Loss of productivity due to work-related traffic accidents in Brazil, 2009-2018

Abstract

The aim is to estimate the prevalence and assess the factors associated with productivity loss among workers injured in traffic accidents while performing work-related activities. A cross-sectional epidemiological study was conducted using data on work-related traffic accidents. Multilevel logistic regression models were applied to analyze factors associated with productivity loss. The overall prevalence of productivity loss was 81.04%. A significant association was observed between working in traffic and productivity loss in the outpatient care stratum. Female sex, age under 30 years, white race, other races, elementary education, urban residence, and being a motorcyclist were identified as risk factors for productivity loss among traffic-exposed workers. Sociodemographic and occupational factors, as well as accident-related characteristics, are associated with productivity loss among workers in traffic settings. Contextual variables influence the strength of these associations, highlighting the severity of occupational and traffic accidents as major public health issues. Understanding these factors is essential for the development of effective prevention and intervention strategies.

Key words:
Work; Traffic accident; Occupational accident; Productivity loss; Workers’ health

Resumo

O objetivo é estimar a prevalência e avaliar os fatores associados à perda de produtividade em trabalhadores que desenvolvem suas atividades no trânsito lesionados por acidentes no trabalho. Realizou-se um estudo epidemiológico transversal analisando dados de acidentes de trânsito relacionados ao trabalho. Foram utilizados modelos de regressão logística multinível para análise dos fatores associados à perda de produtividade. A prevalência geral de perda de produtividade foi de 81,04%. Observou-se associação significativa entre trabalhar no trânsito e perda de produtividade no estrato ambulatorial. Sexo feminino, idade inferior a 30 anos, raça branca, outras raças, ensino fundamental, residência em zona urbana e ser motociclista foram identificados como fatores de risco para perda de produtividade em trabalhadores do trânsito. Fatores sociodemográficos e ocupacionais e características do acidente estão associados à perda de produtividade em trabalhadores do trânsito. Variáveis contextuais influenciam as medidas de associação, destacando a gravidade dos acidentes de trabalho e de trânsito como problemas significativos de saúde pública. A compreensão desses fatores é crucial para desenvolver estratégias eficazes de prevenção e intervenção.

Palavras-chave:
Trabalho; Acidente de trânsito; Acidente de trabalho; Perda de produtividade; Saúde dos trabalhadores

Resumen

El objetivo es estimar la prevalencia y evaluar los factores asociados a la pérdida de productividad en trabajadores que desempeñan sus actividades en el tránsito y resultan lesionados por accidentes viales. Se realizó un estudio epidemiológico transversal analizando datos de accidentes de tránsito relacionados con el trabajo. Se emplearon modelos de regresión logística multinivel para analizar los factores asociados a la pérdida de productividad. La prevalencia general de pérdida de productividad fue del 81,04%. Se observó una asociación significativa entre trabajar en el tránsito y la pérdida de productividad en el estrato ambulatorio. Ser mujer, tener menos de 30 años, ser de raza blanca, pertenecer a otras razas, tener enseñanza primaria, residir en zona urbana y ser motociclista fueron identificados como factores de riesgo para la pérdida de productividad en trabajadores del tránsito. Factores sociodemográficos, ocupacionales y características del accidente están asociados a la pérdida de productividad en trabajadores del tránsito. Las variables contextuales influyen en las medidas de asociación, lo que resalta la gravedad de los accidentes laborales y de tránsito como importantes problemas de salud pública. Comprender estos factores es fundamental para desarrollar estrategias eficaces de prevención e intervención.

Palabras clave:
Trabajo; Accidente de tránsito; Accidente laboral; Pérdida de productividad; Salud de los trabajadores

Introduction

Approximately 1.35 million deaths worldwide each year result from traffic accidents (TAs), while another 20 to 50 million people suffer non-fatal injuries, rendering them disabled due to their involvement in these accidents¹. This proven road safety is multifactorial, characterized by vehicle-related^2,3, environmental^4,5, and human factors^6,7, making it a matter of considerable concern for global public health⁸.

The negative impact of a TA is felt not only by those involved, but also by their families and society at large, imposing significant socioeconomic costs in terms of premature deaths, injuries, and a potential loss of productivity⁴. This event poses a significant burden on public health, social assistance, and pension policies due to its high incidence among men of working age, who are frequently victims while performing their work activities^9-11.

In this context, the need arises to study TA, especially those that are work-related, understanding their interface with labor and the factors related to their potential to generate productivity losses. The emerging national literature on the matter^12,13, similar to that reported in international literature^14,15, shows that most studies on productivity loss due to TA have focused on measuring the costs of this loss. The present study, however, seeks to estimate the prevalence of productivity loss due to TA, thus presenting the magnitude of this problem.

A loss of productivity is considered an indirect cost resulting from illness or accident. It can be investigated under two aspects: temporary or permanent absence from work, whether paid or unpaid, or from other activities, such as studying (absenteeism); or when the individual remains at work or in their usual activities, but their productive capacity is reduced (presenteeism)¹⁶.

Consequently, considering all the repercussions of TAs, their interaction with work, and the scarcity of studies in the scientific literature, investigating the loss of productivity associated with work-related TAs is extremely relevant. Hence, it is hoped that this study will contribute to further discussions on the topic and contribute to the formulation and implementation of public policies aimed at reducing TAs and workplace accidents, seeking to mitigate their impacts on worker health. Given the above, the following objectives are outlined: to estimate the prevalence and evaluate the factors associated with a loss of productivity among workers in traffic injured by TAs.

Method

Study type and participants

This is a cross-sectional, exploratory epidemiological study of Serious Workplace Accidents (SCA) in Brazil between 2009 and 2018. Secondary data from the Notifiable Injuries Information System (SINAN/DATASUS) were used.

The study participants were all workers reported to SINAN for their involvement in an SCA, constituting a TA.

