Asbestos Diseases and Pulmonary Symptoms and Signs in Shipyard Workers and Their Families in Los Angeles Kaye H. Kilburn, MD; Raphael Warshaw; John C. Thornton, PhD \s=b\ Families of 338 male and 81 female shipyard workers (SYW), including 280 wives, 144 daughters, and 81 sons, were examined for diseases resulting from asbestos. The workers were initially exposed to asbestos at least 20 years prior to the study date. Radiographic signs of asbestosis (using standard criteria International Labor Office 1980) were found in 64% of 288 male SYW and 21% of 71 female SYW. After excluding those with any occupational exposure to asbestos, asbestosis prevalence was 11% in wives, 8% in sons, and 2% in daughters. Asbestos disease prevalence in workers and in wives in-

creased with the number of years from initial exposure. Male SYW who had smoked had airway obstruction without volume loss. Nonsmokers had normal pulmonary functions. In SYW households prevalences of respiratory diseases, wheezing on physical examination, and symptoms of asthma and chronic bronchitis, exceeded those in the comparison (Michigan) population, even for the younger daughters and sons. These differences, and airway obstruction and distribution defects reported earlier were not explained by cigarette smoking or by asbestos exposure. Instead, they are tentatively ascribed to long-term exposure to ambient air pollution in Los Angeles. (Arch Intern Med 1986;146:2213-2220)

rPhe purposes of this article are first, to relate the preva* lences of signs of asbestosis in bystander workers and their household members to asbestos exposure in ship¬ yards, age, and cigarette smoking. The presence and de¬ gree of asbestos disease between husband and wife and among various other family members is also compared. Second, the prevalences of cardiopulmonary diseases, pul¬ monary symptoms, and abnormal physical findings in these workers and their household contacts are compared with a reference population from Michigan. Third, the degree of pulmonary functional impairment is compared with the profusion of irregular opacities on male shipyard workers' (SYWs) chest radiographs. The pulmonary function results have been reported elsewhere and will be referred to only briefly in this article.1 Finally, the knowledge of and re¬ sponse to perceived risks of asbestos exposure are as¬ sessed. Accepted for publication March 10, 1986. From the University of Southern California School of Medicine, Los Angeles (Dr Kilburn and Mr Warshaw) and the Mount Sinai School of Medicine, City University of New York (Dr Thornton). Reprint requests to University of Southern California School of Medicine, 2025 Zonal Ave, Los Angeles, CA 90033 (Dr Kilburn).

SUBJECTS AND METHODS This report concerns 338 male SYWs, 81 female SYWs, 280 wives, 144 daughters, and 81 sons of workers. These households were recruited in May 1981, from the Long Beach-South Bay area of Los Angeles by appeals to the unions and radio, television, and newspaper campaigns aimed at SYWs. The wife (mother) or daughter made almost all appointments for family groups. If the worker contact of the family member who responded had begun work prior to May 1961 (20 years before the study), this proband worker and household family members were given appointments for interview and examination. Thus, all subjects were volunteers. They completed occupational, respiratory, and health-attitude questionnaires, and underwent a physical examinaton of the chest and extremities; chest radiographs were performed, and vital capacity and flows were measured by spirometry, and diffusing capacity for carbon monoxide (DLCO,b) was measured. Each participant's questionnaire was completed by a trained interviewer, and covered cardiorespiratory disease history, respi¬ ratory symptoms (from the Medical Research Council Chronic Bronchitis questionnaire), occupational history, and knowledge and attitudes about asbestos-related disease. The chest, lungs, and fingers were examined by chest physicians. The prevalences of health history findings, symptoms of respiratory disease, and physical findings in the workers and family members (wives, daughters, and sons) were compared with those of a stratified random sample of the adult population of Michigan studied during 1978 and 1979.2 This latter population sample included 497 men and 486 women whose average age was 42.6 years. Sixty-eight percent of those invited participated in the Michigan Study. This same population was used as a reference for pulmonary functions and chest radiographs. Prevalences of radiographie signs of asbestosis were also compared with those for 1347 members of a Long Beach census tract population studied in 1975 for the prevalence of effects of air pollution.3 Posteroanterior and lateral chest radiographs were exposed at standard six-foot distance on 14 17-inch film, checked by a physician for suitability for interpretation for pneumoconiosis, and repeated until satisfactory. Chest radiographs were interpreted for pneumoconiosis by three experienced physician "B" readers using ILO criteria for type and profusion of irregular opacities, and presence and extent of pleural abnormalities.4 Readings that were positive were averaged for a consensus reading. Radiographs with discrepancies of one profusion grade or more were reviewed until a consensus was reached by the three readers. In this report, asbestosis was defined as the presence of small or fine irregular opacities on the chest radiograph with a profusion of 1/0 or greater by ILO criteria, which is the ILO threshold for asbestosis, and/or the presence of pleural thickening or plaques, including calcified plaques considered to be consistent with pleural asbestosis by at least two readers.6 It should be noted that such

Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/intemed/16871/ on 01/15/2017

»Percentage with asbestosis signs reflected fine irregular opacities with a profusion of 1/0 or greater, or pleural abnormalities consistent with as¬ bestosis.

radiographie findings were rare in the two comparison popula¬ tions.5 Consequently, attributing instances of nonspecific fibrosis to asbestosis would have little or no effect on prevalences. Spirometry, including forced vital capacity (FVC), forced ex¬ piratory volume in 1 s (FEV,), and flow rates were measured on a waterless rolling seal spirometer with the subjects standing and using a nose clip. At least three satisfactory curves were recorded.6 The best one was digitized on a light box using a stylus and a microcomputer (Hewlett-Packard 87). The chest roentgenographic thoracic gas volumes (TGV) were also digitized with this appara¬ tus. Specific comparisons of cigarette smoking by spirometry were made with the Michigan reference population.7 Diffusing capacity for carbon monoxide, single breath, was measured using a fuel cell carbon monoxide analyzer to modify the technique of Ogilvie et al.8 Alveolar gas volume (ALV) and DLCO,b were compared with those of the reference

population.9 All data were entered into a micro¬

computer (Hewlett-Packard 87), transferred to an IBM computer, and analyzed using standard statistical methods (Statistical Analy¬ sis

System) as previously described.7

RESULTS

The results of the questionnaires and physical examina¬ tions of 924 subjects were analyzed. Of this number, 852 had undergone chest radiographs. Because data to predict pulmonary functions for black people do not exist in the reference populations, we analyzed pulmonary function in whites only. This reduced the population analyzed for pulmonary function by 134 persons to 754, of which 267 were white men who had worked in shipyards beginning 20 years prior to the study. The prevalence of signs of asbestosis on chest radiograph was 64% in white male SYWs and 21% in white female SYWs. In household contacts, asbestosis prevalence was 11% in wives, 8% in sons, and 2% in daughters, all of whom were without occupational exposure to asbestos (Table l).5 The comparison women without occupational exposures in the Long Beach census tract were older, but had a preva¬ lence of asbestosis of only 0.6%. No asbestos signs were found in the chest radiographs of 496 women from the stratified random sample of the Michigan population. The comparison male populations had higher prevalences of radiographie signs, 0.5% in Michigan at a mean age of 42 years, and 3.7% in the Long Beach census tract, mean age 59 years. The subgroup of 134 members of black households in¬ cluded 51 male SYWs, 16 (31%) of whom had signs of asbestosis; 23 female SYWs, 2 (9%) of whom had signs of asbestosis; 34 wives, two (6%) of whom had signs of as¬ bestosis; and 19 daughters and seven sons without evidence of asbestos disease. When male SYWs were grouped by decade of birth, the

