Journal:Introductory evidence on data management and practice systems of forensic autopsies in sudden and unnatural deaths: A scoping review

Full article title	Introductory evidence on data management and practice systems of forensic autopsies in sudden and unnatural deaths: A scoping review
Journal	Egyptian Journal of Forensic Sciences
Author(s)	Prahladh, Salona; van Wyk, Jacqueline
Author affiliation(s)	Inkosi Albert Luthuli Central Hospital, University of KwaZulu-Natal
Primary contact	Email: prahladhs at ukzn dot ac dot za
Year published	2022
Volume and issue	12
Article #	38
DOI	10.1186/s41935-022-00293-3
ISSN	2090-5939
Distribution license	Creative Commons Attribution 4.0 International
Website	https://ejfs.springeropen.com/articles/10.1186/s41935-022-00293-3
Download	https://ejfs.springeropen.com/counter/pdf/10.1186/s41935-022-00293-3.pdf (PDF)

Abstract

Background: The investigation into sudden unexpected and unnatural deaths supports criminal justice, aids in litigation, and provides important information for public health, including surveillance, epidemiology, and prevention programs. The use of mortality data to convey trends can inform policy development and resource allocations. Hence, data practices and data management systems in forensic medicine are critical. This study scoped literature and described the body of knowledge on data management and practice systems in forensic medicine.

Methods: Five steps of the methodological framework of Arksey and O’Malley guided this scoping review. A combination of keywords, Boolean terms, and medical subject headings was used to search PubMed, EBSCOhost (CINAHL with full text and Health Sources), Cochrane Library, Scopus, Web of Science, Science Direct, WorldCat, and Google Scholar for peer review papers in English from June 18–24 of 2020, with an updated search also occurring in November 2021. This study included articles involving unnatural deaths, focused on data practice or data management systems, relating to forensic medicine, all study designs, and published in English. Screening, selection, and data extraction were conducted by two reviewers. Thematic analysis was conducted, and the results were reported using both quantitative and qualitative methods.

Results: Of the examined 23,059 articles, 16 met this study’s inclusion criteria. The included articles were published between 2008 and 2019. Eight of the 16 articles were published between 2017 and 2019. Most of the included studies were conducted in the United States (5) and Australia/New Zealand (4). Only two publications were from lower- and middle-income countries (LMICs; Nigeria and Mexico), and the remaining 14 were from high-income countries (Italy, Denmark, US, Australia, New Zealand, Japan, Switzerland, and Canada). The data management systems found in this study were as follows: Virtopsy, Canadian Coroner and Medical Examiner Database, Infant Injury Database, Ibadan pilot fatal injury surveillance system, Medical Examiners and Coroners Alert System, National Violent Deaths Reporting System, AM/PM Database, Tokyo CDISC/ODM, and National Coronial Information System.

Conclusions: This study’s results revealed limited articles relating to data management and practice systems in forensic medicine—particularly in LMICs—indicating there is a prevalence of unnatural deaths in LMICs. This study, therefore, recommends research on data management and practice systems relating to forensic medicine in LMICs to inform policy decisions.

Keywords: unnatural death, data practice, data management systems, autopsy, post-mortem examination, forensic medicine, pathology

Background

The global burden of trauma, particularly in low- and middle-income countries (LMICs) places a large strain on resources, and therefore, the diagnostic value of autopsies must be reiterated. [Salona Prahladh 2018] The use of autopsies remains the gold standard in assessing standards of medical care. There is a concerning decline in autopsies even though their value to the medical fraternity is acknowledged. [Aase 2013; Bagher et al. 2015] Forensic medicine and forensic pathology apply scientific and medical knowledge to inquests, and the autopsy is frequently regarded as the focus of the death investigation. The investigation into sudden unexpected and unnatural deaths supports criminal justice, aids in litigation, and provides important information for public health, including surveillance, epidemiology, and prevention programs. [Bagher et al. 2015; Tseng et al. 2018; Barbería et al. 2018; Pan et al. 2019; Soto Martinez et al. 2019] The evidence serves to inform policy not only for injury prevention and control, but also to prevent suicide, violence, or substance abuse. [Barbería et al. 2018; Pan et al. 2019; Rao et al. 2005; Grills et al. 2011; Prinsloo 2019; Willcox et al. 2020]

Globally, death investigations are conducted according to prevailing legislation, which differs from country to country. Historically, the coroner system was formalized into law by England’s King Richard I in 1194, with the first coroners being knights. [Koehler 2016] The coroner system from England was introduced in the 1600s by American colonists, becoming an important part of the death investigation system in what would become the United States of America. However, the role of the office was later reduced to the medicolegal examination of a body and the determination of the cause and manner of death. [Koehler 2016] Throughout the Middle Ages, the functions of the coroner included conducting inquests, attending to and inspecting the dead, and investigating suspicious deaths.

