Abstract

This study aims to identify information systems having fatal work-related (ATF) data in Brazil, describing their characteristics, flows and barriers to information quality. Using a documental research approach, we found: the Mortality Information System (SIM), the Hospital Admission Register from the Unified Health System (SIH-SUS), the Notifiable Diseases Information System (SINAN) and the Violence and Injuries Surveillance Program (VIVA) from the Health Ministry; the Work-related Injuries Reporting System (SISCAT) of the Ministry of Social Insurance; and the Annual Report of Social Information (RAIS), Ministry of Labour and Employment. A lack of key common variables limits the construction of a single database composed by all ATF recorded cases. From several barriers identified, the most relevant for data quality was the lack of work-relatedness recognition and recording, a task performed by the health team.

Fatal work-related injuries; Information systems; Brazil

Introduction

Fatal work-related injuries (ATF) have a significant impact on mortality rates, are preventable, and considered an important public health problem. They therefore, require precise recording and related information needs to be employed for planning and management purposes. Every year, approximately two million workers die in work-related injuries globally¹1. Takala J, Hämäläinen P, Saarela KL, Yun LY, Manickam K, Jin TW, Heng P, Tjong C, Kheng LG, Lim S, Lin GS. Global estimates of the burden of injury and illness at work in 2012. J Occup Environ Hyg 2014; 11(5):326-337., while in 2012, the corresponding mortality rate in Brazil was estimated as 7/100,000 workers²2. Souza CAV, Franco Netto G, Machado JMH, Sales LBF, Costa MS, Peres MC, Santana VS. Saúde do trabalhador: informações sobre acidentes, violências e intoxicações exógenas relacionadas ao trabalho, Brasil 2007 a 2012. In: Brasil. Ministério da Saúde (MS). Saúde Brasil 2012: uma análise da situação de saúde e dos 40 anos do Programa Nacional de Imunizações. Brasília: Editora do Ministério da Saúde; 2013. p. 329-344.. These injuries are monitored in several information systems with data based on three conceptual dimensions: 1) the type of injury, characterised as “lesions, poisoning and other consequences of external causes”, which corresponds to the International Classification of Diseases, 10^th Review (ICD-10) codes of Chapter XIX (S and T), replaced by Chapter XX (V, X and Y) named External Causes; 2) a causal relationship with work (typical) or when commuting; and 3) the injury outcome that could be death.

The information systems of interest for ATF identification are those from the social insurance, labour and employment, and health institutions³3. International Labour Organization (ILO). Safety in numbers pointers for global safety culture at work. Geneva: ILO; 2003.. Information systems from social insurance institutions are the most commonly used for monitoring and research of work-related injuries, because of the availability of distinct occupational compensation benefits. In countries where social insurance is universal under the state responsibility or have wide coverage, data from their information systems may represent all workers³3. International Labour Organization (ILO). Safety in numbers pointers for global safety culture at work. Geneva: ILO; 2003.,⁴4. European Agency for Safety and Health at Work (EASHW). European Agency for Safety and Health at Work – annual report, 2003. Bilbao: EASHW; 2004.. However, in countries in which a large number of workers are not covered by social insurance, information from these systems is limited. Information systems from labour and employment protection institutions may contain ATF data records⁵5. Kang S-K, Kwon O-J. Occupational injury statistics in Korea. Saf Health Work 2011; 2(1):52-56.,⁶6. Varakina ZL, Vyazmin AM, Sannikov AL, Nygard CH, Grjibovski AM. Fatal occupational injuries in the Arkhangelsk region, northwest Russia. Occup Med 2010; 60(6):470-475.. National health information systems are commonly universal and may have data that can be used to estimates vital statistics, such as mortality rates. Death certificate are the most well-known source of information, which may contain specific boxes to gather data on the work-relatedness of the cause of death. Other common ATF data sources are repeated surveys conducted on national samples, of complementary nature, which can be used to assess validity of compulsory employers-based information, proven to be commonly unreliable⁷7. England. Health and Safety Executive (HSE). The Health and Safety Executive Annual Report and Accounts 2012/13. London: The Stationery Office; 2013..

