Heart failure describing the underlying cause of death: a misconception, lack of information on the true underlying causes, or both?

AIM

The underlying cause of death represents the most important information on death certificates. Often, conditions that cannot represent a true underlying cause of death are listed as such. This phenomenon affects the quality of vital statistics and results of studies using cause-specific mortality as endpoints. We aimed at exploring the magnitude and factors associated with the use of heart failure to describe the underlying cause of death.

METHODS

In this cross-sectional, register based study we linked data from the Norwegian Cause of Death Registry and the Norwegian Patient Registry. We used logistic regression models to analyse the association between external factors and heart failure listed as the underlying cause of death.

RESULTS

Heart failure was listed as the underlying cause of death in 3.6% of all deaths. The odds of heart failure increased: (a) by 35% for 5-year increment in age; (b) by 78% for deaths occurring at nursing homes (compared with in-hospital deaths); and (c) by 602% for deaths not followed by an autopsy (compared with those followed by an autopsy). Deceased with a previous hospitalisation with heart failure as the discharge diagnosis had 514% higher odds of having heart failure listed as their underlying cause of death. Of the deceased with heart failure listed as the underlying cause of death, 9.4% did not have any, and 69.2% had only irrelevant additional information for assessing the true underlying cause of death in their death certificates.

CONCLUSIONS

Heart failure listed as the underlying cause of death was associated with age, place of death, autopsy and previous hospitalisations - all factors that should not influence coding procedures. Better completion of death certificates in accordance with the World Health Organization rules will help reduce the use of heart failure to describe the underlying cause of death.

Mortality and causes of death in systemic lupus erythematosus in New Zealand: A population-based study

OBJECTIVES

This study aims to assess the mortality of systemic lupus erythematosus (SLE) patients and examine whether there are variations by subgroup.

METHODS

SLE patients from 2005-2021 were identified from the national administrative datasets. The underlying causes of death were examined. Standardised mortality ratio (SMR) was estimated to compare the relative rate of observed deaths in SLE patients to expected deaths in the general population. The hazard ratios (HR) and 95% confidence intervals (CI) of all-cause mortality and SLE specific mortality by ethnicity were estimated after adjustment for age using a Cox proportional hazards model.

RESULTS

Of the 2,802 patients included for analysis, 699 (24.9%) died with 209 (29.9%) SLE deaths. The age-standardised mortality rate of SLE was 0.29 per 100 000 for women and 0.05 for men. The mean age at death was 65.3 ± 17.1 years. Younger patients were more likely to have SLE as the underlying cause of death, from 78.9% for those under 20 years old to 18.7% for those aged 70-79 years. Compared with the general population, SLE patients were four times more likely to die (SMR: 4.0; 95% CI: 3.7-4.3). Young patients had higher SMRs than older patients. Māori had worse all-cause mortality (HR: 1.72; 95% CI: 1.10-2.67) and SLE specific mortality (HR: 2.60; 95% CI: 1.29-5.24) than others.

CONCLUSIONS

The outcomes of SLE in New Zealand were still very poor compared with the general population. Māori with SLE had worse survival than others. Further research is needed to identify the reasons to this disparity.

Survival of infants and children born with severe microcephaly, Texas, 1999-2015

BACKGROUND

Severe microcephaly is a brain reduction defect where the delivery head circumference is <3rd percentile for gestational age and sex with subsequent lifelong morbidities. Our objective was to evaluate survival among 2,704 Texas infants with severe microcephaly delivered 1999-2015.

METHODS

Infants with severe microcephaly from the Texas Birth Defects Registry were linked to death certificates and the national death index. Survival estimates, hazard ratios (HR) and confidence intervals (CI) were calculated using the Kaplan-Meier method and Cox proportional hazards models stratified by presence versus absence of co-occurring defects.

RESULTS

We identified 496 deaths by age 4 years; most (42.9%) occurred in the neonatal period, and another 39.9% died by 1 year of age. Overall infant survival was 84.8%. Lowest infant survival subgroups included those with chromosomal/syndromic conditions (66.1%), very preterm deliveries (63.9%), or co-occurring critical congenital heart defects (44.0%). Among infants with severe microcephaly and a chromosomal/syndromic co-occurring defect, the risk of death was nearly three-fold higher among those with: proportionate microcephaly (i.e., small baby overall), relative to non-proportionate (HR = 2.84, 95% CI = 2.17-3.71); low-birthweight relative to normal (HR = 2.72, 95% CI = 1.92-3.85); critical congenital heart defects (CCHD) relative to no CCHD (HR = 2.90, 95% CI = 2.20-3.80). Trisomies were a leading underlying cause of death (27.5%).