Data collection instrument and procedures

Data on SCAs were collected using the SINAN serious workplace accident investigation form. This form is divided into seven sections, the first three of which provide general information about the individual involved and the municipality where the accident occurred and was reported. The following four sections address epidemiological history, accident data, medical care information, and conclusions, contributing to the investigation of the accident profile. The data used in this study were extracted from the Collaborating Center for Surveillance of Occupational Accidents (Centro Colaborador de Vigilância aos Acidentes de Trabalho - CCVISAT), which maintains a database on SCA in Excel spreadsheets, organized by year of occurrence.

Variables

The study variables were created based on the SINAN serious work accident investigation form.

Outcome

The outcome variable was the loss of productivity resulting from the TA. Therefore, the dichotomization of question 66 in the case conclusion section of the investigation form (yes; no) was considered a productivity loss, when: yes (temporary disability, partial disability, permanent total disability) and no (cure).

Cases resulting in death were not considered in the analysis of productivity loss due to the need for a specific methodological and analytical strategy based on years of life lost, which is not the focus of this study^17-19.

Main exposure variable

The main exposure was considered “Working in traffic.” The “Yes” exposure category included workers who work in traffic, such as drivers, bus conductors, parcel delivery personnel, and traffic agents. Occupations were identified using question 31, which refers to the Brazilian Classification of Occupations (Classificação Brasileira de Ocupações - CBO)²⁰.

Stratification variable

In addition to stratifying the outcome by the main exposure variable, a third variable was added: treatment regimen, categorized as (inpatient; outpatient; both). This enabled a stratified analysis to assess the association between working in traffic and productivity loss due to TA, considering the treatment regimen and the categories of the independent variables.

Independent variables

The covariates used in this study were organized into four groups, as follows:

Group I: sociodemographic variables - sex (male; female); age in years (< 30 years, 30-59 years, ≥ 60 years); race/color (white, black/mixed race, other); education (higher education, high school, elementary school, and no schooling); area (rural, peri-urban, urban).

Group II: occupational variables - type of employment relationship (with employment relationship, without employment relationship, employer/self-employed, and others); length of service in current occupation in full years (more than ten years, two to ten years, and less than two years); economic activity (public administration, defense and security; agriculture, livestock, forestry and logging; fishing; accommodation and food; automotive vehicle repair; construction; education, health, and social services; manufacturing and extractive industries; electricity production and distribution; services; domestic services; and transportation, storage, and communications); work in a third-party company (yes; no). The economic activity classification was based on the National Classification of Economic Activities (Classificação Nacional de Atividades Econômicas - CNAE).

Group III: accident characteristics - accident shift (morning, afternoon, evening, and early morning); time elapsed after the start of the workday in hours categorized as (first 6 hours; 7 to 12 hours of the workday; more than 12 hours of the workday); type of road user (pedestrian, cyclist, motorcyclist, vehicle occupant, and others); type of TA (run-over, collision, and others); type of SCA accident (typical and commuting); care provided in the municipality of the accident (yes; no).

Group IV: contextual variables - HDI (Human Development Index); municipal size (small municipality with up to 50,000 inhabitants; mid-sized municipality with 50,001 to 100,000 inhabitants; large municipality with 100,001 or more inhabitants); data related to contextual variables were extracted from the 2010 census conducted by the Brazilian Institute of Geography and Statistics (IBGE) and made available in the Unified Health System (DATASUS) database²¹.

Data analysis

Data analysis was conducted using STATA statistical software, version 14.2. Initially, a brief characterization of the participants was performed, using simple and relative frequencies for categorical variables.

Subsequently, a bivariate analysis was performed to identify and investigate, respectively, the prevalence of productivity loss due to TA and the association between each of the sociodemographic and occupational variables, accident characteristics, care, and outcome variable. The measure of occurrence used was prevalence (P%), and the measure of association was the prevalence ratio (PR) with its respective 95% confidence intervals (95%CI).

For the multivariate analysis, a Poisson regression model with a multilevel structure was adjusted, considering the municipality as the contextual level (random effect) and the workers as the individual level. The initial selection of variables for the model was based on theoretical and statistical criteria, including those with a p-value < 0.20 in the bivariate analysis, in addition to theoretical relevance. After this initial selection, the permanence of the variables in the final model was determined by the lowest value criterion of the Akaike information criterion (AIC), with the aim of optimizing the adequacy of the model (Figure 1).

Figure 1
Hierarchical model for multilevel logistic regression analysis.

Ethical aspects

This study is a survey of publicly available secondary data; therefore, it was not necessary to submit it to a Research Ethics Committee (REC). It should be noted that the current principles of Resolution No. 466 of the National Health Council of 2012 were observed, and the data do not contain the participants’ personal identification.

Results

The study participants were 81,271 workers involved in SCA, which also constituted as TA. The prevalence of overall productivity loss was 81.04%, and the prevalence of productivity loss by treatment regimen was 85.78%, 76.77%, and 91.33% for inpatient, outpatient, and both, respectively. Regarding the distribution of cases, 68.48% of the workers were treated exclusively in an outpatient setting, 23.52% in an inpatient setting, and 7.99% in both.

Table 1 demonstrates that, overall, an association was observed between work in traffic and productivity loss in the outpatient stratum. In the female category (PR = 1.13; 95%CI = 1.03-1.24), a significant association was identified, indicating that women who work in traffic have a 13% higher prevalence of productivity loss compared to those who do not work in traffic. For the under-30 age group (PR = 1.04; 95%CI = 1.02-1.07), a positive association was also observed, indicating a 4% increase in the prevalence of productivity loss for workers in traffic in this age group.

Table 1
Prevalence of productivity loss due to workplace accidents among workers in traffic, according to categories of sociodemographic variables, stratified by treatment regimen. Brazil, 2009-2018.

