Document OELo6nbVBKbO74RB82VvyZ26j

FILE NAME: Asbestos in Schools & Other Buildings (AIS) DATE: 1991 DOC#: AIS001 DOCUMENT DESCRIPTION: Asbestos-Related Abnormalities in School Maintenance Personnel - Annals NY Academy of Science P A R T 16. D ISEASE RESU LTING FROM ASBESTOS EXPOSURE IN BUILDINGS Asbestos-Related Abnormalities in School Maintenance Personnel L. CHRISTINE OLIVER," NANCY L. SPRINCE," AND REGINALD GREENE* "The Medical Services (Pulmonary and Critical Care Unit) Massachusetts General Hospital, and Department o f Medicine Harvard Medical School Boston, Massachusetts 02114 bDepartment o f Radiology Massachusetts General Hospital, and Harvard Medical School Boston, Massachusetts 02114 The use of asbestos-containing materials (ACM) in public, residential, and com mercial buildings in the United States has been widespread.1Custodians, skilled craftspeople, and occupants are some of the groups at risk for asbestos exposure as a result of deterioration, damage, attrition, or impact of friable ACM, or reen trainment of released and settled fibers. The nature and extent of exposure-related health effects are a function of both the level and duration of exposure. The degree to which fiber type (amphibole versus serpentine) and dimension contrib ute to risk remains a matter of controversy. In 1983, a survey of ACM in Boston public school buildings was carried out in accordance with the regulatory requirements of the EPA.2-3 In all, 129 schools were inspected and categorized on the basis of the presence, type, location, and condition of ACM: Priority I indicated serious disrepair, requiring prompt correc tive action; Priority II, isolated, damaged ACM; and Priority III, intact, stable ACM, Distribution of schools by category is shown in F igure 1. The purpose of .the present study is to investigate the occurrence of asbestos-related disease in building custodians at risk for exposure to asbestos in these schools and to deter mine the proportion of disease attributable to such exposure. METHODS Cross-sectional surveys of Boston public school custodians were carried out in 1987 (n = 52) and 1988 (n = 69). Subjects were active or retired members of the Boston Public School Custodians Association (BPSCA). Eligibility criteria were at least 20 years of service in 1987 and at least 15 years of service in 1988. Medical Evaluation Questionnaire. Detailed occupational, smoking, and respiratory histories and information about current medication were obtained. Occupational history in- 521 522 ANNALS NEW YORK ACADEMY OF SCIENCES eluded questions about asbestos exposure (1) at jobs held prior to the start of work as a Boston public school custodian, (2). at part-time jobs held during the period of work as a custodian, (3) as a result of household contacts, and (4) as a result of geographic proximity to an asbestos source, such as a shipyard or asbestosmanufacturing facility. Information was obtained about the time period and dura tion of employment at individual schools and about the frequency of performance of specific custodial tasks, such as sweeping and dusting of plaster or insulation dust, patching or removing torn insulation on pipes or boilers, and maintenance of boilers. Proximity to major boiler overhaul operations was recorded in terms of number and duration of occurrences. Latency was computed on the basis of date of hire for those custodians with no known exposure to asbestos outside of their work as a Boston public school custodian, and on the basis of date of first reported exposure for those with outside exposure. Where doubt existed, exposure was presumed and subjects were classified accordingly.. ACM Priority Score FIGURE 1. Distribution of inspected schools by asbestos-containing materials (ACM) Pri ority Score: 0 = no ACM; I = serious disrepair; II = isolated damaged ACM; III = intact, stable ACM. (From the Survey for Friable Asbestos-Containing Materials in the City of Boston Public Schools.2) Smoking and respiratory histories were taken from the Epidemiology Stan dardization Project questionnaire.4 Dyspnea was graded according to criteria set forth by the Medical Research Council of Great Britain.5 Pulmonary function tests. Maximal expiratory flow-volume loops were ob tained using the Eagle One Spirometer according to criteria