CONCLUSIONS

Overall, infants with severe microcephaly had high 4-year survival rates which varied by the presence of co-occurring defects. Infants with co-occurring chromosomal/syndromic anomalies have a higher risk of death by age one than those without any co-occurring birth defects.

Secular Trends in Systemic Sclerosis Mortality in the United States from 1981 to 2020

BACKGROUND

Systemic sclerosis (SSc) has the highest mortality rate among autoimmune disorders. Individuals with SSc frequently die from complications or infections related to SSc. Nonetheless, the sex-age-period interaction of SSc is complex and remains unclear. The study aims to analyze the secular trend of SSc mortality based on data regarding underlying cause of death (UCD) and multiple causes of death (MCD) and clarify the sex-age interaction with time.

METHODS

The multiple-cause mortality statistics provided by the National Center for Health Statistics were used to identify all deaths in the United States from 1981 to 2020 in which SSc was indicated anywhere on the death certificates. The age-standardized mortality rate (ASMR) was determined for both sexes, as well as the variations in these rates. Joinpoint regression analysis was utilized to determine the annual percentage change (APC) of ASMR.

RESULTS

A total of 44,672 and 66,259 individuals who died between 1981 and 2020 were identified based on the UCD and MCD data, respectively. According to the UCD data, SSc-related AMSR (SSc-ASMR) of the male and female decedents, respectively, declined from 5.01 and 1.94 in 1981-1990 to 4.77 and 1.32 in 2011-2020, respectively (mortality rate ratio 0.95, 95% confidence interval 0.92-0.98). From 1986 to 1999, the APC of SSc-ASMR in female decedents decreased except for those aged 45-64 years (APC 2.1%, p = 0.002). For MCD analysis, in trend 1, only APC of SSc-ASMR in male decedents aged 45-64 years decreased. The SSc-ASMR of both male and female decedents fell on trend 2 arm. In 2011-2020, the ratio of UCD to MCD increased across all age groups for both sexes compared to 1981-1990. Overall, compared to the male decedents, the SSc-ASMR in female decedents increased significantly before 1999, peaked in 1999, followed by continuous decrease until 2020 according to UCD and MCD statistics.

CONCLUSIONS

Over the past four decades, the SSc deaths based on the MCD data were 1.48 times more than the UCD data, and the proportion of UCD over MCD increased over time. The SSc-ASMRs in all the sex-age groups significantly decreased over the past two decades. Notably, the mortality rate ratio of women to men with SSc increased in the past four decades.

Underlying cause of death at medical facilities in Xaiyabouli Province, Lao People's Democratic Republic

In Lao PDR, Ministry of Health introduced District Health Information Software 2 (DHIS2) in 2013. Although DHIS2 includes cause of death, Lao government did not request to report cause of death through DHIS2, resulting in no information on frequency of underlying cause of death even for the deaths at medical facility. This study aimed to collect the information on the underlying cause of death at all medical facilities in Xaiyabouli province, a rural area in Lao PDR. As well as the point estimate of the proportion, a 95% confidence interval (CI) based on a binomial distribution was calculated for each cause of death. According to the local government request, 226 deaths (128 males and 98 females) in 2019 were reported from all medical facilities in the province. Among them, infectious diseases were the most frequent (33.6%, 95% CI 27.5-40.2%); sepsis (16.8%, 95% CI 12.2-22.3%), pneumonia (8.8%, 95% CI 5.5-13.3%), and meningitis (4.9%, 95% CI 2.5-8.5%). Heart diseases were 15.9% (95% CI 11.4-21.4%) including heart failure and myocardial infarction. Injury was 10.2% (95% CI 6.2-14.4%) including brain injury. Neonatal death was 10.6% (95% CI 6.9-15.4%). Among those, preterm death was common (8.8%, 95% CI 5.3-13.3%). Renal failure was 8.0% (95% CI 4.8-12.3%). According to civil registration covering all deaths both at facilities and outside facilities, deaths at facilities were 16.8% of the whole deaths (1,372 deaths) in Xaiyabouli province. Although deaths outside facilities were not included, this is the first report demonstrating cause of death in one province in Lao PDR.

