluster sample design that produced representative samples of students in year levels that were associated with ages 13–15 years4,6. It uses a self-administered questionnaire to evaluate trends in tobacco use among adolescents and to guide the implementation and evaluation of tobacco prevention and control programs for the Filipino youth6. The Philippines has conducted five rounds of the GYTS since 2000. The Philippine version of the GYTS is a nationally representative survey of students aged 13–15 years and is regularly used to generate data that are comparable to other countries4. It utilizes a global standard to systematically monitor tobacco control indicators that include: data on the prevalence of cigarette and other tobacco use, exposure to secondhand smoke, accessibility and availability of tobacco products, the influence of media and advertising on the use of cigarettes, susceptibility, smoking cessation, and knowledge and attitudes towards tobacco use and policies4. Definition of variables Current cigarette smoking Two questions were used to define current cigarette smoking: 1) ‘Have you ever tried or experimented with cigarette smoking, even one or two puffs?’ with response ‘yes’ or ‘no’; and 2) ‘During the past 30 days (one month), on how many days did you smoke cigarettes?’ where answers were dichotomized (‘1’ for those reporting 0 days in the past 30 days, and ‘2’ for those reporting ≥1 day in the past 30 days). Participants who have answered ‘yes’ in the first question and reported at least 1 day of cigarette use in the past 30 days were considered current cigarette smokers. Intention to continue using tobacco in the next 12 months Participants were asked to indicate their future intention to continue using any form of tobacco with four response options: ‘definitely not’, ‘probably not’, ‘probably yes’, and ‘definitely yes’. For this study, the answers were dichotomized into ‘yes’ (combined ‘probably yes’ and ‘definitely yes’) and ‘no’ (combined ‘probably not’ and ‘definitely not’). Exposure to anti-tobacco media messages Participants were asked to indicate whether they have or have not seen or heard any anti-tobacco media messages on television, radio, internet, billboards, posters, newspapers, magazines, or movies during the past 30 days. Exposure to health warnings on cigarette packages Participants were asked whether they have or have not seen any health warnings on cigarette packages during the past 30 days using the following options: ‘Yes but I didn’t think much of them’, and ‘Yes, and they led me to think about quitting smoking or not starting’. The answers were dichotomized into ‘yes’ and ‘no’. Taught the dangers of tobacco use Participants were asked whether they have or have not Research Paper| Population Medicine Popul. Med. 2021;3(November):30 https://doi.org/10.18332/popmed/143526 3 been taught about the dangers of tobacco use in any of their classes in the past 12 months with the options: ‘yes’, ‘no’, and ‘do not know’. Only participants who answered ‘yes’ or ‘no’ were included in the analysis. Statistical analysis Descriptive statistics were used to identify the demographic characteristics of the participants. Crude and adjusted logistic regression analyses were performed to examine the association between independent variables and intention to continue using tobacco products among current smokers. In the logistic regression analyses, odds ratios and confidence intervals were calculated. The logistic regression models were adjusted for the effects of covariates such as age, sex, and type of school. All analyses were conducted using JASP version 0.14.1, which was released on 17 December 2020. JASP is a free and open-source program for statistical analysis. The statistical software offers standard analysis procedures in c
Resumo Uma questão presente na literatura que trata dos jovens que não estudam nem trabalham é a carência da compreensão de suas características. O objetivo do presente estudo foi encontrar os determinantes do perfil dos jovens que se encontram nas diferentes categorias de condição ocupacional − trabalha e estuda; só trabalha; só estuda; nem estuda nem trabalha, mas procura emprego (“nem-nem” ativo); e nem estuda nem trabalha e nem procura emprego (“nem-nem” inativo) −, bem como buscar a relação entre o fato de os jovens encontrarem-se nas duas últimas categorias e a maior incidência de afazeres domésticos semanais em seus domicílios. Assim, com os dados da Pesquisa Nacional por Amostra de Domicílio de 2015, empregaram-se os métodos logit multinomial e de contagem (regressão de Poisson: incidence rate ratios e zero inflated). Os resultados mostram que o perfil de um jovem pertencente à condição ocupacional “nem-nem” ativo difere-se consideravelmente de outro classificado como “nem-nem” inativo, principalmente quanto ao gênero e à situação de residência. Relativamente às demais categorias, a carga horária de afazeres domésticos incide com maior frequência nos jovens “nem-nem” ativos e inativos, ao passo que estes também estão menos propensos a se pouparem das atividades domésticas. Mostrou-se de suma importância o discernimento entre ativos e inativos para a categoria dos que nem trabalham nem estudam e, a partir dos resultados encontrados, pode-