In Brazil, several information systems include ATF records, which quality and coverage have been improving, particularly in recent decades⁸8. Chagas AMR, Servo LMS, Salim CA. Indicadores da saúde e segurança no trabalho: fontes de dados e aplicações. In: Instituto de Pesquisa Econômica Aplicada (IPEA). Saúde e Segurança no Brasil: aspectos institucionais, sistemas de informação e indicadores. Brasília: IPEA; 2011. p. 289-328., in spite of their limited use in research or surveillance. Findings from a literature review shown that the availability of ATF data is not widely known, causing negligence respect to data quality and low recording levels⁹9. Facchini LA. Sistema de informação em saúde do trabalhador: desafios e perspectivas para o SUS. Cien Saude Colet 2005; 10(4):857-867.. In this study, we intend to improve the knowledge about the available information systems that record ATF data in Brazil, identifying and describing these systems, their quality and coverage limits.

Methods

This is a documentary study carried out with texts about information systems, managed by Brazilian public institutions, in which data about work injuries are recorded. First, a list of institutions of interest was created, comprised by the National Social Insurance Institute, the Ministry of Social Security, the Ministry of Labour and Employment, the Ministry of Health, and the Ministry of Justice. To each of them, official websites were searched for access to information systems, manuals, guides, databases and requirements. We also consulted their data collection instruments, information flows and tools for tabulation and graphic interfaces.

The analytical categories were: the institution; the reference population – people whose data are entitled for the information system; sources – instruments used to input data into the information systems; type of access – public or restricted; formats – types of files available; and time covered – in years. Each source was searched for data useful to identify ATF: 1) ICD code; 2) injury work-relatedness; 3) death as outcome. To facilitate our analysis these systems were classified as: a) non-specific to work injuries; and b) specific or exclusive to work-related injuries. The analysis was based on the organization of information into tables and the creation of a flowchart showing common steps starting with the injury occurrence and ending with the ATF recording in the information systems. To each stage, potential filters and barriers to case identification and recording were identified and described.

The project was registered on the National Research Ethics System, Plataforma Brasil, and approved by the Internal Review Board of the Institute of Collective Health, Federal University of Bahia.

Results and Discussion

Information systems that record ATF data

In Brazil, there are eight information systems that contain ATF data. Five were non-specific for work-related injuries: 1) the Mortality Information System (SIM); 2) the Hospital Medical Records of the Unified Health System (SIH-SUS); 3) two subsystems of the compulsory Notifiable Diseases Information System (SINAN), for Exogenous Poisoning and the other for Violence and Injuries Surveillance Program (VIVA), all managed by the Ministry of Health; 4) under the Ministry of Labour and Employment, the Annual Report on Social Information (RAIS) contains deaths data from registered workers when active (Table 1). Three systems were specific and restricted to work-related injuries: 1) two were from SINAN, Ministry of Health, the Severe Work Injuries and Work Injuries with Potential Exposure to Biological Materials; and 2) from the Social Security Ministry, the Work Injuries Communication System (Siscat) (Table 2). The steps where information is generated and their corresponding flows are presented in Figure 1.

Information systems non-specific to work-related injuries

The Mortality Information System (SIM)

The SIM exclusively provides data about deaths and compose, with other information systems, the country´s vital statistics database from death certificates (DO). These documents enable the identification of ATF by checking ICD codes for the underlying and associated causes of deaths. For external causes of deaths, specifically, data on the “probable circumstances” are required to be registered in the following fields: type – whether death was related to an accident (injuries), suicide, homicide or other; and whether it was work-related (yes/no/unknown) (Table 1). The universal nature of SIM is its main advantage, because all workers regardless their type of job contract are covered, even those having informal jobs, the military and public officers, enabling comparisons of AT mortality estimates across countries. Other SIM advantage is to have data on occupation coded by the Brazilian Occupation Classification (CBO), based on the International Standard Classification of Occupations (ISCO). Unfortunately, trade or type of position in the labour force, whether formal or informal, is not available.

