Magnitude of work accidents among formal and informal workers in Brazil: extrapolation using data from insured individuals of the General Social Security System

Abstract

Understanding the true extent of work-related accidents (WAs) is essential for implementing effective prevention and worker health protection policies. This study aimed to estimate the magnitude of WAs among workers in formal and informal sectors in Brazil. Using an extrapolation method based on the annual incidence of WAs among individuals covered by the General Social Security System (RGPS), the total WA cases were estimated for the economically active and employed population (PEAO) and the informal PEAO for the year 2019. The average incidence was 11.09 per 1,000 employment contracts, more pronounced among men and youth, aged 14 to 24 years, and in the Southern region. An estimated 429,760 WAs occurred among informal workers and 1,059,152 among the overall PEAO. Among informal male workers, the estimate was 308,200, while among informal workers, aged 14 to 24 years, it was 94,581. The estimated underreporting rate was 44.6% (n = 472,295). This study contributes to understanding the magnitude of WAs in Brazil by addressing different characteristics and workers groups. Confirmation studies are needed to validate the obtained estimates.

Key words:
Occupational accidents; Occupational health; Occupational accident registry; Occupational health surveillance; Informal sector

Resumo

Entender a verdadeira extensão dos acidentes do trabalho (AT) é relevante para implementar políticas eficazes de prevenção e proteção à saúde dos trabalhadores. O estudo objetivou estimar a magnitude dos AT entre trabalhadores ocupados e do mercado informal no Brasil. Utilizando o método de extrapolação com base na incidência anual de AT para os segurados pelo Regime Geral de Previdência Social (RGPS), estimou-se os casos totais de AT para a população economicamente ativa e ocupada (PEAO) e para a PEAO informal para o ano de 2019. A incidência média foi de 11,09 por 1.000 vínculos, mais acentuada entre homens, jovens de 14 a 24 anos e na região Sul. Estimaram-se 429.760 AT para a PEAO informal e 1.059.152 para a PEAO em geral. Entre homens da PEAO informal, a estimativa foi de 308.200, e para informais, de 14 a 24 anos, 94.581. A proporção estimada de sub-registro foi de 44,6% (n = 472.295). Este estudo contribui para compreender a magnitude dos AT no Brasil, abordando diferentes características e grupos de trabalhadores. Estudos de confirmação são necessários para validar as estimativas obtidas.

Palavras-chave:
Acidentes de Trabalho; Saúde do trabalhador; Notificação de acidentes de trabalho; Vigilância em saúde do trabalhador; Setor informal

Resumen

Comprender la verdadera magnitud de los accidentes de trabajo (AT) es fundamental para implementar políticas eficaces de prevención y protección de la salud de los trabajadores. Este estudio tuvo como objetivo estimar la magnitud de los AT entre los trabajadores ocupados y del sector informal en Brasil. Utilizando un método de extrapolación basado en la incidencia anual de AT entre los asegurados por el Régimen General de la Seguridad Social (RGPS), se estimaron los casos totales de AT para la población económicamente activa y ocupada (PEAO) y para la PEAO informal en el año 2019. La incidencia media fue de 11,09 por cada 1.000 vínculos laborales, siendo más alta entre hombres, jóvenes de 14 a 24 años y en la región Sur. Se estimaron 429.760 AT para trabajadores informales y 1.059.152 para la PEAO general. Entre los hombres informales, la estimación fue de 308.200, y para los trabajadores informales de 14 a 24 años, de 94.581. La tasa estimada de subregistro fue del 44,6% (n = 472.295). Este estudio contribuye a comprender la magnitud de los AT en Brasil al abordar diferentes características y grupos de trabajadores. Se necesitan estudios de confirmación para validar las estimaciones obtenidas.

Palabras clave:
Accidentes de trabajo; Salud del trabajador; Notificación de accidentes de trabajo; Vigilancia en salud del trabajador; Sector informal

Introduction

Work accidents (WA) are defined by the International Labor Organization (ILO) as unexpected and unplanned events, including acts of violence, arising from or related to work, that result in bodily injury, illness, or even death of one or more workers^1,2. In Brazil, these also include occupational and work-related diseases³. The latter, chronic/insidious in nature, transcend the individual sphere and reverberate in the socioeconomic dynamics of countries. Such events emerge as a challenge to be addressed, given their impact not only on the health and the well-being of workers, families, and communities, but also on public finances and the overall productivity of nations^4-6.

As WAs are preventable, and, for the most part, predictable events, they are also amenable to epidemiological surveillance. In this sense, surveillance of these conditions must encompass the identification, recognition, recording, analysis, dissemination, and, above all, adaptation of care and prevention policies⁷. This monitoring must consider the diversity of possible work arrangements, which, at the national level, include from formal workers with employment relationships governed by the Consolidation of Labor Laws (Consolidação das Leis do Trabalho - CLT) to those who work without due formalization, a scenario that creates a complex context in which actions to promote, prevent, and reduce risks and vulnerabilities require differentiated approaches⁸.

Throughout Brazil, a notable feature of the labor market is the growing presence of informal workers, whose activities often occur outside of normative structures and conventional labor regulations, such as a lack of social protection, lack of access to benefits and labor rights, and a greater exposure to occupational hazards and precarious working conditions^9-11. These workers operate in diverse sectors and constitute a substantial portion of the national workforce, and, in recent years, their numbers have undergone a dramatic increase¹⁰.