In the US, coroners are generally public officials with minimal to no medical training. Some coroners only serve part-time capacities, and they also often have other full-time employment. The medical examiner system was introduced due to public dissatisfaction with the coroner system, accusations of corruption, and an increased need to have highly trained personnel in the death investigation. [Koehler 2016] This led to the emergence of a separate discipline of forensic medicine, which began in the seventeenth century. [Choo and Choi 2012] The first medical examiner system was introduced in Massachusetts in 1877.

In 1959, the medical subspecialty of forensic pathology was formally certified, and medical examiners were trained in pathology. Forensic pathology is viewed as a subspecialty of anatomical pathology in countries such as Canada and the United Kingdom. In countries such as South Africa and Australia, one may train solely in forensic pathology for a minimum of a year (usually more), with additional training in anatomical pathology. In South Africa, the medicolegal death investigation is conducted primarily in terms of the Inquests Act (Act 58 of 1959). The medicolegal autopsies are performed by medical practitioners, but due to the large annual number of unnatural deaths and the small number of qualified forensic pathologists in South Africa, a large number of these autopsies are performed by colleagues with limited formal training in performing autopsies. [du Toit-Prinsloo and Saayman 2012]

The fundamental essence of forensic pathologists’ work is to investigate and report the cause of death. The importance of reporting the cause of death is reiterated and forms the basis of The Global Burden of Disease study. [Roth et al. 2018] This comprehensive worldwide observational epidemiological study describes mortality and morbidity from major diseases, injuries, and risk factors to health at global, national, and regional levels. Mortality reporting systems can help to prioritize health system investments, track progress towards global development goals, and guide scientific research. [Roth et al. 2018] The Global Burden of Disease study acknowledges the need for wider adoption and improvement of these systems because continuous reporting of cause-specific mortality in many countries represents a success for global health.

Information derived from autopsies has historically been paper-documented, filed, and archived. With the current age of technology, this information can be stored and managed electronically to better ensure reporting that is current, relevant, and contributory to training and service delivery, policy implementation, and social interventions. The current COVID-19 pandemic has accentuated the importance of wireless technology and the use of the internet to transcend normal communications. Due to safety reasons, much work has to be conducted remotely in many business sectors, including the medical sector. General practitioners conducted consultations virtually to adhere to social distancing and safety measures, and telephonic communication and telemedicine became a necessity due to the pandemic. At this current point in time, we are forced to be open-minded to integrate technology into our daily work lives.

This scoping review was conducted to map the evidence on data management and practice systems, their use, benefits, and challenges in forensic medicine. The information gained on the use and availability of digital technologies and their strengths and limitations to collect autopsy data can inform models to suit similar purposes in forensic medicine in LMICs.

Methods

This study’s protocol was developed a priori and published. [Prahladh and van Wyk 2020] This study used the Arksey and O’Malley framework to conduct a scoping review, which includes the following: (i) the research question was identified, (ii) relevant studies were identified, (iii) eligible studies were selected, (iv) the data was charted, and (v) the results were collated and summarized. [Arksey and O'Malley 2005; PRISMA 2018]

Identifying the research question

The main research question was “In the last 10 years, what evidence on data management and practice systems and their benefits and challenges in forensic medicine exist globally?" This study’s population, concept, and context were sudden/unnatural deaths, data practices, and forensic medicine (autopsies or post-mortem examinations) globally, respectively. The research sub-questions were as follows:

1. What evidence exists on data management and practice systems in forensic medicine?
2. What are the reported benefits and challenges of the data management and practice systems used in forensic medicine?