set forth by the American Thoracic Society (ATS).6 Maximal values for forced expiratory volume in one second (FEV,), forced vital capacity (FVC), and FEV|/FVC% were used. Prediction equations of Crapo and coworkers were used to derive predicted val ues.7 A restrictive defect was defined as an FVC less than 80% of a value pre dicted on the basis of age, height, and gender together with an FEV,/FVC% of 70 or greater; an obstructive defect was defined as an FEV| less than 80% of the predicted value together with an FEV,/FVC% of less than 70. Single-breath carbon monoxide-diffusing capacity (DLCO) was measured us ing the P. K. Morgan Modular Pulmonary Function Testing System (P. K. Mor OLIVER et al.: DISEASE IN SCHOOL MAINTENANCE PERSONNEL 523 gan Co, Kent, Sussex, UK). The average value of two acceptable maneuvers was computed and used in data analysis.8 Prediction equations of Miller were used to derive predicted values.9 Chest radiographs. Posteroanterior (PA), lateral, and right and left anterior oblique (AO) chest radiographs 41 x 43 cm in size Were taken in a mobile X-ray unit at full inspiration, 110 KVp, and a standard distance of 183 cm. Interpretation was carried out by a chest radiologist (R.G.) who was a NIOSH-certified " B" reader, according to the ILO-1980 system of classification.10 Each chest film was re-read without knowledge of the initial reading by another " B" reader (L.C.O.) or an experienced " A" reader (N.L.S.). Differences were resolved by consensus reading. Definite pleural plaques were determined to be present when plaque-like thickening at the lung-pleura interface situated tangentially along the lateral chest wall or along the enface rib margin was 2 mm or more, or when typical plaque like thickening was present along either hemidiaphragm." When no pleural plaques were detected or when minor or suspect pleural abnormalities were ob served, the pleural findings were categorized as normal. The presence of pleural plaque was judged on the basis of four views of the chest. In order to determine the extent to which AO radiographic views increased the ability of chest X-rays to detect pleural plaques above that of PA and lateral views alone, the two " B" readers re-read 50 randomly selected PA/lateral film sets. The " re-read" was carried out one to two years after the initial reading, without knowledge of the initial 4-view reading. Classification based on two radiographic views was then compared to that based on four views. Physical examination. Chest auscultation was performed by a physician (L.C.O. or N.L.S.). Dry end-inspiratory crackles that failed to clear with cough were recorded at each site at which they were heard. Height was measured without shoes; weight was self-reported. Statistical Analysis Student's t test was us.ed to determine whether asbestos exposure, smoking, age, and body habitus differed significantly between those with and those without radiographic changes, pulmonary function test abnormalities, and dyspnea. Multi variate analyses were then used to examine the associations between exposure and pulmonary outcome, controlling for potential confounders. Logistic regres sion was used in the analysis of each dichotomous outcome and linear regression for each outcome described by a continuous measurement. A significance level of 0.05 was adopted. RESULTS Data analysis was carried out on 120 white males. One subject was excluded because of a history of medication with amiodarone, a drug associated with pul monary fibrosis.12This person had pleural plaques and interstitial fibrosis of pro fusion grade 1/0 on chest roentgenogram. Of 197 eligible active school custodians, 102 (51.8%) were examined. Among those with at least 20 years in service since hire (n = 113), participation rate was 70.8%. The number of retired custodians examined was 18. Information on the number eligible was not available. Age and exposure characteristics of the study population are shown in T able 1. Fifty-seven (47.5%) reported no exposure to asbestos outside of their usual 524 ANNALS NEW YORK ACADEMY OF SCIENCES work as a school custodian (NOE). Most " outside exposures" occurred as a result of work at shipbuilding and ship repair or in building construction. Com pared to the group as a whole, those with NOE were younger, with