Despite this reality, the magnitude of WAs among informal workers remains an insufficiently explored area of research in Brazil^12-15. Moreover, due to the recognized underreporting of these conditions, there is inaccurate information about the occurrence of these incidents among the national working population¹⁵. This reveals a major challenge regarding the need to use alternatives to estimate the totality of these conditions in the country. The present study seeks to fill this research gap, aiming to estimate the magnitude of WAs among formal employed and informal workers in Brazil, based on existing data for those insured under the General Social Security System (RGPS).

Method

This descriptive study used data calculated on the cumulative incidence of work-related accidents (CI-WA) for workers insured under the RGPS, applied to the Economically Active and Employed Population (População Economicamente Ativa e Ocupada - PEAO) and the informal worker subgroup (informal PEAO) in Brazil.

The reference population for this study was the PEAO in Brazil in 2019, represented by the number of individuals, aged 14 or older, employed in the workforce during the reference week of the Continuous National Household Sample Survey (Pesquisa Nacional por Amostra de Domicílios Contínua Trimestral - PNADC/T), 4^th Quarter, 2019, conducted by the Brazilian Institute of Geography and Statistics (IBGE). To obtain the number of individuals that make up the PEAO, by sex, age range, state, and administrative region, PNADC/T data extracted from the IBGE Automatic Recovery System (SIDRA-IBGE) were used. Data collection was carried out on September 19, 2023¹⁶.

Informal workers were considered to be all those who are part of the PEAO but who did not have a formal employment contract, according to the informality proxy defined by the IBGE for the PNADC/T, i.e., employees in the private sector without a formal employment contract, domestic workers without a formal employment contract, employers without a CNPJ registration, self-employed workers without a CNPJ registration, and auxiliary family workers¹⁷.

The concept of WA adopted to calculate the CI-WA was that established by law. According to article 19 of Law No. 8,213/1991, for an accident to be classified as a WA, there must be a direct relationship with a work activity, that is, a causal link between the event and the work. Furthermore, it is essential that this event results in bodily injury or functional impairment that causes death, loss, or reduction, be it temporary or permanent, of the worker’s productive capacity. In addition to the situations mentioned in article 19, the legal concept of WA also covers the conditions described in articles 20 and 21 of the same law, as well as cases of accidents due to a Technical Epidemiological Social Security Nexus (Nexo Técnico Epidemiológico Previdenciário - NTEP).

The CI-WA is an indicator of the intensity of WA occurrences. Its coefficient is defined as the ratio between the number of new WAs each year and the population exposed to the risk of suffering some type of accident. In this article we use to calculate the CI-WA for workers affiliated with the RGPS, the methodology recommended by the Ministry of Social Security (Ministério da Previdência Social - MPS). This method includes both accidents with a registered Work Accident Report (Comunicação de Acidente do Trabalho - CAT) and accidents without a registered CAT, that is, those in which there was a technical characterization of the link between work and the injury by the INSS medical expert¹⁸.

Estimates of the magnitude of WAs for the PEAO and informal PEAO in 2019 were calculated using the extrapolation method to estimate total data (ETD), as described by Soares (2012). ¹⁹ The ETD method consists of applying a known rate to a target population to project the total expected number of events. It is a tool used to estimate absolute values in contexts where actual data are underreported or nonexistent, as is common among informal workers¹⁹. In the present study, this technique was used to estimate the magnitude of WAs within the PEAO and informal PEAO in 2019.

The year of 2019 was selected as the time frame, as it was the last year before the COVID-19 pandemic, an event that had a significant impact on the labor market^20,21. This was also due to the availability of data on WAs among RGPS policyholders at the time of collection in the database used¹⁸.

In the first stage, estimates of the magnitude of WAs were calculated for workers insured by the Degree of Incidence of Work Incapacity due to Environmental Risks at Work (Grau de Incidência de Incapacidade Laborativa decorrente dos Riscos Ambientais do Trabalho - GILRAT, formerly known as Work Accident Insurance). Thus, the CI-WA for Social Security insured individuals was calculated according to the equation standardized by the MPS: CI-WA = (Number of work-related accidents registered in the period / Average annual number of employment contracts) × 1,000.

The number of accidents recorded during the period was obtained using the historical database of work-related accidents (AEAT InfoLog), and the annual average number of employment contracts was extracted from the historical social security database (AEPS InfoLog)¹⁸.

This indicator was stratified by age range, gender, states, and regions of the country. The gender variable was categorized as male or female. The age range was categorized as 14-24, 25-39, 40-59, and 60 or older. The state variable was maintained as the 26 options corresponding to each Brazilian state plus the Federal District, while the administrative region variable was categorized into five categories: North, Northeast, Midwest, Southeast, and South.

Next, the CI-WA coefficient for workers insured by the RGPS corresponding to each stratum of the PEAO and informal PEAO was applied, according to the formula: Estimated AWAT by stratum = (RGPS WA by stratum / RGPS Links by stratum) × PEAO by stratum. This coefficient was used to estimate the expected cases of WA in each specific stratum, that is, the projected magnitude of WA in 2019. After this step, the expected cases of WA in each stratum of the PEAO and, in the case of estimates for informal workers, for the informal PEAO were added. This sum represents the total projected number of WA expected for workers active in the country and for those working informally, in that order.

The underreporting estimate was calculated by the difference between the absolute number of WAs estimated for the PEAO and the number of WAs officially recorded by the INSS (Estimated absolute underreporting = Total estimated WA - WA recorded by the INSS). This calculation allowed us to estimate the discrepancy between the estimated cases and those actually recorded in the INSS records.