Identifying relevant studies

A systematic international search of both gray literature and published literature was done to retrieve articles relating to data practice, use, benefits, and challenges in forensic medicine. A combination of keywords, Boolean terms, and medical subject headings was used to search PubMed, EBSCOhost (Academic Search Complete, CINAHL with full text, and Health Sources), Cochrane Library, Scopus, Web of Science, Science Direct, WorldCat, and Google Scholar for peer review papers in English from June 18–24 of 2020, with an updated search also occurring in November 2021. Study design limitations were removed. The search strategy was piloted to check the appropriateness of keywords and databases. The results were reviewed by the research team to ensure the validity of the search strategy in PubMed. A manual search was conducted of the references of the included studies, and the World Health Organization (WHO) website was also searched. Each search was adequately documented, as illustrated in Supplementary File 1. The Peer Review of Electronic Search Strategies (PRESS) statement guided this study’s electronic search strategy. [McGowan et al. 2016] All citations were managed using the EndNote X9 reference manager.

Selection of articles and eligibility criteria

The principal investigator conducted the database searches and title screening using this study’s eligibility criteria. The search strategy and screening tools were piloted to calibrate operators, increase consistency, and fine-tune the methods. A second reviewer reviewed the retrieved titles to ensure completeness before the abstract screening. Subsequently, the cleaned EndNote library was shared among the review team after the removal of duplicate titles. Using an electronic screening tool developed in Google forms, two reviewers independently screened the abstracts and full texts and categorized them into “include” or “exclude” categories based on this study’s eligibility criteria. The review team met throughout the screening process and resolved the discrepancies between the two reviewers at the abstract screening stage through discussions until a consensus was reached, though there were no significant disagreements among the reviewers. It was decided that the articles would be selected on a minimum agreement of at least 50% between the two reviewers due to the complex and specialized field the review would entail. The second reviewer, however, resolved the discrepancies between the principal investigator and the third reviewer at the full-text screening phase. The PRISMA flow diagram was used to account for all the articles. The eligibility criteria used in this study are outlined below:

Inclusion criteria

• Studies that involved unnatural deaths
• Studies that focused on forensic medicine (autopsies/post-mortem)
• Articles that reported data practices such as use, benefits, and challenges
• Articles published from 2008 to 2021
• Articles published in the English language
• All study designs

Exclusion criteria

• Articles that do not involve forensic medicine, pathology, and/or autopsies
• Studies with no clear targeted population
• Studies where full-text articles could not be obtained
• Articles reporting photo capture/imaging programs
• Non-English publications

Charting the data

The data for this study were collected using a spreadsheet comprising of the following: bibliographic details, publication year, study design, study setting, data practices relating to forensic medicine, uses, benefits, challenges, and conclusion and recommendations. The form was pilot tested by two reviewers independently, and all discrepancies were resolved before its usage. Finally, two reviewers performed the data extraction using both inductive and deductive approaches. Subsequent discrepancies were resolved through discussion by the review team.

Collating, summarizing, and reporting the results

Thematic content analysis was conducted for this study. The emerging themes and subthemes relating to data practices in forensic medicine were collated, summarized, and reported narratively. However, the bibliographic details of the included studies, such as design and publication year, were reported quantitatively and presented as percentages.

Results

Of the total of 23,059 search yields, 144 were deemed relevant from the title screening phase based on this study’s eligibility criteria. Of the 144 titles, 43 duplicates were identified and removed, and then 74 and 16 articles were removed at the abstract and full-text screening phases, respectively. The 16 articles removed at the full-text screening phase focused on software for clinical forensic medicine, and graph-based document representation models and others did not report the data storage method. Therefore 15 articles were finally deemed eligible for inclusion for data extraction (Fig. 1). Following reviewer indication, an additional article was added to the results after the full-text screening for a total of 16 articles to be included in the study.

Figure 1. PRISMA 2009 flow diagram of the article selection process.

