Risk factors for infant mortality from diseases in Japan 1995-98

OBJECTIVE: Since 1995, additional information (i.e. birth weight, singleton/multiple births, gestational weeks, maternal age, maternal parity and stillbirth experience) has been required for certificates of infant (less than 1 year of age) death from diseases in Japan. The present study examined the effects of biological, demographic and social variables, as reported on birth and death certificates, on infant, neonatal and postneonatal mortality in Japan. METHODS: Using data from vital statistics between 1995 and 1998, more than 4,787,000 livebirths and 16,000 infant deaths from diseases were analyzed. Univariate and multivariate analyses with the Poisson regression model were employed to assess the effects of variables on infant, neonatal and postneonatal mortality by singleton and multiple livebirths separately. RESULTS: The infant mortality rates from diseases were 3.2/1000 for singleton livebirths and 17.7/1000 for multiple livebirths. In singleton livebirths, low birth weight, infant born in earlier years, being a male infant, employment status as "unemployed or unknown", short gestational weeks, late birth in multiparity and maternal stillbirth experience were all significantly related to increased risk of neonatal and postneonatal deaths. Teenage mother were also at high risk of postneonatal deaths. Regional differences were observed. Compared with singleton livebirths, birthweight-specific mortality rates in multiple livebirths were relatively low among infants weighing under 2500 g. In multiple livebirths, elevated risk of death was associated with low birth weight, infant born in earlier years, employment status as "unemployed or unknown" and short gestational weeks. However, late birth in multiparity was related to a reduced risk of death, and maternal stillbirth experience was not a significant variable. CONCLUSION: This study provided the first quantitative estimate of risk of infant mortality from diseases in Japan. Since a more detailed elucidation of actual conditions and risk factors of infant deaths by vital statistics has become possible, efficient measures for improvement of infant mortality are to be expected.     

A study of selected death certificates from three Dublin teaching hospitals

One hundred death certificates were compared over two time periods with the corresponding autopsy reports on the cases to ascertain if the causes of death from the two sources were similar. There was poor concordance between the two and it seems likely that reasons for requesting an autopsy did not extend to using the information to complete the death certificates. There were 55 errors on 45 certificates, and 19 certificates were so inaccurate as to warrant a change in the underlying cause of death. In only 10 cases was the certificate signed before the autopsy report was available; however, should clinicians have wished to add autopsy findings to the certificate later, there is no facility on the Irish death certificate to do so. A revision of the format of the certificate is recommended. An examination of death certificates from varied medical sources would be welcome to see if the serious errors identified in this study are more widespread.     

Decline of acute myocardial infarction death rates not due to cause of death coding

The recent decline in coronary heart disease (CHD) mortality has been attributed to reduction in risk factors, improved management and the possibility of statistical artifacts. The purpose of this study is to assess the impact of geographic and time variation in the coding of cause of death from death certificates into ICD codes upon CHD mortality rates in Canada. Equal samples of death certificates were recoded for Nova Scotia and Saskatchewan for each of the years 1970 and 1984: 1) a first set of 1,600 death certificates originally coded as acute myocardial infarction (AMI) and 2) a second set of 800 death certificates from all causes of death. The coding error rates increased with age and with the number of contributing and underlying causes of death reported on the death certificates. The net effect of false positive and false negative AMI codes on death certificates did not vary significantly by province or year. Thus, variation of death certificate coding over time and geographic regions do not contribute toward the explanation of the AMI mortality rate decline.     

Validity of maternal and perinatal risk factors reported on fetal death certificates

We sought to estimate the accuracy, relative to maternal medical records, of perinatal risk factors recorded on fetal death certificates. We conducted a validation study of fetal death certificates among women who experienced fetal deaths between 1996 and 2001. The number of previous births, established diabetes, chronic hypertension, maternal fever, performance of autopsy, anencephaly, and Down syndrome had very high accuracy, while placental cord conditions and other chromosomal abnormalities were reported inaccurately. Additional population-based studies are needed to identify strategies to improve fetal death certificate data.     

