Errors In Pediatric Death Certificates Issued By Emergency Departments

Abstract

Background: Errors in pediatric death certificates (DCs) are rarely reported. We analyzed the errors in the pediatric DC issued by an emergency department (ED).

Methods: The DCs issued at the ED to patients below the age of 18 years were retrospectively analyzed. Major and minor errors in the DC were investigated based on the WHO International Statistical Classification of Diseases and Related Health Problems (ICD-10) guidelines and evaluated based on the review of the medical records by four emergency physicians. Manner of death was classified into disease (disease group) and external causes (external group) and that two groups were compared.

Results: Among the 87 cases of DCs issued by the ED, 98% of errors were confirmed in the disease group and 100% in the external group. The total number of errors in the analyzed DC was 2.3 in the disease group and 3.3 in the external group. Blank space in the cause of death (COD) or duplicated COD (18%) were most common major error in the disease group and two or more causes in a single line of COD (36%) in the external group. Among minor errors, the cases without a time interval record for the COD were the most common in 37 cases (93%) and 42 cases (89%) of the two groups, respectively.

Conclusion: One or more errors were identified in 99% of pediatric DCs issued by ED, and the total number of errors was higher in the external group than in the disease group.

Trial registration: none 

Background

Multiple errors have been found in death certificates (DCs).^1-11 A large number of errors in the DC reduces its value as a medical document, and adversely affects the quality of studies and statistics related to the cause of death (COD). Simple training in DC writing and various types of intervention can reduce such errors.^1,2,4,6-11 Most of the studies investigating defective DCs targeted adults.^1-4,6-8 These studies did not consider the characteristics of pediatric DC. In one study that only reviewed the errors in pediatric DC, no medical records were reviewed, thereby reducing the accuracy of determining the COD and the analysis of errors in the DC. Inability to analyze various errors comprehensively is a limitation.⁵ A pediatric DC is also an indicator of health care policy and child mortality statistics, suggesting the need for further investigation, in this study, we analyzed the errors in the pediatric DC issued by an emergency department (ED).

Methods

This study was a retrospective analysis of DCs involving pediatric patients aged below 18 years at the ED of a university training hospital located on the southeastern coast of Korea from 2005 to 2020. It was reviewed by an institutional review board. Errors in the DC were investigated based on the WHO International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) guidelines.¹² Errors in the DC were evaluated based on the review of the medical records of the deceased by three emergency physicians who are currently working in the ED and experienced in writing a DC. An experienced senior emergency physician reviewed the conflicting evaluations. In the case of disagreement even after the second review, a final consensus was reached by 4 emergency physicians. DCs issued with an unknown or unclear COD even after the final consensus were excluded from the study. Among the 2309 DCs issued from 2005 to 2020 at the ED, 87 DCs involving pediatric cases under the age of 18 years were finally included after excluding 13 DCs with unknown COD.

Errors were divided into major and minor errors according to the criteria described in the previous study.^3-5,8,9,13 The major errors were defined as follows. (1) The mode of death, such as cardiac arrest, heart failure, or respiratory failure, was recorded as the underlying cause of death (UCOD). (2) Secondary conditions such as sepsis or gastrointestinal bleeding are recorded as the UCOD without an explanation of the preceding cause. (3) Only uncertain conditions (ICD-10 codes R00-R94, R96-R99.4) such as senility, cachexia, abnormalities, symptoms and signs of clinical or laboratory results are recorded as the UCOD. (4) Incompatible causal relationship exists between two or more causes. (5) Two or more CODs are recorded in a single COD. (6) A blank space is left between the COD or repeated CODs. (7) A case of erroneous judgement involving the manner of death was classified as disease and external causes. (8) Unacceptable COD was supported by evidence based on illogical decision (Table 1).