This study assumed that the annual CI-WA specific to each group of descriptor variables was the same for workers insured under the RGPS and the PEAO/informal PEAO. A previous study supports this assumption and provides evidence of similarities between the CI-WA between informal and formal workers in the country²². This methodological choice was adopted given the lack of recent and comprehensive data on this indicator for the informal PEAO nationwide.

The chi-square (χ²) test was used to investigate statistically significant associations between the WA records/estimates and the variables region, sex, age range, and state. The CI-WA values were also compared between administrative regions using analysis of variance (ANOVA), and a boxplot was generated to view regional distributions. All analyses were conducted using Jamovi 2.6.26 software.

Due to the secondary nature of the data, there was no need to submit the protocol for review by a Research Ethics Committee.

Results

The PEAO projected by the IBGE for 2019 was 95,515,000 workers. Of these, 38,756,000 (40.57%) were informal workers. The average number of formal RGPS contributors, according to the MPS, was 52,923,137. The estimated overall CI-WA for workers insured by Social Security was 11.09 per 1,000 employment contracts in Brazil.

Among RGPS insured workers, the groups with the highest CI-WA per 1,000 employment contracts were male workers, with a coefficient of 13.59; the 14-24 age range, with an CI-WA of 13.84; and the South region, with a coefficient of 13.98. Among RGPS members, 586,857 WA cases were recorded in 2019. The number of WA cases for informal workers was estimated at 429,760. For the PEAO, it was 1,059,152.

For male workers affiliated with the RGPS, 386,601 WA cases were recorded. Among informal workers of the same sex, 308,200 cases were estimated. For male PEAO, the estimate was 740,007 WA cases.

Among workers affiliated with the RGPS, the 14-24 age range recorded the highest incidence (13.84 WA/1,000 contracts). Among informal workers in the same age range, the number of expected cases was 94,581, while for the PEAO, it was 196,439. The age range with the highest number of absolute cases was that of 25-39 years of age, with 271,632 WAs among INSS insured individuals and 458,757 and 161,780 WAs estimated for the PEAO and informal PEAO, in that order (Table 1).

Table 1
Number of cases and annual cumulative incidence of work-related accidents (WA) per 1,000 RGPS insured persons and estimates for PEAO, informal PEAO, and underreporting, according to sociodemographic characteristics. Brazil, 2019.

Variable*	RGPS insured persons			PEAO			PEAO-I			Estimate of underreporting
	WA	Relationships	CI- WA	N**	WA	%	N**	WA	%	N	%
Region
Midwest	47,608	4,354,506	10.93	7,934	86,743	8.4	3,010	32,908	8.2	39,135	45.1
Northeast	65,011	8,828,885	7.36	21,744	160,111	15.6	11,677	85,983	21.5	95,100	59.4
North	24,016	2,402,016	10.00	7,493	74,917	7.3	4,283	42,823	10.7	50,901	67.9
Southeast	314,550	27,500,692	11.44	43,086	492,813	47.9	14,977	171,305	42.8	178,263	36.2
South	135,672	9,701,883	13.98	15,257	213,355	22.7	4,810	67,263	18.3	77,683	36.4
Sex
Male	386,601	28,455,665,33	13.59	54,468	740,007	68.7	22,685	308,200	70.0	353,406	47.8
Female	200,111	24,353,982,68	8.22	41,047	337,274	31.3	16,072	132,060	30.0	137,163	40.7
Age range
14 to 24 years	89,556	6,472,840	13.84	14,198	196,439	18.5	6,836	94,581	22.3	106,883	54.4
25 to 39 years	271,632	22,550,873	12.05	38,086	458,757	43.3	13,431	161,780	38.1	187,125	40.8
40 to 59 years	209,208	20,716,293	10.10	36,465	368,250	34.8	14,669	148,138	34.9	159,042	43.2
60 years or more	16,452	3,120,710	5.27	6,766	35,670	3.4	3,822	20,149	4.70	19,218	53.9

• RGPS: General Social Security System. PEAO: economically active employed population. PEAO-I: informal economically active employed population. Estimate of underreporting: calculation of the difference between estimated and recorded work-related accidents. WA: work-related accident. Relationships: formal employment relationships (employment relationships). CI-AT: cumulative incidence of work-related accidents. N: total number of cases. %: percentage of work-related accidents in relation to the total or subgroup. * Differences in the sum of categories in relation to the total number of accidents in the country are due to cases with missing answers. ** Values expressed per 1,000 workers.
• Source: Authors.

The state of São Paulo presented the highest absolute estimates of WA. According to Table 2, in 2019, 288,917 cases were projected for the PEAO and 89,924 for informal PEAO. The estimated number of underreporting cases in the state was 87,886. Among the states, the state with the highest CI-WA among RGPS insured individuals was Mato Grosso do Sul, with a coefficient of 15.04, while the state with the lowest was Tocantins, with 5.02, all per 1,000 employment contracts (Table 2).

Table 2
Number of cases and annual cumulative incidence of work-related accidents (WA) per 1,000 RGPS insured persons and estimates for PEAO, informal PEAO and underreporting, by state. Brazil, 2019.

