Document vVVbBRKb0qOz8GL9ap72De3v6

A Retrospective Cohort Study of Disability Among Chemists Shelia K. Hoar, Sc.D., and Sidney Pell, Ph.D. This retrospective cohort study examines disability among 3,686 men and 75 women who were employed as chemists in 1959 and who were followed through 1977. Overall, the chemists experienced fewer absences of eight or more consecutive days than expected on the basis of the experience of other salaried employees of the chemical company. Among the men, fewer claims than expected were observed for absences due to benign and unspecified neoplasia, heart disease, and peptic ulcer, and to diseases of the urinary system, bones and joints, and skin. Also, the chemists had lower than expected incidence rates of first myocardial infarction (90 observed, 149.8 expected). The deficits may be attributable to exposure misdassification or to confounding by smoking or socioeconomic status. A slight excess was seen for absences due to mental dis orders. In response to reports of excess cancer mortality among chemists,18 a retrospective cohort study of chem ists employed by Du Pont was undertaken.' In addition to mortality and cancer incidence data, the investigation yielded information regarding nonmalignant, nonfatal ill nesses. These diseases may also be related to occupa tional chemical exposures.10 This paper compares the frequency of illnesses result ing in disability of eight days or more among chemists and among other salaried employees of E. I. du Pont de Nemours and Company, Inc. Materials and Methods The populations studied and the methods of investiga tion are identical to those described in an associated report, "A Retrospective Cohort Study of Mortality and Cancer Incidence Among Chemists."11 Disability Frequency. -- Disability was ascertained by review of the Medical Division accident and health insur- From the Department of Epidemiology, Harvard School of Public Health, 677 Huntington Ave. Boston. MA 02115 (Dr. HoarX and Epidemiology Section, Medical Division. E, \. du Pont de Nemours and Co., Inc., Wilmington. DE 19898 (Dt. Pell) Address communications to Dr. Hoar, Environmental Epidemiology Branch, National Cancer Institute. Landow 3C06, Bethesda, MD 20205. Supported by grants from E. I, du Pont de Nemours and Co.. Inc., the Exxon Corporation, and the United States National Cancer Institute (5 P01 06373). ance (AHI) claim file. Under the group insurance plan, claims are submitted for benefits for nonoccupational in juries and for illnesses that result in eight or more con secutive days of disability. To enroll in the plan, employ ees must have at least six months of service and must pay a monthly premium. The participation rate among eligi ble employees is about 97%. Coverage ceases after em ployee termination from Du Pont. AHI claims may not be submitted for injuries or illnesses currently acknowledged to be occupationally related, which are covered by the Workmen's Compensation Insurance Program. The AHI plan has been described in detail elsewhere.15 The AHI claims contain the diagnosis for the illness, the number of days of disability, the treatment given, and some employee demographic data. The diagnosis, cer tified by the attending physician, has been coded accord ing to the International Classification of Diseases, Adapted for Use in the United States.'3 M The Seventh Revision was used from 1964 through 1968, and the Eighth Revision from 1969 through 1977. The claims of the study subjects were identified by matching on Social Security number and name. For all diseases combined and for 46 diagnostic cate gories two outcomes were investigated: number of claims and average number of days of disability per claim. The estimated expected frequency of claims was generated by multiplying the nonchemist age- and calendar yearspecific rates by the appropriate number of chemist per son-years and summing over the strata. Person-years were accumulated from January 1,1964, to termination of Du Pont employment, or, if earlier, to the close of the obser vation period, December 31,1977. The person-years con tinue to accrue after submission of the first claim because the subjects can have repeated absences due to one or several illnesses. The expected average number of days per claim was obtained by dividing the expected number of days of disability by the expected number of claims. Ratios (O/Es) were calculated by dividing the observed numbers by the expected numbers and multiplying by 100. The effects of age and decade hired and of usual com pany department group were evaluated. For each value of the factors, the expected number was generated by the experience of the nonchemists with the same value for 495 DBZ0200901 Table 1. -- Observed and Expected Frequency ot Disability* by Cause Among Male Chemists, Aged 20-64 Years,. 