Minor errors were defined as follows. (1) When an appropriate COD is recorded, the mode of death is recorded as a direct COD. (2) In case the appropriate cause (ICD-10 codes V01-Y89) corresponding to the accident was recorded as the external cause, the accident mode, such as traffic accident, fall, and hanging (ICD-10 codes S00-T98), was not included. (3) When surgery was performed for COD, the surgical findings were not recorded. (4) The date of onset was not recorded. (5) When the date of onset was compared with the medical records, the time was recorded incorrectly. (6) The time to death was not recorded for each COD. (7) Other physical conditions are not recorded or recorded inappropriately. (8,9,10) In case of external causes, the type, the intention, the date and time, and the location of the accident were not recorded or were recorded inappropriately (Table 1).

The investigations included general characteristics of the DC, the number of CODs recorded in the DC, the detailed analysis of each error, the number of major and minor errors, and the total number of errors. Manner of death was classified into disease (disease group) and external causes (external group) and that two groups were compared. Frequency analysis, chi-square test, Fisher's exact test, and Student's t-test were used to compare the two groups of disease and external causes. IBM SPSS 24.0 (IBM, Armonk, NY, USA) was used for statistical analysis, and statistical significance was defined as p<0.05.

Results

The 87 cases of DCs issued by the ED included 40 cases (46%) analyzed according to the manner of death by disease (disease group), and 47 cases (54%) analyzed according to the manner of death by external causes (external group). The disease group revealed 98% of errors and the external group showed 100% errors. The disease group included 17 males (43%) and the external group comprised 36 males (77%). The average age of disease group was 5.2 years, which was lower than 10.6 years in the external group. The most common cause of death in children aged less than 1 year was death by disease (53%). Death by external causes accounted for 53% of all deaths in those above 13 years. 

Among the 4 COD record lines, an average of 2.0 record lines were filled in the disease group, and an average of 2.4 record lines were filled in the external group. Only one COD was most common in 16 cases (40%) in the disease group, whereas two CODs were the most common in 16 cases (34%) belonging to external group. The total number of errors in the analyzed DC was 2.3 in the disease group and 3.3 in the external group. The number of major and minor errors in the disease group were 0.6 and 1.7, respectively, compared with 0.7 and 2.7 in the external group (Table 2).

The most frequent major errors in the disease group occurred in 7 cases (18%) reporting a blank space in the COD or a format error such as duplication. Only a secondary condition was recorded as the UCOD without antecedent causes in 6 cases (15%). In the external group, 17 cases (36%) involved writing two or more causes in a single line of COD, and 10 cases (21%) of blank space in the COD or a format error such as duplication were reported. Only the mode of death was recorded in 4 cases (10%) in the disease group and 2 cases (4%) in the external group (Table 3).

In minor errors, the cases of no time interval record for the COD were the most common in both groups with 37 cases (93%) and 42 cases (89%), respectively. Other frequent minor errors such as not recording the onset date and time involved 11 cases (28%) in the disease group and 13 cases (28%) in the external group. The mode of death as the direct COD was recorded despite an appropriate UCOD in 8 cases (20%) of disease group and 9 cases (19%) of external group. Ten cases (21%) in the external group did not include accident-related information as the UCOD although the type of accident was recorded (Table 4).

Discussion

Previous studies failed to analyze various errors and instead only mentioned errors in the pediatric DC.⁵ By comparison, this study divided the manner of death in children into two groups depending on disease and external causes and analyzed various types of error in the DC by closely examining the medical records. One or more errors were found in 86 DCs except for one out of 87 DCs (Table 2). Previous studies confirmed multiple errors involving DCs.^1-11 Several errors were similarly reported in this study. 

A review of some of the main errors indicates that writing of two or more causes in single COD was the most common in 17 cases (36.1%) under the external group. This finding suggests that the principles of writing a DC were not fully understood by doctors. However, no further investigation into DCs written by physicians was conducted. Because it is difficult to determine the priority or importance of the direct causes leading to death due to other reasons, it is presumed that all of them were written down.