Territoriality	RGPS insured persons			PEAO			PEAO-I			Estimate of Underreporting
	WA	Relationships	CI-WA	N**	WA	%	N**	WA	%	N	%
Brazil	586,857	52,923,137	11.09	95,515	1,059,152	100	38,756	429,760	100	472,295	44.6
São Paulo	201,031	16,013,365	12.55	23,014	288,917	27.3	7,163	89,924	20.9	87,886	30.4
Minas Gerais	62,346	6,005,995	10.38	10,363	107,574	10.2	4,093	42,488	9.9	45,228	42.0
Rio Grande do Sul	48,971	3,397,925	14.41	5,819	83,864	7.9	1,936	27,902	6.5	34,893	41.6
Paraná	45,145	3,531,671	12.78	5,671	72,492	6.8	1,865	23,840	5.5	27,347	37.7
Rio de Janeiro	38,516	4,414,752	8.72	7,761	67,710	6.4	2,941	25,658	6.0	29,194	43.1
Santa Catarina	41,556	2,772,286	14.99	3,767	56,467	5.3	1,009	15,125	3.5	14,911	26.4
Bahia	17,428	2,468,501	7.06	5,867	41,422	3.9	3,209	22,656	5.3	23,994	57.9
Goiás	16,261	1,651,444	9.85	3,431	33,783	3.2	1,392	13,706	3.2	17,522	51.9
Pernambuco	14,396	1,626,020	8.85	3,646	32,280	3.0	1,818	16,096	3.7	17,884	55.4
Pará	9,303	1,031,879	9.02	3,521	31,744	3.0	2,195	19,789	4.6	22,441	70.7
Ceará	11,475	1,521,623	7.54	3,790	28,581	2.7	2,066	15,580	3.6	17,106	59.9
Espírito Santo	12,657	1,066,579	11.87	1,948	23,117	2.2	779	9,244	2.2	10,460	45.2
Amazonas	7,052	504,629	13.97	1,654	23,114	2.2	965	13,485	3.1	16,062	69.5
Mato Grosso	12,717	939,385	13.54	1,703	23,054	2.2	673	9,111	2.1	10,337	44.8
Mato Grosso do Sul	10,175	676,305	15.04	1,336	20,100	1.9	514	7,733	1.8	9,925	49.4
Maranhão	3,806	630,942	6.03	2,330	14,055	1.3	1,413	8,524	2.0	10,249	72.9
Distrito Federal	8,455	1,087,371	7.78	1,463	11,376	1.1	431	3,351	0.8	2,921	25.7
Rondônia	4,136	314,589	13.15	810	10,649	1.0	381	5,009	1.2	6,513	61.2
Rio Grande do Norte	4,865	617,500	7.88	1,335	10,518	1.0	641	5,050	1.2	5,653	53.7
Piauí	3,039	455,019	6.68	1,303	8,703	0.8	770	5,143	1.2	5,664	65.1
Alagoas	3,798	453,416	8.38	1,028	8,611	0.8	473	3,962	0.9	4,813	55.9
Paraíba	3,719	648,362	5.74	1,489	8,541	0.8	777	4,457	1.0	4,822	56.5
Sergipe	2,485	407,497	6.10	957	5,836	0.6	510	3,110	0.7	3,351	57.4
Tocantins	1,401	279,308	5.02	656	3,290	0.3	303	1,520	0.4	1,889	57.4
Amapá	575	85,106	6.76	331	2,236	0.2	186	1,257	0.3	1,661	74.3
Roraima	796	77,967	10.21	217	2,215	0.2	104	1,062	0.2	1,419	64.1
Acre	753	108,534	6.94	304	2,109	0.2	149	1,034	0.2	1,356	64.3

• RGPS: General Social Security System. PEAO: economically active employed population. PEAO-I: informal economically active employed population.
• Estimate of underreporting: calculation of the difference between estimated and recorded work-related accidents. WA: work-related accident. Relationships: formal employment relationships (employment relationships). CI-AT: cumulative incidence of work-related accidents. N: total number of cases. %: percentage of work-related accidents in relation to the total or subgroup. ** Values expressed per 1,000 workers.
• Source: Authors.

The estimated proportion of underreporting was 44.6% (n = 472,295) nationwide. The highest percentage estimates of underreporting were observed in the North (67.9%) and Northeast (59.4%) regions, within the age ranges of 14 to 24 years (54.4%) and 60 years or older (53.9%). Regarding sex, the percentage was higher among men (47.8%) than among women (40.7%) (Table 1). In the breakdown by state, the highest percentages were estimated for the states of Amapá (74.3%), Maranhão (72.9%), Pará (70.7%), and Amazonas (69.5%).

The χ² test revealed statistically significant associations between WA records/estimates and the variables of region, sex, age range, and state, in all three data sources analyzed: WAs among RGPS insured individuals, WA estimates for PEAO, and WA estimates for PEAO-I (Table 3). Territoriality, in particular, presented χ² values above 1 million for RGPS and PEAO, showing strong heterogeneity in the occurrence of WA among the country’s states.

Table 3
Association between sociodemographic variables and the absolute number of cases of work-related accidents (WA) among RGPS insured individuals and with the absolute estimates of occurrences for PEAO and informal PEAO. Brazil, 2019.

Variable	WA (N)	χ²	gl	p-value
Region	RGPS	473,182.45	4	< 0.001
	PEAO	563,388.14	4	< 0.001
	PEAO-I	151,573.15	4	< 0.001
Sex	RGPS	59,276.99	1	< 0.001
	PEAO	150,558.55	1	< 0.001
	PEAO-I	70,470.40	1	< 0.001
Age range	RGPS	270,906.49	3	< 0.001
	PEAO	398,426.62	3	< 0.001
	PEAO-I	116,686.78	3	< 0.001
State	RGPS	1,059,517.23	26	< 0.001
	PEAO	1,059,517.23	26	< 0.001
	PEAO-I	438,928.67	26	< 0.001

• RGPS: General Social Security System. PEAO: economically active employed population. PEAO-I: informal economically active employed population. WA (N): number of work-related accidents. χ² = chi-square test value. df = degrees of freedom.
• Source: Authors.