Variables	TT*	Treatment regime
		Hospital		Outpatient		Both
		Loss of productivity		Loss of productivity		Loss of productivity
		%	PR (95%CI)	%	PR (95%CI)	%	PR (95%CI)
Sex
Female	Yes	84.2	1.00 (0.94-1.06)	84.2	1.13 (1.03-1.24)	90.9	1.00 (0.87-1.14)
	No	83.9		74.4		90.9
Male	Yes	84.8	0.98 (0.97-0.99)	78.8	1.02 (0.99-1.04)	91.3	1.00 (0.98-1.02)
	No	86.3		77.2		91.6
Age
< 30 years	Yes	85.7	1.00 (0.99-1.01)	79.0	1.04 (1.02-1.07)	91.4	1.00 (0.98-1.03)
	No	85.7		75.7		91.2
30-59 years	Yes	83.9	0.97 (0.95-0.99)	80.2	1.02 (0.98-1.06)	91.9	1.00 (0.97-1.03)
	No	86.4		78.8		91.9
≥ 60 years	Yes	76.2	0.93 (0.87-1.00)	74.3	1.01 (0.87-1.19)	78.8	0.88 (0.73-1.06)
	No	81.9		75.4		89.5
Race
White	Yes	82.1	0.98 (0.97-1.00)	74.0	1.05 (1.01-1.10)	88.3	1.00 (0.96-1.03)
	No	83.4		70.2		88.5
Black/Brown	Yes	85.2	0.99 (0.98-1.01)	72.3	0.98 (0.94-1.02)	91.2	0.99 (0.96-1.02)
	No	85.8		73.6		92.3
Others	Yes	84.2	0.94 (0.84-1.06)	92.9	1.27 (1.05-1.52)	100.0	1.16 (1.00-1.34)
	No	89.4		73.3		86.2
Educational level
Higher	Yes	82.2	1.02 (0.93-1.12)	65.5	0.95 (0.72-1.24)	80.0	0.91 (0.66-1.24)
	No	80.6		69.1		88.0
High School	Yes	83.9	0.97 (0.95-0.99)	75.2	0.98 (0.93-1.03)	90.7	1.00 (0.96-1.04)
	No	86.5		76.7		90.9
Elementary	Yes	87.3	1.01 (1.00-1.03)	81.7	1.08 (1.04-1.11)	88.1	0.98 (0.94-1.01)
	No	86.2		75.7		90.1
No schooling	Yes	85.7	0.99 (0.87-1.13)	85.8	1.13 (0.81-1.56)	66.7	0.75 (0.33-1.70)
	No	86.1		75.9		88.5
Zone
Rural	Yes	88.5	1.04 (1.00-1.08)	64.6	0.99 (0.82-1.19)	82.8	0.95 (0.80-1.13)
	No	85.1		65.2		87.0
Periurban	Yes	92.1	1.05 (0.97-1.15)	88.5	0.99 (0.85-1.16)	100.0	1.07 (0.97-1.18)
	No	87.3		89.0		93.1
Urban	Yes	84.6	0.98 (0.95-0.99)	79.4	1.03 (1.01-1.06)	8.0	1.00 (0.98-1.02)
	No	85.9		76.8		8.4

• * Worker in traffic.
• Source: Authors.

Regarding race, in the White category (PR = 1.05; 95%CI = 1.01-1.10), a significant association was found, suggesting that White transit workers have a 5% higher prevalence of productivity loss compared to those who do not work in traffic. In the other race category (PR = 1.27; 95%CI = 1.05-1.52), the association was even more significant, indicating a 27% increase in the prevalence of productivity loss for workers in traffic of other races.

Similarly, the educational level showed a significant association for the elementary school category (PR = 1.08; 95%CI = 1.04-1.11), revealing an 8% increase in the prevalence of productivity loss for workers in traffic with this educational level. Finally, for urban residents (PR = 1.03; 95%CI = 1.01-1.06), the association was also statistically significant, indicating a 3% increase in the prevalence of productivity loss for workers in traffic residing in urban areas.

These results highlight the importance of considering these sociodemographic variables when analyzing the association between work in traffic and productivity loss, especially in the outpatient setting.

Analyzing the results presented in Table 2, it is clear that the relationship between work in traffic and productivity loss is still more pronounced in the outpatient setting. Among the occupational variables, statistically significant associations were observed in several categories:

Table 2
Prevalence of productivity loss due to workplace accidents among workers in traffic, according to categories of occupational variables, stratified by treatment regimen. Brazil, 2009-2018.