Over the last decade, SIM coverage has been increasing, varying from 87.0% in 2000 to 96.1% in 2011¹⁰10. Brasil. Ministério da Saúde (MS). Sistema de Informações sobre Mortalidade – SIM – Consolidação da base de dados de 2011. Brasília: MS; 2012., classified as of intermediary quality by the World Health Organization¹¹11. .Pan American Health Organization (PAHO). Health statistics from the Americas 2006, edition - special topic the ten leading causes of death in countries of the Americas - registered mortality. Washington: PAHO; 2006.. The SIM recording quality is also improving, as shown by the decreasing of deaths registered with ill-defined causes, falling from 7.2% in 2009 to 6.7% in 2011¹⁰10. Brasil. Ministério da Saúde (MS). Sistema de Informações sobre Mortalidade – SIM – Consolidação da base de dados de 2011. Brasília: MS; 2012.. The introduction of a specific section to register data about non-natural deaths (external causes) on death certificates in 1999, with a field for data on work-relatedness, represented a considerable advance for ATF recording. However, this field completion is low, 20% average between 2000 and 2010¹²12. Santana VS, Dias EC, Oliveira GL, Moura MCP, Nobre LCC, Machado JMH. Accidentes de trabajo fatales y violência interpersonal em Brasil, 2000-2010. Salud Colect 2013; 9(2):139-149.. A field for work-relatedness was also incorporated into the death certificates in the United States¹³13. Feyer AM, Williamson AM, Stout N, Driscoll T, Usher H, Langley JD. Comparison of work related fatal injuries in the United States, Australia, and New Zealand: method and overall findings. Inj Prev 2001; 7(1):22-28.,¹⁴14. Marsh SM, Jackson LL. A comparison of fatal occupational injury event characteristics from the Census of Fatal Occupational Injuries and the Vital Statistics Mortality System. J Safety Res 2013; 46:119-125., a strategy that could be used worldwide, especially in countries with large ATF underreporting.

Hospital Medical Records of the Unified Health System (SIH-SUS)

The SIH/SUS is other non-specific information system that takes records of ATFs¹⁵15. Brasil. Portaria GM/MS Nº 1969, de 25 de outubro de 2001. Dispõe sobre o preenchimento de Autorização de Internação Hospitalar - AIH, em casos de quadro compatível com causas externas e com doenças e acidentes relacionados ao trabalho. Diário Oficial da União 2001; 26 out.,¹⁶16. Brasil. Ministério da Saúde (MS). Manual técnico operacional do sistema SIH – Sistema de Informação Hospitalar do SUS. Brasília: MS; 2015.. Although limited to SUS hospital care, excluding private ones, the SIH-SUS comprises 70% of the country total hospital admissions¹⁷17. Bittencourt AS, Camacho LAB, Leal MC. O sistema de Informação Hospitalar e sua aplicação na saúde coletiva. Cad Saude Publica 2006; 22(1):19-30.,¹⁸18. Melione LPR, Mello-Jorge MHP. Gastos do Sistema Único de Saúde com internações por causas externas em São José dos Campos, São Paulo, Brasil. Cad Saude Publica 2008; 24(8):1814-1824.. Its data source is a required document, for funding/reimbursement purposes, named Hospital Admittance Authorization (AIH), in which are registered: 1) ICD codes of the main and secondary diagnosis; 2) the work-relatedness and type (whether it occurred “in the workplace/on duty” or “commuting”); and 3) death occurred at the hospital. In 2001, these fields were updated¹⁵15. Brasil. Portaria GM/MS Nº 1969, de 25 de outubro de 2001. Dispõe sobre o preenchimento de Autorização de Internação Hospitalar - AIH, em casos de quadro compatível com causas externas e com doenças e acidentes relacionados ao trabalho. Diário Oficial da União 2001; 26 out. (Table 1) and, of interest to workers’ health, these other ones were added: 1) occupation coded with the Summarized Brazilian Occupations Classification (CBOR), industries, which are coded with the Industry National Classification (CNAE), the firm registration number at the National Register of Juridical Person (CNPJ), and the “social insurance coverage status” (employee, employer, self-employed, unemployed, retired and “uninsured”)¹⁶16. Brasil. Ministério da Saúde (MS). Manual técnico operacional do sistema SIH – Sistema de Informação Hospitalar do SUS. Brasília: MS; 2015., which instead means labour market status.

