Quality assurance review of death certificates: a pilot study

BACKGROUND: Although quality assurance programs for medical examiners are required by the National Association of Medical Examiners' Inspection and Accreditation Checklist, quality assurance programs specifically targeting death certificate completion have not been addressed. The Fulton County Medical Examiner, Atlanta, GA, has implemented a pilot quality assurance program for death certificate information, and this report contains information about 1 year's experience with the program. METHODS: All death certificates are reviewed by the case medical examiner(s) and chief medical examiner prior to their release to funeral homes. Death certificates with errors are retained for quality assurance and review purposes, and needed corrections are made before death certificates are released. During a 1-year period, death certificates with errors were collected and then reviewed and tabulated by type of error. RESULTS: Between May 26, 2003, and May 25, 2004, the Fulton County Medical Examiner certified 1267 deaths. Of these, 47 (4%) were found to contain errors that were corrected and an additional 52 (4%) had been amended for various reasons. The most common errors were misspellings in causes of death or poor or incomplete wording in injury-related information. Forty-seven percent of errors involved omitted, incomplete, or incorrect information that was potentially significant. The most common reason for amended certificates was unexpected detection of acute intoxications among people with significant cardiovascular disease. CONCLUSIONS: Quality assurance review of death certificates can assist in preventing the release of death certificates with incomplete, erroneous, or omitted information and may also be useful as an educational forum regarding completion of the death certificate.     

Use of "therapeutic complication" as a manner of death

Analyses of deaths due to therapeutic complications (TCs) provide important quality of care information for medical providers. In New York City, 463 deaths were investigated by the Office of Chief Medical Examiner and certified with TC as the manner of death in 2003. The TC manner of death is used for fatalities due to predictable complications of appropriate medical therapy. All death certificates and select autopsy, hospital, and investigation reports were reviewed. Data concerning cause of death, contributing conditions, age, race, and sex were extracted. The types of complications and the causes of death were classified into various types of surgical and nonsurgical categories of complications. These included: postoperative infections, pulmonary emboli, and technical and medication complications. The use of TC as a manner of death has benefits and limitations. Without the TC option, one is forced to certify certain deaths (e.g., penicillin anaphylaxis) either as natural or accident. The TC option allows easy identification and tracking of medical complications for public health purposes and also allows more consistent reporting of natural and medical-accidental deaths. In general, complications that occur during emergency surgeries/procedures for natural disease, tend to be certified with a natural manner. The "but for" test may be used to distinguish natural from TC deaths. There are criteria for distinguishing TC from accidents and homicides. TCs that occur during treatment of a potentially life-threatening injury, are superseded by the manner dictated by the circumstances of the initiating injury. The certification of TC usually does not address errors of omission, clinical judgement/management, or missed diagnoses.     

Variability in cancer death certificate accuracy by characteristics of death certifiers

Death certificates are the source for mortality statistics and are used to set public health goals. Accurate death certificates are vital in tracking outcomes of cancer. Deaths may be certified by physicians or other medical professionals, coroners, or medical examiners. Idaho is one of 3 states that participated in a Centers for Disease Control and Prevention-funded study to assess the concordance between cancer-specific causes of death and primary cancer site among linked cancer registry/death certificate data. We investigated variability in the accuracy of cancer death certificates by characteristics of death certifiers, including certifier type (physician vs coroner), physician specialty, years of experience as death certifier, and number of deaths certified. This study showed significant differences by certifier type/physician specialty in the accuracy of cancer mortality measured by death certificates. Nonphysician coroners had lower accuracy rates compared with physicians. Although nonphysician coroners certified less than 5% of cancer deaths in Idaho, they were significantly less likely to match the primary site from the cancer registry. Results from this study may be useful in the future training of death certifiers to improve the accuracy of death certificates and cancer mortality statistics.     

