Work-related diseases and health-related compensation claims , Northeastern Brazil , 2000

OBJECTIVE: To estimate the contribution of work-related diseases to sick leaves due to general and occupational health problems. METHODS: Sociodemographic, occupational and health data from 29,658 records of temporary disability benefits, granted on account of health problems by the Instituto Nacional do Seguro Social (National Institute of Social Security) in the state of Bahia (Northeastern Brazil), were analyzed. All constant ICD-10 clinical diagnoses were taken into consideration, except for those referring to external causes and factors that infl uence contact with health services. The link between diagnosis and occupation was based on the ICD-10 code and whether the type of compensation was due to a “work-related accident/disease” or not. RESULTS: From all the benefi ts, 3.1% were granted due to work-related diseases: 70% were musculoskeletal system and connective tissue diseases, while 14.5% were related to the nervous system. In general, benefi ts granted at more than two times the expected frequency were as follows: tenosynovitis in the manufacturing sector (Proportion Ratio-PR=2.70), carpal tunnel syndrome in the fi nancial intermediation sector (PR=2.43), and lumbar disc degeneration in the transportation, postal service and telecommunications sectors (PR=2.17). However, no causal connection could be established for these diseases, in these activity sectors, in a signifi cant percentage of benefi ts. CONCLUSIONS: Results suggest the existence of possible occupational risk factors for diseases in these fi elds of activity, as well as the underreporting of the link between diseases and work, thus disguising the responsibility of companies and the perspective of prevention through work reorganization. DESCRIPTORS: Occupational Diseases. Causality. Diseases Registries. Insurance Benefi ts. Social Security. Occupational Health.


INTRODUCTION
Despite the fact that work-related diseases (WRD) are recognizably avoidable, they are responsible for most of workers' morbidity and, may cause disability and even death. 8In Brazil, WRD claims have increased from 5,025 in 1988 to 30,334 in 2005, among workers under the Regime Geral da Previdência Social (Social Security General System), a resulting in an increment of occupational benefi ts for lost wage compensation.On the other hand, there has been a decreasing tendency for the occurrence of these diseases in developed countries.In the United States, the 2002 Bureau of Labor Statistics a registered 294,500 WRD claims, a number that dropped to 242,500 in 2005.b In Ontario, Canada, the estimated WRD rates, based on benefi t payments, decreased by approximately 50% in the last decade. 10e causes suggested for the decrease in WRDs are many, from underreporting 9 to macro-economic factors, such as demographic changes in the workers' population and in the employment distribution among economic trades. 10Another determinant factor could be the primary prevention in the workplace.Controversies over explanations for these changes in the occupational morbidity profi le have been observed worldwide and sparked debates about the required changes in the approach towards workers' healthcare and prevention.
Overall, information on the WRD magnitude and the establishment of diagnoses come from governmental institutions.In developed countries, despite restrictions, the most frequently used sources to learn about WRDs are the Workers' Compensation Insurance data, a specifi c type of insurance to compensate workers' lost wages when affected by work-related diseases/accidents.Driscoll et al 5 mention that one of the advantages of these databases is the existence of standard systems to confi rm cases and register information.On the other hand, they have important limitations: they only include employees; they may exclude trades such as the armed forces and agriculture workers, who are usually self-employed; and also WRD underreporting, due to the diffi culty to establish an occupational causal connection.
In Brazil, workers under the Regime Geral da Previdência Social who may be granted WRD benefi ts are employed workers, self-employed professionals, and special insurance holders.The specialized medical doctors from the Instituto Nacional do Seguro Social -INSS (National Institute of Social Security), a governmental institution linked to the Ministry of Social Security, are the ones responsible for establishing a causal connection between diseases and work.Among employed workers, this only happens in case of harmful situations involving sick leaves longer than 15 days, when the INSS pays for those unable to work.The establishment of a causal connection is based on the expert's clinical and epidemiological knowledge, according to a WRD list prepared by the Ministry of Social Security.Until March of 2007, the INSS required that a Comunicação de Acidente de Trabalho (CAT -Work-related accident communication) was issued to characterize a work-related disease.However, beginning in April of 2007, the causal connection is also established based on the Nexo Técnico Epidemiológico (Epidemiological-Technical Connection), which means, an excess of risk related to the workers' industrial trade, regardless of the CAT c being issued.
To be aware of the contribution of WRDs to the sum of leaves due to general and occupational health problems, as well as their distribution among the several economic trades and respective, most frequent diagnoses, may subsidize decisions aimed towards the adoption of preventive measures.Moreover, this will help to promote public debate over the impact of these health problems on social insurance.
The objective of the present study was to estimate the proportion of temporary disability compensation claims on account of work-related diseases.