middle of the birth decade plus 20 years approximated their initial occupational asbestos exposure. The prevalence of asbestosis increased moving back in time from 31% in 1931-1940 to 62% in 1921-1930, 70% in 1911-1920, and 86% in 1900-1910 (including five men born before 1900) (Table 2). Increasing prevalence in the earlier decades of birth was also found for wives, which were 10%, 8%, 12%, and 32% for the same decades. The onset of household contact exposure was assumed to match the worker's initial exposure. Preva¬ lence of signs of asbestos disease in women who had worked in shipyards clustered overwhelmingly (43%) for those born in 1911-1920; 24% of those born in 1900-1910 showed such signs; only 4% born in 1921-1930 had signs, and none born between 1931 and 1940 had evidence of asbestosis. Pleural disease was present in 74% of 170 white male SYWs who had asbestos signs and was equally divided into 37% who had pleural disease only and 37% who also had fine irregular opacities (1/0 greater) in the lung fields (Table 3). Only 26% had fine irregular opacities without any pleural thickening or plaques. Of the 31 wives with signs of as¬ bestosis, 12 (39%) had pleural disease only, 3 (10%) had both fine irregular opacities and pleural signs, and 16 (52%) had parenchymal irregular opacities only. Of female SYWs with asbestos signs, 47% had pleural disease alone, 7% had both fine irregular opacities and pleural disease, and 47% had only fine irregular opacities. Families

Wives.—Seven (22%) of 31 wives with signs of asbestosis did not have husbands in the study. Of the remaining 24 wives with signs of asbestosis, 18 (75%) had husbands with asbestosis signs, while six were without such signs. Eight husbands had pleural disease alone, and one of these had

Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/intemed/16871/ on 01/15/2017

calcified

pleural plaques. Ten husbands had parenchymal disease, and eight of these had pleural disease as well. Two had calcified pleural plaques. The trades of these 18 in¬

cluded: 12 welders (four of whom had also been sandone mechanic and one sheet metal worker, four insulators or pipefitters, and two sandblasters, one of whom was also a shipfitter. None of the sandblasters had fine regular rounded opacities typical of silicosis on chest ra¬ diographs. Although six of 24 husbands were insulators or pipe coverers, only two of these six had signs of asbestosis by chest radiograph. Daughters and Sons.—The six sons and three daughters with asbestosis were from six different families. Two in¬ sulators' families accounted for five of these individuals. The husband (father) of family 1, a 77-year-old insulator, had fine irregular opacities (FIO) with a profusion of 1/1 and pleural calcifications. The wife had FIO with a profusion of 1/0 and diaphragmatic plaques. Two daughters, aged 41 and 39 years, respectively, and one son, aged 43 years, had pleural thickening with calcifications; the 41-year-old daughter also had FIO with a profusion of 1/0. The father of family 2, a shipyard and powerhouse insulator, died of asbestosis two years before the study. The wife was nega¬ tive (ILO 0/1) but three of four sons, aged 46, 43, and 40 years, respectively, had FIO with a profusion of 1/0, and two had pleural calcifications. The 40-year-old son had occupa¬ tional exposure and was not counted as a household contact. A fourth son, aged 37 years, had no signs of asbestosis. The shipyard welder who headed family 3 died in 1973; his wife and two of three sons, aged 41 and 38 years, respec¬ tively, had pleural thickening; however, because the sons had occupational exposure to asbestos, they were not counted as household contacts. The father of family 4, aged 71 years, was first exposed to asbestos in 1927 in the US Navy. His chest radiographs showed a profusion of FIO of 1/1. The eldest daughter, aged 49 years, had FIO with a profusion of 1/0, but his wife and four younger children, aged 47, 44, 33, and 26 years, respectively, had no signs of asbestosis. In family 5, the wife was the shipyard worker, but had no signs of asbestosis; a 31-year-old son had pleural thickening, but a 24-year-old son had no signs. The father of family 6 had FIO with a profusion of 1/0. His wife had pleural plaques; a son, aged 43 years, had FIO with a profusion of 1/0, but a daughter, aged 42 years, showed no signs. The father of family 7 had been a SYW from 1941 to 1945; one son had FIO with a profusion of 1/0 and pleural calcifications; no other family members were examined. Three sons from these seven families worked in ship¬ yards. This is consistent with the "hand-me-down" nature of

blasters),

occupations among skilled craftsmen and may have contrib¬ uted to underestimation of the effect of household exposure in this study. None of these three sons was 20 years or more from initial work exposure to asbestos, but each was 24 to 30 years from initial household exposure. Medical