Errors on death certificates requiring amendments: the Broward County experience

The medical examiner's office in Broward County is responsible for determining the cause and manner of death in cases falling under its jurisdiction and issuing death certificates on these decedents. Amendments are occasionally required to correct misinformation on death certificates or within the autopsy reports. The purpose of this study was to investigate the major causes for the amendments and to develop strategies to avoid future errors. We found 128 cases from 2006 to 2007 that required amendments; 103 contained sufficient data in the file for further analysis. Over this time period, 3790 death certificates were issued over that same period, resulting in a 3.37% amendment rate. In this study, the cohort included both males and females with a ratio of 2:1. Their ages ranged from newborn to 103 years, with a mean age of 49 years. Of the 103 amended cases, amendments were made to the cause (n = 30) and often the manner (n = 21) of death listed on the death certificate; the remaining changes were limited to the autopsy report. The most common reasons for amendments included reception of delayed laboratory findings (35%), acquisition of additional medical history (22.5%), and typographic errors (15.5%). Typographic errors mainly occurred because of inaccuracies in the names originally provided to our office, the use of aliases by decedents, incorrect personal/demographic history, or various misspellings by funeral homes or medical examiner staff. The most significant reclassifications involved changing certified natural deaths to accidental overdoses and vice versa, based on toxicological analysis. Because of delays in specimen turnaround, these amendments often were made months after the original death certificate was issued. STAT urine drug screening has been helpful in reducing the number of amendments made, but certain drugs of significance are missed by rapid screens. Given that our office performed complete toxicological analysis on all cases over this period, it seems likely that we detected several overdoses that would have been missed if natural deaths were not routinely screened for potential toxins.     

Medical examiner/coroner records: uses and limitations in occupational injury epidemiologic research

Epidemiologic research often relies on existing data, collected for nonepidemiologic reasons, to support studies. Data are obtained from hospital records, police reports, labor reports, death certificates, or other sources. Medical examiner/coroner records are, however, not often used in epidemiologic studies. The National Institute for Occupational Safety and Health's Division of Safety Research has begun using these records in its research program on work-related trauma. Because medical examiners and coroners have the legal authority and responsibility to investigate all externally caused deaths, these records can be used in surveillance of these deaths. Another use of these records is to validate cases identified by other case ascertainment methods, such as death certificates. Using medical examiner/coroner records also allows rapid identification of work-related deaths without waiting several years for mortality data from state offices of vital statistics. Finally, the records are an invaluable data source since they contain detailed information on the nature of the injury, external cause of death, and results of toxicologic testing, which is often not available from other sources. This paper illustrates some of the ways that medical examiner/coroner records are a valuable source of information for epidemiologic studies and makes recommendations to improve their usefulness.     

Investigation of elderly deaths in nursing homes by the medical examiner over a year

Despite death being one of the most common reasons for discharge from a nursing home, fewer than 1% of nursing home resident deaths are autopsied. To evaluate our role as medical examiner in nursing home deaths, we conducted a retrospective review of all decedents in Jefferson County, Alabama, for the year 2001. Death certificate data indicate that 995 deaths occurred in nursing homes in Jefferson County in 2001. Of those 995 deaths, 119 (12%) were reported to the Jefferson County Coroner/Medical Examiner Office. Jurisdiction was accepted in 5 cases in which the circumstances already made clear that the death was a nonnatural event. In the remaining 96% of nursing home deaths reported to the medical examiner, the statements of the reporting person were taken to be true concerning the expected nature of the death. An independent scene evaluation was provided by a police officer or paramedic in 82% of the cases reported to the medical examiner's office. Elderly individuals, as a group, are expected to die, but the death of a particular elder may or may not be expected. In our jurisdiction, only 12% of all nursing home deaths are reported to our office, and only 4% of reported deaths are actively investigated. Actively investigating each nursing home death would overwhelm the resources currently available to our office. We advocate the study and development of criteria to aid in determining whether the death of an individual elder is sudden and unexpected.     

Sudden epilepsy deaths and the forensic pathologist

Sudden unexpected deaths in epileptic persons are not rare events, most commonly encountered by the forensic pathologist rather than the clinician. Such deaths may represent 1-1.5% of all "natural" deaths certified by the medical examiner or coroner. The typical victim is a black male about 30 years of age who tends to abuse alcohol, with a history of generalized epilepsy for more than 1 year and likely for more than 10 years. There are a lack of obvious anatomic causes for the death at autopsy, but 60-70% of cases will have a lesion in the brain (most commonly old trauma) to explain the epilepsy. Most victims have no blood levels of anticonvulsant medications at the time of death. We have evolved a form for use by medical examiner/coroner's investigators at the scene to collect relevant information which will be of assistance to the pathologist in interpreting the case. Estimated prevalence of sudden epilepsy death, mechanisms, and other features of such cases are reviewed briefly.     