<b>Background</b>: Recent kinship models focus on the age structures of kin as a function of the age of the focal individual. However, variables in addition to age have important impacts. Generalizing age-specific models to multistate models including other variables is an important and hitherto unsolved problem. <b>Objective</b>: The aim is to develop a multistate kinship model, classifying individuals jointly by age and other criteria (generically, "stages"). <b>Methods</b>: The vec-permutation method is used to create multistate projection matrices including age- and stage-dependent survival, fertility, and transitions. These matrices operate on block-structured population vectors that describe the age×stage structure of each kind of kin, at each age of a focal individual. <b>Results</b>: The new matrix formulation is directly comparable to, and greatly extends, the recent age-classified kinship model of Caswell (2019a). As an application, a model is derived including age and parity. It provides, for all types of kin, the joint age×parity structure, the marginal age and parity structures, and the (normalized) parity distributions, at every age of the focal individual. The age×parity distributions provide the distributions of sibship sizes of kin. As an example, the model is applied to Slovakia (1960-2014). The results show a dramatic shift in the parity distribution as the frequency of low-parity kin increased and that of high-parity kin decreased. <b>Contribution</b>: This model extends the formal demographic analysis of kinship to age×stage-classified models. In addition to parity, other stage classifications, including marital status, maternal age effects, and sex are now open to analysis.
Mehdi Feyzolahpour, Jamshid Einali, Hassan Gasemlu
Geotourism and geoparks provide good opportunities for rural development and reduce unemployment and migration. It attracts local communities for employment in geoparks and tourism marketing in the form of investments in ecotourism, rural tourism and health geotourism. Geotourism is closely related to the geology of treatment. One of these is the spa spring that has the therapeutic potential and plays a significant role in attracting domestic and foreign tourists. For this purpose the status of 11 hot springs in the geographical range between Ardabil city and Sarein city has been investigated in terms of parameters such as discharge, temperature, pH and anions and cations. However, this geographical area also has a number of natural and human hazards the most important of which is the occurrence of killing earthquakes. The earthquake of February 28, 1997 killed nearly 1000 people and destroyed many villages. The presence of spa springs on or near the faults and the establishment of recreational facilities in the area have threatened investment in the area. In this regard the status of clay, silt and sand percentage, Liquefaction Limit and Plasticity PI index in the study area were studied. It was observed that if in clay formations the amount of dough is about 40 to 50 percent and soil moisture reaches 25 to 35 percent there is a possibility of demolition of buildings and asphalt. According to the results of soil physics analysis, Sarein city is susceptible to soil infiltration and liquefaction during earthquake. In the final section the sanitary condition of the pools was investigated in terms of the possibility of dermatophyte fungi and 284 samples were examined. It was observed that due to the complete disinfection of pools with chlorine no dermatophytes were observed.
El objetivo de esta investigación es determinar la existencia y magnitud de ciclos de variación estacional en la mortalidad de Costa Rica de 1970 a 2016. Metodológicamente, el estudio se basa en microdatos de 630,000 defunciones entre esos años. La naturaleza cíclica de la variación estacional se modela con regresión sinusoide y parámetros estimados con regresión de Poisson. Los resultados muestran que hay variación estacional significativa con un máximo en enero y un mínimo en mayo. La mortalidad tiende a ser 7 % mayor en su ápex que en su nadir. Este patrón está determinado por accidentes, especialmente de transporte, alcoholismo, enfermedades cardio- y cerebrovasculares, e infecciones respiratorias. La mortalidad por diarreas presenta un patrón diferente de estacionalidad. En conclusión, la mayor mortalidad de enero estaría asociada con la temperatura menor, ausencia de pluviosidad, menor luz solar y comportamiento durante días festivos. Algunos picos de mortalidad podrían deberse a fluctuaciones en la calidad de los servicios de salud y de atención de emergencias.