shorter duration of asbestos exposure and latency. Distributions by smoking category and pack-years of cigarette smoking were similar. Chest Radiographs Chest radiographs revealed pleural plaques (PP) in 40 subjects (33.3%). In 39 cases, PP were bilateral; in one case, unilateral. Diffuse pleural thickening oc curred unilaterally in one individual with a history of pleural effusion. This case table I. Age and Exposure Characteristics of Study Participants Total Having No Outside Exposure n Age Mean Range Years employed as a custodian Mean Range Latency (yr) Mean Range Duration of asbestos exposure (yr) Mean Range Pack-years Mean Range Smoking category Nonsmoker Ex-smoker Current smoker 120 56.9 (7.8f . 38-77 26.1 (7.1) 6-46 31.3 (8.3) 17-49 28.6 (8.3) 6-48 30.1 (31.5) 0-184 25 [20.81 63 [52.5] 32 [26.7] 57 55.2 (8.7) 38-77 25.7 (7.8) 6-46 26.8 (7.7) 17-46 25.7 (7.8) 6-46 28.2 (29.0) 0-141 11 [19.3] 33 [57.9] 13 [22.8] " No outside exposure vs. outside exposure. b Standard deviations are in parentheses; percentages are in brackets. p Value" 0.0241 0.5375 0.0001 0.0003 .0.5337 0.514 X2 = 1.331 was not included in the group with PP. Exposure associations are shown in Table 2. Of the group with NOE, 12 (21%) had PP. Duration of asbestos exposure and latency were greater in those with PP compared to those without PP. Multivariate analysis revealed a significant association between the occurrence of PP and duration of asbestos exposure (p = 6.0265) and latency {p = 0.0146) for the group as a whole, controlling for smoking. For those with NOE, PP were associated with duration of work as a school custodian (p = 0.0155) and latency (p = 0.0163). Pack-years of cigarette smoking were similar in the groups with and without PP (p > 0.7) in univariate analysis and .were not associated with PP (p > 0.9) in multivariate analysis. Other factors were examined as potential confounders. Among those with PP, there were no reports of pneumonia, pleurisy, chest injury or surgery, or medica tion with agents known to be associated with pleural fibrosis. Weight (kg)/height OLIVER et al.: DISEASE IN SCHOOL MAINTENANCE PERSONNEL 525 table 2. Chest Radiograph Results in Boston Public School Custodians: Exposure Associations Outside exposure Yes No Years employed as custodian Total group NOE* Duration of asbestos exposure (yr) Total group NOE Latency (yr) Total group NOE Pleural Plaques n = 40, [33]" 28 [44.4] 12 121.0] 27.7 (7.4) 30.9 (7.9) 31.0(7.4) 30.9 (7.9) 34.4 (8.6) '31.9 (8.2) No PleuralPlaques = 80, [67] 35 155.6] 45 [78.9] 25.4 (6.9) 24.3 (7.2) 27.4 (8.5) 24.3 (7.2) 29.7 (9.8) 25.5 (7.0) p Value 0.007 y 2 = 7.368 0.0891 0.0082 0.0221 0.0082 0.0118 0.0088 " Percentages are in brackets, standard deviations are in parentheses. b NOE = no outside exposure to asbestos. (cm) ratios were similar in subjects with PP compared to those without PP--both for the group as a whole (0.49 vs. 0.48, p = 0.4604) and for the group with NOE (0.51 vs. 0.48, p = 0.1447). Three subjects (2.5%) had small irregular opacities on chest radiograph of profusion grade 1/0 by the ILO-1980 system. All reported outside exposure to asbestos. Pulmonary Function Tests Mean values for measured variables of pulmonary function are shown in Ta ble 3. The group with a history of outside asbestos exposure had a trend toward table 3. Pulmonary Function Test Results in Boston Public School Custodians Outside Exposure Total No Yes p Value" FVC Liters % Predicted FEV| Liters % Predicted FEV,/FVC, % DLCO ml/min per mmHg % Predicted Restriction Obstruction 3.8 (0.8)'' 85 (13) 2.9 (0.7) 82 (17) 77 (8) 27.2 (5.5) 87 (17) 22 [18.3] 19 [15.8] 3.9 (0.8) 86.6 (13) 3.7 (0.7) 82.9 (13.9) 3.0 (0.7) 85.4(15.4) 78.5 (7.3) 28.3 (4.8) 90.1 (16.4) 10 [17.51 2.8 (0.7) 79.8 (17.2) 75.9(9.1) 26.1 (6.0) 84.2(18) 12 [19] 8 [141 11 [17.5]- 0.0873 0.1328 0.0354 0.0652 0.0934 0.0278 0.0622 0.832 X2 = 0.045 0.608 X2 = 0.263 " No outside exposure vs. outside exposure. b Standard deviations are in parentheses; percentages are in brackets. 