Level of agreement between opinions of medical examiner investigators and forensic pathologist medical examiners regarding the manner of death

To evaluate the level of agreement between medical examiner investigators' opinion of the manner of death and what the manner of death was as certified by forensic pathologist medical examiners (MEs), we reviewed the case records stored in a database of all deaths reported to the office of the medical examiner in Fulton County, Georgia. Of 15,771 deaths reported to the office during a 10-year period, a difference exists in 1908 cases. In 900 natural deaths, the investigators recorded 135 accident, 10 homicide, 10 suicide, and 745 undetermined manners of death. In 755 accidental deaths, the investigators recorded 16 natural, 8 homicide, 13 suicide, and 718 undetermined manners of death. In 107 homicides, the investigators recorded 12 natural, 8 accident, 0 suicide, and 87 undetermined manners of death. In 70 suicides, the investigators recorded 9 natural, 9 accident, 3 homicide, and 49 undetermined manners of death. In 61 deaths classified as undetermined, the investigators recorded 25 natural, 13 accident, 17 homicide, and 6 suicide manners of death. In 15 deaths, the discrepancy exists due to an apparent error in the database information. This study confirms a high concordance between investigator and ME opinion regarding manner of death but also documents the need for case review and autopsies by forensic pathologists to confirm the investigators' opinion of the manner of death, determine the manner of death when the investigator selects undetermined, and on occasion, refute the investigators' opinion regarding the manner of death.     

Recent status of the medical examiner system in Japan: demographic variation of medicolegal deaths in Hyogo Prefecture and uncertainty in medicolegal investigations conducted by medical practitioners.

The medical examiner system has been steadily abolished in Japan. Instead, medicolegal investigations are entrusted by the police to medical practitioners, who are not permitted to perform autopsies. The necessity for the medical examiner system was assessed through inquest records in Hyogo, one of the three prefectures which still have medical examiner systems. Standardized mortality ratios (SMRs) for accidents and suicides were negatively associated with population density, being high in rural areas with a large proportion of elderly citizens, while the SMR for natural deaths was high in urbanized areas and associated with the proportion of inquests to total resident deaths. The high proportion of inquests, however, did not always mean that inquest records were of good quality. Significant differences in the quality of medicolegal investigations seemed to exist between medical examiners and medical practitioners. That is, in order to certify the cause-of-death, medical examiners performed autopsies in about half of their cases, while only 2% of medical practitioner cases were subjected to autopsies. Medical practitioners, who certified the cause-of-death as "heart failure" without advising an autopsy, were regularly entrusted with inquests. It is likely that the causes-of-death for medicolegal cases may be questionable since more than 85% of all medicolegal deaths were investigated by medical practitioners, which may cause inaccuracy in at least 3-7% of mortality statistics. It is necessary to educate medical practitioners concerning the importance of mortality statistics and ICD and on the validity of autopsies, in order to obtain accurate mortality statistics from medicolegal cases.     

Post-mortem examination prior to cremation--an instrument to verify the quality of medical post-mortems and uncover non-natural deaths?