Characteristics of included publications

Table 1 provides a summary of the main characteristics of the articles included. The included articles were published between 2008 and 2019. Eight of the 16 articles were published between 2017 and 2019. [Soto Martinez et al. 2019; Dunstan 2019; Fowler et al. 2018; Hargrove et al. 2018; Dennis et al. 2018; Saar et al. 2017; Hofmeister et al. 2017; Ottaviani et al. 2017] Most of the studies were published in the United States [Soto Martinez et al. 2019; Fowler et al. 2018; Hargrove et al. 2018; Levy 2015; Blair et al. 2016] and Australia/New Zealand [Willcox et al. 2020; Dunstan 2019; Saar et al. 2017; Lyndal et al. 2016; Pearse 2012], with three from European countries. Two publications were from LMICs (Nigeria and Mexico), and the remaining were from high-income countries (Italy, Denmark, United States, Australia, New Zealand, Japan, Switzerland, and Canada (Fig. 1). The article types included brief communication [Ottaviani et al. 2017; Kiuchi et al. 2013], original studies [Dunstan 2019; Saar et al. 2017; Hofmeister et al. 2017; Levy 2015; Aghayev et al. 2008], author manuscript (preprint) [Blair et al. 2016], retrospective descriptive studies [Soto Martinez et al. 2019; Fowler et al. 2018; Hargrove et al. 2018; Dennis et al. 2018; Kipsaina et al. 2015; Colville-Ebeling et al. 2014], a single systematic review [Lyndal et al. 2016], and an annual/government report. [Canada 2012]

Table 1. General characteristics of included studies. Criterion Number Percentage Publication year 2008–2012 2 12.50% 2013–2016 6 37.50% 2017–2019 8 50.00% Publication type Brief communication 2 12.50% Original article/review 5 31.25% Retrospective descriptive 6 37.50% Systematic review 1 6.25% Annual/government report 1 6.25% Author manuscript 1 6.25% Sector/country Nigeria 1 6.25% Multiple countries including Mexico, Central America 1 6.25% Tokyo/Japan 1 6.25% Canada 1 6.25% Australia/New Zealand 4 25.00% Switzerland 1 6.25% Denmark 1 6.25% United States 5 31.25% Italy 1 6.25% Data system coverage National 6 37.50% Almost complete national coverage 4 27.00% Provincial/states 5 31.25% Other/review 1 6.25% Uses Diagnostic models/education 3 18.75% Research 15 93.75% Statistics 6 37.50% Inform policy 5 31.25% Information sharing 5 31.25% Benefits Wide coverage 9 56.25% Contribution to policies and epidemiology 10 62.50% Data standardization 4 27.00% Information security 2 13.00% Anonymization of data 3 20.00% Data quality control 7 47.00% Information sharing 6 37.50% Challenges/limitations Underreporting/low-case numbers 10 67.00% Data quality/validity (missing cases) 8 53.00% Poor information exchange systems 8 53.00% No standardization of data 5 33.00% Ethical considerations 2 13.00% Reporting bias/error/selection bias 4 27.00% Implications of results Beneficial for research, policy making, prevention strategies, information exchange 16 100.00% Level of sophistication: Simple beneficial database 1 7.00% Sophisticated database: Multiple stakeholder involvement therefore beneficial outside forensic medicine-private companies, etc. 5 31.25%

Findings

Table 2 present a summary of these findings. A brief narrative of the finding is reported under the following main themes: data management and practice systems, benefits/uses of the data management and practice systems, and challenges/limitations of the data management and practice systems.