Social Sciences, Demography. Population. Vital events
Monique Kilkenny, Joosup Kim, Nadine Andrew
et al.
Introduction
Recent advances in data linkage infrastructure in Australia mean that data can be linked based on various identifiers across datasets. In a first for Australia, we tested the feasibility of linking data between a clinical quality disease registry with Australian and state government health data across multiple jurisdictions.
Objectives and Approach
To determine whether high quality linked data for stroke can be obtained using a non-government managed registry (Australian Stroke Clinical Registry, AuSCR), national death registry data (Australian government), and hospital admission and emergency presentation data (state governments) to assess the accuracy of consistent variables across the different datasets. We used a cohort design with probabilistic data linkage to merge patient-level records. Descriptive statistics presented for matching concordance and Cohen’s kappa for concordance across demographic variables. The sensitivity and specificity of in-hospital deaths collected in the AuSCR was assessed against national death registrations.
Results
There were 16,214 registrants in the study cohort. Their identifiers in the AuSCR from 2009-2013 were linked with death, emergency department and hospital discharge data from April 2004 to December 2016. In total, 99% of the AuSCR registrants were linked to one or more datasets; 98\% were linked with emergency presentation (80%) and/or admission (95%) data. Linkage to national death registrations identified 4,183 death; 1440 of these were identified as in-hospital deaths in both data sets demonstrating that in-hospital death classification in AuSCR had a 98.7% sensitivity and 99.6% specificity. Concordance between common demographic variables was excellent (kappa 0.84 for aboriginal status and kappa 0.99 for sex).
Conclusion/Implications
The majority of AuSCR registrants were accurately linked to the Australian and state government datasets. Linkage quality was excellent and there was high concordance between common variables. The ability to reliably merge the datasets assures future comprehensive analyses of stroke care, ongoing health care resource utilisation and patient outcomes.
Introduction
The Australian 2016 census reported Aboriginal population of Goulburn Murray Region (GMR) as being 3% of the total GMR population is considered a significant under ascertainment. This is supported by the Freemantle et al population data linkage project that reported an underestimation of 51% of Aboriginal births in Hume region.
Objectives and Approach
The main objective of the Gana Burrai project was to undertake a Proof of Concept (POC) project to address the critical issue the under-ascertainment of data describing Aboriginal maternal and child health outcomes within the GMR. The PoC explored the feasibility of linking a number of population administrative datasets that included information describing Aboriginal babies born and mothers and fathers living in, the GMR. The PoC also enabled an evaluation of the cultural, organisational, practical and technical issues critical to the development of this comprehensive population data set. The linkage was undertaken using GRHANITETM privacy protecting software.
Results
Population data from seven administrative datasets describing birth years 2005-2015, representing 111862 records were successfully extracted. These data represented 37,338 patients. Following the linkage process, 28,016 patient records were successfully linked. The ever-Aboriginal concept was applied to the linked data. After linkage, there was a significant under ascertainment of Aboriginal births in records maintained by the Shire of Campaspe (16%), less so by Shire of Greater Shepparton (50%); Goulburn Valley Heath (92%), and Echuca Regional Health (86%) demonstrated a more complete ascertainment. However, the linkage showed a strong correlation of patient Indigenous identity collected by the shires’/hospitals’ and the mothers’/fathers’ chance of having a record at the Aboriginal Community Controlled Health Organisations. Selected birth outcomes were also analysed and will be reported.
Conclusion/Implications
1st Nations Governance combined with a privacy-protecting linkage model underpinned PoC ensuring strong community trust, engagement and control. It is 1st time such extensive data-linkage has been achieved demonstrating feasibility in developing a comprehensive dataset with the potential for evidence-based, targeted population health initiatives to be developed, implemented and evaluated.