METHODS
This is a cross-sectional study that used records of benefi ts granted by the INSS due to general and work-related health problems which led to temporary disability, in the state of Bahia, in 2000.These records were obtained from the INSS' Sistema Único de Benefícios (SUB -Unifi ed System of Benefi ts).A total of 29,663 benefi ts were granted on account of temporary occupational disability in this state.Of these, fi ve were excluded from the analysis as they did not include the clinical diagnosis.The total number of benefi ts studied was 29,658.
The benefi ts granted due to work-related accidents and diseases are coded as B91 by the INSS; whereas those due to health problems that are not related to work are coded as B31.The total study population involved all insured workers in the period who were eligible to receive these two types of benefi ts.They were, thus, under the Regime Geral da Previdência Social, from which domestic workers, individual contributors (self-employed professionals), civil servants and non-contributing workers are excluded.The study observation units were all paid B91 and B31 benefi t records, except for the clinical diagnoses corresponding to chapter XV of the ICD-10 (pregnancy, childbirth and puerperium).
The SUB is the INSS' data registry system, processed by the DATAPREV, the Ministry of Social Security's technology and information company, where each social security event is registered, thus originating the concession of a benefi t.These registers include the company's and the employee's data: company's trade, according to the Classifi cação Nacional de Atividade Econômica (CNAE -National Economic Activity Classifi cation); health problem diagnosis, according to the 10 th revision of the International Classifi cation of Diseases (ICD-10); and data on the beginning, duration and type of benefi t.
The disease variable was created (1=yes, 0=other diagnoses), defi ned as all the ICD-10 clinical diagnoses, except for those that characterize accidents, included in chapters XIX (injuries, poisoning, and certain other consequences of external causes) and XXI (factors infl uencing health status and contact with health services), which were considered as other diagnoses.Diseases were classifi ed in accordance with ICD-10 groups.Three specifi c diagnoses were taken into consideration in the analysis, the carpal tunnel syndrome (G56 and G56.0), tenosynovitis (M65, M65.8, M65.9) and intervertebral disc degeneration (M51).Other variables analyzed were as follows: type of benefi t, whether it was related to work (B91) or not (B31), and company's trade (according to the CNAE).
Proportion comparisons were performed, but statistical inference was not, as all benefi ts were taken into consideration, rather than a sample.To calculate the proportion ratios (PR), the sum of benefi ts was considered as reference.Prevalence was not estimated because denominators were the sum of benefi ts, instead of the population from which cases come from.The databases were provided by the Ministry of Social Security and the analysis was performed using the SAS 8.1 statistical software.There were no individual identifi cation data about benefi ciaries or companies.As the data were administrative in nature, the study protocol was not submitted to a Research Ethics Committee.