Table 2. Summary of study findings. Author, year Country Data reporting systems Benefits/uses of reporting system Challenges/limitations of reporting system [Aghayev, 2008] Switzerland Virtopsy-repository of autopsy and radiological data Digital and standardized documentation tool for forensic-radiological and -pathological findings and comparison, epidemiological tool, manages archiving and distribution of data, manages continuing research and education, includes tracking tool for quality control, capable of telemedicine, anonymous data passes to the central server Tedious, requires a large amount of time to enter data, and is language-limited [Canada, 2012] Canada Canadian Coroner and Medical Examiner Database, National Coronial Database Centralized source of data, closed cases, thorough analysis of data quality and integrity, standardization of cases, enhances information exchange for policy making Under-coverage (minimal), not completely reflective (not natural cases), may not link toxicology [Kiuchi, 2013] Japan Institutional Database System located in each institute and containing personal information, and the Central Anonymous Database System located in the University Hospital Medical Information Network Pathologists can retain, check, or search personal database use by authorized users; provides a proofreading system, enhancing quality and comprehensiveness of data; supports anonymization of data; hard copy of death certificates; real-time preservation; storage Cases may be identified even after anonymization, informed consent or ethical approval not required for data submission or analysis [Colville-Ebeling, 2014] Denmark Database with authorized access only belonging to Department of Forensic Medicine Security confidentiality maintained with authorized confidential (i.e., social security numbers used) access only; detailed information including autopsy reports, police reports, crime scene reports, and almost complete coverage of region; can be cross-referenced No current validity methods, some information varies on a case-by-case basis [Kipsaina, 2015] Nigeria Ibadan pilot fatal injury surveillance system First known fatal injury surveillance system in Nigeria to formulate injury prevention policies, created using available resources, standard data collection form by one data collector Underreporting due to social circumstances, cultural influences, etc.; collection of data during a short period [Levy, 2015] United States Varied systems: Medical Examiners and Coroners Alert System, ME/Coroner Information Sharing System, NVDRS, and the National Missing and Unidentified Persons System Supports MECAP-9000 product recalls or standard development; MECISP-tried to standardize data; NVDRS-improved research into preventable deaths; NamUs-resolved 9000 cases of missing or unidentified persons Voluntary entry of information; manual or semi-automated; limited resources, no standardization of data collection, no communication or information sharing methods [Lyndal, 2016] Australia and New Zealand National Coronial Database Comprehensive coverage of cases with in-depth detail; reliable, with high-quality data due to consistency with other data sources; a useful tool in death investigation and research on public health and safety, as it helped identify trends in specific death types; valuable to researchers and injury prevention practitioners/policy makers; can identify hazards, inform the development of prevention strategies, and assess their effectiveness; provides accurate estimation of mortality Unavailability of data due to open cases; missing information; coding errors: underreporting of relevant cases, incomplete datasets, and misclassification; inability to detect trends: erroneous reporting of decreasing trends due to small number of cases; selection bias or reporting bias (due to interest in closing the certain high-profile case) [Ottaviani, 2017] Italy Creation of a web portal for a national data bank registry Enhance epidemiological correlations with risk factors to provide further insight into sudden infant death syndrome (SIDS) Requires consent from family [Hofmeister, 2017] Multiple countries including Mexico and Central America Standard reporting form, a software application AM/PM Database, single or multi-user, in two languages Assists in the identification of missing persons in armed conflicts and migration, acts as one consolidated centralized system Coordination of data exchange is difficult; needs training and technical staff because multiple countries involved; infrastructure and funding required; difficult to standardize data [Saar, 2017] Australia and New Zealand Internet database maintained by an IT support team, with nightly or weekly uploading of data Standardization of data, more than 100 ethically approved research or monitoring projects as an ongoing data source, 215 publications, informed manufacturing changes, regulatory changes, awareness campaigns, suicide prevention initiatives, and coronial recommendations Funding required for maintenance and support of multiple agencies (allocation of court resources) [Hargrove, 2018] United States Data is manually entered or imported into Epi Info V.7 with five data entries: death certificate data, coroner report data, autopsy report data, toxicology report data, and prescription drug report data Enhanced surveillance data, data quality improvement, intervention and policy implementation, multi-stakeholder involvement Underreporting of cases by hospital physicians was identified [Fowler, 2018] United States State-based surveillance system; data collected from individual information sources are entered into the NVDRS online data entry system with quality checks, training, and and more quality checks Used to define public health priorities, develop and evaluate programs and policies, and conduct research; online platform with simplified system operations and management, improved timeliness of data entry and reporting, and enhanced flexibility Not nationally representative; availability, completeness, and timeliness of data dependent on partnerships among state health departments; has sharing and communication challenges and incomplete data, with toxicology data not consistently collected; uses different classifications of deaths and different coding; protective factors not collected [Dennis, 2018] Australia National Coronial Database funded by a governmental association Study showed an opportunity to institute preventative measures (CPR training and defibrillators); outcomes improved, possibly due to an increase in witnessed events; looked at preparticipation screening Low number of cases, retrospective, non-standardized data, missed cases [Soto Martinez, 2019] United States The web-based Research Electronic Data Capture (REDCap) platform: migration of the IID to a REDCap platform provided an opportunity to redesign the database to capture internal, external, and skeletal injuries with greater detail Used to develop statistically sound diagnostic models, is reliable, and uses autopsy gold standard Has small number of cases, errors, and complexity of observations; the autopsy sample therefore not a complete complement of injuries [Dunstan, 2019] Australia and New Zealand National Coronial Information System Data sharing with statistical and research data; able to identify mortality trends and formulate effective recommendations in the prevention of death and injury; contributes to health policy and prevention; provides access levels; has ethics application process to obtain data Reports only on closed cases, although nationally standardized information availability may vary; not all fatalities reported; no transcripts, photographic evidence, witness statements, non-fatal injury data, or information on the perpetrator [Blair, 2016] United States NVDRS, created in response to a 1999 Institute of Medicine report outlining the need for a national fatal intentional injury system; the first multistate system to provide detailed information on circumstances precipitating violent deaths, the first to link multiple source documents on violence-related deaths to enable researchers to understand each death more completely, and the first to link multiple deaths that are related to one another (e.g., multiple homicides, multiple suicides, and cases of homicide followed by the suicide of the suspected perpetrator) Detailed circumstantial information regarding homicides and suicides enables research to be conducted to provide evidence basis for prevention programs. State health departments utilize information from the systems to identify areas of need, evaluate state policies and areas requiring intervention to produce targeted solutions or interventions. Improved elder abuse and neglect surveillance and targeted intervention programs As of the publication of this article, the system was in place in 32 states and not national for complete surveillance.

