Original articleDiversity in death certification: A case vignette approach
Introduction
Death certificates are the building blocks of cause-of-death statistics, which guide decisionmaking for allocation of medical resources. High-quality cause-of-death statistics are therefore essential to ensure that resources are used effectively. Although great effort has been devoted to teaching physicians how to fill out death certificates correctly 1, 2, 3, 4, 5, 6, 7, Maudsley and Williams' review showed high rates of errors and discrepancies between the original certifiers and evaluators in death certification worldwide [8]. Nevertheless, the nature of these errors has not been widely studied.
Most evaluations to date have defined “accurate” death certification according to a single standard, the International Classification of Disease (ICD) code. However, as Moriyama suggested, when evaluating the quality of death certification we should consider not only the extent of disagreement, but also the level of agreement and the reasons for discrepancies [9].
The cause-of-death section of the death certificate is designed according to the concept of underlying cause of death (UCOD). To prevent death due to specific causes, it is necessary to break the chain of events or to effect a cure at some point. From this standpoint the most effective public health objective is to prevent the precipitating cause of death from operating. Therefore, the UCOD has been defined as “(a) the disease or injury which initiated the train of morbid events leading directly to death, or (b) the circumstances of the accident or violence which produced the fatal injury” [10]. The medical practitioner signing the death certificate is responsible for indicating which morbid conditions led directly to death and stating any antecedent conditions giving rise to this cause [10]. Nevertheless, in everyday clinical situations, determining the causal sequence of death is not this simple.
In this study, we looked at the nature of errors in death certificate completion. To explore the pattern of variation in death certification by level of complexity of the UCOD, we asked a group of physicians attending a continuing medical education course to complete the cause-of-death section on dummy death certificates of four case vignettes. We further explored the associations between the characteristics of the physicians (experience with and knowledge of death certification) and two important measures of certification quality: correct certification format and concordance with referent UCOD. Our objective was to determine if variations in certification among physicians are due to lack of knowledge, as many previous studies assumed, or to differences in interpreting the information and selecting the UCOD.
Section snippets
Case vignettes
The four case history vignettes used in this study were modified from the Physicians' Handbook on Medical Certification of Death [2]. The vignettes represented four levels of complexity in determining the UCOD: Case A, an acute condition without comorbidity; Case B, a chronic disease with multiple complications; Case C, an acute condition of one prominent chronic disease with two mild comorbidities; and Case 4, two competing prominent diseases.
Results
Of the 145 physicians, 124 (85.5%) agreed to participate in the study; however, only 121 physicians filled out all four dummy death certificates. Comparison of the characteristics of participating physicians and general physicians in central Taiwan revealed that the study participants were older and more likely to have non-medical school training (i.e., they had received their medical training in the military).
Table 2 shows the characteristics, including experience in and knowledge of death
Discussion
The physicians in this study showed great variation in wording and diagnostic semantics in death certification, given the same case history information. The rates of correct certification format did not show great diversity; nevertheless, the rates of agreement with the referent UCOD varied, to a large extent, with the level of complexity in UCOD determination. Physicians showed the greatest diversity in choosing between an acute condition of a chronic disease and the chronic disease per se,
Conclusion
We conclude that primary source of diversity in death certification, especially selection of UCOD, was differences in interpreting the information rather than differences in knowledge of death certification. Thus, the traditional concept of UCOD tabulation and using a single standard ICD code in evaluating the quality of death certification may be an oversimplification of a complex situation. Physicians showed the greatest diversity in choosing between an acute condition of a chronic disease
References (29)
- et al.
Factors associated with errors in death certificate completiona national study in Taiwan
J Clin Epidemiol
(2001) - et al.
Problems in the classification of cause of death diagnoses affecting the reliability of mortality statistics for rheumatoid arthritis
J Chron Dis
(1985) - et al.
The WHO principles for registering causes of deathsuggestions for improvement
J Clin Epidemiol
(1990) Medical Certification of Cause of DeathInstructions for Physicians on Use of International Form of Medical Certificate of Cause of Death
(1979)Physicians' Handbook on Medical Certification of Death
(1987)The medical cause of death manualinstructions for writing cause of death statements for deaths due to natural causes
(1994)- et al.
Cause of deathproper completion of the death certificate
JAMA
(1987) - et al.
Death certificationpurposes, procedures, and pitfalls
West J Med
(1989) Protocol for writing cause-of-death statements for deaths due to natural causes
Arch Intern Med
(1996)- et al.
Certification of death by family physicians
Am Fam Phys
(1997)
‘Inaccuracy’ in death certification—where are we now?
J Public Health Med
Evaluation of medical certification of causes of deathreview and critique of alternative approaches for a national program
Manual of the international statistical classification of diseases, injuries, and causes of death. Vol. 1
1998 directory of physicians in Taiwan
Cited by (43)
Forensic pathology: overview
2022, Principles of Forensic Pathology: From Investigation to CertificationPalliative Local Surgery for Locally Advanced Breast Cancer Depending on Hormone Receptor Status in Elderly Patients
2019, Clinical Breast CancerCitation Excerpt :The primary outcome was overall survival (OS) estimated using the Kaplan–Meier method. Breast cancer-specific mortality was not included as an outcome because of its questionable validity as reported by SEER.3,4,23-25 Unadjusted associations of variables among no surgery, lumpectomy, and mastectomy groups were tested using Pearson χ2.
Medical death certification by forensic physicians in the Netherlands: Validity and interdoctorvariation
2016, Journal of Forensic and Legal MedicineOmission of surgery in elderly patients with early stage breast cancer
2013, European Journal of CancerCitation Excerpt :Furthermore, patients were frequently referred back to their primary care physician for follow-up and thus, we had to rely on the death certificate to ascertain the clinical relevance of the breast cancer at the time of death. In addition, establishing the cause of death in elderly patients with multiple interacting comorbidities can be susceptible to errors.32,33 Despite these issues, we believe that this real-life patient cohort provides an interesting and relevant supplement to clinical trial data.
Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: A systematic analysis for the Global Burden of Disease Study 2010
2012, The LancetCitation Excerpt :Vital registration systems that include medical certification of the cause of death captured about 18·8 million deaths of an estimated annual total of 51·7 million deaths in 2005, which is the latest year for which the largest number of countries (100) reported deaths from a vital registration system. Even for these deaths, the comparability of findings on the leading causes of death is affected by variation in certification skills among physicians, the diagnostic and pathological data available at the time of completing a death certificate, variations in medical culture in choosing the underlying cause, and legal and institutional frameworks for governing mortality reporting.1–5 For the remaining deaths that are not medically certified, many different data sources and diagnostic approaches must be used from surveillance systems, demographic research sites, surveys, censuses, disease registries, and police records to construct a consolidated picture of causes of death in various populations.
Trends in death rate from diabetes according to multiple-cause-of-death differed from that according to underlying-cause-of-death in Taiwan but not in the United States, 1987-2007
2012, Journal of Clinical EpidemiologyCitation Excerpt :However, as the population ages, many patients with diabetes die with coexisting cardiovascular diseases or other chronic diseases, which makes determination of the UCOD difficult. Questionnaire survey studies have revealed that, given the same case history, some physicians report diabetes in part I of the death certificate but others enter diabetes in part II of the death certificate [2–4]. One nationwide survey study in Taiwan indicated that the percentage of cases in which diabetes was assigned as the UCOD varied when diabetes coexisted with different complications or comorbidities among internists [4].