The most common minor errors involved the absence of recorded time interval to death for each cause in 37 cases (93%) belonging to the disease group and 42 cases (89%) under the external group. The next most common error was the lack of recorded onset date in 11 cases (28%) belonging to the disease group and 13 cases (28%) under the external group. In the analysis of errors in DC involving older adults, the number of cases without recorded date of onset or the time to death was significant.^3-5,8,9 Doctors usually record the date of death but not the date of onset in many cases of pediatric DCs, probably because it is difficult to know the onset date and time in the case of children and also careless records of the time taken for each COD.⁷ In the case of the time interval, although it is not a critical item in the DC, it is important to fill in the details comprehensively in order to determine the temporal sequence of the causes leading to death and to infer a reasonable causal relationship.

A recent study analyzing errors in DC related to external causes reported that the number of errors increased with the degree of description of COD.³ The number of errors was 2.3 when the COD was written in 2.0 lines in the disease group, and 3.2 when written in 2.4 lines in the external group. In this study, in the absence of substantial number of cases, no significant correlation was detected.

In the case of pediatric death, considering the absence of a substantial number of deaths due to long-term chronic diseases or complicated causal relationships unlike adults, errors involved cases of recording only an uncertain condition as the COD (2 cases (5%) in disease group) or cases of incompatible causal relationship between two or more cases (1 case (2%) in disease group) were few. 

The 21 cases in disease group involved more than 5 cases belonging to the external group of infants aged below one year. Unlike other age groups, the large number of cases in the disease group dying within one year of birth suggests Sudden Infant Death Syndrome (SIDS).¹⁴ A total of 8 cases were due to SIDS among 21 children in the disease group in this study. SIDS refers to unexplained sudden death in infants under 12 months of age, and the COD cannot be explained even after autopsy, a thorough investigation at the time of death, or patient history. A total of 5 infants aged less than one year had unknown cause of COD in the DC and these cases were excluded from this study. However, these 5 cases can also be considered as SIDS. In the case of a DC recorded as unknown, it is classified as a possible COD during the final compilation of the national COD statistics, which may contribute to erroneous statistics. In the case of medical doctors who are authorized to write a DC, based on direct review of medical information related to the COD, it is necessary not to mention an unknown COD and minimize inaccuracies.

In the age group of 13 years or older, there were 25 cases (53%) in the external group, which had a higher than 10 cases (25%) in the disease group. Given the multiple causes of death in school-aged children, the higher number of pediatric DCs related to external causes of death in the hospital is associated with a greater number of errors. Compared with the disease group, external causes such as the location, time, type and intention of the accident contribute to the increased number of secondary errors.

This study included only pediatric DCs issued by the ED of a university hospital, and it is difficult to generalize the results of this study. However, additional studies investigating pediatric DC errors based on the results of this study could have a positive effect on the statistics of the COD and related health policy. Another limitation was that no investigation of factors affecting the writing of DC was conducted, such as the level of physician education, the number of days spent working in the ED, and whether or not an experienced physician was involved in determining the COD.

One or more errors were identified in 99% of pediatric DCs issued by ED, and the total number of errors was higher in the external group than in the disease group. Among major errors, writing two or more signs in a single line of COD was the most common in the external group, and errors such as blank space under the COD or duplication were the most common in the disease group. Among minor errors, failure to record the time interval for each COD was the most common in both groups. The mode of death such as cardiac arrest as the UCOD was recorded in 10% of the patients in the disease group and 4% in the external group. Errors in DC can be minimized by physician education and continuous feedback.

Abbreviations

ED: Emergency department; ICD-10: International Statistical Classification of Diseases and Related Health Problems 10th Revision; DC: Death certificate; COD: Cause of death; UCOD: Underlying cause of death; 

Declaration

Ethics approval and consent to participate

This study was reviewed and approved by Ulsan University Hospital Institutional Review Board (UUH-IRB-2021-06-072). Informed consent was exempted by the Ulsan University Hospital Institutional Review Board due to retrospective study. The data used in this study was anonymised before its use. We confirm that all methods were performed in accordance with the relevant guidelines and regulations. 