Variables	TT*	Treatment regimen
		Hospital		Outpatient		Both
		Productivity loss		Productivity loss		Productivity loss
		%	PR (95%CI)	%	PR (95%CI)	%	PR (95%CI)
Type of relationship
With work relationship	Yes	84.3	0.99 (0.98-101)	82.1	1.05 (1.03-1.08)	89.9	0.99 (0.97-1.02)
	No	85.0		77.5			90.7
Without work relationship	Yes	86.0	1.00 (0.98-1.03)	65.4	0.95 (0.86-1.06)	93.8	1.07 (0.99-1.15)
	No	85.7		68.4			87.6
Employer/self-employed	Yes	88.3	1.01 (0.99-1.02)	74.8	1.04 (0.97-1.14)	90.6	0.97 (0.92-1.03)
	No	87.5		71.3			93.0
Other	Yes	83.7	0.93 (0.89-0.98)	56.1	0.79 (0.63-1.00)	96.2	1.04 (0.96-1.12)
	No	89.7		70.3			92.7
Time at job
More than 10 years	Yes	86.0	0.96 (0.94-0.99)	75.9	0.97 (0.90-1.06)	91.3	1.01 (0.95-1.07)
	No	89.2		74.0			90.2
from 2 to 10 years	Yes	87.6	1.01 (0.99-1.02)	71.6	0.98 (0.93-1.04)	89.3	1.00 (0.96-1.05)
	No	87.2		72.7			89.0
Less than 2 years	Yes	86.6	1.00 (0.98-1.01)	73.6	1.04 (1.01-1.09)	88.3	0.98 (0.94-1.02)
	No	86.7		70.4			90.0
Economic activity
Public adm., defense, and security	Yes	98.2	1.06 (1.03-1.08)	88.9	1.11 (1.01-1.23)	90.0	1.02 (0.81-1.28)
	No	92.9		79.8			88.2
Agriculture, livestock, and fishing	Yes	72.8	0.88 (0.76-1.01)	70.0	0.89 (0.66-1.19)	100.0	1.03 (1.01-1.06)
	No	83.0		78.8			96.7
Accommodation and food	Yes	89.2	0.99 (0.93-1.06)	73.9	0.91 (0.76-1.09)	100.0	1.03 (0.99-1.08)
	No	89.9		81.3			96.7
Commerce; vehicle repair	Yes	86.2	1.00 (0.96-1.04)	86.0	1.06 (1.01-1.12)	92.2	0.97 (0.91-1.04)
	No	86.1		80.6			94.7
Construction	Yes	74.8	0.86 (0.74-1.01)	82.9	0.96 (0.82-1.13)	100.0	1.07 (1.02-1.12)
	No	86.0		85.8			93.7
Education, health, and social services	Yes	73.5	0.93 (0.78-1.10)	70.4	1.08(0.84-1.39)	88.9	0.99 (0.78-1.26)
	No	79.0		65.2			89.6
Transformation and extractive industries	Yes	82.7	0.97 (0.90-1.05)	82.9	1.03 (0.93-1.14)	95.6	1.04 (0.94-1.14)
	No	84.9		80.4			92.1
Electricity production/distribution	Yes	66.7	0.80 (0.35-1.79)	0.0	-	100.0	1.00 (1.00-1.00)
	No	83.7		55.6	-		100.0
Services	Yes	87.3	0.99 (0.92-1.06)	87.4	1.01 (0.92-1.10)	95.4	1.02 (0.92-1.12)
	No	88.3		86.5			93.6
Domestic services	Yes	90.0	1.11 (0.86-1.43)	75.0	1.78 (1.06-2.99)	0.0	0.00 (-)
	No	80.9		42.0			100.0
Trans., storage, and communications	Yes	83.4	0.93 (0.89-0.98)	83.9	0.97 (0.91-1.03)	93.4	0.99 (091-1.07)
	No	89.1		86.7			94.7
Outsource
Yes	Yes	83.4	1.01 (0.95-1.07)	80.2	1.08 (0.97-1.19)	90.9	0.99 (0.89-1.10)
	No	82.4		74.4			91.8
No	Yes	86.0	0.99 (0.98-1.01)	79.2	1.04 (1.02-1.07)	88.7	0.99 (0.96-1.02)
	No	86.4		75.8			89.5

• * Worker in traffic.
• Source: Authors.

Workers in traffic with a formal job revealed a stronger association (PR = 1.05; 95%CI = 1.03-1.08), indicating a 5% higher prevalence of productivity loss compared to workers who do not work in traffic. Those with less than two years of work in traffic experience were 4% more likely to experience productivity loss, as compared to workers with the same amount of experience who did not work in traffic.

The economic activity of public administration, defense, and security showed a higher association (PR = 1.11; 95%CI = 1.01-1.23) for workers in traffic, reflecting an 11% higher prevalence of productivity loss in this sector. In commerce and vehicle repair, the association was significant (PR = 1.06; 95%CI = 1.01-1.12), indicating a 6% higher prevalence of productivity loss for workers in traffic.

In the context of domestic services, the association was substantially higher (PR = 1.78; 95%CI = 1.06-2.99), demonstrating a 78% higher prevalence of productivity loss for workers in traffic in this segment. Finally, in non-outsourced organizations, there was also an association (PR = 1.04; 95%CI = 1.02-1.07), indicating a 4% higher prevalence of productivity loss for workers in traffic. These results highlight the importance of considering specific nuances of occupational variables when analyzing the association between work in traffic and productivity loss, especially in the outpatient setting.

Table 3 shows the association between work in traffic and productivity loss due to TA, considering different treatment regimens. A higher prevalence in the outpatient setting stands out, with specific categories of independent variables associated with the outcome. In cases of accidents involving workers in traffic in the morning (PR = 1.05; 95%CI = 1.02-1.09) and afternoon (PR = 1.05; 95%CI = 1.02-1.10) shifts; in the first six hours of the workday (PR = 1.05; 95%CI= 1.02-1.08); and for road users. such as motorcyclists (PR = 1.03; 95%CI = 1.01-1.05), collisions (PR = 1.05; 95%CI = 1.02-1.09) and other types of TA (PR = 1.04; 95%CI = 1.01-1.08). Furthermore, a strong association was found in cases of typical SCA (PR = 1.10; 95%CI = 1.06-1.14) and when care was provided in the same municipality as the accident (PR = 1.04; 95%CI = 1.01-1.06), as compared to the same categories of workers not involved in traffic.

Table 3
Prevalence of productivity loss due to workplace accidents among workers in traffic, according to categories of variables related to accident characteristics, stratified by treatment regimen. Brazil, 2009-2018.