526 ANNALS NEW YORK ACADEMY OF SCIENCES lower FEV, and single-breath DLCO compared to those with NOE. The propor tion with restriction and obstruction'was not significantly different. In logistic regression analysis, restriction was positively associated with duration of asbes tos exposure, both for the total group. (p = 0.0044) and for the group with NOE (p - 0.0420), controlling for cigarette smoking. The single-breath DLCO, % pre dicted, was lower in the group with restriction (78.5 21 vs. 88.9 16, p =' 0.0108). Linear regression analysis revealed a negative association between DLCO, % predicted, and restriction (p < 0.02), taking into account asbestos exposure and smoking. Other Twelve subjects (10.1%) reported symptoms of dyspnea of grade 2 or greater. The percentage was higher in the group with outside exposure (12.9%) compared to those with NOE (7%), but the difference was not significant (p = 0.287, x2 1.135). In multivariate analysis, dyspnea was associated with duration of asbestos exposure (p = 0.0346), taking into account pack-years of cigarette smoking (p < 0.7861) and pleural plaques (p = 0.3651). Crackles at two sites or more on chest auscultation occurred in 6.7% (n = 8) of the total group: 11.3% of the group with . outside exposure, 1.8% of the group with NOE (p = 0.038, x2 = 4.307). There was no significant association between crackles and asbestos exposure (p = 0.6647), smoking (p = 0.5327), or restriction (p = 0.2556) in logistic regression analysis, controlling for each of the other variables. One study participant subsequently developed malignant mesothelioma. In addition to his work as a school custodian for 32 years, he reported employment as a sheet-metal worker in a naval shipyard for 6 months in 1942. His chest radiograph revealed PP. DISCUSSION Pleural plaques revealed by chest radiography are a marker of asbestos expo sure, both occupational and environmental.89 Reported background prevalence figures vary from 0.003% in Australian adults over the age of 40, to 1.6% in nonexposed university employees in Paris, to 1.8% in a group of university labo ratory technicians, maintenance personnel, and grounds workers in the northeast ern United States.13"15 In .the present study, asbestos-related PP as revealed on chest radiographs occurred in 33% of a group of public school custodians who worked in buildings with friable ACM. Thirty percent (n = 12) of those with PP had no known exposure to asbestos outside of their usual work as a school custodian. PP were positively associated with duration of work as a custodian. . Pulmonary restriction occurred in 10 (17.5%) of the group with NOE. The health risks associated with exposure to asbestos from friable ACM in buildings are undetermined. Levels of exposure depend upon (1) the nature of the association with ACM (direct vs. indirect), (2) the nature of the ACM (friable vs. nonfriable), (3) the condition of the AfcM (stable vs. damaged, deteriorating), and (4) accessibility. In general, custodians in buildings with ACM are likely to accrue higher doses of asbestos than are bujlding occupants such as teachers and stu dents. Exposure assessment in a university library with a deteriorating sprayed asbestos ceiling using phase contrast microscopy (PCM) revealed that dusting produced an average level of 4.0 f/cm3 for custodians, compared to 0.3 f/cm3 for OLIVER et al.: DISEASE IN SCHOOL MAINTENANCE PERSONNEL 527 proximate library users.16Cleaning and moving books in the stack area resulted in levels of 15,5 f/cm3; removing a ceiling section resulted in levels of 12.7 f/cm3; and sweeping and dry dusting resulted in levels of 1,6 f/cm3. Our finding of radiographic evidence of asbestos-related pleural plaques in school custodians without other known exposure is consistent with these levels of asbestos exposure and with their mean duration of school employment of 26 years. An excess of PP compared to background was observed by Young and associ ates in maintenance workers in nonasb.estos factory buildings with ACM around steam and hot water pipes and boilers (6.1% [n - 13] compared to 0.003%).13Only one of the 13 with PP had prior asbestos exposure; 12 had worked at least 20 years. Lilis and colleagues observed asbestos-related PP on chest radiography in 37.5% of maintenance workers in a large chemical plant in New Jersey.17 Al though the group studied included skilled tradespeople, 58% (n = 108) reported " bystander" exposure only. Thirty percent of this group had abnormal chest radiographs. There was no significant difference in outcome in those with and without a history of