Types of data management and practice systems

All the reviewed studies described the use of electronic systems, which ranged in complexity from a simplified Excel spreadsheet [Kipsaina et al. 2015] to a more complex system based on the use of a web portal [Kiuchi et al. 2013], with varying use at the state and national levels, as with the National Violent Death Reporting System [Blair et al. 2016] and the National Coronial Information System. [Dunstan 2019; Saar et al. 2017; Lyndal et al. 2016] The national system received full technical support. In some cases, such as mentioned in the Nigerian study, data was collected by a single user and was based on tools developed at the same time as the WHO-Monash University fatal injury surveillance manual. [Kispaina 2015] It was found to be acceptable and timely with good data quality, which was representative and specific due to a single experienced user. Other systems sourced data from multiple users and departments such as a forensic medicine department, the police, and sychologists. Many of the systems were consolidated centralized systems that were used in forensic medicine departments (n= 10; 62.5%). [Dunstan 2019; Saar et al. 2017; Hofmeister et al. 2017; Ottaviani et al. 2017; Blair et al. 2016; Lyndal et al. 2016; Pearse 2012; Aghayev et al. 2008; Canada 2012] Access to all the systems was strictly controlled either through single or multi-user authorization [Dunstan 2019; Hofmeister et al. 2017; Lyndal et al. 2016; Kiuchi et al. 2013; Aghayev et al. 2008; Colville-Ebeling et al. 2014], or data are captured in a spreadsheet created in Excel. [Kipsaina et al. 2015] The articles based on systems that reported the best utility were those that allowed greater flexibility and lend themselves to state-based use. [Fowler et al. 2018; Blair et al. 2016] In these cases, the systems received government funding and technical support. [Hofmeister et al. 2017; Blair et al. 2016; Lyndal et al. 2016; Aghayev et al. 2008; Colville-Ebeling et al. 2014; Canada 2012]

Benefits/uses of the reported data management and practice systems

In nine (56%) of the articles, the wide coverage of cases increased the value of the system for those who used the data. [Dunstan 2019; Fowler et al. 2018; Hargrove et al. 2018; Saar et al. 2017; Hofmeister et al. 2017; Ottaviani et al. 2017; Blair et al. 2016; Colville-Ebeling et al. 2014; Canada 2012] The benefits reported included the formulation of injury prevention policies and enhancement of epidemiological studies (n= 10; 62.5%). [Dunstan 2019; Fowler et al. 2018; Hargrove et al. 2018; Saar et al. 2017; Ottaviani et al. 2017; Levy 2015; Blair et al. 2016; Lyndal et al. 2016; Colville-Ebeling et al. 2014; Canada 2012] Many of the cited systems contributed and supported research (n= 10; 62.5%). [Arksey and O'Malley 2005; Dunstan 2019; Fowler et al. 2018; Saar et al. 2017; Ottaviani et al. 2017; Levy 2015; Blair et al. 2016; Lyndal et al. 2016; Colville-Ebeling et al. 2014; Canada 2012] Although only a few systems had standardization of data as an attribute (n= 4; 27%) [Fowler et al. 2018; Hofmeister et al. 2017; Aghayev et al. 2008; Kipsaina et al. 2015], two reported the benefits of information security (n= 2; 13%) [Kiuchi et al. 2013; Colville-Ebeling et al. 2014] and three (20%) included the anonymization of data as a characteristic. [Kiuchi et al. 2013; Aghayev et al. 2008; Colville-Ebeling et al. 2014] Some of the systems mentioned the use of their own data quality control measures (n= 7; 47%). [Dunstan 2019; Hargrove et al. 2018; Saar et al. 2017; Blair et al. 2016; Lyndal et al. 2016; Kiuchi et al. 2013; Canada 2012] Only six (37.5%) of the reported systems allowed for ease of data sharing/information exchange for research and policy implementation purposes. [Dunstan 2019; Saar et al. 2017; Blair et al. 2016; Lyndal et al. 2016; Aghayev et al. 2008]