Variables	TT*	Treatment regimen
		Hospital		Outpatient		Both
		Productivity loss		Productivity loss		Productivity loss
		%	PR (95%CI)	%	PR (95%CI)	%	PR (95%CI)
Shift
Morning	Yes	84.8	1.00 (0.98-1.02)	79.4	1.05 (1.02-1.09)	90.0	0.99 (0.96-1.03)
	No	84.9		75.2		90.8
Afternoon	Yes	85.4	1.00 (0.98-1.01)	78.8	1.05 (1.02-1.10)	92.1	1.01 (0.97-1.04)
	No	85.7		74.6		91.7
Night	Yes	86.7	0.99 (0.97-1.01)	77.0	0.97 (0.91-1.03)	91.8	1.00 (0.95-1.04)
	No	87.3		79.2		91.9
Early morning	Yes	83.3	0.97 (0.94-1.01)	72.7	0.93 (0.85-1.01)	88.3	0.96 (0.89-1.04)
	No	85.6		78.2		91.6
Time of shift
First 6 hours	Yes	86.7	1.01 (0.99-1.02)	79.2	1.05 (1.02-1.08)	90.0	0.99 (0.96-1.02)
	No	86.1		75.2		90.9
From 7 to 12 hours	Yes	86.3	0.98 (0.96-0.99)	77.8	0.99 (0.94-1.04)	92.4	1.01 (0.97-1.05)
	No	88.2		78.4		91.4
More than 12 hours	Yes	86.0	0.98 (0.93-1.04)	70.2	0.95 (0.78-1.15)	88.9	0.96 (0.86-1.07)
	No	87.4		73.6		92.7
Type of transportation
Pedestrian	Yes	81.5	1.01 (0.96-1.06)	79.6	1.04 (0.94-1.15)	81.6	0.91 (0.78-1.06)
	No	80.6		76.6		90.0
Cyclist	Yes	84.2	1.02 (0.97-1.09)	74.4	1.11 (0.98-1.28)	82.8	0.92 (0.77-1.09)
	No	81.8		66.5		90.0
Motorcyclist	Yes	89.0	1.01 (1.00-1.01)	81.5	1.03 (1.01-1.05)	93.2	1.01 (0.98-1.03)
	No	88.3		79.1		92.5
Vehicle occupant	Yes	76.9	0.98 (0.96-1.01)	75.0	1.04 (0.98-1.10)	87.2	1.01 (0.95-1.07)
	No	78.2		72.0		86.1
Others	Yes	71.3	0.91 (0.86-0.98)	67.9	0.94 (0.84-1.05)	94.1	1.04 (0.96-1.12)
	No	77.5		72.2		90.5
Type of TA
Run over	Yes	81,5	1,01 (0,95-1,07)	79,6	1,04 (0,93-1,15)	81,6	0,91 (0,78-1,06)
	No	80.6		76.6		89.6
Colision	Yes	87.7	1.01 (0.99-1.01)	80.3	1.05 (1.02-1.09)	92.6	1.01 (0.99-1.04)
	No	87.4		76.1		91.6
Others	Yes	86.5	1.01 (0.99-1.03)	81.4	1.04 (1.01-1.08)	89.7	0.97 (0.94-1.01)
	No	85.8		77.6		92.0
Type of SCA
Typical	Yes	84.4	1.01 (0.99-1.02)	81.9	1.10 (1.06-1.14)	93.8	1.04 (1.01-1.08)
	No	84.1		74.3		89.9
Trajectory	Yes	85.6	0.99 (0.98-1.01)	77.0	1.01 (0.97-1.03)	88.7	0.97 (0.93-1.02)
	No	86.1		76.8		91.6
Care provided in the municipality where the accident occurred
Yes	Yes	84.55	0.99 (0.98-1.00)	78.6	1.04 (1.01-1.06)	90.6	0.99 (0.97-1.02)
	No	85.45		75.6		91.1
No	Yes	85.46	0.98 (0.97-1.00)	81.2	1.01 (0.96-1.05)	93.1	1.01 (0.97-1.04)
	No	86.70		80.7		92.4

• * Worker in traffic.
• Source: Authors.

These results point to a significant relationship between work in traffic and productivity loss, especially in outpatient settings and in specific accident and care situations.

Table 4 presents the results of the multilevel analysis for productivity loss due to workplace accidents among workers in traffic in Brazil, according to the treatment regimen. The model presented obtained the lowest AIC value (68155.38) among the models tested, thus showing the best performance in terms of fit.

Table 4
Prevalence ratio (PR) and 95% confidence intervals (95%CI) values of the multilevel analysis for productivity loss due to workplace accidents among workers in traffic. Brazil, 2009-2018.

Variables	Treatment regimen
	Hospital	Outpatient	Both
	PR (95%CI)	PR (95%CI)	PR (95%CI)
Contextuaal
HDI
Average	1.05 (0.95-1.16)	0.91 (0.81-1.02)	0.96 (0.86-1.07)
Low	0.95 (0.83-1.08)	0.81 (0.70-0.95)	0.97 (0.85-1.10)
Size of municipality
Mid-sized	1.18 (01.08-1.29)	1.43 (1.26-1.61)	1.24 (1.12-1.37)
Small	1.31 (1.21-1.43)	1.45 (1.30-1.60)	1.24 (1.11-1.37)
Main exposure
Work in Traffic	1.01 (0.09-1.02)	1.08 (1.04-1.13)	0.97 (0.92-1.01)
Individuals
Educational level
High school	1.04 (1.01-1.07)	1.07 (1.01-1.15)	1.02 (0.95-1.12)
Elementary	1.04 (1.01-1.08)	1.13 (1.05-1.22)	1.05 (0.98-1.14)
No schooling	1.03 (0.97-1.09)	1.13 (0.97-1.32)	0.98 (0.81-.1.19)
Time at job
2-10 years	0.97 (0.96-0.99)	0.95 (0.92-0.99)	0.98 (0.94-1.02)
< 2 years	0.97 (0.95-0.99)	0.93 (0.89-0.97)	0.99 (0.95-1.03)
Type of work relationship
No work relationship	1.01 (0.98-1.03)	1.04 (0.99-1.10)	1.02 (0.96-1.08)
Employer/Self-employed	1.02 (1.01-1.04)	0.97 (0.92-1.03)	1.06 (1.01-1.10)
Others	0.95 (0.89-1.01)	1.05 (0.91-1.20)	1.01 (0.89-1.13)
Outsource
No	0.96 (0.92-0.99)	1.02 (0.96-1.08)	1.05 (1.00-1.10)
Shift
Afternoon	1.01 (0.99-1.02)	0.99 (0.96-1.02)	1.01 (0.99-1.04)
Night	1.02 (1.01-1.04)	0.98 (0.93-1.03)	1.02 (0.98-1.06)
Early morning	1.01 (0.99-1.04)	0.99 (0.94-1.04)	0.99 (0.91-1.07)
Type of traffic user
Cyclist	1.05 (0.99-1.12)	0.97 (0.90-1.04)	1.09 (1.02-1.17)
Motorcyclist	1.07 (1.03-1.12)	1.05 (1.00-1.12)	1.09 (1.02-1.16)
Vehicle occupant	1.01 (0.97-1.04)	0.99 (0.93-1.05)	0.96 (0.85-1.08)
Others	1.01 (0.90-1.04)	1.11 (0.95-1.30)	1.00 (0.70-1.41)
Care provided in the municipality where the accident occurred
No	1.02 (1.00-1.05)	1.03 (0.99-1.09)	1.00 (0.96-1.04)