Congenital malformations as a cause of neonatal and postnatal death in Massachusetts (1970-1980)

Massachusetts birth and death certificate tapes for the years 1970-1980 were linked and analyzed to determine causes of death in the neonatal and postneonatal periods and to identify any related sociodemographic factors. Our analysis suggests that, although the neonatal mortality rate declined by about 43 percent, the postneonatal mortality rate remained relatively unchanged. Perinatal problems remained the principal cause of death during the neonatal period, throughout the decade. In the postneonatal period, congenital malformations became a leading cause of death toward the end of the decade because of a reduction in mortality from infectious diseases and perinatal problems. Infants born to mothers under 18 and over 34 years of age had the highest death rates from congenital birth defects. Higher mortality rates caused by congenital malformations were found in the more industrialized areas of Massachusetts. Further declines in infant mortality rates in Massachusetts will depend on preventive measures to reduce the incidence of congenital malformations.     

Underreporting of diabetes on death certificates, King County, Washington.

We estimated the reporting of diabetes on death certificates for persons known to have diabetes. Surveillance of 19 hospitals and two paramedic emergency medical services during 12 months in Seattle and King County, Washington, ascertained acute ischemic heart disease events for persons with diabetes and yielded 1235 persons with suspected ischemic heart disease. Mortality was 23.6%, and 41% of death certificates listed diabetes. The reporting of diabetes on the death certificate was not random, and it varied by patient and physician characteristics. Diabetes is strongly linked to fatal ischemic heart disease, but its importance is underrepresented by death certificates for some subgroups.     

Reporting of meticillin-resistant and -susceptible Staphylococcus aureus on death certificates in Irish hospitals

The documentation of infection with meticillin-resistant Staphylococcus aureus (MRSA) on death certificates has been the subject of considerable public discussion. Using data from five tertiary referral hospitals in Ireland, we compared the documentation of MRSA and meticillin-susceptible S. aureus (MSSA) on death certificates in those patients who died in hospital within 30 days of having MRSA or MSSA isolated from blood cultures. A total of 133 patients had MRSA or MSSA isolated from blood cultures within 30 days of death during the study period. One patient was excluded as the death certificate information was not available; the other 132 patients were eligible for inclusion. MRSA and MSSA were isolated from blood cultures in 59 (44.4%) and 74 (55.6%) cases respectively. One patient was included as a case in both categories as both MRSA and MSSA were isolated from a blood culture. In 15 (25.4%) of the 59 MRSA cases, MRSA was documented on the death certificate. In nine (12.2%) of the 74 patients with MSSA cases, MSSA was documented on the death certificate. MRSA was more likely to be documented on the death certificate than MSSA (odds ratio: 2.46; 95% confidence interval: 1.01-6.01; P < 0.05). These findings indicate that there may be inconsistencies in the way organisms and infections are documented on death certificates in Ireland and that death certification data may underestimate the mortality related to certain organisms. In particular, there appears to be an overemphasis by certifiers on the documentation of MRSA compared with MSSA.     

Analysis of unlinked infant death certificates from the NIMS project

The National Infant Mortality Surveillance (NIMS) project used linked birth and infant death certificates to calculate birth weight-specific infant mortality risks for the 1980 U.S. birth cohort. Record linkage depends on complete registration of vital events, interstate exchange of vital records, accurate information on certificates, and a comprehensive linkage system. States reported 2,604 unlinked infant death certificates for 1980, ranging from 0 to 397 per State. Age at death for these infants ranged from 1 minute to 11 months. More than 41 percent of the unlinked death certificates were for postneonates, compared with 32.5 percent found in the cohort's total infant death experience. Only 38.2 percent of the unlinked infant death certificates showed strictly intrastate events (birth and death occurrence, and residence at death all in one State), compared with 92.9 percent in the cohort's total infant death experience. Estimates of the percentage successfully linked by State ranged from 86.0 to 100.0. After adjusting for the certainly unlinked infant death certificates, nine States' infant mortality risks increased by more than 0.2 per 1,000 live births. Improvements are needed both within and between States to ensure more complete birth and infant death certificate linkage.     