The SIH-SUS is an important additional source of ATF data, although not always fatal work injuries require hospital or emergency care. Its coverage and accuracy are presumable high because of its own nature, although work-relatedness may be missed or misreported in association with vested interests¹⁹19. Mendes ACG, Silva Junior JB, Medeiros KR, Lyr TM, Sá DA. Avaliação do Sistema de Informações Hospitalares-SIH/SUS como fonte complementar na vigilância e monitoramento de doenças de notificação compulsória. Inf Epidemiol Sus 2000; 9(2):67-86.. A study about the quality of SIH/SUS external cause records showed a moderate level of agreement between recorded and gold-standard diagnoses²⁰20. Mathias TAF, Andrade SM, Tomimatsu MFAI, Soares DFPP, Sapata MPM, Frascarelli AS, Souza RKT. Confiabilidade dos diagnósticos de internações por causas externas financiadas pelo Sistema Único de Saúde em dois municípios do estado do Paraná, Brasil. Cien Saude Colet 2014; 19(10):4257-4265., but it did not address work-related injuries. Hospital records have been used and recommended for ATF surveillance in the United States²¹21. Massachusetts. Department of Public Health. Inpatient hospitalizations for work-related injuries and illnesses in Massachusetts, 1996-2000. Department of Public Health Massachusetts. [acessado 2015 Mar 10]. Disponível em: http://www.mass.gov/eohhs/docs/dph/occupational-health/hospitalization-report-05.pdf.
http://www.mass.gov/eohhs/docs/dph/occup... and Finland, where national hospital admissions databases are available²²22. Finland. Official Statistics of Finland (OSF). Occupational accident statistics, quality description: accidents at work statistics. Helsinki: Statistics Finland; 2010. [acessado 2015 Mar 10]. Disponível em: http://www.stat.fi/til/ttap/2013/ttap_2013_2015-11-27_laa_001_en.html
http://www.stat.fi/til/ttap/2013/ttap_20... .

Notifiable Diseases Information System (SINAN)

The SINAN comprises several information subsystems concerning compulsory notifiable diseases and injuries, some of them having useful data for ATF identification. For instance, in the two subsystems for Exogenous Poisoning and for Violence there is in each one a box to be filled with data on work-relatedness (yes/no/unknown) and two other fields contain data about death as the outcome (Table 1). Similarly, these data are also recorded in the Accidental Tetanus, Venomous and Poisonous Animal-Related Injuries and Viral Hepatitis subsystems. A major advantage of these SINAN subsystems is having coded “occupation” although not always recorded.

Annual Report on Social Information (RAIS)

The Ministry of Labour and Employment manages the RAIS information system, which monthly receives data compulsorily provided by firms to the Employed and Unemployed General Register (CAGED). To each active worker, employment status and every change in job contracts such as wages and occupation, any maternity or sick leaves, and work-related injuries, either “typical” or “when commuting”, are recorded and reported to CAGED. This database access is granted upon request (Table 1). We did not find ATF studies based on the RAIS or information about its quality or coverage, but certainly it can be used as an additional source of information in research or surveillance. It is worth noticing that the RAIS is limited to formal registered workers, approximately 51% of the employed Brazilian population in 2010²³23. Departamento Intersindical de Estatística e Estudos Socioeconômicos (DIEESE). Anuário do Sistema Público de Emprego, Trabalho e Renda 2010/2011: mercado de trabalho. 3ª ed. São Paulo: DIEESE; 2011..