RESULTS
Of all the benefi ts, 17,282 (58.3%) were granted due to diseases; 6,499 (21.9%) due to injuries, poisoning and other external causes; and 5,877 (19.8%) due to factors that infl uenced health status and contact with health services, according to the ICD-10 th Review.
Work-related diseases represented 3.1% (935) of all the temporary disability benefi ts granted on account of health problems in general.In terms of trade, this percentage varied from 1.2%, referring to undeclared CNAE, to 7.4% in the manufacturing trade, where the proportion of benefi ts per WRD was more than double the expected rate (PR=2.31).Similarly, the trade entitled "fi nancial intermediation, real estate, leases, services and public administration" showed 7.2% of WRDs (PR=2.25).Workers in the retail, storage and food trades showed the lowest percentages of WRDs (2.7%), among the benefi ts related to health problems in general.A low percentage of WRDs (2.7%) was also observed among benefi ts with undeclared CNAE (54%, 16,014) (Table 1).
Of all the temporary disability benefi ts granted due to work-related accidents and diseases, the WRDs contributed with 27.6%.Only in the "fi nancial intermediation, real estate, leases, services and public administration" trade did the WRDs comprise the majority of benefi ts (53.5%), a percentage almost two times greater than the total (PR=1.93).The group entitled "retail, storage and food trades" had the lowest WRD proportion (17.8%).The percentage of undeclared CNAE for benefi ts on account of work-related health problems was signifi cant (26.8%) (Table 1).
Only 935 (5.4%) temporary disability benefi ts granted due to diseases were related to work.The ICD-10 groups that ranked high in the sum of benefi ts were the following: musculoskeletal system and connective tissue diseases (37.2%), diseases of the circulatory system (19.3%) and behavioral and mental disorders (10.9%).By analyzing WRD benefi ts exclusively, it could be observed that musculoskeletal system and connective tissue diseases (70%) ranked high, followed by diseases of the nervous system (14.5%)(Table 2).
Table 3 shows the proportion of both types of benefi ts, granted to people suffering from carpal tunnel syndrome, tenosynovitis, and lumbar intervertebral disc degeneration, out of the sum of benefi ts granted due to temporary disability, according to the industrial trades.The proportion of benefi ts for tenosynovitis varied from 1.2% in the "construction, electricity, and gas" activity group to 6.2% among workers from the manufacturing trade.By taking the proportion of benefi ts for tenosynovitis from the total number of benefi ts granted on account of diseases (2.3%) as reference, almost three times more benefi ts for tenosynovitis were given to the manufacturing trade (PR=2.70)than what was expected.Important estimates were also calculated among workers from the fi nancial intermediation, real estate, leases, services, and public administration trades (PR=2.09);transportation, postal service, and telecommunications trade (PR=1.61);and health and social service trade (PR=1.61).Groups from the "construction, electricity and gas" and "retail, storage, and food" trades showed the smallest proportions of benefi ts for tenosynovitis, 1.2% and 2.2%, respectively.
Benefi ts for carpal tunnel syndrome represented 1.4% of those granted for diseases in general.The proportions of benefi ts for this syndrome, regardless of the relationship with work, were greater than those from the comparison group in the following trades: "fi nancial intermediation, real estate, leases, services, and public administration" (PR=2.43),"health and social services" (PR=2.43),"education, and recreational, cultural and sport activities" (PR=1.86),"manufacturing trade" (PR=1.78),"retail, storage, and food" (PR=1.28),and "transportation, postal service, and telecommunications" (PR=1.14).Only workers from the "construction, electricity and gas trades" and the "undeclared CNAE" group had smaller proportions for carpal tunnel syndrome than what was expected, 0.5 and 0.7, respectively (Table 3).
Of the three health problems assessed, benefi ts due to lumbar intervertebral disc degeneration had the greatest contribution: 3% of the sum of temporary disability benefi ts granted on account of diseases.In the "transportation, postal service, and telecommunications" trade, the proportion of benefi ts on account of this disorder was greater (6.5%).Other trades showed proportions that were a little above the comparison group, except for "fi nancial intermediation, real estate, leases, services, and public administration", which had estimates of 2.4%, and "undeclared CNAE", with 2.7% (Table 3).
Table 4 shows the proportions of temporary disability benefi ts granted on account of carpal tunnel syndrome, tenosynovitis, and lumbar intervertebral disc degeneration, per type of benefi t and trade.Only in four situations were greater proportions of work-related benefi ts observed, compared to those that were not related: carpal tunnel syndrome in the "fi nancial intermediation, real estate, leases, services and public administration" group of activities (69.5%), and tenosynovitis among workers from the "manufacturing trade" (60%), "transportation, postal service, and telecommunications" (53.8%) and "fi nancial intermediation, real estate, leases, services, and public administration" trade groups (65.8%).Even in these situations, at least one third of benefi ts were granted as not work-related.