Population-Based Study on the All-Cause and Cause-Specific Risks of Mortality among Long-Term Opioid Analgesics Users without Cancer in Taiwan

(1) Background: The prevalence of opioid use in Taiwan increased by 41% between 2002 and 2014. However, little is known regarding the risk of mortality among long-term opioid analgesics users who do not have cancer. This study investigated this mortality risk with an emphasis on the calendar year and patients’ age and sex. (2) Methods: This retrospective cohort study included 12,990 adult individuals without cancer who were long-term users of opioid analgesics and were randomly selected from the data set of Taiwan’s National Health Insurance program from 2000 to 2012. They were then followed up through 2013. Information on the underlying causes of death was retrieved from the Taiwan Death Registry. Age, sex, and calendar year-standardized mortality ratios (SMRs) of all-cause and cause-specific mortality were calculated with reference to those of the general population. (3) Results: With up to 14 years of follow-up, 558 individuals had all-cause mortality in 48,020 person-years (cumulative mortality: 4.3%, mortality rate: 11.62 per 1000 person-years). Compared with the general population, the all-cause SMR of 4.30 (95% confidence interval (95% CI): 3.95–4.66) was significantly higher: it was higher in men than in women, declined with calendar year and age, and was significantly higher for both natural (4.15, 95% CI: 3.78–4.53) and unnatural (5.04, 95% CI: 3.88–6.45) causes. (4) Conclusions: Long-term opioid analgesics use among individuals without cancer in Taiwan was associated with a significantly increased risk of mortality. The notably increased mortality in younger adults warrants attention. Strategies to reduce long-term opioid analgesics use, especially their overuse or misuse, are in an urgent need.

Overall and Cause-Specific Mortality in Patients With Type 1 Diabetes Mellitus: A Population-Based Cohort Study in Taiwan From 1998 Through 2014

Background

To investigate all-cause and cause-specific mortality in Taiwanese patients with type 1 diabetes.

Methods

A cohort of 17,203 patients with type 1 diabetes were identified from Taiwan’s National Health Insurance claims in the period of 1998–2014. Person-years were accumulated for each individual from date of type 1 diabetes registration to date of death or the last day of 2014. Age, sex, and calendar year standardized mortality ratios (SMRs) were calculated with reference to the general population.

Results

In up to 17 years of follow-up, 4,916 patients died from 182,523 person-years. Diabetes (30.15%), cancer (20.48%), circulatory diseases (13.14%), and renal diseases (11.45%) were the leading underlying causes of death. Mortality rate (26.93 per 1,000 person-years) from type 1 diabetes in Taiwan was high, the cause of death with the highest mortality rate was diabetes (8.12 per 1,000 person-years), followed by cancer (5.52 per 1,000 person-years), and circulatory diseases (3.54 per 1,000 person-years). The all-cause SMR was significantly elevated at 4.16 (95% confidence interval, 4.04–4.28), with a greater all-cause SMR noted in females than in males (4.62 vs 3.79). The cause-specific SMR was highly elevated for diabetes (SMR, 16.45), followed by renal disease (SMR, 14.48), chronic hepatitis and liver cirrhosis (SMR, 4.91) and infection (SMR, 4.59). All-cause SMRs were also significantly increased for all ages, with the greatest figure noted for 15–24 years (SMR, 8.46).

Conclusions

Type 1 diabetes in both genders and all ages was associated with significantly elevated SMRs for all-cause and mostly for diabetes per se and renal disease.