Information Systems specific to work-related injuries

SINAN - Subsystems for Severe Work Injuries and for those Involving Exposure to Biological Material

The Severe Work Injuries subsystem of SINAN comprises data from compulsory notification of fatal or non-fatal cases, when involving mutilations or hospitalizations among adults, and regardless injury severity when victims are children or adolescents. The other SINAN subsystem is for Work Injuries involving Potential Exposure to Biological Materials. In both subsystems, deaths are recorded in two fields of the source form (Table 2), which also has data on occupation, industry, and the firm or employer name and the CNPJ number. Unfortunately, almost ten years since its inception in 2007, these subsystems remain under implementation with high underreporting. Up to 2011, 71.7%²⁴24. Bastos-Ramos TP, Santana VS, Ferrite S. Estratégia Saúde da Família e notificações de acidentes de trabalho. Epidemio Serv Saude 2015. 24(4):641-650. of municipalities did not report any work-related injury, despite its increasing coverage²⁵25. Machado JMH, Souza CAV, Santana VS, Ferrite S, Campos A, Vasconcelos Neto R. 2º Inventário de saúde do trabalhador, 2010-2011: acompanhamento da Rede Nacional de Atenção Integral em Saúde do Trabalhador, 2010-2011. Brasília, Salvador, Rio de Janeiro: Ministério da Saúde (MS), Universidade Federal da Bahia, Fundação Oswaldo Cruz; 2013.. It can be a result of the initial strategy to limit notifications from the sentinel units’ network. This is a system composed by health services especially prepared to input data into the information system which was modified in 2014, when every health care facility became entitled to notify occupational injuries²⁶26. Brasil. Portaria GM/MS Nº 1.271, de 06 de junho de 2014. Define a Lista Nacional de Notificação Compulsória de doenças, agravos e eventos de saúde pública nos serviços de saúde públicos e privados em todo o território nacional. Diário Oficial da União 2014; 7 jun.. The access to these databases is upon requirement to the Ministry of Health.

Injuries Communication Information System - SISCAT

Siscat is an information system exclusive for work-related diseases and injuries, the most utilized to estimate national epidemiological indicators. Its feed source is the Work Injury Communication (CAT), and it is not limited to injuries covering work-related illnesses as well. CAT is compulsory issued by firms, in addition to health services or workers themselves, regardless the injury/disease severity or work disability. Besides identification of the worker and firm, it contains data on type of injury (1-typical work-related injury/ 2-occupational illness / 3-work injury when commuting); the ICD code; and whether death had occurred (1-yes/ 2-no). The Statistical Yearbook of Labour Injuries (AEAT), available in the National Social Insurance Institute website, provides some summarized SISCAT data, while an infologo allows drawing tables, although limited to a few indicators and descriptors (Table 2). Other information system under the National Social Insurance Institute is the Compensation Benefits Information System (SUIBE), not limited to occupational-related injuries and diseases. Based on records of compensation benefits granted in cases of work disability due to sickness, pensions or retirement, these data are also presented in the AEAT. For ATF, however, the SUIBE is limited because benefits are only granted to eligible relatives. Although SISCAT records have good quality, underreporting has been found, particularly for cases having work disability for 15 or more days, time required for compensation benefits eligibility²⁷27. Waldvogel BC. Quantos acidentes do trabalho ocorrem no Brasil? Proposta de integração de registros administrativos. In: Minayo Gomez CM, Machado JMH, Pena PGL. Saúde do Trabalhador na Sociedade Brasileira Contemporânea. Rio de Janeiro: Editora Fiocruz; 2011. p. 227-244..

