Dataset Information

Public health utility of cause of death data: applying empirical algorithms to improve data quality.

ABSTRACT:

Background

Accurate, comprehensive, cause-specific mortality estimates are crucial for informing public health decision making worldwide. Incorrectly or vaguely assigned deaths, defined as garbage-coded deaths, mask the true cause distribution. The Global Burden of Disease (GBD) study has developed methods to create comparable, timely, cause-specific mortality estimates; an impactful data processing method is the reallocation of garbage-coded deaths to a plausible underlying cause of death. We identify the pattern of garbage-coded deaths in the world and present the methods used to determine their redistribution to generate more plausible cause of death assignments.

Methods

We describe the methods developed for the GBD 2019 study and subsequent iterations to redistribute garbage-coded deaths in vital registration data to plausible underlying causes. These methods include analysis of multiple cause data, negative correlation, impairment, and proportional redistribution. We classify garbage codes into classes according to the level of specificity of the reported cause of death (CoD) and capture trends in the global pattern of proportion of garbage-coded deaths, disaggregated by these classes, and the relationship between this proportion and the Socio-Demographic Index. We examine the relative importance of the top four garbage codes by age and sex and demonstrate the impact of redistribution on the annual GBD CoD rankings.

Results

The proportion of least-specific (class 1 and 2) garbage-coded deaths ranged from 3.7% of all vital registration deaths to 67.3% in 2015, and the age-standardized proportion had an overall negative association with the Socio-Demographic Index. When broken down by age and sex, the category for unspecified lower respiratory infections was responsible for nearly 30% of garbage-coded deaths in those under 1 year of age for both sexes, representing the largest proportion of garbage codes for that age group. We show how the cause distribution by number of deaths changes before and after redistribution for four countries: Brazil, the United States, Japan, and France, highlighting the necessity of accounting for garbage-coded deaths in the GBD.

Conclusions

We provide a detailed description of redistribution methods developed for CoD data in the GBD; these methods represent an overall improvement in empiricism compared to past reliance on a priori knowledge.

SUBMITTER: Johnson SC

PROVIDER: S-EPMC8170729 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundCoverage and quality of cause-of-death (CoD) data varies across countries and time. Valid, reliable, and comparable assessments of trends in causes of death from even the best systems are limited by three problems: a) changes in the International Statistical Classification of Diseases and Related Health Problems (ICD) over time; b) the use of tabulation lists where substantial detail on causes of death is lost; and c) many deaths assigned to causes that cannot or should not be considered underlying causes of death, often called garbage codes (GCs). The Global Burden of Disease Study and the World Health Organization have developed various methods to enhance comparability of CoD data. In this study, we attempt to build on these approaches to enhance the utility of national cause-of-death data for public health analysis.MethodsBased on careful consideration of 4,434 country-years of CoD data from 145 countries from 1901 to 2008, encompassing 743 million deaths in ICD versions 1 to 10 as well as country-specific cause lists, we have developed a public health-oriented cause-of-death list. These 56 causes are organized hierarchically and encompass all deaths. Each cause has been mapped from ICD-6 to ICD-10 and, where possible, they have also been mapped to the International List of Causes of Death 1-5. We developed a typology of different classes of GCs. In each ICD revision, GCs have been identified. Target causes to which these GCs should be redistributed have been identified based on certification practice and/or pathophysiology. Proportionate redistribution, statistical models, and expert algorithms have been developed to redistribute GCs to target codes for each age-sex group.ResultsThe fraction of all deaths assigned to GCs varies tremendously across countries and revisions of the ICD. In general, across all country-years of data available, GCs have declined from more than 43% in ICD-7 to 24% in ICD-10. In some regions, such as Australasia, GCs in 2005 are as low as 11%, while in some developing countries, such as Thailand, they are greater than 50%. Across different age groups, the composition of GCs varies tremendously - three classes of GCs steadily increase with age, but ambiguous codes within a particular disease chapter are also common for injuries at younger ages. The impact of redistribution is to change the number of deaths assigned to particular causes for a given age-sex group. These changes alter ranks across countries for any given year by a number of different causes, change time trends, and alter the rank order of causes within a country.ConclusionsBy mapping CoD through different ICD versions and redistributing GCs, we believe the public health utility of CoD data can be substantially enhanced, leading to an increased demand for higher quality CoD data from health sector decision-makers.

Project description:Many states and local areas are affected by the national epidemic of drug-related mortality, which recently has shown signs of a rising "licit-to-illicit drug" death ratio. Appropriate local public health surveillance can help monitor and control this epidemic.Using our state as an example, we sought to illustrate how to describe the changes in drug death rates, causes, and circumstances. In contrast to most other surveillance reports, our approach includes both drug-induced and drug-related deaths as well as both demographic and socioeconomic characteristics of decedents.Cross-sectional study.All residents of the state of Wisconsin.Decedents from 1999 to 2008.Annual numbers and population-based rates of deaths due to drugs, including both identified and unidentified drugs. Information was obtained from death certificates with any of approximately 270 underlying, immediate, or contributing cause-of-death codes from the International Classification of Diseases, 10th Revision.Drug-related death rates increased during much of the 10-year study period, and the male-to-female death ratio rose. The median age at death from drug-related causes was 43 years. Opioid analgesic poisoning surpassed cocaine and heroin poisoning as the most frequent type of fatal drug poisoning. Of all 4828 deaths from drug-related causes--virtually all of which were certified by a county medical examiner or coroner--3410 (71%) were unintentional and 1053 (22%) were suicides. The unintentional-to-suicide death rate ratio grew from 1.6 to 3.5 during the study period. Methadone-related deaths increased from 10 in 1999 to 118 in 2008 (1080%), while benzodiazepine-related deaths rose from 23 to 106 (361%).Although premature deaths from drug use and abuse continue to rise, even surpassing deaths due to motor vehicle crash in some states, "multiple causes of death" information from death certificates is available to monitor their occurrence and nature and to inform selection of prevention strategies.