1964-1977, Based on the Male Nonchemist Cohort. Ciusaf ' Toial Inactive and parasitic diseases Malignant neoplasia ' Neoplasia, benign and nature N0S Mental disorders Diseases ot the nervous system and sense organs Diseases of the cardiovascular system, except heart disease Diseases ot the heart Infections ol the respiratory system Other diseases of the respiratory system Peptic ulcer Appendicitis Hernia of the abdominal cavity Other disease of the digestive system Diseases of the urinary system Diseases of the genital organs Arthritis and rheumatism Diseases of bones and joints Accidents, poisonings, and violence Diseases of the skin Residual Claim Frequency U53n Exp H7If 1,989 83 72 18 65 102 2,939.5 85.0 99.6 36.4 57.7 140.7 68 98 72 49 113 72 182 251.1 72 158 274.7 513 223 324.2 61) 63 84.4 7!i 53 98.1 54 35 50.6 6!) 213 243.9 87 172 244.4 71) 68 125.1 54 70 101.1 69 37 49.3 75 171 351.2 49 103 144.9 71 16 52.7 30 85 124.4 Averaqe Days Per Claim Obs Exp 0/E 34.1 35.8 20.0 27.6 73.4 78.0 21.8 25.3 51.7 49.2 38.4 32.3 95 72 94 86 105 119 31.8 36.6 69.0 73.2 17.8 17.0 23.2 23.7 32.7 30.2 25.0 28.2 27.8 31.3 33.4 32.1 24.8 26.9 28.2 28.4 44.4 39.6 34.0 35.8 33.5 34.2 24.3 20.2 87 94 104 98 109 89 89 104 92 99 112 95 98 120 Disability of eight or more consecutive days fSee appendix A for code from International Classification of Diseases, Adapted, Eighth Revision14 io/E a (observed/expected) x 100 N0S = Not otherwise specified the factor. The authors also examined the effects of alco holism." The diagnostic criteria for alcoholism are given in the associated report on mortality and cancer inci dence.' Myocardial Infarction Incidence, -- Since 1956, the Medical Division has maintained a file with data on all employees who were diagnosed with a first acute myocar dial infarction or who died suddenly from coronary heart disease. The Division receives notification of these events by two methods: through disability claims submitted under the accident and health insurance plan described previously and through claims for benefits from a group life insurance plan. All employees with at least one year of service are enrolled in the life insurance plan. The file is restricted to records of active employees whose disease was confirmed by electrocardiograms, lab oratory findings, autopsy records, or clinical histories. Per sons whose disease was limited to angina pectoris, coro nary insufficiency, or chronic myocardial ischemia are ex cluded. The diagnostic criteria were not strictly defined, but have been dependent on the clinical acumen of the physicians at the Du Pont Medical Division. Although the criteria may have changed over time, there is no reason to suspect variation from the chemists to the nonchemists. The surveillance system covers the entire company and was established over 20 years prior to the initiation of this cohort study. The file contains information on blood pressure and body weight obtained during annual physical examina tions preceding the myocardial infarction and found in the employee medical records. A case is classified as hy pertensive if, during the previous ten years, there were at least two successive annual readings of 150 mm Hg or more systolic, or 94 mm Hg or more diastolic. Subjects are also classified by the percentage that the body weight exceeds the ideal weight during the year before the at tack. The ideal weight is determined from the tables of desirable weights from the Metropolitan Life Insurance Company. The file has been described in detail in other studies.'6 It was not feasible to retrieve the study subjects' medical records from the plants to obtain comparable risk factor data on the non-cases. Cases among the study subjects were ascertained by review of the myocardial infarction file, matching on Social Security