Other information systems

Because of their violent nature, ATF are also registered by public security institutions such as the National System of Public Security and Criminal Justice Statistics (SINESPJC) and the National Information System for Public Security, Prisons and Drugs (SINESP). Several data sources fed the SINESPJC, such as the Police Occurrence Report (BO) compulsory issued by the police authority to each claimed violent event. Therefore, each time a death related to external cause occurs the local police department might be immediately informed. Following, a registration document, BO, and an authorization to the technical police are issued. This last one permit to carry out a local investigation, the collection of proofs and the body removal to the closer forensic legal medicine institute (IML), in which the needed exams are conducted to release a coroner report and the death certificate²⁸28. Brasil. Ministério da Justiça (MJ). Fórum Brasileiro de Segurança Pública. Anuário Brasileiro de Segurança Pública. Brasil: MJ; 2012.. When death occurs during the transportation to the health care unit, or afterward when under hospital treatment, regardless its duration, procedures are similar: the district security police department might be informed, which issues a BO, the authorization for the technical police and the following already described steps, accordingly. The SINESP is computerized and decentralized, represents an advance to SINESPJC, and it is aimed at to create a single database network to ensure efficient interoperability between security and defence institutions²⁹29. Brasil. Lei Nº 12.681, de 4 de julho de 2012. Institui o Sistema Nacional de Informações de Segurança Pública, Prisionais e sobre Drogas - SINESP; altera as Leis nos 10.201, de 14 de fevereiro de 2001, e 11.530, de 24 de outubro de 2007, a Lei Complementar no 79, de 7 de janeiro de 1994, e o Decreto-Lei no 3.689, de 3 de outubro de 1941 - Código de Processo Penal; e revoga dispositivo da Lei no 10.201, de 14 de fevereiro de 2001. Diário Oficial da União 2012; 29 jul.. However, the SINESP is not fully implemented yet and data are not available. Throughout the country, IMLs³⁰30. Brasil. Lei Federal nº 3.689, de 3 de outubro de 1941. Dispõe sobre o Código de Processo Penal. Diário Oficial da União 1941; 13 out. have distinct information systems which may also register forensic and BO data that enable ATF identification. These data can be used in research or surveillance³¹31. MacAskill P, Discoll TR. National occupational injury statistics: what can the data tell us? In: Flyer A-M, Williamson A, editors. Occupational Injury: Risk, Prevention and Intervention. London: Taylor & Francis e-Library; 2004. p. 5-13..

In sum, there are multiple ATF records in several information systems available, similarly to other countries. This is a consequence of distinct interests and responsibilities of institutions that need to create and keep ATF records. Originally created for administrative purpose, such as those under the Ministries of Labour and Employment and Social Security, they have been utilised for monitoring and even for prevention. However, they are limited to formal workers or those having work injuries insurance, leaving the unregistered out of the occupational health statistics. In addition, multiple databases without a common individual key variable require complex procedures to be integrated in a single one. Although some databases have limited coverage and poor recording quality, they enable missing data imputation or misclassification correction. Also, overlap of responsibilities across multiple institutions may pose difficulties to set up an efficient and effective coordination and integration of practices, a still major challenge for workers’ health policies³²32. Minayo-Gomez C. Campo da saúde do trabalhador: trajetória, configuração e transformações In: Minayo Gomez CM, Machado JMH, Pena PGL. Saúde do Trabalhador na Sociedade Brasileira Contemporânea. Rio de Janeiro: Editora Fiocruz; 2011. p. 23-34.. The recent creation of a single information system under the national social insurance, the E-social, integrating life-long workers’ data as for occupation, trade, maternity or sick leaves among others, is an optimistic perspective for both surveillance and research³³33. Brasil. Ministério da Previdência Social (MPS). Manual de Orientação do e-social para o empregador doméstico Versão 1.6.1. Brasília: MPS; 2016.. The SUS is creating the E-saude, alike the E-social, rendering data for the much needed interdisciplinary and intersectorial approaches³⁴34. Brasil. Ministério da Saúde (MS). Estratégia e-saúde para o Brasil. Brasília: MS; 2014..

Barriers to ATF recording

The Figure 1 aims at to provide a better understanding of the connections between the several information systems involved in ATF recording, showing their steps, flows and potential barriers or filters that presumably limit data coverage and quality. The flow chart starts with the work-related injury death which may occur under the following circumstances: 1) at the scene, in the workplace or outdoors, when performing a job task – an immediate police communication to the Public Safety Authority is required, and a recording document, the BO is issued, in addition to a legal request to remove the corpse by the Technical Police. An investigation about the circumstances is performed by hearing witnesses at the location where materials are collected for evidence. The corpse is taken to the closest IML for necropsy and laboratory exams when needed. Finally, a coroner report (LC) and death certificates (DO) are released, containing information that may be useful to establish the injury work-relatedness; 2) death occurs during the transport or under treatment in emergency or hospital care – the institution in charge informs the Public Safety Authority, and all steps described previously might be followed, and BO, LC and DO are also released. If the health care was provided by the SUS, the ATF is recorded in the SIH/SUS and in several SINAN subsystems where applicable; 3) death occurs in remote, distant places, where there is no police office, IML or health care – procedures change according to the context but in every case all legal documents, BO, LC and DO might be issued.