Project description:BackgroundThe Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach has been adopted by many national and international organisations as a systematic and transparent framework for evidence-based guideline development. With reference to an ongoing debate in the literature and within public health organisations, this study reviews current experience with the GRADE approach in rating the quality of evidence in the field of public health and identifies challenges encountered.MethodsWe conducted semi-structured interviews with individuals/groups that have applied the GRADE approach in the context of systematic reviews or guidelines in the field of public health, as well as with representatives of groups or organisations that actively decided against its use. We initially contacted potential participants by email. Responses were obtained by telephone interview or email, and written interview summaries were validated with participants. We analysed data across individual interviews to distil common themes and challenges.ResultsBased on 25 responses, we undertook 18 interviews and obtained 15 in-depth responses relating to specific systematic reviews or guideline projects; a majority of the latter were contributed by groups within the World Health Organization. All respondents that have used the GRADE approach appreciated the systematic and transparent process of assessing the quality of the evidence. However, respondents reported a range of minor and major challenges relating to complexity of public health interventions, choice of outcomes and outcome measures, ability to discriminate between different types of observational studies, use of non-epidemiological evidence, GRADE terminology and the GRADE and guideline development process. Respondents' suggestions to make the approach more applicable to public health interventions included revisiting terminology, offering better guidance on how to apply GRADE to complex interventions and making modifications to the current grading scheme.ConclusionsOur findings suggest that GRADE principles are applicable to public health and well-received but also highlight common challenges. They provide a starting point for exploring options for improvements and, where applicable, testing these across different types of public health interventions. Several public health organisations are currently testing GRADE, and the GRADE Working Group is eager to engage with these groups to find ways to address concerns.

Project description:BackgroundHyperbilirubinemia is emerging as an increasingly common problem in newborns due to a decreasing hospital length of stay after birth. Jaundice is the most common disease of the newborn and although being benign in most cases it can lead to severe neurological consequences if poorly evaluated. In different areas of medicine, data mining has contributed to improve the results obtained with other methodologies.Hence, the aim of this study was to improve the diagnosis of neonatal jaundice with the application of data mining techniques.MethodsThis study followed the different phases of the Cross Industry Standard Process for Data Mining model as its methodology.This observational study was performed at the Obstetrics Department of a central hospital (Centro Hospitalar Tâmega e Sousa--EPE), from February to March of 2011. A total of 227 healthy newborn infants with 35 or more weeks of gestation were enrolled in the study. Over 70 variables were collected and analyzed. Also, transcutaneous bilirubin levels were measured from birth to hospital discharge with maximum time intervals of 8 hours between measurements, using a noninvasive bilirubinometer.Different attribute subsets were used to train and test classification models using algorithms included in Weka data mining software, such as decision trees (J48) and neural networks (multilayer perceptron). The accuracy results were compared with the traditional methods for prediction of hyperbilirubinemia.ResultsThe application of different classification algorithms to the collected data allowed predicting subsequent hyperbilirubinemia with high accuracy. In particular, at 24 hours of life of newborns, the accuracy for the prediction of hyperbilirubinemia was 89%. The best results were obtained using the following algorithms: naive Bayes, multilayer perceptron and simple logistic.ConclusionsThe findings of our study sustain that, new approaches, such as data mining, may support medical decision, contributing to improve diagnosis in neonatal jaundice.

Project description:BackgroundBirth registration is a child's first right. Registration of live births, stillbirths and deaths is foundational for national planning. Completeness of birth registration for live births in low- and middle-income countries is measured through population-based surveys which do not currently include completeness of stillbirth or death registration.MethodsThe EN-INDEPTH population-based survey of women of reproductive age was undertaken in five Health and Demographic Surveillance System sites in Bangladesh, Ethiopia, Ghana, Guinea-Bissau and Uganda (2017-2018). In four African sites, we included new/modified questions regarding registration for 1177 stillbirths and 11,881 livebirths (1333 neonatal deaths and 10,548 surviving the neonatal period). Questions were evaluated for completeness of responses, data quality, time to administer and estimates of registration completeness using descriptive statistics. Timing of birth registration, factors associated with non-registration and reported barriers were assessed using descriptive statistics and logistic regression.ResultsAlmost all women, irrespective of their baby's survival, responded to registration questions, taking an average of < 1 min. Reported completeness of birth registration was 30.7% (6.1-53.5%) for babies surviving the neonatal period, compared to 1.7% for neonatal deaths (0.4-5.7%). Women were able to report age at birth registration for 93.6% of babies. Non-registration of babies surviving the neonatal period was significantly higher for home-born children (aOR 1.43 (95% CI 1.27-1.60)) and in Dabat (Ethiopia) (aOR 4.11 (95% CI 3.37-5.01)). Other socio-demographic factors associated with non-registration included younger age of mother, more prior births, little or no education, and lower socio-economic status. Neonatal death registration questions were feasible (100% women responded; only 1% did not know), revealing extremely low completeness with only 1.2% of neonatal deaths reported as registered. Despite > 70% of stillbirths occurring in facilities, only 2.5% were reported as registered.ConclusionsQuestions on birth, stillbirth and death registration were feasible in a household survey. Completeness of birth registration is low in all four sites, but stillbirth and neonatal death registration was very low. Closing the registration gap amongst facility births could increase registration of both livebirths and facility deaths, including stillbirths, but will require co-ordination between civil registration systems and the often over-stretched health sector. Investment and innovation is required to capture birth and especially deaths in both facility and community systems.