The data management and practice systems served as repositories for information to inform about international best practices and supported the development of diagnostic models. [Soto Martinez et al. 2019; Levy 2015; Fowler et al. 2018] The systems were also seen as a conduit for information sharing once privacy challenges were tackled. [Levy 2015; Colville-Ebeling et al. 2014; Aghayev et al. 2008; du Toit-Prinsloo and Saayman 2012] The provision of data for research was a large contribution to most of the databases' establishment. [Soto Martinez et al. 2019; du Toit-Prinsloo and Saayman 2012; Dennis et al. 2018; Dunstan 2019; Fowler et al. 2018; Lyndal et al. 2016; Saar et al. 2017; Levy 2015; Pearse 2012; Kiuchi et al. 2013; Aghayev et al. 2008; Kipsaina et al. 2015; Colville-Ebeling et al. 2014] The systems contributed to the institution of preventative measures by identifying risk factors and by predicting the possible outcomes by using the data available in the reporting systems. [du Toit-Prinsloo and Saayman 2012; Fowler et al. 2018; Blair et al. 2016; Lyndal et al. 2016] Improvement of surveillance and statistics related to epidemiology was notable in some studies [Saar et al. 2017; Ottaviani et al. 2017; Blair et al. 2016], and this further informed policy changes, as well as contributed to the institution of new policies and the reformation of laws. [Bagher et al. 2015; Lyndal et al. 2016; Saar et al. 2017; Blair et al. 2016; Pearse 2012]

Challenges/limitations of the data management and practice systems

In one of the systems included for review, the authors described a voluntary data capture process. This however led to inconsistencies, underreporting, and time lag that impacted the usability of the data for research and epidemiology. [PRISMA 2018; Dunstan 2019; Fowler et al. 2018; Hofmeister et al. 2017; Kipsaina et al. 2015] Some of the other challenges reported (n= 8; 53%) in service work and research included underreporting, low-case numbers, or use of closed cases, only which generally impacted on epidemiology and research. [Soto Martinez et al. 2019; Hargrove et al. 2018; Dennis et al. 2018; Saar et al. 2017; Blair et al. 2016; Lyndal et al. 2016; Kipsaina et al. 2015; Canada 2012] In the articles that cited data quality and validity challenges (n= 8; 53%), missing information was also reported as a limitation. [Dunstan 2019; Fowler et al. 2018; Saar et al. 2017; Levy 2015; Blair et al. 2016; Lyndal et al. 2016; Colville-Ebeling et al. 2014] Data exchange and entry limitations were also reported as issues (n= 8; 53%) [Dunstan 2019; Fowler et al. 2018; Saar et al. 2017; Hofmeister et al. 2017; Levy 2015; Lyndal et al. 2016; Aghayev et al. 2008], and these barriers may in part be due to an inability to institute standardization practices for data capturing (n= 5; 33%). [Fowler et al. 2018; Dennis et al. 2018; Hofmeister et al. 2017; Levy 2015; Blair et al. 2016] Ethical considerations such as consent for autopsy or submitting data for analysis and research also impacted the use of the systems (n= 2; 13%). [Ottaviani et al. 2017; Kiuchi et al. 2013] Factors such as case-reporting bias, reporting errors, and bias in case selection also impacted case numbers and research (n= 4; 27%). [Soto Martinez et al. 2019; Blair et al. 2016; Lyndal et al. 2016; Canada 2012]