The comparison of the CI-WA values between regions, performed using the ANOVA test, revealed an F-statistic of 6.1462 and a p-value of 0.0018, indicating significant differences in the CI-WA means between regions. The boxplot by region highlighted that the South region had the highest median CI-WA values, while the Northeast region had the lowest, and the Midwest and North regions showed greater variability among the country’s states (Figure 1).

Figure 1
Distribution of work-related accident (WA) incidence rates in Brazil by administrative region, 2019.

Discussion

Obtaining estimates of WA occurrences in different work arrangements is essential for defining the morbidity profile among workers and allows for the identification of priority areas for the implementation of preventive strategies tailored to the actual needs and demands of the working population. Furthermore, delineating the magnitude of WAs according to strata of the working population is a necessary step to properly evaluate occupational health and safety (OHS) policies in the country.

Regarding the PEAO, estimated by the IBGE at 95,515,000 workers in 2019, it was observed that 40.57% were informal workers, highlighting the complexity of the country’s occupational structure and the relevance of this segment of professionals in the national labor market. This scenario confirms the importance of investigating WA occurrence among informal workers, as well as understanding the dynamics of workforce distribution within different work arrangements. Assessing the magnitude of WAs among formal and informal workers is, therefore, essential to support the formulation of specific public policies in the field of workers’ health and OHS.

The CI-WA rate obtained in this study was 11.09 WA per 1,000 employment contracts nationwide. However, this indicator varies depending on the period analyzed, as demonstrated by previous studies based on Social Security information systems, which report different values: 9.85 per 1,000 insured persons between 2008 and 2014 and 14.6 per 1,000 insured persons between 1998 and 200826. Population surveys, in turn, have shown that the annual cumulative incidence varies between 3% and 6%²⁵, a fact confirmed by recent population-based studies²⁶. Despite these differences, the literature reveals that this indicator has been decreasing, especially for less severe WAs^25,27.

WAs recorded in the RGPS represented approximately 55.4% of the expected cases for the national PEAO. Analysis by age range revealed that workers, aged 25 to 39 years, had the highest absolute incidence of WA, with 458,757 estimated cases for the PEAO. This result may be related to the concentration of the economically active population in this age range²⁸, as well as specific occupational factors associated with this group. This age range represents a significant portion of the younger, more active workforce, with less experience and professional training, which may increase the risk of accidents in the workplace²⁹.

The highest concentration of accidents in São Paulo, in absolute terms, is explained by the large number of workers and economic activities present in that state, which is considered the most populous and industrialized state in Brazil^30,31. Another factor that may contribute to these high rates is the greater formalization of the labor market, notably in the Southeast region. In relative terms, Rio Grande do Sul and Santa Catarina lead, with CI-WA rates of 14.41 and 14.99 per 1,000 employees, respectively, followed by Mato Grosso do Sul, Paraná, and São Paulo, all with rates above 12 WAs per 1,000 employees. This high prevalence of accidents in the Midwest and South regions, according to Malta et al., is directly related to the economic profile of this region, where agriculture plays a predominant role. These activities, by their nature, involve the intensive use of machinery and heavy equipment, increasing the risk of accidents. WAs involving machinery and equipment result in amputations and other serious injuries 15 times more frequently than other causes, generating three-fold more fatal accidents than the general average²⁶.

Discrepancies were found in the magnitude of WAs by gender. Men continued to have a higher incidence of WAs when compared to women. This finding corroborates inequalities already widely described in national literature^26,32-34. This pattern of occurrence, combined with disparities by age range, reinforces the importance of occupational health and safety policies that consider these differences and seek to reduce occupational risks equitably.

The underreporting estimate suggests a scenario marked by gaps in WA records. A total of 472,295 unreported cases were projected, with the North and Northeast regions experiencing the highest number of omissions. WA underreporting is a widely recognized problem in Brazil, and this negatively impacts the national WA reporting and recording systems^35-39. This phenomenon not only distorts the true understanding of the extent of WAs in the country, but it also undermines the ability to implement effective prevention measures. A recent study also indicates that the true incidence of WAs may be significantly underestimated in states with a high prevalence of informal work, a characteristic of many states in the North and Northeast regions. In these areas, spatial clusters with a lower concentration of WAs were identified in states with greater socioeconomic inequality, as measured by the Theil-L, Gini, HDI-M, and IVS indices⁴⁰.

The association identified in this study reinforce the hypothesis that WAs in Brazil are not evenly distributed, but rather are conditioned by structural and sociodemographic factors. The territoriality variable, with particularly high χ² values, indicates that regional factors strongly influence the occurrence and reporting of WAs. Disparities were found not only in the absolute number of events, but also in the relative frequency of these occurrences, expressed by the statistically significant differences in the average CI-WA rates between the country’s regions. These findings suggest that the heterogeneity in the distribution of WAs in the country is influenced by territorial, structural, and occupational factors, among others, reinforcing the need for differentiated approaches in the field of occupational health and OHS surveillance in response to WA occurrences and the disparities reported here. This premise has been discussed in recent studies, which highlights the importance of considering structural socioeconomic inequalities within territories when formulating public policies in this area^40,41.