Mortality Associated with Idiopathic Pulmonary Fibrosis in Northeastern Italy, 2008-2020: A Multiple Cause of Death Analysis

Mortality from idiopathic pulmonary fibrosis (IPF) is increasing in most European countries, but there are no data for Italy. We analysed the registry data from a region in northeastern Italy to assess the trends in IPF-related mortality during 2008–2019, to compare results of underlying vs. multiple cause of death analyses, and to describe the impact of the COVID-19 epidemic in 2020. We identified IPF (ICD-10 code J84.1) among the causes of death registered in 557,932 certificates in the Veneto region. We assessed time trends in annual age-standardized mortality rates by gender and age (40–74, 75–84, and ≥85 years). IPF was the underlying cause of 1310 deaths in the 2251 certificates mentioning IPF. For all age groups combined, the age-standardized mortality rate from IPF identified as the underlying cause of death was close to the European median (males and females: 3.1 and 1.3 per 100,000/year, respectively). During 2008–2019, mortality rates increased in men aged ≥85 years (annual percent change of 6.5%, 95% CI: 2.0, 11.2%), but not among women or for the younger age groups. A 72% excess of IPF-related deaths was registered in March–April 2020 (mortality ratio 1.72, 95% CI: 1.29, 2.24). IPF mortality was increasing among older men in northeastern Italy. The burden of IPF was heavier than assessed by routine statistics, since less than two out of three IPF-related deaths were directly attributed to this condition. COVID-19 was accompanied by a remarkable increase in IPF-related mortality.

Animal-Encounter Fatalities, United States, 1999-2016: Cause of Death and Misreporting

OBJECTIVES

Errors and misreporting on death certificates are common, along with potential inaccuracies in cause-of-death coding. We characterized and compared fatalities by animal-encounter mentions reported as underlying cause of death (UCD) with animal-encounter mentions reported as multiple cause of death (MCD) to determine factors associated with misreporting UCD.

METHODS

We analyzed fatality data from 1999-2016 from the Centers for Disease Control and Prevention Wide-ranging ONline Data for Epidemiologic Research by UCD and MCD animal-encounter mentions (International Classification of Diseases, 10th Revision codes W53-59, X20-27 and X29, T63.0-63.6, T63.8-63.9, and T78.2-78.4). We examined differences in reporting by age, sex, race, autopsy (yes, no, unknown), allergic reactions, and toxicities.

RESULTS

The number of animal-encounter mentions by UCD was 3638 (202 average per year) and by MCD was 4280 (238 average per year), a difference of 18% (n = 642; 36 average per year) by MCD analysis. The number of nonvenomous animal-encounter mentions increased 20% (from 2138 UCD to 2567 MCD), and the number of venomous animal-encounter mentions increased 14% (from 1500 UCD to 1713 MCD). Decedents aged ≥65 had the highest additional number of animal-encounter mentions among all age groups, primarily encounters with other reptiles (n = 113), other mammals (n = 71), and dogs (n = 42). Of 642 MCD additional animal-encounter mentions, heart disease (n = 211, 33%) and infections (n = 146, 23%) represented more than half of the UCD. Of 553 dog-encounter fatalities, 165 (30%) were among children aged ≤4.

CONCLUSIONS

Animal-encounter fatalities, analyzed by UCD alone, may be underreported. An initiating animal injury, complicated by comorbidities and fatality, may obscure the causal chain, resulting in misreporting UCD. Ongoing training for medical certifiers is recommended, highlighting accurate identification of UCD and contributing causes in the causal chain of death.

Unraveling Specific Causes of Neonatal Mortality Using Minimally Invasive Tissue Sampling: An Observational Study

BACKGROUND

Postmortem minimally invasive tissue sampling (MITS) is a potential alternative to the gold standard complete diagnostic autopsy for identifying specific causes of childhood deaths. We investigated the utility of MITS, interpreted with available clinical data, for attributing underlying and immediate causes of neonatal deaths.

METHODS

This prospective, observational pilot study enrolled neonatal deaths at Chris Hani Baragwanath Academic Hospital in Soweto, South Africa. The MITS included needle core-biopsy sampling for histopathology of brain, lung, and liver tissue. Microbiological culture and/or molecular tests were performed on lung, liver, blood, cerebrospinal fluid, and stool samples. The "underlying" and "immediate" causes of death (CoD) were determined for each case by an international panel of 12-15 medical specialists.