Discussion

This study sought to map the evidence on data management and practice systems in forensic medicine. The results show that several data management and practice systems exist. However, most of the existing systems were from high-income countries, with few found in LMICs based on this study’s eligibility criteria. This review was informed by the need to collect systematic evidence on the data management and practice systems being used in forensic medicine and the possible lessons and applicability to LMIC contexts where paper reporting is still being used. The lack of information on data management and practice systems is visible in that only two of the articles included for review reported on studies from the LMIC context. Half the publications (50%) were published during and after 2017. This trend may be due to the greater access, use, and reporting of electronic data reporting systems. The inclusion of forensic pathology reports on databases also captures information that is not routinely captured in vital registration statistics. The existing data management systems in forensic medicine reported in the included articles have several benefits. Nonetheless, the included articles also reported several challenges about those existing data management systems (see Fig. 2).

Figure 2. Countries where publications originated.

To the best of our knowledge, this scoping review is the first present evidence on data management and practice systems in forensic medicine. Therefore, we cannot compare our findings. Nonetheless, the literature shows that electronic data reporting systems are relevant and were developed from the recognition of coronial data as not only a part of the death investigation but as a contributor to preventable death research and public health initiatives. For instance, in an article by Bruce Levy, which discussed the United States' systems currently in place to support forensic pathology and death investigation, the Centers for Disease Control and Prevention (CDC) implemented the National Violent Death Reporting System in response to a report describing the need for a national fatal intentional injury surveillance system. Initially, the system started in six states but later was expanded to 18 states, as stated in the publication. [Fowler et al. 2018] According to the CDC website, and an article published in 2019 regarding its future directions, the program has now expanded to all 50 states and is constantly being updated and improved for data sharing. [Blair et al. 2016; CDC, 2012] Other systems developed in the US included a state-wide comprehensive multisource drug overdose fatality surveillance system in Kentucky (developed in response to drug overdoses cited as a public health crisis) and a database related to infant and child abuse. [Soto Martinez et al. 2019; Hargrove et al. 2018] The burden of sudden infant death syndrome (SIDS) and the large number of cases that remain unexplained led to the passing of legislation in Italy (2006) that fetuses and infants, from 25 weeks of gestation to one postnatal year, who died suddenly and unexpectedly should be sent to the University of Milan, Italy, for a postmortem, with parental consent. [Ottaviani et al. 2017] An Italian research center developed a web portal for a national bank registry which has been set up to centralize records retrieved from regions across Italy, which hopes to contribute data for epidemiology and study into risk factors for SIDS and other sudden unexpected deaths in infants.

Other examples were found. For example, Canada instituted a National Coroner and Medical Examiner database to detect emerging trends and hazards for the prevention of avoidable deaths. [Canada 2012] Tokyo is recognized as a technology hub, and the latest inclusion of a legal medicine information system for forensic systems is discussed in an included article. [Kiuchi et al. 2013] Using the information system, forensic pathologists and other staff can register and search for institutional autopsy information, print death certificates, and extract data for research and analysis. Switzerland created a tool called Virtopsy, a centralized database in forensic medicine for analysis and comparison of radiological and autopsy findings. [Aghayev et al. 2008] It is a database currently created but not in routine use, as it has not been validated. The database compares autopsy and radiological data with photograph storage. In the Nigerian study, recognition of the poorly representative mortality injury surveillance system prompted the authors to institute an electronic injury surveillance system. [Kipsaina et al. 2015] The system included features based on a separate South African initiative, namely the National Injury Mortality Surveillance System (NIMSS), which due to logistical reasons and lack of funding was deemed unfeasible and discontinued. (Prinsloo 2019) It should be noted that articles related to NIMSS did not meet the inclusion criteria of the current study. While the NIMSS-based tools that were utilized were already existing, making the system feasible and sustainable, an appropriate infrastructure needed to be in place to maintain the system. [Prinsloo 2019]

Notes

This presentation is faithful to the original, with only a few minor changes to presentation, grammar, and punctuation. In some cases important information was missing from the references, and that information was added. The original article lists references in alphabetical order; this version lists them in order of appearance, by design.