Dataset Information

Estimating a Set of Mortality Risk Functions with Multiple Contributing Causes of Death.

ABSTRACT:

Background

There are few methodologic examples of how multiple causes of death may be summarized in cause-specific mortality analyses to address limitations of attributing death to a single underlying cause. We propose a cause-of-death weighting approach to estimate the set of risk functions of specific causes of mortality using both underlying and contributing cause-of-death information.

Methods

We constructed weights according to a user-specified function. Using data from four southern US human immunodeficiency virus (HIV) clinics, we constructed a cause of death-weighted Aalen-Johansen estimator of the cumulative incidence function to estimate risks of five specific causes of mortality in the full sample and by injection drug use history.

Results

Among 7740 HIV-positive patients initiating antiretroviral therapy between 1999 and 2014, the 8-year risk of all-cause mortality was 17.5% [95% confidence interval (CI) = 16.5, 18.4]. The cause of death-weighted risk of HIV-related mortality was 6.7% (95% CI = 6.0, 7.3) and accounted for 39% (95% CI = 35, 42) of total mortality risk. This compared with 10.2% (95% CI = 9.2, 11.2) using only the underlying cause, in which case HIV-related deaths accounted for nearly 60% of total mortality risk. The proportion attributable to cardiovascular disease among those whose HIV risk factor was injection drug use was twice as high using cause-of-death weights compared with only the underlying cause (8%; 95% CI = 5, 11 vs 4%; 95% CI = 1, 6).

Conclusion

Using cause of death-weighted estimators to incorporate multiple causes of death may yield different conclusions regarding the importance of certain causes of mortality. See video abstract: http://links.lww.com/EDE/B706.

SUBMITTER: Breger TL

PROVIDER: S-EPMC7822729 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundAll-cause mortality risk and causes of death in bronchiectasis patients have not been fully investigated. The aim of this study was to compare the mortality risk and causes of death between individuals with bronchiectasis and those without bronchiectasis.MethodsPatients with or without bronchiectasis determined based on chest computed tomography (CT) at one centre between 2005 and 2016 were enrolled. Among the patients without bronchiectasis, a control group was selected after applying additional exclusion criteria. We compared the mortality risk and causes of death between the bronchiectasis and control groups without lung disease. Subgroup analyses were also performed according to identification of Pseudomonas or non-tuberculous mycobacteria, airflow limitation, and smoking status.ResultsOf the total 217,702 patients who underwent chest CT, 18,134 bronchiectasis patients and 90,313 non-bronchiectasis patients were included. The all-cause mortality rate in the bronchiectasis group was 1608.8 per 100,000 person-years (95% confidence interval (CI), 1531.5-1690.0), which was higher than that in the control group (133.5 per 100,000 person-years; 95% CI, 124.1-143.8; P?<?0.001). The bronchiectasis group had higher all-cause (adjusted hazard ratio (aHR), 1.26; 95% CI, 1.09-1.47), respiratory (aHR, 3.49; 95% CI, 2.21-5.51), and lung cancer-related (aHR, 3.48; 95% CI, 2.33-5.22) mortality risks than the control group. In subgroup analysis, patients with airflow limitation and ever smokers showed higher all-cause mortality risk among bronchiectasis patients. Therefore, we observed significant interrelation between bronchiectasis and smoking, concerning the risks of all-cause mortality (P for multiplicative interaction, 0.030, RERI, 0.432; 95% CI, 0.097-0.769) and lung cancer-related mortality (RERI, 8.68; 95% CI, 1.631-15.736).ConclusionIndividuals with bronchiectasis had a higher risk of all-cause, respiratory, and lung cancer-related mortality compared to control group. The risk of all-cause mortality was more prominent in those with airflow limitation and in ever smokers.

Project description:BackgroundPapua New Guinea (PNG) is a diverse country with high mortality and evidence of increased prevalence of non-communicable diseases (NCDs), but there is no reliable cause of death (COD) data because civil registration is insufficient and routine health data comprise only a small proportion of deaths. This study aims to estimate cause-specific mortality fractions (CSMFs) for five broad groups of causes (endemic infections, emerging infections, endemic NCDs, emerging NCDs and injuries), by sex for each of PNG's provinces.MethodsCSMFs are calculated as the average of estimates obtained from: (1) Empirical cause method: Utilising available Verbal Autopsy (VA) data and Discharge Health Information System (DHIS) data, and applying statistical models of community versus facility CODs; and (2) Expected cause patterns method: Utilising existing estimates of mortality levels in each province and statistical models of the relationship between all-cause and cause-specific mortality using Global Burden of Disease (GBD) data.ResultsAn estimated 41% of male and 49% of female deaths in PNG are due to infectious, maternal (female only), neonatal and nutritional causes. Furthermore, 45% of male and 42% of female deaths arise from NCDs. Infectious diseases, maternal, neonatal and nutritional conditions account for more than half the deaths in a number of provinces, including lower socioeconomic status provinces of Gulf and Sandaun, while provinces with higher CSMFs from emerging NCDs (e.g. ischemic heart disease, stroke) tend to be those where socioeconomic status is comparatively high (e.g. National Capital District, Western Highlands Province, Manus Province, New Ireland Province and East New Britain Province). Provinces with the highest estimated proportion of deaths from emerging infectious diseases are readily accessible by road and have the highest rates of sexually transmitted infections (STIs), while provinces with the highest CSMFs from endemic infectious, maternal, neonatal and nutritional causes are geographically isolated, have high malaria and high all-cause mortality.ConclusionsInfectious, maternal, neonatal and nutritional causes continue to be an important COD in PNG, and are likely to be higher than what is estimated by the GBD. Nonetheless, there is evidence of the emergence of NCDs in provinces with higher socioeconomic status. The introduction of routine VA for non-facility deaths should improve COD data quality to support health policy and planning to control both infectious and NCDs.