SINAN is the system responsible for collecting information on various types of health problems relevant to surveillance, including WAs, regardless of the worker’s employment status. By contrast, the RGPS focuses primarily on Social Security beneficiaries, and therefore encompasses only a portion of all workers. Until 2019, however, the concept of WA in SINAN was limited to the reporting of serious WAs. Thus, only WAs that resulted in serious injuries or significant consequences for the worker’s health were reported to SINAN. In 2019, the number of new accidents registered in this system was 111,989, representing only 10.6% of the total of 1,059,152 WAs estimated for the national PEAO. Effective September 1^st, 2019, however, Information Note N^o. 94/2019-DSASTE/SVS/MS established new definitions of work-related injuries and diseases within the scope of the Ministry of Health, and all typical and commuting WA accidents, regardless of severity, became mandatory to be reported to SINAN. This change was intended to improve the sensitivity of WA case identification and to enable comparability of cases recorded in SINAN with other relevant systems, such as the CAT web and the Unified Benefits System (Sistema Único de Benefícios - SUB).

The aforementioned changes represent an important step in improving the quality and comprehensiveness of WA data in Brazil. It is important to highlight, however, that, in SINAN, work-related diseases (WRD), i.e., work-related mental disorders, work-related cancer, occupational dermatoses, pneumoconiosis, noise-induced hearing loss, and repetitive strain injuries/work-related musculoskeletal disorders, due to their chronic-insidious nature, are considered to be a category of injuries that is different from WA⁴². By contrast, for the INSS/RGPS, the definition of WA covers both WA resulting from sudden and unexpected causes, called “typical/type,” and diseases that develop as a result of work processes, establishing themselves insidiously, known as occupational diseases (professional and work-related) or atypical accidents, in addition to commuting accidents and other similar conditions³.

This conceptual difference between the Ministry of Health and the INSS/RGPS results in discrepancies in the system for recording and reporting WA cases. Although the INSS/RGPS only publishes data on RGPS insured individuals, and the Ministry of Health considers WA and DRT to be all conditions affecting workers in both the formal and informal markets, a uniform classification of these conditions, which considers both the specificities of the RGPS and the definitions of the Ministry of Health, would be beneficial in order to harmonize data collection and recording, and would promote a unified approach and joint data analysis, which would allow for a more accurate assessment of the impact of these events on ST. Another equally relevant path would be interoperability between the Ministry of Health’s own information systems (Sinan, SIM, SIH) and, ambitiously, between these systems and e-Social, which would facilitate the epidemiological monitoring of these conditions. It is important to highlight that the overall WA estimates for the PEAO and the informal PEAO present a specific order of magnitude, while the estimates stratified by sociodemographic characteristics, such as sex and age range, present a different order of magnitude. This difference is expected and methodologically justifiable, as the direct application of the RGPS’s CI-WA coefficient to a heterogeneous population, such as the total PEAO, results in a weighted average that smooths out internal variations. However, when estimates are disaggregated by specific strata, such as sex, the CI-WA more accurately reflects the magnitude among subgroups, resulting in higher (or lower) estimates relative to the overall average, consequently impacting the order of magnitude of the estimated total. These differences reinforce the relevance of stratified analyses, which allow for greater sensitivity in detecting underreporting and reveal more accurate magnitudes for more vulnerable groups.

It is important to highlight some limitations of this study. The main one lies in the assumption that the risk of WA among informal workers is equivalent to that of those covered by the RGPS, a hypothesis supported by the specialized literature. This premise may underestimate the reality, considering that informal workers often perform their activities in more precarious and dangerous conditions, in addition to being less subject to formal accident reports. However, even given this limitation, the estimates presented here can be considered minimum values, which does not compromise their relevance. On the contrary, they reinforce their role as a useful tool to assess the problem and guide more inclusive and effective public policies to protect the health of workers, especially those in more vulnerable situations. Therefore, it is possible that those not covered by the RGPS may be exposed to even greater work-related risks than those covered by Social Security, and that, as a result, the estimates presented here have underestimated the true extent of WA in the Brazilian population. Additionally, the present study focused on the comprehensive concept of WA used by Social Security, without specifically disaggregating the data into accident types (typical, atypical, and commuting), nor specific coefficients by cause, such as those found in the International Classification of Diseases. While the results provide a general overview of WA incidence, the lack of data disaggregation may limit a complete understanding of patterns and trends. Therefore, a more detailed and segmented analysis, considering the different types of accidents and territorial divisions, is encouraged.

Despite these considerations, the lack of comprehensive information on WAs among informal workers and the recognized underreporting of WAs by SINAN make this research relevant, as it employs a methodology recognized in the specialized literature to estimate the magnitude of WAs among formal employed and informal workers in Brazil. The study, therefore, contributes to confirming the need for policies and actions aimed at the safety and health of the entire Brazilian workforce, including those not covered by the general social security system.

Final considerations

This study estimated the magnitude of WAs in Brazil, considering different characteristics and groups of workers who make up the formal and non-formal workers.

A higher magnitude of WAs, in absolute terms, was observed among men and young individuals, aged 25 to 39 years, and in the South region. A total of 472,295 unreported WA cases were estimated. The underreporting rate was even higher in some subgroups, exceeding 70% in certain regions of the country. The findings further reinforced the hypothesis that WAs in Brazil are not evenly distributed, with a particular regional influence.

It is important to note that the numbers reported here may not fully capture the entire phenomenon of interest, especially among informal workers. Therefore, it is recommended that confirmatory studies be conducted to compare the magnitudes estimated here and that the technique used be replicated for regional estimates.