RESULTS

We enrolled 153 neonatal deaths, 106 aged 3-28 days. Leading underlying CoD included "complications of prematurity" (52.9%), "complications of intrapartum events" (15.0%), "congenital malformations" (13.1%), and "infection related" (9.8%). Overall, infections were the immediate or underlying CoD in 57.5% (n = 88) of all neonatal deaths, including the immediate CoD in 70.4% (58/81) of neonates with "complications of prematurity" as the underlying cause. Overall, 74.4% of 90 infection-related deaths were hospital acquired, mainly due to multidrug-resistant Acinetobacter baumannii (52.2%), Klebsiella pneumoniae (22.4%), and Staphylococcus aureus (20.9%). Streptococcus agalactiae was the most common pathogen (5/15 [33.3%]) among deaths with "infections" as the underlying cause.

CONCLUSIONS

MITS has potential to address the knowledge gap on specific causes of neonatal mortality. In our setting, this included the hitherto underrecognized dominant role of hospital-acquired multidrug-resistant bacterial infections as the leading immediate cause of neonatal deaths.

A Deep Artificial Neural Network-Based Model for Prediction of Underlying Cause of Death From Death Certificates: Algorithm Development and Validation

Background

Coding of underlying causes of death from death certificates is a process that is nowadays undertaken mostly by humans with potential assistance from expert systems, such as the Iris software. It is, consequently, an expensive process that can, in addition, suffer from geospatial discrepancies, thus severely impairing the comparability of death statistics at the international level. The recent advances in artificial intelligence, specifically the rise of deep learning methods, has enabled computers to make efficient decisions on a number of complex problems that were typically considered out of reach without human assistance; they require a considerable amount of data to learn from, which is typically their main limiting factor. However, the CépiDc (Centre d’épidémiologie sur les causes médicales de Décès) stores an exhaustive database of death certificates at the French national scale, amounting to several millions of training examples available for the machine learning practitioner.

Objective

This article investigates the application of deep neural network methods to coding underlying causes of death.

Methods

The investigated dataset was based on data contained from every French death certificate from 2000 to 2015, containing information such as the subject’s age and gender, as well as the chain of events leading to his or her death, for a total of around 8 million observations. The task of automatically coding the subject’s underlying cause of death was then formulated as a predictive modelling problem. A deep neural network−based model was then designed and fit to the dataset. Its error rate was then assessed on an exterior test dataset and compared to the current state-of-the-art (ie, the Iris software). Statistical significance of the proposed approach’s superiority was assessed via bootstrap.

Results

The proposed approach resulted in a test accuracy of 97.8% (95% CI 97.7-97.9), which constitutes a significant improvement over the current state-of-the-art and its accuracy of 74.5% (95% CI 74.0-75.0) assessed on the same test example. Such an improvement opens up a whole field of new applications, from nosologist-level batch-automated coding to international and temporal harmonization of cause of death statistics. A typical example of such an application is demonstrated by recoding French overdose-related deaths from 2000 to 2010.

Conclusions

This article shows that deep artificial neural networks are perfectly suited to the analysis of electronic health records and can learn a complex set of medical rules directly from voluminous datasets, without any explicit prior knowledge. Although not entirely free from mistakes, the derived algorithm constitutes a powerful decision-making tool that is able to handle structured medical data with an unprecedented performance. We strongly believe that the methods developed in this article are highly reusable in a variety of settings related to epidemiology, biostatistics, and the medical sciences in general.

Obscuring effect of coding developmental disability as the underlying cause of death on mortality trends for adults with developmental disability: a cross-sectional study using US Mortality Data from 2012 to 2016

OBJECTIVE

To determine whether coding a developmental disability as the underlying cause of death obscures mortality trends of adults with developmental disability.

DESIGN

National Vital Statistics System 2012-2016 US Multiple Cause-of-Death Mortality files.

SETTING

USA.

PARTICIPANTS

Adults with a developmental disability indicated on their death certificate aged 18 through 103 at the time of death. The study population included 33 154 adults who died between 1 January 2012 and 31 December 2016.

PRIMARY OUTCOME AND MEASURES

Decedents with a developmental disability coded as the underlying cause of death on the death certificate were identified using the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision code for intellectual disability, cerebral palsy, Down syndrome or other developmental disability. Death certificates that coded a developmental disability as the underlying cause of death were revised using a sequential underlying cause of death revision process.

RESULTS

There were 33 154 decedents with developmental disability: 7901 with intellectual disability, 11 895 with cerebral palsy, 9114 with Down syndrome, 2479 with other developmental disabilities and 1765 with multiple developmental disabilities. Among all decedents, 48.5% had a developmental disability coded as the underlying cause of death, obscuring higher rates of choking deaths among all decedents and dementia and Alzheimer's disease among decedents with Down syndrome.