• Source: Authors.

In the hospital regimen, a higher prevalence of productivity loss was observed among workers in small municipalities (PR = 1.31; 95%CI = 1.21-1.43), with elementary (PR = 1.04; 95%CI = 1.01-1.08) or high school (PR = 1.04; 95%CI = 1.01-1.07) education, and among those employed or self-employed (PR = 1.02; 95%CI = 1.01-1.04). The positive association with the night shift (PR = 1.02; 95%CI = 1.01-1.04), with motorcyclists (PR = 1.07; 95%CI = 1.03-1.12), and with care outside the municipality of the accident (PR = 1.02; 95%CI = 1.00-1.05) also stood out. In the outpatient setting, work in traffic (PR = 1.08; 95%CI= 1.04-1.13), living in small (PR = 1.45; 95%CI= 1.30-1.60) or mid-sized municipalities (PR = 1.43; 95%CI = 1.26-1.61), and having elementary (PR = 1.13; 95%C I= 1.05-1.22) or high school (PR = 1.07; 95%CI = 1.01-1.15) education were associated with a higher prevalence of the outcome. Motorcyclists also showed a higher prevalence (PR = 1.05; 95%CI = 1.00-1.12).

In the combined regimen, the highest prevalence was observed among workers from small municipalities (PR = 1.24; 95%CI = 1.11-1.37), who were self-employed or had a similar form of employment (PR = 1.06; 95%CI = 1.01-1.10), and among motorcyclists (PR = 1.09; 95%CI = 1.02-1.16) and cyclists (PR = 1.09; 95%CI = 1.02-1.17).

Discussion

Based on a cross-section of SCAs in Brazil, this study investigated the factors associated with productivity loss due to TA among workers in traffic. Despite its relevance, this topic is rarely discussed at a global level, a fact evidenced by the scarce literature on the subject¹³.

This research used an unusual methodological approach for data analysis: the multilevel logistic regression model. This analytical approach allowed us to examine the effects and interactions of group (contextual) and individual characteristics with the outcome under study²².

The high prevalence of productivity loss due to work-related TA reflects the process of productive restructuring that has occurred since the beginning of last century, characterized by the flexibilization and precariousness of work processes in Brazil. This phenomenon has been intensified in recent years by changes in labor laws, which reduce the control and monitoring of workers’ health, leaving them more vulnerable to workplace accidents, illnesses, and disabilities²³.

Regarding sociodemographic characteristics, it was observed that younger workers in traffic, those with an elementary education, and those living in urban areas experienced greater productivity losses, especially when undergoing outpatient treatment. Although previous studies indicate that men are more frequently involved in workplace accidents²⁴ and TAs²⁵, in our study, productivity losses were more prevalent among women. It is important to note, however, that the gender variable was not included in the final multilevel analysis, which limits robust inferences about this association. Furthermore, the literature still lacks studies that provide in-depth analyses of the repercussions of TAs, such as productivity losses, specifically among women. Therefore, it is recommended that future studies further explore the relationship between gender and workplace and traffic accidents and their impact on productivity.

The results found, according to occupational variables, demonstrated an association between productivity losses and outpatient treatment. Regarding the type of employment relationship, studies on serious work-related TAs indicate that workers with formal employment contracts tend to be more frequently involved in workplace accidents²⁶, which corroborates the findings of this study, which identified a higher prevalence of productivity loss among workers with formal employment contracts. However, it is important to consider that this association may partly reflect differences in access to labor rights and social security coverage between formal and informal workers. Workers without formal employment contracts, lacking the same rights (such as paid leave and the issuance of a CAT), often remain active even after the accident, which may underestimate their actual unproductivity. Therefore, the results presented may go beyond the prevalence of productivity loss due to work-related accidents and actually reveal inequalities in the social protection of informal workers.

Regarding the economic activity variable, Table 2 shows that the categories that showed a statistically significant association with productivity loss tend to belong to historically more formalized sectors, such as public administration and commerce. This formalization implies a greater likelihood of official accident reports and issuance of a Workplace Accident Report (Comunicação de Acidente de Trabalho - CAT), which increases the statistical visibility of these events. By contrast, markedly informal sectors, such as agriculture, construction, and some services, showed lower or non-significant prevalence rates, which may partly reflect the underreporting of workplace accidents in these groups²⁷.

It was also observed that productivity loss was more prevalent and associated with workers in traffic with shorter tenure. Among the behavioral factors that most influence the occurrence of TA, driver inexperience stands out. This is related to reckless behavior, such as speeding, as well as a lack of knowledge of the vehicle, the road, and traffic laws²⁸.