References

• ¹ International Labor Organization (ILO). Occupational Safety and Health Statistics (OSH database) - ILOSTAT [Internet]. 2023 [cited 2025 maio 2]. Available from: https://ilostat.ilo.org/resources/concepts-and-definitions/description-occupational-safety-and-health-statistics/
  » https://ilostat.ilo.org/resources/concepts-and-definitions/description-occupational-safety-and-health-statistics
• ² Özdemir M, Save D. Self-reported occupational accidents and affecting factors among the adult population. Cad Bras Ter Ocup 2020; 28(4):1193-1202.
• ³ Oliveira SG. Indenizações por acidente do trabalho ou doença ocupacional. São Paulo: Juspodium; 2024.
• ⁴ Pega F, Hamzaoui H, Náfrádi B, Momen NC. Global, regional and national burden of disease attributable to 19 selected occupational risk factors for 183 countries, 2000-2016: a systematic analysis from the who/ilo joint estimates of the work-related burden of disease and injury. Scand J Work Environ Health 2022; 48(2):158-168.
• ⁵ Rushton L. The global burden of occupational disease. Curr Environ Health Rep 2017; 4(3):340.
• ⁶ Tompa E, Mofidi A, van den Heuvel S, van Bree T, Michaelsen F, Jung Y, Porsch L, van Emmerik M. Economic burden of work injuries and diseases: a framework and application in five European Union countries. BMC Public Health 2021; 21(1):49.
• ⁷ Brasil. Ministério da Saúde (MS). Guia de Vigilância em Saúde. Brasília: MS; 2022.
• ⁸ Neves D. A exploração do trabalho no Brasil contemporâneo. Rev Katálysis 2022; 25(1):11-21.
• ⁹ Nogueira MO, Carvalho SS. Trabalho precário e informalidade: desprecarizando suas relações conceituais e esquemas analíticos [Internet]. 2021 [acessado 2025 maio 2]. Disponível em: https://doi.org/10.38116/td2707
  » https://doi.org/10.38116/td2707
• ¹⁰ Fundo Brasil. Mapeamento do trabalho informal no Brasil. São Paulo: Fundo Brasil; 2022.
• ¹¹ Siqueira JS, Werneck GL, Pena PGL, Fernandes RCP. Delivery workers' subordinate work in delivery platform companies and street accidents: magnitude and associated factors. Cad Saude Publica 2025; 41(3):e00204723.
• ¹² Costa MS. Trabalho informal: um problema estrutural básico no entendimento das desigualdades na sociedade brasileira. Cad CRH 2010 ;23(58):171-190.
• ¹³ Rios MA, Nery AA, Rios PAA, Casotti CA, Cardoso JP. Fatores associados a acidentes de trabalho envolvendo trabalhadores informais do comércio. Cad Saude Publica 2015; 31(6):1199-1212.
• ¹⁴ Iriart JAB, Oliveira RP, Barreto ML, Lima LD, Moreira MR, Carvalho LC, Paim JS. Representações do trabalho informal e dos riscos à saúde entre trabalhadoras domésticas e trabalhadores da construção civil. Cien Saude Colet 2008; 13(1):165-174.
• ¹⁵ Cortez EA, Jasmim JS, Silva LM, Queluci GC. Análise da mortalidade por acidentes de trabalho no Brasil: revisão integrativa. Rev Enferm UFPE On Line 2017; 11(Supl. 1):429-437.
• ¹⁶ Instituto Brasileiro de Geografia e Estatística (IBGE). Sistema IBGE de Recuperação Automática - SIDRA [Internet]. 2019 [acessado 2025 maio 2]. Disponível em: https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas
  » https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas
• ¹⁷ Instituto Brasileiro de Geografia e Estatística (IBGE). Indicadores IBGE: Principais destaques da evolução do mercado de trabalho no Brasil 2012-2020. Rio de Janeiro: IBGE; 2021.
• ¹⁸ Brasil. Ministério da Previdência Social (MPS). Base de dados histórica do Anuário Estatístico da Previdência Social (AEPS InfoLogo) [Internet]. 2022 [acessado 2025 maio 2]. Disponível em: http://www3.dataprev.gov.br/infologo/inicio.htm
  » http://www3.dataprev.gov.br/infologo/inicio.htm
• ¹⁹ Soares JFS. Incidência cumulativa anual de acidentes de trabalho não fatais - estimativas nacionais para o Brasil [tese]. Salvador: Universidade Federal da Bahia; 2012.
• ²⁰ Júnior CJS, Garcia KKS, Antunes JLF, Fischer FM. Impact of the COVID-19 pandemic on the incidence of occupational accidents among beneficiaries of the National Institute of Social Security in Brazil: an interrupted time series study. Cien Saude Colet 2025; 30(1):e10282023.
• ²¹ Santos Júnior CJ, Antunes JLF, Fischer FM. Como a pandemia de COVID-19 afetou a notificação de acidentes do trabalho em diferentes atividades econômicas e ocupações no Brasil? Um estudo ecológico usando o p-score. Rev Bras Saude Ocup 2024; 49:e11.
• ²² Cordeiro R, Lima F, Moreira M, Silva D, Oliveira A. Incidência de acidentes do trabalho não fatais em localidade do Sudeste do Brasil. Cad Saude Publica 2006; 22(2):387-393.
• ²³ Bezerra JC, Arantes LJ, Shimizu HE, Merchán-Hamann E, Ramalho WA. A saúde do trabalhador no Brasil: acidentes registrados pela Previdência Social de 2008 a 2014. Rev Bras Enferm 2020; 73(6):e20180471.