BACKGROUND Disability-adjusted life years (DALYs) provide a summary measure of health and can be a critical input to guide health systems, investments, and priority-setting in Ethiopia. We aimed to determine the leading causes of premature mortality and disability using DALYs and describe the relative burden of disease and injuries in Ethiopia. METHODS We used results from the Global Burden of Diseases, Injuries, and Risk Factors Study 2015 (GBD 2015) for non-fatal disease burden, cause-specific mortality, and all-cause mortality to derive age-standardized DALYs by sex for Ethiopia for each year. We calculated DALYs by summing years of life lost due to premature mortality (YLLs) and years lived with disability (YLDs) for each age group and sex. Causes of death by age, sex, and year were measured mainly using Causes of Death Ensemble modeling. To estimate YLDs, a Bayesian meta-regression method was used. We reported DALY rates per 100,000 for communicable, maternal, neonatal, and nutritional (CMNN) disorders, non-communicable diseases, and injuries, with 95% uncertainty intervals (UI) for Ethiopia. RESULTS Non-communicable diseases caused 23,118.1 (95% UI, 17,124.4-30,579.6), CMNN disorders resulted in 20,200.7 (95% UI, 16,532.2-24,917.9), and injuries caused 3781 (95% UI, 2642.9-5500.6) age-standardized DALYs per 100,000 in Ethiopia in 2015. Lower respiratory infections, diarrheal diseases, and tuberculosis were the top three leading causes of DALYs in 2015, accounting for 2998 (95% UI, 2173.7-4029), 2592.5 (95% UI, 1850.7-3495.1), and 2562.9 (95% UI, 1466.1-4220.7) DALYs per 100,000, respectively. Ischemic heart disease and cerebrovascular disease were the fourth and fifth leading causes of age-standardized DALYs, with rates of 2535.7 (95% UI, 1603.7-3843.2) and 2159.9 (95% UI, 1369.7-3216.3) per 100,000, respectively. The following causes showed a reduction of 60% or more over the last 25 years: lower respiratory infections, diarrheal diseases, tuberculosis, neonatal encephalopathy, preterm birth complications, meningitis, malaria, protein-energy malnutrition, iron-deficiency anemia, measles, war and legal intervention, and maternal hemorrhage. CONCLUSIONS Ethiopia has been successful in reducing age-standardized DALYs related to most communicable, maternal, neonatal, and nutritional deficiency diseases in the last 25 years, causing a major ranking shift to types of non-communicable disease. Lower respiratory infections, diarrheal disease, and tuberculosis continue to be leading causes of premature death, despite major declines in burden. Non-communicable diseases also showed reductions as premature mortality declined; however, disability outcomes for these causes did not show declines. Recently developed non-communicable disease strategies may need to be amended to focus on cardiovascular diseases, cancer, diabetes, and major depressive disorders. Increasing trends of disabilities due to neonatal encephalopathy, preterm birth complications, and neonatal disorders should be emphasized in the national newborn survival strategy. Generating quality data should be a priority through the development of new initiatives such as vital events registration, surveillance programs, and surveys to address gaps in data. Measuring disease burden at subnational regional state levels and identifying variations with urban and rural population health should be conducted to support health policy in Ethiopia.
Background We believe that it is important that governments understand the reliability of the mortality data which they have at their disposable to guide policy debates. In many instances, verbal autopsy (VA) will be the only source of mortality data for populations, yet little is known about how the accuracy of VA diagnoses is affected by the reliability of the symptom responses. We previously described the effect of the duration of time between death and VA administration on VA validity. In this paper, using the same dataset, we assess the relationship between the reliability and completeness of symptom responses and the reliability and accuracy of cause of death (COD) prediction. Methods The study was based on VAs in the Population Health Metrics Research Consortium (PHMRC) VA Validation Dataset from study sites in Bohol and Manila, Philippines and Andhra Pradesh, India. The initial interview was repeated within 3–52 months of death. Question responses were assessed for reliability and completeness between the two survey rounds. COD was predicted by Tariff Method. Results A sample of 4226 VAs was collected for 2113 decedents, including 1394 adults, 349 children, and 370 neonates. Mean question reliability was unexpectedly low (kappa = 0.447): 42.5 % of responses positive at the first interview were negative at the second, and 47.9 % of responses positive at the second had been negative at the first. Question reliability was greater for the short form of the PHMRC instrument (kappa = 0.497) and when analyzed at the level of the individual decedent (kappa = 0.610). Reliability at the level of the individual decedent was associated with COD predictive reliability and predictive accuracy. Conclusions Families give coherent accounts of events leading to death but the details vary from interview to interview for the same case. Accounts are accurate but inconsistent; different subsets of symptoms are identified on each occasion. However, there are sufficient accurate and consistent subsets of symptoms to enable the Tariff Method to assign a COD. Questions which contributed most to COD prediction were also the most reliable and consistent across repeat interviews; these have been included in the short form VA questionnaire. Accuracy and reliability of diagnosis for an individual death depend on the quality of interview. This has considerable implications for the progressive roll out of VAs into civil registration and vital statistics (CRVS) systems. Electronic supplementary material The online version of this article (doi:10.1186/s12963-016-0104-2) contains supplementary material, which is available to authorized users.