History, Respiratory Symptoms, and Physical Findings

All of the volunteers for this study had considerably higher prevalences of heart and lung disease than the men (mean age, 42 years) and women (mean age, 43 years) who participated in the Michigan survey (Tables 4 and 5). While differences for angina, heart disease, and emphysema would be expected because of the 15-year difference in mean age, this explanation does not hold true for pneumonia, pleurisy, chronic bronchitis, and asthma. Nor does it ex¬ plain why asthma and chronic bronchitis are more prevalent in sons of SYWs who are 14 years younger than the Michigan men. Volunteers frequently differ from the entire popula¬

tion in awareness and health concern. These volunteers were recruited on the basis of concern about asbestosrelated diseases, as opposed to respiratory disease in general. This concern about asbestosis should not, in itself, be selective for those individuals with respiratory illness. The wife (mother) was the contact person for almost all of the families. Even if she had a lung illness as motivation for entry into the study, this would only affect the wives (mothers) disease rates. With the exception of asthma, these lung diseases are not familial, the rates for other family members should not be affected. There is no connec¬ tion known between asbestosis and asthma or chronic bronchitis. While cigarette smokers, including ex-smokers, had increased prevalences of chronic bronchitis, pleurisy, and pneumonia compared with nonsmokers, asthma was not different among smoking categories. There were similar trends in group comparisons of respi¬ ratory symptoms for men and women (Tables 6 and 7). Thus, dyspnea on having climbed two flights of stairs, sputum production, chronic bronchitis (sputum for three months of two succeeding years), wheezing, and normal breathing between bouts of wheezing, were all more prevalent in the older men and women. Wheezing was more prevalent in these family groups than in the Michigan population. Dysp¬ nea on having climbed two flights was more frequent in sons and daughters than in the ten- to 14-years older Michigan comparison population; but chronic bronchitis diagnosed from the questionnaire criteria was less common. Finally, more chest pain was reported by the family members than by the Michigan men and women.

Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/intemed/16871/ on 01/15/2017

The physical findings of rales, wheezes, and reduced breath sounds had a similarly low prevalence in Michigan men and in sons of SYWs, but were greatly increased in male SYWs (Table 8). All three findings were more preva¬ lent in ex-smokers than in nonsmokers, and most prevalent in current cigarette smokers. All the Los Angeles women, female workers, wives, and daughters, had more wheezing. Rales were most prevalent in female SYWs who were exsmokers. Finger clubbing in both men and women was more prevalent in cigarette smokers. Abnormalities of all types on chest roentgenograms, including asbestosis and emphy¬ sema, were most common in the older groups.

Pulmonary

Function

Pulmonary function results in 45 nonsmoking white male SYWs were expressed as percentage predicted, and grouped by increasing profusion of fine irregular opacities. Functional loss did not increase as the profusion increased (Table 9). In fact, those with irregular opacities (1/0 or greater) had better function than those without radiographic abnormalities (0/0, 0/1) or with pleural changes only (0/0p). These data show that just detectable asbestosis alone has minimal effects on function. Did this degree of asbestosis adversely affect function to a greater degree in cigarette smokers? As expected, the nonsmokers (Table 9)

had better function than those who had ever smoked (Table 10). This was the case for all ILO profusion categories. In cigarette smokers vital capacity, FEVt, FEF2M6, and FEF7M6 was significantly lower (P