Use of death certificates for mesothelioma surveillance.

Data from the Massachusetts Cancer Registry and death certificates were linked for mesothelioma cases reported to the registry from 1982 through 1987 to determine the extent to which the cause of death information that is given on the death certificate is useful in identifying mesothelioma cases for disease surveillance. Only 12 percent of all persons reported with mesothelioma who had died were detected using underlying cause of death codes for cancers of the peritoneum and pleura, which are commonly used to identify mesothelioma cases. The rate increased to 83 percent when death certificates were reviewed manually for any mention of mesothelioma. Surveillance using only the coded cause of death data currently available will result in a large underascertainment of mesothelioma cases.     

Agreement of occupation and industry data on Rhode Island death certificates with two alternative sources of information

There is increasing interest in documenting the putative health effects of occupational hazards, prompting Federal and State efforts that rely primarily on occupational information obtained from the death certificate. Previous studies have assessed the agreement of occupational data on death certificates with actual lifetime employment by using current employment data from census records for comparisons. Such analyses have largely been confined to males. We compared lifetime occupational information obtained from a panel survey for both sexes with death certificate data for 446 deceased panel members. After adjusting for inadequate information, the occupation recorded on the death certificates of the men agreed with the occupation recorded in the survey 66 percent of the time. The comparable percentage for the industry where the deceased had been employed was 78 percent. Among the women's records, agreement on occupation was 65 percent, and on industry, 69 percent. Using another sample of death certificates, comparisons of the information for 322 decedents with city directory data produced similar results. The higher level of agreement for women was due in part to the large number who were reported as "housewives." In a separate analysis, the agreement rate for nonhousewives declined. Suggestions for improvements in the recording of occupational data and the constraints imposed by the use of death certificate data in occupational epidemiology are presented.     

Out-of-hospital coronary death in an urban population--validation of death certificate diagnosis. The Minnesota Heart Survey

To assess the validity of death certificate diagnoses of out-of-hospital coronary heart disease deaths, the authors studied a one-third random sample of out-of-hospital deaths occurring in 1979 in Minneapolis-St. Paul, Minnesota, residents. Death certificates with diagnoses possibly containing coronary heart disease deaths were enumerated, and cause of death was recorded from the certificate in two ways: as the first listed ("immediate") cause and as the "underlying cause" assigned by a trained nosologist. Validation was performed by standardized physician review of information obtained about the death, which included one or more of the following: an interview with a relative or friend, physician report, autopsy report, medical record, and/or nursing home record. Missing information was frequent, but cases with at least an informant interview and/or autopsy report (82%) were representative and could be used for validation. The sensitivity and specificity of the underlying cause of coronary heart disease (International Classification of Diseases, Ninth Revision, codes 410-414, 427) on the death certificate were 90.3% and 82.7%, respectively, compared with the physician-assigned diagnosis. For the immediate cause, sensitivity and specificity were 90.3% and 67.9%, respectively. These findings suggest that the validity of death certificates for out-of-hospital coronary heart disease death is high, as assessed by this method of retrospective physician review.     

Comparing Swedish hospital discharge records with death certificates: implications for mortality statistics

BACKGROUND: The quality of mortality statistics is of crucial importance to epidemiological research. Traditional editing techniques used by statistical offices capture only obvious errors in death certification. In this study we match Swedish hospital discharge data to death certificates and discuss the implications for mortality statistics. METHODS: Swedish death certificates for 1995 were linked to the national hospital discharge register. The resulting database comprised 69 818 individuals (75% of all deaths), 39 872 (43%) of whom died in hospital. The diagnostic statements were compared at Basic Tabulation List level. RESULTS: The last main diagnosis and the underlying cause of death agreed in 46% of cases. Agreement decreased rapidly after discharge. For hospital deaths, the main diagnosis was reported on 83% of the certificates, but only on 46% of certificates for non-hospital deaths. Malignant neoplasms and other dramatic conditions showed the best agreement and were often reported as underlying causes. Conditions that might follow from some other disease were often reported as contributory causes, while symptomatic and some chronic conditions were often omitted. In 13% of cases, an ill-defined main condition was replaced by a more specific cause of death. CONCLUSIONS: There is no apparent reason to question the death certificate if the main diagnosis and underlying cause agree, or if the main diagnosis is a probable complication of the stated underlying cause. However, cases in which the main diagnosis cannot be considered a complication of the reported underlying cause should be investigated, and assessments made of the feasibility and cost-effectiveness of routinely linking hospital records to death certificates.     