CONCLUSION

Death certificates that recorded the developmental disability in Part I of the death certificate were more likely to code disability as the underlying cause of death. While revising these death certificates provides a short-term corrective to mortality trends for this population, the severity and extent of this problem warrants a long-term change involving more precise instructions to record developmental disabilities only in Part II of the death certificate.

Use of Verbal Autopsy to Determine Underlying Cause of Death during Treatment of Multidrug-Resistant Tuberculosis, India

Of patients with multidrug-resistant tuberculosis (MDR TB), <50% complete treatment. Most treatment failures for patients with MDR TB are due to death during TB treatment. We sought to determine the proportion of deaths during MDR TB treatment attributable to TB itself. We used a structured verbal autopsy tool to interview family members of patients who died during MDR TB treatment in India during January-December 2016. A committee triangulated information from verbal autopsy, death certificate, or other medical records available with the family members to ascertain the underlying cause of death. For 66% of patient deaths (47/71), TB was the underlying cause of death. We assigned TB as the underlying cause of death for an additional 6 patients who died of suicide and 2 of pulmonary embolism. Deaths during TB treatment signify program failure; accurately determining the cause of death is the first step to designing appropriate, timely interventions to prevent premature deaths.

Underlying Cause of Death Recorded during 2013 to 2015 at a Tertiary General Hospital in Vientiane Capital, Lao PDR

In Lao People’s Democratic Republic (Lao PDR), the cause of death is not registered in death reports. As a result, the government cannot produce official reports that show mortality according to cause of death. This study aimed to report the underlying cause of death in a tertiary general hospital (Mittaphab Hospital) in Vientiane capital. Mittaphab hospital is a governmental teaching hospital with 300 beds for inpatient services specialized in orthopedics, neurology, and hemodialysis. Since a children hospital exists beside Mittaphab Hospital, severe pediatric cases are referred to the child hospital. HIV-positive cases and sputum positive tuberculosis are also transferred to the other specialized hospitals. All of the subjects in this study were patients who died in 2013–2015 at Mittaphab Hospital. Paper-based medical charts were examined by a medical doctor and staff from the medical records division. This chart review revealed that 1,509 patients (1,006 males and 503 females) died in this hospital during the study period. Of those, the number of patients aged <20 years and >80 years was small (6.2% and 7.7%, respectively). The most common underlying causes were injury (29.7%), cerebrovascular diseases (26.8%), renal disease (13.3%), infectious diseases (12.4%), and malignant neoplasm including brain tumor (4.8%). Among those aged 20–59 years, these percentages were 37.9%, 23.7%, 12.3%, 10.2%, and 5.0%, respectively. Although the patients visiting the hospital did not represent the whole population, the distribution of cause of death in the hospital was the only available information reported in Lao PDR.

Do Death Certificates Underestimate the Burden of Rare Diseases? The Example of Systemic Lupus Erythematosus Mortality, Sweden, 2001-2013

OBJECTIVES

Mortality due to rare diseases, which are substantial sources of premature mortality, is underreported in mortality studies. The objective of this study was to determine the completeness of reporting systemic lupus erythematosus (SLE) as a cause of death.

METHODS

In 2017, we linked data on a Swedish population-based cohort (the Swedish Lupus Linkage, 2001-2013) comprising people with SLE (n = 8560) and their matched general population comparators (n = 37 717) to data from the Cause of Death Register. We reviewed death records of deceased people from the cohort (n = 5110) and extracted data on patient demographic characteristics and causes of death. We estimated odds ratios (ORs) and 95% confidence intervals (CIs) for not reporting SLE as a cause of death by using multivariable-adjusted logistic regression models.

RESULTS

Of 1802 deaths among SLE patients in the study, 1071 (59%) did not have SLE reported on their death records. Most SLE decedents were aged 75-84 at death (n = 584, 32%), female (n = 1462, 81%), and born in Nordic countries (n = 1730, 96%). Decedents aged ≥85 at death were more likely to have SLE not reported on their death records than were decedents aged <50 (OR = 2.34; 95% CI, 1.48-3.68). Having renal failure listed as a cause of death decreased the likelihood of SLE not being reported on the death record (OR = 0.54; 95% CI, 0.40-0.73), whereas having cancer listed as a cause of death increased this likelihood (OR = 2.39; 95% CI, 1.85-3.07).