Independently of the death circumstance of occurrence, records of ATF on SINAN will be made whether health care units are available with skilled teams, given that universality of compulsory notification is recent. In addition, for any case of registered formal worker, the National Social Insurance Institute (INSS) records ATF in the SISCAT when CAT is issued and, in the SUIBE when relatives are granted with indemnities or pensions (Figure 1). Records on both SISCAT and SUIBE will depend: on the victim’s family will and awareness about their rights to receive these benefits; and, on the knowledge about the importance of ATF recording by the health care staff or employers. Indeed, employers need to inform about ATF in the monthly forms of CAGED and, consequently, into the RAIS (Figure 1).

Barriers and filters in the information flows are common and described in other countries³⁵35. Azaroff LS, Levenstein C, Wegman DH. Occupational injury and illness surveillance: conceptual filters explain underreporting. Am J Public Health 2002; 92(9):1421-1429.,³⁶36. Webb GR, Redman S, Wilkinson C, Sanson-Fisher RW. Filtering effects in reporting work injuries. Accid Anal Prev 1989; 21(2):115-123.. In this study, the Model of Filtering Effects in Reporting Work Injuries described by Webb et al.³⁶36. Webb GR, Redman S, Wilkinson C, Sanson-Fisher RW. Filtering effects in reporting work injuries. Accid Anal Prev 1989; 21(2):115-123. is used, in which filters that may impede or compromise the needed flow, creating underreporting detectable in the following steps. Therefore, filters are partial barriers for recording. They vary and are influenced by individual aspects of the victim, of the professionals in charge of recording, and the information system management as well³⁶36. Webb GR, Redman S, Wilkinson C, Sanson-Fisher RW. Filtering effects in reporting work injuries. Accid Anal Prev 1989; 21(2):115-123.. This Model facilitates the understanding of the reasons for underreporting in each step of information flows³⁵35. Azaroff LS, Levenstein C, Wegman DH. Occupational injury and illness surveillance: conceptual filters explain underreporting. Am J Public Health 2002; 92(9):1421-1429., and demonstrates the feasibility of multiple data sources usage for imputation, when needed to enhance records quality³⁵35. Azaroff LS, Levenstein C, Wegman DH. Occupational injury and illness surveillance: conceptual filters explain underreporting. Am J Public Health 2002; 92(9):1421-1429.,³⁷37. Boden LI, Ozonoff A. Capture-recapture estimates of nonfatal workplace injuries and illnesses. Ann Epidemiol 2008; 18(6):500-506.. Strategies to improve data quality need to be based on the causes and effects to each filter and barriers in their respective contexts³⁵35. Azaroff LS, Levenstein C, Wegman DH. Occupational injury and illness surveillance: conceptual filters explain underreporting. Am J Public Health 2002; 92(9):1421-1429..

To make it simple, filters are classified in two types: missed death reporting in the information system; death is recorded but not its work-relatedness (Figure 1). Studies addressing causes of these filters show: 1) insufficient training of professionals involved³⁸38. Oliveira MLC, Souza LAC. Causas externas: investigação sobre a causa básica de óbito no Distrito Federal, Brasil. Epidemio Serv Saude 2007; 16(4):245-250.; 2) poor motivation and awareness about the relevance of work-relatedness recording³⁹39. Mendonça FM, Drumond E, Cardoso AMP. Problemas no preenchimento da declaração de óbito: estudo exploratório. Rev Bras Estud Popul 2010; 27(2):285-295.,⁴⁰40. Scheid R, Gressler MA, Martins D, Fanfa LS, Krug SBF. Agravos relacionados ao trabalho como causa de Internações hospitalares. Rev Epidemiol Control Infect 2012; 2(3):82-84.; 3) concerns about legal implications, particularly among those in charge of ATF registration⁴¹41. Laurenti R, Mello-Jorge MHP, Gotlieb SLD. Estatísticas de mortalidade e seus usos. Rev Eletron de Comum Inf Saúde 2013; 7(2). [acessado 2015 Mar 10]. Disponível em: www.reciis.icict.fiocruz.br/index.php/reciis/article/view/500.
www.reciis.icict.fiocruz.br/index.php/re... ; 4) lack of equipment or other needed resources for notification or to investigate the death work-relatedness⁴²42. Cavalcante CAA, Cossi MS, de Oliveira Costa RR, de Medeiros SM, de Menezes RMP. Análise crítica dos acidentes de trabalho no Brasil. Rev Aten Saude 2015; 13(44):100-109.; and 5) pressures from employers, lawyers, colleagues, health professionals and even family members to omit the work-relatedness recording due to pecuniary interests⁴³43. Probst TM, Estrada AX. Accident under-reporting among employees: testing the moderating influence of psychological safety climate and supervisor enforcement of safety practices. Accident Anal Prev 2010; 42(5):1438-1444..