Regarding accident characteristics, the results show that, for the outpatient treatment regimen, there was an association between morning and afternoon shifts and productivity loss among workers in traffic. This finding differs from the study by Cho et al. (2020)²⁹, which identified an association between productivity loss and night shifts. One possible explanation for this discrepancy is that during the day there is a greater traffic volume and, consequently, greater exposure of workers in traffic to accidents. Furthermore, studies indicate that, although accident severity is greater during the early morning hours due to such factors as speeding and alcohol consumption, the absolute number of accidents is higher during the day, which may result in greater productivity loss during this period³⁰.

The association with productivity loss can also be observed in workers who were using motorcycles at the time of the accident. In this sense, one must consider the increase in motorcycles on public roads, their use as a work tool, especially for passenger transport and package delivery, as well as the conditions these two-wheeled workers are exposed to, which are vulnerable to accidents and injuries³¹.

Due to poor public transportation infrastructure, or in some cases, the lack thereof, motorcycles are used as the primary means of transportation in many small municipalities and on the outskirts of large capital cities. In most cases, the motorcycles used for this purpose are low-displacement vehicles, as they are cheaper and consume less fuel; they are purchased by people with lower purchasing power, who use them as a work tool, such as motorcycle taxi drivers and delivery drivers^32,33. The findings of this study corroborate this information, which points to the association of motorcycle use with productivity loss in post-accident workers.

Furthermore, it is important to highlight the low adherence to the proper use of personal protective equipment (PPE), such as helmets, reinforced jackets, and appropriate sidewalks, as well as the frequency of risky and reckless maneuvers in traffic among professional motorcyclists. These practices significantly increase the risk of serious injuries, which can contribute to longer time away from work and lost productivity³⁴.

Regarding the multilevel analysis, it was observed that the contextual variables exerted a more intensifying effect on the outcome, since the measure of association was increased in the independent variables selected for the final analysis model. A greater proportion of productivity loss was also observed for these variables. As a result, productivity loss due to TA can be considered a socioeconomic problem that requires social and economic solutions³⁵.

Regarding the limitations of this study, despite all the methodological care adopted, it is worth noting that the use of secondary data from information systems, whose quality has often been questioned, and the possible inadequate recording of the SCA investigation form (SINAN), underreporting, especially in small municipalities where workplace accidents occur without proper notification, are part of this process. However, the size of the database, with a significant number of incidents, helps minimize potential biases in the research. In conclusion, important sociodemographic, occupational, and accident-related factors were associated with productivity loss due to TA, with a greater prevalence of this event found among women, younger individuals, those of mixed race or black descent, those with a high school education, those with a formal employment relationship, those with less work experience, motorcyclists, and those involved in commuting accidents. Contextual variables modified the measures of association between covariates and the outcome in the final multilevel analysis model, suggesting the influence of socioeconomic factors on the prevalence of this study’s outcome.

Thus, the results of this study confirm that both SCA and TA are serious public health problems. For both, productivity loss is a consequence that goes beyond individual losses, but also affects institutions and overburdens organizations, the healthcare system, social assistance, and pension systems. Therefore, investments in urban mobility and infrastructure policies are required, as are educational initiatives that make commuting safer for all involved.