CONCLUSIONS

SLE was greatly underreported as a cause of mortality on death records of SLE patients, particularly in older decedents and those with cancer, thereby underestimating the true burden of this disease. Public health resources need to focus on improving the recording of rare diseases in order to enhance the epidemiological utility of mortality data.

Do Death Certificates Underestimate the Burden of Rare Diseases? The Example of Systemic Lupus Erythematosus Mortality, Sweden, 2001-2013

OBJECTIVES

Mortality due to rare diseases, which are substantial sources of premature mortality, is underreported in mortality studies. The objective of this study was to determine the completeness of reporting systemic lupus erythematosus (SLE) as a cause of death.

METHODS

In 2017, we linked data on a Swedish population-based cohort (the Swedish Lupus Linkage, 2001-2013) comprising people with SLE (n = 8560) and their matched general population comparators (n = 37 717) to data from the Cause of Death Register. We reviewed death records of deceased people from the cohort (n = 5110) and extracted data on patient demographic characteristics and causes of death. We estimated odds ratios (ORs) and 95% confidence intervals (CIs) for not reporting SLE as a cause of death by using multivariable-adjusted logistic regression models.

RESULTS

Of 1802 deaths among SLE patients in the study, 1071 (59%) did not have SLE reported on their death records. Most SLE decedents were aged 75-84 at death (n = 584, 32%), female (n = 1462, 81%), and born in Nordic countries (n = 1730, 96%). Decedents aged ≥85 at death were more likely to have SLE not reported on their death records than were decedents aged <50 (OR = 2.34; 95% CI, 1.48-3.68). Having renal failure listed as a cause of death decreased the likelihood of SLE not being reported on the death record (OR = 0.54; 95% CI, 0.40-0.73), whereas having cancer listed as a cause of death increased this likelihood (OR = 2.39; 95% CI, 1.85-3.07).

CONCLUSIONS

SLE was greatly underreported as a cause of mortality on death records of SLE patients, particularly in older decedents and those with cancer, thereby underestimating the true burden of this disease. Public health resources need to focus on improving the recording of rare diseases in order to enhance the epidemiological utility of mortality data.

Errors in cause-of-death statement on death certificates in intensive care unit of Kathmandu, Nepal

BACKGROUND

Death certificates (DC) are one of the most important medico-legal documents that physicians work through. DCs are extensively used in health statistics for epidemiological studies, and in health policy planning as a public health resource tool. Cause-of-death (COD) statement, which is vulnerable to various errors, is the vital part of a DC that has the potential to mislead the policy makers and statisticians. Hence, we evaluated and analyzed the errors prevalent in COD statement of DC.

METHODS

A retrospective observational study was conducted at medical Intensive Care Unit (ICU) of Blue Cross Hospital, Kathmandu, Nepal within two years of study period. A total of 204 medical records of the deceased patients were reviewed. Three sub-headings of COD statement of DC- Part I Immediate COD (ICOD), Part I Underlying COD (UCOD), and Part II Other significant conditions (OSC) were extensively evaluated for the major medical errors.

RESULTS

The study found errors in 78.4 % of DCs. The highest number of errors was in UCOD (83 %). Most common errors were "Mechanism of Death- terminal event" in ICOD, "More than one competing causes" in UCOD, and "OSC present but not listed" in OSC. The error in DC was found to be statistically significant with the severity of sepsis (p = 0.003), and presence of chronic organ failures (p = 0.034). Age, time of death, source of admission, and duration of ICU stay were not found to be statistically associated with the errors in DC.

CONCLUSION

Prevalence of errors in DC was quite high. Most errors were committed in underlying cause of death, which is the most important part of DC. Complexity of the cases was the key factor that increased the risks of committing errors. Specific education should supersede general educational interventions to minimize the errors considerably in writing DC in complex cases.