DISCUSSION
The fi ndings from the present study must be viewed with caution because of database limitations.One piece of evidence of such limitations is the great number of benefi ts without a record of the company's economic trade, which varied from 54% for temporary disability benefi ts related to health problems in general to 26.8% for those related to work.In addition, these benefi ts correspond to leaves from work longer than 15 days, which presume greater severity of the diseases studied.Thus, diseases that do not potentially lead to leaves from work and those involving shorter leave periods are not entitled to benefi ts.
Another limitation to the study was that information on possible risk factors for health was not available.Such information, which includes age, sex, time of employment and occupational background, could interfere with the CNAE registered when the benefi t was granted.Work-related diseases with an occupational causal connection recognized, according to the INSS characterization, did not stand out in the group of general health problems that resulted in leaves from work, representing 3.1% of all the health problems.However, this percentage varied according to the trade assessed.The difference between the WRD percentage of the "manufacturing trade" and "fi nancial intermediation and others" (7%) and that of "retail" and "health and social services" (4%) is probably due to distinct occupational exposures in the workplace and the workers' level of knowledge of occupational diseases, their rights and benefi ts.Workers' awareness of health and safety rights is one of the factors known to determine more records of occupational health problems. 13In the state of Bahia, the bank and (chemical, oil, metal) manufacturing trade unions are recognizably more active in the area of occupational health than others.
Work-related diseases represent less than 30% of all the temporary disability benefi ts granted on account of health problems related to work, as expected, once it is less diffi cult to establish a causal connection for typical accidents in relation to diseases.The only exception occurred in the "fi nancial intermediation and others" trade, where the WRDs were the majority.In this fi eld, risk factors for occupational diseases, such as ergonomic ones, are more prevalent in the workplace, compared to those that cause typical accidents.
The main causes of benefi ts granted due to diseases in general, regardless of their relationship with work, were musculoskeletal system and connective tissue diseases, diseases of the circulatory system, and behavioral and mental disorders.In a study performed with social security benefi ts in the city of Porto Alegre (Southern Brazil) restricted to the B31 code, the most common diseases were the same found in this study, except for the rank order, with diseases of the circulatory system in place of the behavioral and mental disorders. 2e WRDs that predominated (84.5%) were musculoskeletal system and connective tissue diseases and those in the nervous system, corresponding, in their majority, to cumulative trauma disorders or osteomuscular work-related diseases (CTD/OWRD): muscle, nerve, tendon, joint, cartilage and intervertebral disc dysfunctions.The proportion of CTD/OWRD found in this study is substantially greater than those from other locations where musculoskeletal disorders are also prevailing.In outpatient services for occupational health in the state of São Paulo, musculoskeletal system diseases represented the main cause of service (56%), followed by diseases of the nervous system and sense organs (20.8%). 15However, the percentage of CTD/ OWRD was greater than the 56% reported, if carpal tunnel syndrome is included, as this is a peripheral nervous system disease that could also be considered as CTD/OWRD.In Canada, data from the Workers'  12 The reasons for this underreporting have already been discussed by several authors, 3,11,14 and range from the changes in the social security legislation, thus extending the period without insurance payment to 15 days, to the employer's failure to issue the CAT to avoid responsibilities, such as the employee's guarantee of stability, withdrawal from the Fundo de Garantia por Tempo de Serviço -FGTS (social security fund based on time of service), in addition to maintaining the company's reputation.
As regards WRDs, another factor that may increase underreporting is the controversy over causality.Unlike typical occupational accidents, whose connection with work is less subjective, it is more diffi cult to establish a causal connection with work when it comes to occupational diseases.Among diseases whose origin is exclusively due to work, such as asbestosis, silicosis, and mesotelioma, the diffi culty to recognize the causal connection is primarily related to the very long latency period between exposure and the fi rst disease symptoms, rather than the uncertainties about occupational causality.Currently, however, the predominant diseases in the workplace are those whose causal agent is not only work, frequently occurring in the non-working REFERENCES population as well or having non-occupational causes, such as CTD/OWRD, hearing losses, upper respiratory tract diseases, asthma, and mental disorders, among others.Thus, the causal connection for these health problems has caused confl icts to arise among companies, insurance companies, and workers all over the world.Underreporting harms workers who, by not having the disease characterized as work-related, do not have their rights recognized.On the other hand, it hinders the making of public policies to prevent occupational diseases and accidents, once there is no reliable information about these health problems available.
Thus, the Ministry of Social Security's initiative to adopt the Nexo Técnico Epidemiológico to establish a causal connection between the health problem and work is positive.This methodology considers a health problem having higher incidence among workers of a certain trade in relation to the average morbidity of the remaining working population as the identifi cation criterion for WRDs.The fi rst effects of the Nexo Técnico Epidemiológico can already be seen when compensations due to work-related diseases or accidents are granted.In April of 2007, when the Nexo Técnico Epidemiológico became effective, a total of 28,594 benefi ts were granted all over Brazil, a number 147.8% greater than the previous month's.a Thus, it is expected that the making of public and private policies to prevent occupational health problems will be enhanced, based on more consistent information.Thus, studies performed after the implementation of the Nexo Técnico Epidemiológico could reveal the gap between what has been considered to be WRD and the actual occupational morbidity among workers covered by the Regime Geral da Previdência Social.