Use of death certificates in epidemiological studies, including occupational hazards: discordance with clinical and autopsy findings.

There has long been evidence of frequent inaccuracy of death certificates, with significant discordance between such designations and clinical and autopsy data. This exists for occupational diseases as well. The use of statistical rates based on death certificates has been seriously questioned despite their utility for total mortality. Programs to supplement death certificate data, particularly in occupational disease studies, may be helpful, and are reviewed.     

Agreement on cause of death between proxies, death certificates, and clinician adjudicators in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study

Death certificates may lack accuracy and misclassify the cause of death. The validity of proxy-reported cause of death is not well established. The authors examined death records on 336 participants in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study, a national cohort study of 30,239 community-dwelling US adults (2003-2010). Trained experts used study data, medical records, death certificates, and proxy reports to adjudicate causes of death. The authors computed agreement on cause of death from the death certificate, proxy, and adjudication, as well as sensitivity and specificity for certain diseases. Adjudicated cause of death had a higher rate of agreement with proxy reports (73%; Cohen's kappa (κ) statistic = 0.69) than with death certificates (61%; κ = 0.54). The agreement between proxy reports and adjudicators was better than agreement with death certificates for all disease-specific causes of death. Using the adjudicator assessments as the "gold standard," for disease-specific causes of death, proxy reports had similar or higher specificity and higher sensitivity (sensitivity = 50%-89%) than death certificates (sensitivity = 31%-81%). Proxy reports may be more concordant with adjudicated causes of death than with the causes of death listed on death certificates. In many settings, proxy reports may represent a better strategy for determining cause of death than reliance on death certificates.     

The association of maternal smoking with age and cause of infant death

Linked birth certificate and infant death certificate data from Missouri for 1979-1983 were used to explore the association of maternal smoking with age and cause of infant death. The data included 305,730 singleton white livebirths, of which 2,720 resulted in infant deaths. Using multiple logistic regression to control for the confounding effects of maternal age, parity, marital status, and education, the authors found that smoking was associated with both neonatal and post-neonatal mortality and with each cause of death except congenital anomalies. The adjusted odds ratio for smoking was higher for postneonatal deaths than neonatal deaths and was particularly high for two causes: respiratory disease (odds ratio = 3.4) and sudden infant death syndrome (odds ratio = 1.9). A moderate odds ratio (about 1.4) was found for causes attributed to the International Classification of Diseases, 9th Revision Perinatal Conditions Chapter. Although the associations for neonatal deaths and perinatal conditions were partially attributable to the effect of maternal smoking in lowering birth weight, virtually none of the excess respiratory mortality and sudden infant death syndrome mortality among the offspring of smokers was attributable to birth weight differences between the infants of smokers and nonsmokers. This suggests that respiratory deaths and sudden infant death syndrome deaths may be related to the effect of passive exposure of the infant to smoke after birth.     

How accurately is euthanasia reported on death certificates in a country with legal euthanasia: a population-based study

Background

Death certificates are the main source of information on the incidence of the direct and underlying causes of death, but may be unsuitable for monitoring the practice of medical assistance in dying, e.g. euthanasia, due to possible underreporting. This study examines the accuracy of certification of euthanasia.

Methods

Mortality follow-back survey using a random sample of death certificates (N?=?6871). For all cases identified as euthanasia we checked whether euthanasia was reported as a cause of death on the death certificate. We used multivariable logistic regression analysis to evaluate whether reporting varied according to patient and decision-making characteristics.