prior asbestos exposure. The occurrence of PP in our study was comparable to that of Lilis and in excess of that reported by Young and Cordier.13-1417 Our finding of prevalence was based on four radiographic views of the chest, while those of Cordier and Lilis were based on a PA view alone. Young obtained both PA and lateral views. table 4. Pleural Plaque Prevalence in Boston Public School Custodians: Contribution of Anterior Oblique Chest Radiographs Pleural Plaques Radiographic Views Yes No Prevalence Posteroanterior, lateral 10 40 0.20 Posteroanterior, lateral anterior obliques 19 31 0.38 The extent to which AO chest radiographs aid in the diagnosis of asbestosinduced pleural disease varies, depending upon the diagnostic criteria applied and the underlying risk of the population examined.1118 In the present study, the addition of AO radiographs resulted, in a finding of PP prevalence which was greater by a factor of 1.9 in 50 film sets (Table 4). Prevalence which was greater by a factor of 1.9 in 50 film sets (Table 4). Prevalence of PP was 38% using four views and 20% using two views in the subset that was reexamined. In 18% of our study population, pulmonary function tests revealed a restric tive defect. Its incidence was similar in those with and without outside exposure to asbestos and was associated with duration of asbestos exposure in both groups. This prevalence of restriction was greater than that observed by Lilis and associ ates, who found restriction (FVC % predicted < 79) in 5.4% of maintenance workers examined.17 Cordier and colleagues found FVC and FEV| to be signifi cantly lower in the group with occupational exposure to asbestos .in buildings, but neither they nor Young reported the occurrence of restriction in their study popu lations. 14J3 The significance of our finding is unclear. Single-breath DLCO was lower in the group with restriction and was negatively associated with restriction in multivariate analysis. Restriction occurred more commonly in those with crackles, but the number with both, abnormalities was small. These findings sug gest the presence of occult interstitial fibrosis, possibly related to exposure to 528 ANNALS NEW YORK ACADEMY OF SCIENCES asbestos and/or some other as yet unidentified substance in the work environment of school custodians. Potential weaknesses of the present study are (1) the voluntary nature of participation, (2) lack of an unexposed comparison group, and (3) possible misclassification with regard to outside exposure status. Voluntary participation can introduce selection bias. To address this issue, we characterized nonparticipants by age and seniority at the time of the study. Participants and nonparticipants were similar in age: 55.4 7.2 years vs. 54.5 9.4 years, p --0.4648. Years since hire (seniority) were 19.5 5.8 for nonparticipants, while years worked as a custodian were 25.4 6.8 for participants (p = 0.0001). This difference was largely attributable to the difference in the number whose eligibility was deter mined on the basis of 15 years' seniority. For participants the number was 22 (18.3%); for nonparticipants, 61 (64.9%). For those with at least 20 years' senior ity, the participation rate was higher and participants resembled nonparticipants with regard to both age (56 and 59 years, respectively; p = 0.0587) and years of custodial work (28 and 26 years, respectively; p = 0.1536). An external referent group was not available for comparison. Because of a high degree of interschool mobility and homogeneity of the study population with regard to performance of custodial tasks, the identification of an internal compari son group was not possible. Fortunately, data were available on prevalence of PP in a group of 717 laboratory, maintenance, and grounds personnel in a large university in Boston, Massachusetts.