It is possible that these barriers and filters have been intensified due to transformations in the world of work that have been taking place in Brazil, such as the reduction on outsourcing restrictions and the resulting increase in job precariousness⁴⁴44. Costa D, Lacaz FAC, Jackson Filho JM, Vilela RAG. Saúde do Trabalhador no SUS: desafios para uma política pública. Rev. bras. Saúde Ocup. 2013; 38(127):11-30.. Other barriers and filters, however, may be indirect. For instance, the Technical Epidemiological Nexus of Prevention (NTEP) was created in 2007 by the INSS, to promote the identification and recording of the work-relatedness of injuries and illnesses for registered workers by qualified physicians. Consequently, work-related compensation benefits could be granted, independently of CAT emission²⁷27. Waldvogel BC. Quantos acidentes do trabalho ocorrem no Brasil? Proposta de integração de registros administrativos. In: Minayo Gomez CM, Machado JMH, Pena PGL. Saúde do Trabalhador na Sociedade Brasileira Contemporânea. Rio de Janeiro: Editora Fiocruz; 2011. p. 227-244.,⁴⁵45. Todeschini R, Lino D, de Melo LEA. O Ministério da Previdência Social e a institucionalidade no campo da saúde do trabalhador. In: Instituto de Pesquisa Econômica Aplicada (IPEA). Saúde e Segurança no Brasil: aspectos institucionais, sistemas de informação e indicadores. Brasília: IPEA; 2011. p. 77-88.. Unfortunately, because the amount of such benefits is used to define the value of the company’s payments to the compulsory Workplace Personal Injury Insurance, this may cause the development of strategies to cover up ATF. In addition, it is noticeable the weakening of health-related guidelines in policies adopted by many unions and worker movements over recent decades³²32. Minayo-Gomez C. Campo da saúde do trabalhador: trajetória, configuração e transformações In: Minayo Gomez CM, Machado JMH, Pena PGL. Saúde do Trabalhador na Sociedade Brasileira Contemporânea. Rio de Janeiro: Editora Fiocruz; 2011. p. 23-34..

Conclusion

This study verified that ATF data in Brazil can be identified in several information systems of distinct government institutions. But the lack of key unique across these systems, which hinders the use of multiple sources to improve coverage and data quality and, consequently, compromising accuracy and completeness of epidemiological estimates. However, the complexity of the identification and recording of ATF is clear. In many cases, their non-recognition may be intentional, resulting from pecuniary interests and/or to avoid legal penalties. The existence of important barriers and filters is presumable, which needs studies focusing the quality and coverage of information systems used for ATF investigation. We emphasize that an appropriate ATF recording precedes and allows the planning of efficient preventive public initiatives.

Acknowledgements

This article is based on the doctoral thesis of Adriana Galdino Batista “Quality of fatal work injuries data records in Brazil”, Graduate Program on Collective Health, Federal University of Bahia. This study was partially supported by the Collaborator Center on Workers’ Health Surveillance to the Health Ministry, Federal University of Bahia. National Research Council, CNPq, granted Vilma Sousa Santana with a PQ 1C award.

References

Publication Dates

History