• ²⁴ Almeida FS, Morrone LC, Ribeiro KB. Tendências na incidência e mortalidade por acidentes de trabalho no Brasil, 1998 a 2008. Cad Saude Publica 2014; 30(9):1957-1964.
• ²⁵ Santana V, Nobre L, Waldvogel BC. Acidentes de trabalho no Brasil entre 1994 e 2004: uma revisão. Cien Saude Colet 2005; 10(4):841-855.
• ²⁶ Malta DC, Bernal RTI, Vasconcelos NM, Ribeiro AP, Vasconcelos LLC, Machado EL. Acidentes no deslocamento e no trabalho entre brasileiros ocupados: Pesquisa Nacional de Saúde 2013 e 2019. Rev Bras Epidemiol 2023; 26(Supl. 1):e230019.
• ²⁷ Pinto JM. Tendência na incidência de acidentes e doenças de trabalho no Brasil: aplicação do filtro Hodrick-Prescott. Rev Bras Saude Ocup 2017; 42:e10.
• ²⁸ Brasil. Ministério do Trabalho (MT). Cadastro Geral de Empregados e Desempregados (Caged) [Internet]. 2022 [acessado 2025 maio 2]. Disponível em: http://pdet.mte.gov.br/novo-caged
  » http://pdet.mte.gov.br/novo-caged
• ²⁹ Mascarenhas MDM, Barros MBA, Silva MMAD, Malta DC, Oliveira MM, Morais Neto OL. Emergency room visits for work-related injuries: characteristics and associated factors - Capitals and the Federal District, Brazil, 2011. Cien Saude Colet 2015; 20(3):667-678.
• ³⁰ Neto AM, Silva RO, Severian D. Região e indústria no Brasil: ainda a continuidade da "desconcentração concentrada"? Econ Soc 2020; 29(2):581-607.
• ³¹ Abel L, Filho S, Queiroz SN, Livramento M, Clementino M. Mercado de trabalho nas regiões metropolitanas brasileiras. Mercat 2016; 15(1):37-54.
• ³² Malta DC, Stopa SR, Silva MMA, Szwarcwald CL, Franco MS, Santos FV, Machado EL, Minayo-Gómez C. Acidentes de trabalho autorreferidos pela população adulta brasileira: Pesquisa Nacional de Saúde, 2013. Cien Saude Colet 2017; 22(1):169-178.
• ³³ Ferreira MJM, Correa FGS, Lacerda EM, Hajat S, Araújo LF. Analysis of risk factors in occupational accidents in Brazil: a population-based study. J Occup Environ Med 2020; 62(3):E46-E51.
• ³⁴ Alves AMS, Filho CG, Santos NM, Souki GQ. Factors influencing occupational accidents: a multidimensional analysis in the electricity sector. Gest Produçao 2020; 27(1):e4609.
• ³⁵ Rodrigues AB, Santana VS. Acidentes de trabalho fatais em Palmas, Tocantins, Brasil: oportunidades perdidas de informação. Rev Bras Saude Ocup 2019; 44(1):e8.
• ³⁶ Galdino A, Santana VS, Ferrite SF. Fatores associados à qualidade de registros de acidentes de trabalho no Sistema de Informações sobre Mortalidade no Brasil. Cad Saude Publica 2019; 36(1):e00218318.
• ³⁷ Batista AG, Santana VS, Ferrite S. Registro de dados sobre acidentes de trabalho fatais em sistemas de informação no Brasil. Cien Saude Colet 2019; 24(2):693-704.
• ³⁸ Galdino A, Santana VS, Ferrite S. Quality of the record of data on fatal workplace injuries in Brazil. Rev Saude Publica 2017; 51(1):72.
• ³⁹ Oliveira PRA, Portela MC, Filho HRC, Souza WRN. Nexo Técnico Epidemiológico Previdenciário (NTEP): risco das sete atividades econômicas e condições incapacitantes mais frequentes, Brasil, 2000-2016. Cad Saude Publica 2021; 37(1):e00191119.
• ⁴⁰ Santos Júnior CJ, Fischer FM. Spatiotemporal patterns and structural inequalities associated with the incidence of common workplace accidents in Brazil. Rev Bras Med Trab 2025; 23(1):e2024.
• ⁴¹ Santos Júnior CJ, Fischer FM. Iniquidades socioeconômicas e mortalidade por acidentes do trabalho entre segurados da Previdência Social no Brasil. Cien Saude Colet [periódico na internet]. 2025 abr. [acessado 2025 Set 23]. [cerca de 10 p]. Disponível em: https://cienciaesaudecoletiva.com.br/artigos/socioeconomic-inequities-and-mortality-from-workplace-accidents-among-social-security-beneficiaries-in-brazil/19589?id=19589
  » https://cienciaesaudecoletiva.com.br/artigos/socioeconomic-inequities-and-mortality-from-workplace-accidents-among-social-security-beneficiaries-in-brazil/19589?id=19589
• ⁴² Zack BT, Ross C, Gouvêa LAVN, Tonini NS. Acidente de trabalho grave: perfil epidemiológico em um município do oeste do Paraná. Saude Debate 2021; 44(125):1036-1052.

• Funding
  FM Fischer is a CNPq productivity scholarship holder A1 (process number 306963/2021-3).
• Data availability statement
  The data sources used in the research are indicated in the body of the article.

• Chief editors:
  Maria Cecília de Souza Minayo, Romeu Gomes, Antônio Augusto Moura da Silva, Vânia de Matos Fonseca

Data availability

The data sources used in the research are indicated in the body of the article.

Publication Dates

• Publication in this collection
  10 Nov 2025
• Date of issue
  Oct 2025

History

• Received
  01 Mar 2024
• Accepted
  03 July 2025
• Published
  05 July 2025