Consent for publication 

Not applicable 

Availability of data and materials

The data that support the findings of this study can be provided as a supplementary file.  

Competing Interests

The authors declare that they have no competing interests

None

Author’s contribution

Conceptualization: SHK, BJP 

Data curation: SHK, BJP

Formal analysis: SHK, BJP

Investigation: SHK, BJP

Methodology: SHK, BJP

Project administration: SHK, BJP

Resources: SHK, BJP

Software: SHK, BJP

Supervision: SHK

Validation: SHK, BJP

Visualization: SHK, BJP

Writing—original draft preparation: SHK, BJP 

Writing—review and editing: SHK, BJP

All authors have read and agreed to the contents of the manuscript. 

Acknowledgements

The authors grateful to Dr. Park Soobeom and Dr. Jung Sangyeop for their contribution of collecting data.  

References

1.   Akakpo PK, Awuku YA, Derkyi-Kwarteng L, Gyamera KA, Eliason S. Review of errors in the issue of medical certificates of cause of death in a tertiary hospital in Ghana. Ghana Med J. 2017;51(1):30-35.

2.   Cambridge B, Cina SJ. The accuracy of death certificate completion in a suburban community. Am J Forensic Med Pathol. 2010;31(3):232-235.

3.   Chang JH, Kim SH, Lee H, Choi B. Analysis of Errors on Death Certificate for Trauma Related Death. J Trauma Inj. 2019;32(3):127-135.

4.   Filippatos G, Andriopoulos P, Panoutsopoulos G, et al. The quality of death certification practice in Greece. Hippokratia. 2016;20(1):19-25.

5.   Gupta N, Bharti B, Singhi S, Kumar P, Thakur JS. Errors in filling WHO death certificate in children: lessons from 1251 death certificates. J Trop Pediatr. 2014;60(1):74-78.

6.   Haque AS, Shamim K, Siddiqui NH, Irfan M, Khan JA. Death certificate completion skills of hospital physicians in a developing country. BMC Health Serv Res. 2013;13:205.

7.   Hazard RH, Chowdhury HR, Adair T, et al. The quality of medical death certification of cause of death in hospitals in rural Bangladesh: impact of introducing the International Form of Medical Certificate of Cause of Death. BMC Health Serv Res. 2017;17(1):688.

8.   Maharjan L, Shah A, Shrestha KB, Shrestha G. Errors in cause-of-death statement on death certificates in intensive care unit of Kathmandu, Nepal. BMC Health Serv Res. 2015;15:507.

9.   Myers KA, Farquhar DR. Improving the accuracy of death certification. CMAJ. 1998;158(10):1317-1323.

10.   Schuppener LM, Olson K, Brooks EG. Death Certification: Errors and Interventions. Clin Med Res. 2020;18(1):21-26.

11.   Wood KA, Weinberg SH, Weinberg ML. Death Certification in Northern Alberta Error Occurrence Rate and Educational Intervention. Am J Foren Med Path. 2020;41(1):11-17.

12.   Organization WH. International statistical classification of diseases and related health problems : ICD-10. In. Vol vol.2 instruction manual. 5th ed. Geneva2016.

13.   Lee H, Kim SH, Choi B, Ock M, Park EJ. Concordance between the underlying causes of death on death certificates written by three emergency physicians. Clin Exp Emerg Med. 2019;6(3):218-225.

14.   Office KNS. Infant death and mortality by sex, by age group. In: Korea National Stastical Office; 2005-2019.

Tables

Table 1. The definition of major and minor errors on death certificate

UCOD, underlying cause of death

COD, cause of death

Table 2. Characteristics of pediatric death certificates by manner of death

UCOD, underlying cause of death

COD, cause of death

Table 3. Major errors in pediatric death certificates by manner of death

UCOD, underlying cause of death

COD, cause of death

Table 4. Minor errors in pediatric death certificates by manner of death

UCOD, underlying cause of death

COD, cause of death