Table 1 .
Proportion of benefi ts for work-related diseases in the sum of temporary disability benefi ts due to general and workrelated health problems per trade.State of Bahia, Northeastern Brazil, 2000.
WRD: work-related diseases PR: proportion ratio * According to the Classifi cação Nacional de Atividade Econômica -CNAE (National Economic Activity Classifi cation); ** Diseases with occupational causal connection established by the Instituto Nacional do Seguro Social -INSS (National Institute of Social Security)

Table 2 .
Distribution of temporary disability benefi ts granted due to diseases, according to ICD-10 chapter and type of benefi t.

Table 3 .
Proportion of benefi ts granted due to carpal tunnel syndrome, tenosynovitis and lumbar intervertebral disc degeneration, regardless of the relationship with work, among all of the temporary disability benefi ts granted on account of diseases and proportion ratio according to fi elds of activity.State of Bahia, Northeastern Brazil, 2000.

Table 4 .
Proportion of temporary disability benefi ts granted due to carpal tunnel syndrome, tenosynovitis, and lumbar intervertebral disc degeneration, according to the type of benefi t and trade.State of Bahia, Northeastern Brazil, 2000.According to the Classifi cação Nacional de Atividade Econômica -CNAE (National Economic Activity Classifi cation).However, they represented one of the main WRDs in a survey conducted by the INSS, using CAT data and not taking into consideration leaves from work. a Noise-induced hearing loss is the most common and severe work-related disease, whose treatment is limited.Workers who are thus diagnosed are usually sent to the INSS for medical evaluation.Thus, the CAT is issued without a leave and, as a result, does not generate a benefi t, which explains the low reporting of this health problem observed in the present study.Related Diseases, included in the Annex II of the Regulamento da Previdência Social (Social Security Regulation).b Even though it is not included in this list, lumbar intervertebral disc degeneration is present in an indirect way, once its symptomatology (dorsalgia, sciatica, lumbago with sciatica) is present.Thus, the health problems above mentioned were not characterized as work-related on account of legal impediment.It is implausible to assume that these fi ndings are restricted to the location of study, namely the state of Bahia, even if they are limited to these health problems exclusively.On the contrary, information from the Ministry of Social Security shows that in 2005, Bahia was the state with highest incidence of workrelated diseases, 2.0 per 1,000 records, c whereas in the country it was 1.2 per 1,000 records.d However, as this may better refl ect the register capacity and the level of implementation of occupational health practices in the country, whereas the risk of health problems would not be as well refl ected, it is possible that the underreporting situation in other states could be even greater.Despite the increase in occurrence of WRDs in Brazil, underreporting of occupational health problems remains high, similarly to what has been observed in terms of typical work-related accidents due to external causes. * Biddle et al 1 compared compensation insurance data on work-related health problems in Michigan, United States, to occupational diseases notifi ed by health professionals, and found that 55% of workers with these diseases did not seek insurance.Morse et al 9 observed that only 7% of work-related musculoskeletal diseases were registered in the compensation insurance records as work-related health problems in Connecticut, United States, between 1995 and 2001.