Results

Through the death certificates, 0.7% of all deaths were identified as euthanasia, compared with 4.6% through the mortality follow-back survey. Only 16.2% of the cases identified from the survey were reported on the death certificate. Euthanasia was more likely to be reported on the death certificate where death was from cancer (14% covered), neurological diseases (22%) and stroke (28%) than from cardiovascular disease (7%). Even when the recommended drugs were used or the physician self-labelled the end-of-life decision as euthanasia, euthanasia was only reported on the death certificate in 24% of cases.

Conclusions

Death certificates substantially underestimate the frequency of euthanasia as a cause of death in Belgium. Mortality follow-back studies are essential complementary instruments to examine and monitor the practice of euthanasia more accurately. Death certificate forms may need to be modified and clear guidelines provided to physicians about recording euthanasia to ensure more accurate certification.

     

Use of death certificates to study ethnic-specific mortality

OBJECTIVES: The Hispanic population in the United States represents more than 40 million individuals, with Mexican Americans (MA) as the largest subgroup. To assess the utility of death certificates and medical records as the source of race/ethnicity data for epidemiologic studies, we compared self-reported race/ ethnicity to race/ethnicity recorded on death certificates and medical records in a bi-ethnic, non-immigrant U.S. community with a significant MA population. METHODS: This study utilized data collected from a subset of 1,856 participants of the Brain Attack Surveillance in Corpus Christi (BASIC) project. In-person interviews were conducted to determine self-reported race/ethnicity. Of those interviewed, 480 subsequently expired. Using self-reported race/ethnicity as the gold standard, we determined percent agreement, sensitivity, and specificity of the death certificate and medical record. RESULTS: Of the 480 subjects, 259 self-reported their race/ethnicity as non-Hispanic white (NHW), 195 self-reported as MA, and 26 self-reported as non-Hispanic black. Median age was 78.5 years and 55.8% were female. Percent agreement between self-reported race/ethnicity and race/ethnicity recorded on the death certificate and medical record was 97.1% and 96.3% respectively. Five percent of MAs were misclassified as NHW on their death certificates and 3% on their medical records. CONCLUSIONS: Results indicated that Hispanic designation recorded on death certificates and medical records in this community was largely consistent with that of self-report. This study suggests that vital statistics data in non-immigrant U.S. Hispanic communities can be used with confidence to investigate ethnic-specific aspects of disease and mortality. Similar studies in other multi-racial communities should be conducted to confirm and generalize these results.     

Vibrio vulnificus infection reporting on death certificates: the invisible impact of an often fatal infection.

This study assessed accuracy of (a) recording Vibrio vulnificus infection on death certificates and (b) International Classification of Disease (ICD)-9 codes for V. vulnificus. Patients with microbiologically confirmed V. vulnificus infection were identified as part of co-ordinated surveillance in four USA Gulf Coast states between 1989 and 1993. Of 60 deaths, 51 death certificates were reviewed and V. vulnificus was recorded as the immediate cause of death on 11 (22%). There was no ICD-9 code for V. vulnificus infection, thus no patients had an ICD-9 code indicating V. vulnificus infection. Of 23 certificates where V. vulnificus was recorded on the death certificate, only 5 (22%) were coded for Gram-negative, septicaemia. This study highlights the importance of teaching physicians how to provide epidemiologically meaningful data on death certificates and the need for accurate ICD mortality codes.     

Errors in reporting education on the death certificate: some findings for older male decedents from New York State and Utah

This paper examines possible biases in death certificate data on education that may be used in studies of mortality and socioeconomic status. By means of a matching study conducted in 1987, self-reported education level in a large-scale survey (Cancer Prevention Study II) is compared with education as subsequently reported by next of kin on death certificates in upstate New York and in Utah. These are the only two states that presently have an education item on the certificates. In both state samples, agreement was highest for high school graduates and for those with graduate school education. Overall agreement was 68%, indicating serious problems in the accuracy of education reporting. Possible systematic biases in the new death certificate item on education are indicated.