Project description:Background and aimsA substantial share of fatal drug overdoses is missing information on specific drug involvement, leading to under-reporting of opioid-related death rates and a misrepresentation of the extent of the opioid epidemic. We aimed to compare methodological approaches to predicting opioid involvement in unclassified drug overdoses in US death records and to estimate the number of fatal opioid overdoses from 1999 to 2016 using the best-performing method.DesignThis was a secondary data analysis of the universe of drug overdoses in 1999-2016 obtained from the National Center for Health Statistics Detailed Multiple Cause of Death records.SettingUnited States.CasesA total of 632 331 drug overdose decedents. Drug overdoses with known drug classification comprised 78.2% of the cases (n = 494 316) and unclassified drug overdoses (ICD-10 T50.9) comprised 21.8% (n = 138 015).MeasurementsKnown opioid involvement was defined using ICD-10 codes T40.0-40.4 and T40.6, recorded in the set of contributing causes. Opioid involvement in unclassified drug overdoses was predicted using multiple methodological approaches: logistic regression and machine learning techniques, inclusion/exclusion of contributing causes of death and inclusion/exclusion of county-level characteristics. Having selected the model with the highest predictive ability, we calculated corrected estimates of opioid-related mortality.FindingsLogistic regression and random forest models performed similarly. Including contributing causes substantially improved predictive accuracy, while including county characteristics did not. Using a superior prediction model, we found that 71.8% of unclassified drug overdoses in 1999-2016 involved opioids, translating into 99 160 additional opioid-related deaths, or approximately 28% more than reported. Importantly, there was a striking geographic variation in undercounting of opioid overdoses.ConclusionsIn modeling opioid involvement in unclassified drug overdoses, highest predictive accuracy is achieved using a statistical model-either logistic regression or a random forest ensemble-with decedent characteristics and contributing causes of death as predictors.

Project description:Background: Neurological impairment (NI) and disability are associated with reduced life expectancy, but the risk and magnitude of premature mortality in children vary considerably across study settings. We conducted a systematic review to estimate the magnitude of premature mortality following childhood-onset NI worldwide and to summarize known risk factors and causes of death. Methods: We searched various databases for published studies from their inception up to 31st October 2020. We included all cohort studies that assessed the overall risk of mortality in individuals with childhood-onset epilepsy, intellectual disability (ID), and deficits in hearing, vision and motor functions. Comparative measures of mortality such as the standardized mortality ratio (SMR), risk factors and causes were synthesized quantitatively under each domain of impairment. This review is registered on the PROSPERO database (registration number CRD42019119239). Results: The search identified 2,159 studies, of which 24 studies were included in the final synthesis. Twenty-two (91.7%) studies originated from high-income countries (HICs). The median SMR was higher for epilepsy compared with ID (7.1 [range 3.1-22.4] vs. 2.9 [range 2.0-11.6]). In epilepsy, mortality was highest among younger age groups, comorbid neurological disorders, generalized seizures (at univariable levels), untreatable epilepsy, soon after diagnosis and among cases with structural/metabolic types, but there were no differences by sex. Most deaths (87.5%) were caused by non-epilepsy-related causes. For ID, mortality was highest in younger age groups and girls had a higher risk compared to the general population. Important risk factors for premature mortality were severe-to-profound severity, congenital disorders e.g., Down Syndrome, comorbid neurological disorders and adverse pregnancy and perinatal events. Respiratory infections and comorbid neurological disorders were the leading causes of death in ID. Mortality is infrequently examined in impairments of vision, hearing and motor functions. Summary: The risk of premature mortality is elevated in individuals with childhood-onset NI, particularly in epilepsy and lower in ID, with a need for more studies for vision, hearing, and motor impairments. Survival in NI could be improved through interventions targeting modifiable risk factors and underlying causes.