References

• ¹ World Health Organization (WHO). Global status report on road safety 2018. Geneva: WHO; 2018.
• ² Bhalla K, Gleason K. Effects of vehicle safety design on road traffic deaths, injuries, and public health burden in the Latin American region: a modelling study. Lancet Glob Health. 2020; 8(6):e819-e828.
• ³ Martín-de-Los Reyes LM, Martínez-Ruiz V, Lardelli-Claret P, Moreno-Roldán E, Molina-Soberanes D, Jiménez-Mejías E. Association between type of vehicle and the risk of provoking a collision between vehicles. Gac Sanit 2020; 34(4):350-355.
• ⁴ Hadaye RS, Rathod S, Shastri S. A cross-sectional study of epidemiological factors related to road traffic accidents in a metropolitan city. J Family Med Prim Care 2020; 9(1):168-172.
• ⁵ Jalilian MM, Safarpour H, Bazyar J, Keykaleh MS, Malekyan L, Khorshidi A. Environmental related risk factors to road traffic accidents in Ilam, Iran. Med Arch 2019; 73(3):169-172.
• ⁶ Mahajan K, Velaga NR. Effects of partial sleep deprivation on braking response of drivers in hazard scenarios. Accid Anal Prev 2020; 142:105545.
• ⁷ Choudhary P, Velaga NR. Effects of phone use on driving performance: a comparative analysis of young and professional drivers. Saf Sci 2019; 111:179-187.
• ⁸ Boulagouas W, García-Herrero S, Chaib R, Febres JD, Mariscal MA, Djebabra M. An investigation into unsafe behaviors and traffic accidents involving unlicensed drivers: a perspective for alignment measurement. Int J Environ Res Public Health 2020; 17(18):6743.
• ⁹ Byler C, Kesy L, Richardson S, Pratt SG, Rodríguez-Acosta RL. Work-related fatal motor vehicle traffic crashes: matching of 2010 data from the Census of Fatal Occupational Injuries and the Fatality Analysis Reporting System. Accid Anal Prev 2016; 92:97-106.
• ¹⁰ Charbotel B, Martin JL, Chiron M. Work-related versus non-work-related road accidents, developments in the last decade in France. Accid Anal Prev 2010; 42(2):604-611.
• ¹¹ Souto CC, Reis FKW, Bertolini RPT, Lins RSMA, Souza SLB. Perfil das vítimas de acidentes de transporte terrestre relacionados ao trabalho em unidades de saúde sentinelas de Pernambuco, 2012-2014. Epidemiol Serv Saude 2016; 25(2):351-360.
• ¹² Cardoso JP, Mota ELA, Ferreira LN, Rios PAA. Custos de produtividade entre pessoas envolvidas em acidentes de trânsito. Cien Saude Colet 2020; 25(2):749-760.
• ¹³ Cardoso JP, Mota ELA, Rios PAA, Ferreira LN. Fatores associados à perda de produtividade em pessoas envolvidas em acidentes de trânsito: um estudo prospectivo. Rev Bras Epidemiol 2020; 23:e200015.
• ¹⁴ Carozzi S, Elorza ME, Moscoso NS, Ripari NV. Metodologías para estimar los costos indirectos de los accidentes de tránsito. Rev Med Inst Mex Seguro Soc 2017; 55(4):441-451.
• ¹⁵ Chantith C, Permpoonwiwat CK, Hamaide B. Measure of productivity loss due to road traffic accidents in Thailand. IATSS Res 2021; 45(1):131-136.
• ¹⁶ Holko P, Kawalec P, Mossakowska M, Pilc A. Health-related quality of life impairment and indirect cost of Crohn's disease: a self-report study in Poland. PLoS One 2016; 11(12):e0168586.
• ¹⁷ Miškulin I, Ambroš I, Ambroš HAAB. Productivity losses from road traffic deaths in Croatia. Inter Manag Res 2014; 10:732-741.
• ¹⁸ Naci H, Baker TD. Productivity losses from road traffic deaths in Turkey. Int J Inj Contr Saf Promot 2008; 15(1):19-24.
• ¹⁹ Zhou Y, Baker TD, Rao K, Li G. Productivity losses from injury in China. Inj Prev 2003; 9(2):124-127.
• ²⁰ Brasil. Ministério do Trabalho e Emprego (MTE). Portaria nº 397, de 9 de outubro de 2002. Aprova a Classificação Brasileira de Ocupações - CBO/2002, para uso em todo território nacional e autoriza a sua publicação. Diário Oficial da União 2002; 10 out.
• ²¹ Brasil. Ministério da Saúde (MS). Banco de dados do Sistema Único de Saúde - DATASUS, Informações de Saúde (TABNET) [Internet]. [acessado 2020 out 29]. Disponível em: http://www2.datasus.gov.br/DATASUS/index.php?area=02
  » http://www2.datasus.gov.br/DATASUS/index.php?area=02
• ²² Puente-Palacios KE, Laros JA. Análise multinível: contribuições para estudos sobre efeito do contexto social no comportamento individual. Estud Psicol (Campinas) 2009; 26(3):349-361.
• ²³ Costa BS, Costa SS, Cintra CLD. Os possíveis impactos da reforma da legislação trabalhista na saúde do trabalhador. Rev Bras Med Trab 2018; 16(1):109-117.
• ²⁴ Souza ACD, Barbosa IR, Souza DLB. Prevalence of occupational accidents and associated variables in the Brazilian workforce. Rev Bras Med Trab 2021; 18(4):434-443.
• ²⁵ Rios PAA, Mota ELA, Ferreira LN, Cardoso JP, Ribeiro VM, Souza BS. Factors associated with traffic accidents among drivers: findings from a population-based study. Cien Saude Colet 2020; 25(3):943-955.
• ²⁶ Santos ER, Bertolin DC, Santos LL, Fucuta PS, Pompeo DA, André JC. Relação entre qualidade de vida, transtornos mentais menores e resiliência entre profissionais de enfermagem. In: Alves GSB, Oliveira E, organizadoras. Tópicos em Ciências da Saúde - Volume 24. Belo Horizonte: Editora Poisson; 2021. p. 121-134.
• ²⁷ Gonçalves MR, Gonçalves MR, Ito FY, Mizoguti NN, Hirota MM, Hayashida MR, Oliveira JLC. Acidentes de trabalho graves notificados em uma unidade sentinela, no período entre 2008 e 2018. Rev Bras Med Trab 2021; 19(3):299-306.
• ²⁸ Fortes AG, Mamudo AA, Chau MJ, Fernando ET. Estudo sobre os fatores que contribuem no acidente de trânsito na cidade de Nampula. RECIMA21 2021; 2(1):267-287.
• ²⁹ Cho SS, Lee DW, Kang MY. The association between shift work and health-related productivity loss due to either sickness absence or reduced performance at work: a cross-sectional study of Korea. Int J Environ Res Public Health 2020; 17(22):8493.
• ³⁰ Rocha GS, Silva CA, Crispim LV. Gravidade e lesões traumáticas em vítimas de acidente de trânsito internadas em um hospital público. Rev Enferm Cent O Min 2021; 11:e3870.
• ³¹ Corgozinho MM, Montagner MA, Rodrigues MAC. Vulnerabilidade sobre duas rodas: tendência e perfil demográfico da mortalidade decorrente da violência no trânsito motociclístico no Brasil, 2004-2014. Cad Saude Colet 2018; 26(1):92-99.
• ³² Mendonça BMP, Souza NKT, Borges JHS, Neto JSA. Perfil do condutor de moto vítima de acidente de trânsito no Distrito Federal. Rev Bras Med 2021; 58:1-6.
• ³³ Souza CDF, Machado MF, Quirino TRL, Leal TC, Paiva JPS, Magalhães APN, Silva Júnior AG. Padrões espaciais e temporais da mortalidade de motociclistas em estado do nordeste brasileiro no século XXI. Cien Saude Colet 2021; 26(4):1501-1510.
• ³⁴ Souto RMCV, Corassa RB, Lima CM, Malta DC. Uso de capacete e gravidade de lesões em motociclistas vítimas de acidentes de trânsito nas capitais brasileiras: uma análise do Viva Inquérito 2017. Rev Bras Epidemiol 2020; 23(Supl. 1):e200011.supl.1.
• ³⁵ Haghighi MRR, Sayari M, Ghahramani S, Lankarani KB. Social, economic, and legislative factors and global road traffic fatalities. BMC Public Health 2020; 20:1413.

• 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
  10 Nov 2023
• Accepted
  14 May 2025
• Published
  16 May 2025