Carren Ginsburg, Philippe Bocquier, Donatien Beguy
et al.
<b>Background</b>: Education, as a key indicator of human capital, is considered one of the major determinants of internal migration, with previous studies suggesting that human capital accumulates in urban areas at the expense of rural areas. However, there is fragmentary evidence concerning the educational correlates of internal migration in sub-Saharan Africa. <b>Objective</b>: The study questions whether more precise measures of migration in Health and Demographic Surveillance System (HDSS) populations support the hypothesis that migrants are self-selected on human capital and more educated people are more likely to leave rural areas or enter urban areas within a geographical region. <b>Methods</b>: Using unique longitudinal data representing approximately 900,000 people living in eight sub-Saharan African HDSS sites that are members of the INDEPTH Network, the paper uses Event History Analysis techniques to examine the relationship between formal educational attainment and in- and out-migration, over the period 2009 to 2011. <b>Results</b>: Between 7Š and 27Š of these local populations are moving in or out of the HDSS area over this period. Education is positively associated with both in- and out-migration in the Kenyan HDSS areas; however, the education effect has no clear pattern in the HDSS sites in Burkina Faso, Mozambique, and South Africa. <b>Conclusions</b>: Empirical results presented in this paper confirm a strong age profile of migration consistent with human capital expectation, yet the results point to variability in the association of education and the propensity to migrate. In particular, the hypothesis of a shift of human capital from rural to urban areas is not universally valid.
BackgroundWe present a method for reclassifying external causes of death categorized as “event of undetermined intent” (EUIs) into non-transport accidents, suicides, or homicides. In nations like Russia and the UK the absolute number of EUIs is large, the EUI death rate is high, or EUIs comprise a non-trivial proportion of all deaths due to external causes. Overuse of this category may result in (1) substantially underestimating the mortality rate of deaths due to specific external causes and (2) threats to the validity of studies of the patterns and causes of external deaths and of evaluations of the impact of interventions meant to reduce them.MethodsWe employ available characteristics about the deceased and the event to estimate the most likely cause of death using multinomial logistic regression. We use the set of known non-transport accidents, suicides, and homicides to calculate an mlogit-based linear score and an estimated classification probability (ECP). This ECP is applied to EUIs, with varying levels of minimal classification probability. We also present an optional second step that employs a population-level adjustment to reclassify deaths that remain undetermined (the proportion of which varies based on the minimal classification probability). We illustrate our method by applying it to Russia. Between 2000 and 2011, 521,000 Russian deaths (15 % percent of all deaths from external causes) were categorized as EUIs. We used data from anonymized micro-data on the ~3 million deaths from external causes. Our reclassification model used 10 decedent and event characteristics from the computerized death records.ResultsResults show that during this period about 14 % of non-transport accidents, 13 % of suicides, and 33 % of homicides were officially categorized as EUIs. Our findings also suggest that 2011 levels of non-transport accidents and suicides would have been about 24 % higher and of homicide about 82 % higher than that reported by official vital statistics data.ConclusionsOveruse of the external cause of death classification “event of undetermined intent” may indicate questionable quality of mortality data on external causes of death. This can have wide-ranging implications for families, medical professionals, the justice system, researchers, and policymakers. With our classification probability set as equal to or higher than 0.75, we were able to reclassify about two-thirds of EUI deaths in our sample. Our optional additional step allowed us to redistribute the remaining unclassified EUIs. Our method can be applied to data from any nation or sub-national population in which the EUI category is employed.