The external post-mortem examination, its deficient quality and possible causes have been the subject of numerous political and professional discussions. The external post-mortem examination is the basis for the decision whether further criminal investigations are required to clarify the cause of death. It is thus an essential instrument to ensure legal certainty. Before cremation, a second external post-mortem examination is performed by a public medical officer to make sure that errors of the first post-mortem are corrected. In the present study, cases were retrospectively analyzed in which a forensic autopsy had been ordered on the basis of the results of the post-mortem examination performed before cremation. The entries on the death certificate regarding the manner and cause of death were compared with the autopsy results. Between 1998 and 2007, 387 autopsies were ordered after external examination before cremation. In 55 cases (14.2%), the autopsy revealed a non-natural death, although a natural death had been attested on the death certificate. In descending order, a wrong manner of death was attested by clinicians, general practitioners and emergency physicians. With regard to the place where the first external post-mortem had been performed the lowest error rate was seen in nursing homes. Concerning the cause of death, discrepancies between the first post-mortem and autopsy were found in 59.4% of the cases. In this respect, general practitioners and clinicians were ranking first, whereas in nursing homes the cause of death was wrongly assessed in over 70% of cases. At present, the medical post-mortem does not meet the required quality standards, especially with regard to legal certainty. Determination of the cause of death on the basis of the external post-mortem examination is a challenging task even for the experienced medical examiner. As to the categorization of the manner of death it has to be stated that non-natural deaths are often not recognized or that the possibility to certify a death as unclear is not sufficiently used. As a result, it seems important to demand intensive, qualified, additional training in external post-mortem examinations for physicians.     

Fatal Diabetic Ketoacidosis and Antipsychotic Medication

Hyperglycemia and new onset diabetes have been described with certain antipsychotic medications and some of the initial presentations are fatal diabetic ketoacidosis (DKA). We report 17 deaths due to DKA in psychiatric patients treated with second generation antipsychotic medications. Death certificates and toxicology data were searched for DKA and hyperglycemia. We reviewed the medical examiner records which included the autopsy, toxicology, police, and medical examiner investigators' reports. The decedents ranged in age from 32 to 57 years (average 48 years). There were 15 men and two women. The immediate cause of death was DKA in all. The psychiatric disorders included: 10 schizophrenia, three bipolar/schizophrenia, two bipolar, and two major depression. The most frequent atypical antipsychotic medications found were quetiapine and olanzapine followed by risperidone. In 16 deaths, we considered the medication as primary or contributory to the cause of death.     

Accuracy of cause of death determination without forensic autopsy examination

Medical examiners and coroners commonly determine cause and manner of death without an autopsy examination. Some death certificates generated in this way may not state the correct cause and manner of death. From the case files of the Department of Forensic Medicine in Sydney, Australia, the authors retrospectively reviewed investigative information of all cases in a 6-month period that were initially considered natural deaths (429). The authors, blinded to autopsy results, accepted 261 cases as appropriate for certification without autopsy and assigned a cause of death to each. Per standard local practice, all cases had been autopsied. The actual causes of death as determined by autopsy were then revealed and compared with the presumed causes of death. Most presumed and actual causes of death were cardiovascular (94% and 80%, respectively). The majority of presumed causes of death were listed as ASCVD as the cases lacked features of a more specific cardiovascular process. A large majority of cases had a presumed cause of death of ischemic heart disease based on individual case details. The actual causes of death demonstrated a large breadth of cardiovascular and noncardiovascular disease processes, even though ischemic heart disease accounted for 62% of deaths. The presumed cause of death was completely wrong in 28% of cases. A nonnatural manner of death was present in 3% of cases. This study demonstrates that experienced forensic pathologists may generate erroneous death certificates for cases that are not autopsied.     

Differentiating cause-of-death terminology for deaths coded as sudden infant death syndrome, accidental suffocation, and unknown cause: an investigation using US death certificates, 2003-2004

We compared written text on infant death certificates for deaths coded as sudden infant death syndrome (R95), unknown cause (R99), and accidental suffocation (W75). Using US mortality files supplemented with the death certifiers' written text for all infant deaths with International Classification of Diseases (ICD)-10 assigned codes R95, R99, and W75, we formed cause-of-death subcategories from common themes identified from the written text. Among all infant deaths in 2003-2004, the underlying cause of death was listed as R99 for 2128 deaths, R95 for 4408 deaths, and W75 for 931 deaths. Among the postneonatal deaths, the differences in subcategories varied between assigned ICD-10 codes: for R99-coded deaths, 45.8% were categorized as "Unknown" and 48.6% as "Pending"; for R95-coded deaths, 67.7% were categorized as "sudden infant death syndrome (SIDS)"; and for W75-coded deaths, 76.4% were categorized as "Suffocation." Examination of the written text on the death certificates demonstrates variability in the assigned ICD-10 codes which could have an important effect on the estimates of SIDS cases in the United States.     