*15 None had known of any exposure to as bestos at the university. Prevalence of PP was 0.7% if 8 subjects with prior occupational asbestos exposure were .excluded. Using this group for comparison, we conclude that the occurrence of PP in our study population was excessive, taking into account the increment in observed occurrence attributable to the use of AO radiographs. In order to minimize misclassification with regard to outside asbestos expo sure, we contacted those subjects who reported " Don't know" in response to specific questions about asbestos exposure or who reported work at jobs for which the likelihood of asbestos exposure was uncertain. In these cases, tele phone interviews were conducted by one of the authors (L.C.O.) to obtain more detailed information about job responsibilities and workplace exposures. When doubt about asbestos exposure status could not be resolved, subjects were consid ered to have had outside exposure for purposes of data analysis. In summary, the present study reveals evidence of asbestos-related disease in a group of public school custodians. Our results suggest (1) that pleural plaques as revealed by chest radiography are in excess of background, taking into account the incremental effect of anterior oblique radiographs on findings, and (2) that pleural plaques are attributable to asbestos exposure in public schools in a subset of the study population. Our findings indicate that prudent management or re moval of friable ACM in buildings is important to the prevention of asbestosrelated disease. ACKNOWLEDGMENTS We gratefully acknowledge the support of the Boston Public School CustodU ans Association and the Boston School Department. We further acknowledge the technical assistance of Barry Callahan, Paul Pappagianopoulos, and Henry Clausen, the clerical and administrative assistance of Beverly Snell, and the work of our computer programmer, Mary Hyde. OLIVER et al.: DISEASE IN SCHOOL MAINTENANCE PERSONNEL 529 REFERENCES 1. Environmental Protection Agency. 1988. EPA Study of Asbestos-Containing Materials in Public Buildings. A Report to Congress. Washington, DC. 2. 1983. Survey for Friable Asbestos-Containing Materials in the City of Boston Public Schools. Hygeia, Inc. Boston, MA. 3. Federal Register. 1982. 47 RF 23360-23389. Washington, DC. 4. Ferris, B. G. 1978. Epidemiology standardization project. Am. Rev. Resp. Dis. 118: 1-87. 5. F letcher, C. M ., P. C. E lmes, A .-S. Fairbairn & C, H. Wood. 1959. The signifi cance of respiratory symptoms and the diagnosis of chronic bronchitis in a working population. Br. Med. J. 2: 257-266. 6. 1979. ATS Statement: Snowbird workshop on standardization of spirometry. Am. Rev. Resp. Dis. 119: 831-838. 7. Crapo, R. O., A. H. Morris & R. M. Gardner. 1981. Reference spirometric values using techniques and equipment that meet ATS recommendations. Am. Rev. Resp. Dis. 123: 659-664. 8. 1987. ATS Statement: Single-breath carbon monoxide diffusing capacity (transfer fac tor). Am. Rev. Resp. Dis. 136: 1299-1307, 9. Miller, A., J. C. Thornton, R. Warshaw, H. An.derson, A. S. Teirstein & I. J. Selikoff. 1983. Single breath diffusing capacity in a representative sample of the population of Michigan, a large industrial state. Am. Rev. Resp. Dis. 127: 270-277. 10. Classification of radiographs of pneumoconioses 1981. Med. Radiogr. Photogr. 57: 1-17. 11. Greene, R., C. Boggis & S. H. Jantsch. 1984. Asbestos-related pleural thickening: Effect of threshold criteria on interpretation. Radiology 152: 569-573. 12. Adams, G. D., R. Kehoe, M. Lesch & J. Glassroth. 1988. Amiodarone-induced pneumonitis. Assessment of risk factors and possible risk reduction. Chest 93(2): 254-262. 13. Young, I.. S. West, J. Jackson & P. Cantrell. 1981. Prevalence of asbestos-related lung disease among employees in non-asbestos industries. Med. J. Aust. 1:464-467. 14. Cordier, S., P. Lazar, P. Brochard, J. Bignon, J. Amejlle & J. Proteau. 1987. Epidemiologic investigation of respiratory effects related to environmental exposure to asbestos inside insulated buildings. Arch. Environ. Health 42: 303-309. 15. Epler, G. R., T. C. McLoud & E. A. Gaensler. 1982. Prevalence and incidence of benign asbestos pleural effusion in a working population. JAMA 247: 617-622. 16. Sawyer, R. N. & C. M. Spooner. 1978. Sprayed Asbestos-Containing Materials in Buildings. A Guidance Document-. U.S. Environmental Protection Agency. EPA 450/2-78-014. 17. L in s, R., S. Daum, H. Anderson, M. Sinota, G. Andrews & I. J, Selikoff. 1979. Asbestos disease in maintenance workers of the chemical industry. Ann. N.Y. Acad. Sci. 330: 127-135. 18. Baker, E. L. & R. Greene. 1982. Incremental value of oblique chest radiographs in the diagnosis of asbestos-induced pleural disease. Am. J. Ind. Med. 3: 17-22. .