Standardisation of information submitted to an endpoint committee for cause of death assignment in a cancer screening trial – lessons learnt from CAP (Cluster randomised triAl of PSA testing for Prostate cancer)

BACKGROUND

In cancer screening trials where the primary outcome is target cancer-specific mortality, the unbiased determination of underlying cause of death (UCD) is crucial. To minimise bias, the UCD should be independently verified by expert reviewers, blinded to death certificate data and trial arm. We investigated whether standardising the information submitted for UCD assignment in a population-based randomised controlled trial of prostate-specific antigen (PSA) testing for prostate cancer reduced the reviewers' ability to correctly guess the trial arm.

METHODS

Over 550 General Practitioner (GP) practices (>415,000 men aged 50-69 years) were cluster-randomised to PSA testing (intervention arm) or the National Health Service (NHS) prostate cancer risk management programme (control arm) between 2001 and 2007. Assignment of UCD was by independent reviews of researcher-written clinical vignettes that masked trial arm and death certificate information. A period of time after the process began (the initial phase), we analysed whether the reviewers could correctly identify trial arm from the vignettes, and the reasons for their choice. This feedback led to further standardisation of information (second phase), after which we re-assessed the extent of correct identification of trial arm.

RESULTS

1099 assessments of 509 vignettes were completed by January 2014. In the initial phase (n = 510 assessments), reviewers were unsure of trial arm in 33% of intervention and 30% of control arm assessments and were influenced by symptoms at diagnosis, PSA test result and study-specific criteria. In the second phase (n = 589), the respective proportions of uncertainty were 45% and 48%. The percentage of cases whereby reviewers were unable to determine the trial arm was greater following the standardisation of information provided in the vignettes. The chances of a correct guess and an incorrect guess were equalised in each arm, following further standardisation.

CONCLUSIONS

It is possible to mask trial arm from cause of death reviewers, by using their feedback to standardise the information submitted to them.

TRIAL REGISTRATION

ISRCTN92187251.

Utility of death certificate data in predicting cancer incidence

BACKGROUND

Studies often rely on death certificates to identify cancer occurrence. This research assessed the death certificate's ability to reflect cancer incidence and factors that influence agreement with cancer registry data.

METHODS

This study compared death certificates to cancer incidence data for an occupational cohort of 1,795 deceased workers who were registered by the Iowa Cancer Registry (ICR) between 1973 and 2005. Logistic regression models examined the effects of factors such as survival time, age at diagnosis, and gender on the odds of agreement between death certificate and incidence data.

RESULTS

Death certificates under-reported cancer incidence by 10-100%, depending on site. A 1-year increase in survival decreased the odds of agreement between death certificate and ICR data by 18%. Younger and female workers had increased odds of agreement.

CONCLUSIONS

Death certificates can be useful predictors of cancer incidence, particularly for diseases with shorter survival and among subjects diagnosed earlier in life.

Determining stroke's rank as a cause of death using multicause mortality data

BACKGROUND AND PURPOSE

Stroke has fallen from the second to the fourth leading cause of death in the United States without large declines in stroke incidence or case fatality. We explored whether this decline may be attributable to changes in mortality attribution methodology.

METHODS

Multicause mortality files from 2000 to 2008 were used to compare changes in reporting of stroke as underlying cause of death (UCOD) with changes in death certificates reporting any mention (AMCOD) of stroke. In addition, the UCOD/AMCOD ratio was calculated for the 6 leading organ and disease-specific causes of death. If stroke mortality is underestimated by the system of mortality attribution, we hypothesized that we would find: (1) a greater decline in stroke as UCOD than as AMCOD; and (2) a decline in the UCOD/AMCOD ratio compared with other causes of death.

RESULTS

Age-adjusted death rates for stroke as UCOD (61 per 100,000 in 2000 versus 41 in 2008) and AMCOD (102 per 100,000 versus 68) both declined by 33%. The ratio of UCOD to AMCOD for stroke did not change over time (0.595 in 2000 versus 0.598 in 2008). Changes in UCOD/AMCOD ratio for the diagnoses that surpassed stroke as UCOD were too small (no change for lung cancer and a slight increase from 0.49 to 0.52 for chronic lower respiratory diseases) to explain stroke's decline as UCOD.

CONCLUSION

Changes in mortality attribution methodology are not likely responsible for stroke's decline as a leading cause of death. The discordant trends in incidence, case fatality, and mortality require further study.