Identification of the unidentified deceased: turnaround times, methods, and demographics in Fulton County, Georgia

CONTEXT: Multiple methods may be employed in an attempt to identify deceased individuals who are unidentified when reported to the medical examiner or coroner. The success and turnaround times of various methods differ. OBJECTIVE: To determine the number of deaths initially involving unidentified individuals, their demographics, the methods used and turnaround times for cases in which identification was successful and the portion of cases that remain unidentified for significant periods of time. DESIGN, SETTING, AND PARTICIPANTS: Retrospective case review of all decedents who were initially unidentified when death was reported to the Fulton County Medical Examiner in Atlanta, Georgia, during a 1-year period between May 2003 and May 2004. MAIN OUTCOME MEASURES: Rate per 1000 deaths which involved an unidentified decedent, tabulation of methods used to make identifications, and rate of cases remaining unidentified. RESULTS: Of 2279 deaths reported to the office, 100 were originally unidentified, resulting in a rate of 44 unidentified decedents per 1000 deaths reported. Those who remained unidentified involved 6 cases, resulting in a "cold case" rate of 2.6 per 1000 deaths. Seventy-eight percent were identified within 2 days, and the most common successful methods were visual identification (52%) and fingerprints (31%). Dental (10%), x-ray (4%), and other methods (6%) were least commonly used. Of the 94 persons identified, all identifications occurred within 29 days. DNA procedures were not needed in most cases, and in the few cases where such tests were needed, results were either unavailable or failed to show a match with known samples. CONCLUSIONS: The majority of unidentified deceased individuals were identified within 2 days, with visual verification or fingerprints accounting for about 83% of methods successfully employed. The medical examiner noted a rate of 44 unidentified deceased persons per 1000 death reports, with an ultimate "cold case" (long-term unidentified) rate of about 2.6 cases per 1000 death reports. These data may be useful in developing office policy and procedure regarding the procedural and temporal aspects of investigations centered on achieving identification and making dispositions of unidentified deceased bodies.     

Methadone-related deaths in Hennepin County,Minnesota: 1992-2002

Methadone maintenance therapy (MMT) is the only currently established medical therapy for heroin addiction. However, MMT still remains controversial. In Hennepin County, Minnesota, methadone is one of the top ten drugs reported in medical examiner investigated deaths and one of the most commonly diverted pharmaceuticals. This report reviews the role of methadone in medical examiner deaths over a 10-year period, 1992-2002. We compare cause and manner of death (accidental, natural, suicide) and methadone blood concentrations for decedents who were members of MMT programs with illicit users and those prescribed methadone for chronic pain. Findings reveal that 65% of decedents with measurable blood methadone concentrations were not participating in MMT programs. A total of 96 cases were identified, with the majority white (90.5%) and male (76.8%). MMTP program members were the minority (34.7%) of the methadone positive deaths and 39% were illicit users. Fifteen percent were chronic pain patients with almost half of this group dying from overdose. Methadone concentrations of drug caused/related deaths (0.18-3.99 mg/L) overlapped with those of deaths not attributable to methadone (0.18-3.03 mg/L) with no definable lethal level. Interpretation of methadone blood concentrations must be done in the context of the clinical history for determining cause of death, and may be confounded by postmortem redistribution.     

Evaluation of certifier practices regarding alcohol-related deaths: Fulton County Medical Examiner's Center, Atlanta, Georgia, 2004

CONTEXT: Alcohol can contribute to various manners of death by acute intoxication that places a person at risk for fatal injury, acute fatal alcohol poisoning, or the various fatal complications of chronic abuse with or without superimposed acute intoxication. The reporting of alcohol use on the death certificate may vary with office policy or procedure, certifier judgment, and the timing of information received during investigation. OBJECTIVE: To determine the number of deaths including mention of alcohol use in the investigative case file, the number of death certificates on which alcohol use is reported, the number of discrepancies between the 2, and the possible reasons for observed discrepancies. DESIGN, SETTING, AND PARTICIPANTS: Retrospective case review of all deaths where alcohol use was mentioned in the investigative case file and/or on the death certificate for deaths investigated by the Fulton County Medical Examiner in Atlanta, Georgia, during a 1-year period between January 1, 2004, and December 31, 2004. MAIN OUTCOME MEASURES: Percentage of deaths with alcohol use reported on the death certificate, tabulation of where and how alcohol use is reported on the death certificate, and tabulation of the differences between the investigative case file and death certificate regarding alcohol's possible role in causing death. RESULTS: Among the 1324 deaths certified by the office, 105 (8%) had alcohol use reported on the death certificate. The majority (67%) of these cases were natural deaths. Sixty-nine (5%) deaths had mention of alcohol use in the investigative case notes but did not include it on the death certificate. Twenty-five (2%) deaths had mention of alcohol on the death certificate but did not have mention of it in the investigative case file based on our search criteria. However, subsequent review of additional case follow-up information disclosed a history of alcohol use or acute intoxication in each case. CONCLUSIONS: The data show that more natural deaths are considered to be directly caused by alcohol than other manners of death. For the unnatural manners of death (excluding acute alcohol poisoning), alcohol use is often viewed by medical examiners as an incidental, associated finding or risk factor surrounding the circumstances of death rather than being an actual cause of death. In such cases, alcohol use is often omitted from the death certificate. For deaths directly caused by alcohol, the proportion of cases involving possible underreporting or overreporting of alcohol involvement was relatively small and usually involved the omission of chronic alcohol use from the death certificate. Researchers need to be aware of potential limitations of death certificate data for studying alcohol-related deaths.     

Sudden infant death syndrome: diagnostic practices and investigative policies, 2004

Using a 2004 population-based survey of all US medical examiner and coroner offices, we examined the characteristics of offices accepting an infant death case and calculated the percentage of offices that had death scene investigation or autopsy policies for the investigation of sudden unexpected infant death (SUID). We also calculated the percentage of offices that used and did not use sudden infant death syndrome (SIDS) as a cause of death, and we compared differences in characteristics among those offices.Of medical examiner and coroner offices, 52% did not report an infant death in 2004. Of the 7957 infant deaths reported, 43% occurred in jurisdictions that experienced 1 or 2 infant deaths. Of the offices that used SIDS as a classification, 34% did not have policies for conducting death scene investigations and autopsies for SUID. At least 5% of offices that reported an infant death did not use SIDS as a cause of death classification. These findings have important implications for understanding recent trends in SIDS and SUID. Supporting the implementation of national standards for investigating and certifying infant deaths could provide guidelines for consistent practices in medical examiner and coroner offices.     

Forensic medicine experiences in cremation autopsy in Bonn

2016 coroner's crematorial inquests of the Bonn Institute of Legal Medicine over a 3 years' period were evaluated and the statements in the death certificates were compared to those ones of the Statistical Country Office of Northrhine-Westphalia. No significant differences were found with respect to age, gender and reason of death. However, the number of insufficiently classified deaths with respect to the reason of death were overrepresented especially in cases of older women, while the number of unnatural deaths was generally underrepresented. According to legal medical criteria 2 1/4 per cent of the deaths were falsely classified as natural. Only in 2 cases an autopsy was accomplished. Other author's statements on the validity of death certificates were confirmed.     

Death in sauna

Bathing in sauna is common in Finland, where there are approximately 2 million saunas among the population of 5.2 million. In this paper, deaths occurring while in a sauna in 1990-2002 in Finland were studied by analyzing police and forensic autopsy reports, death certificates, and toxicological results. The annual rate of death occurring while in a sauna was less than 2 per 100,000 inhabitants. Close to half (51%) of the cases were determined to be natural deaths and exposure to heat was the cause of death in 25%. Overall, 50% of all cases were under the influence of alcohol. The main conclusion is that death in the sauna is a rare event even in Finland where the frequency of sauna bathing is high. The role of alcohol as a risk factor has grown. The prevention of these deaths should focus on less drinking of alcohol and avoid leaving a drunken bather alone in the sauna.