number and name. Using a computer pro gram developed by Monson," the observed number of myocardial infarction cases diagnosed during the years 1964 to 1977 was compared to the estimated expected number based on the experience of the nonchemists. For this analysis, person-years accrued from January 1,1964, to the earliest of three dates: date of first acute myocar dial infarction, date of termination from Du Pont, or December 31,1977. Standardized incidence ratios (SIRs) and 95% confidence intervals were calculated, assuming a Poisson distribution for the observed number of events.11 The effects of the following factors were investi gated: age hired, decade hired, diagnosed alcoholism, and ever and usual department affiliation. Results Disability Frequency. -- Table 1 presents the observed and expected frequencies of disability, by cause, for the male chemists. While actively employed at Du Pont, the 3,686 male chemists accrued 44,839.3 person-years of ob servation and filed 1,989 claims for benefits for non-occupational illnesses and injuries. Based on the experience of the nonchemist cohort, 2,939.5 claims were expected, yielding a ratio of 68. The observed average number of 496 DBZ0200902 Table 2. -- Observed and Expected Frequency of Disability* Due to Mental Disorders (ICDA 290-315)f Among Male Chemists, Aged 20-64 Years, 1964-1977, Based on the Male Nonchemist Cohort. Total Age hired 16-19 20-24 25-29 30-64 Decade hired 1920-1929 1930-1939 1940-1949 1950-1959 Usual department group Plastic resins and fibers Basic chemicals and finishes Petrochemicals Elastomers Research, developmentand engineering Atomic energy . Claim Frequency Obs Exp O/Ef 65 57.7 113 6 5.2 115 22 19.3 114 24 20.0 120 13 14.6 89 5 .6 17 37 1.3 385 6.1 98 16.0 106 38.7 96 28 25.0 112 15 13.2 114 4 7.1 56 3 6.8 44 12 3.8 316 3 2.6 115 Average Days Per Claim Obs Exp 0/E 51.7 49.2 1G5 69.3 52.1 47.6 38.9 54.8 47.3 44.6 47.2 133 122 116 94 80.4 66.9 120 73.8 55.7 133 46.1 52.5 88 46.7 39.3 117 53.5 54.5 98 ' 61.1 39.0 157 29.0 41.8 69 26.0 34.7 75 53.9 48.6 111 33.3 60.7 55 Disability of eight or more consecutive days (International Classification of Diseases, Adapted, Eighth Revision'* jo/E = (observed/expected) x 100 days per claim vyas 34.1, as compared to 35.8 expected (O/E = 95). Absences due to mental disorders are slightly more fre quent among the chemists than expected (65 claims observed, 57.7 expected, O/E = 113; average number of days per claim was 51.7,49.2 expected, O/E = 105). The O/Es appear excessive upon consideration of the overall ratio of 68 for claims due to all causes. Table 2 presents the frequency of claims due to mental disorders accord ing to work history characteristics. The relative frequency of claims is elevated among chemists hired during the period 1920 to 1929 or before age 30. Chemists usually employed in research, development and engineering had a threefold increase in claims filed for disability due to mental disorders (12 observed, 3.8 expected, O/E = 316). Slight excesses also were seen in the following depart ments; plastic resins and fibers, basic chemicals and finishes, and atomic energy. Alcoholic chemists sub mitted more claims than expected for neuroses (11 ob served, 2.6 expected, O/E = 421) and other mental dis orders (19 observed, 16.5 expected, O/E = 115), com pared to alcoholic: nonchemists. Nonalcoholic chemists experienced 22 absences due to neuroses (27.6 expected, O/E = 80), and 13 absences due to other mental disorders (12.7 expected, O/E = 102). Fewer claims than expected were submitted for ab sences due to benign and unspecified neoplasia, heart disease, and peptic ulcer, and to diseases of the urinary system, bones and joints, and skin. The frequencies of all causes of disability among chemists and nonchemists were compared by department Table 4. -- Observed and Expected Frequency of Disability,* for Selected Causes, Among Male Chemists Usually Employed in Basic Chemicisls and Finishes, Aged 20-64 Years, 1964-1977, Based on the Male Nonchemist Cohort. Tabls 3. - Observed and Expected Frequency of Disability,* for Selected Causes, Among Male Chemists Usually Employed In Plastic Resins and Fibers, Aged 20-64 Years, 1964-1977, Based on the Male Nonchemist Cohort. Causef Total Other intective and parasitic diseases Unspecified viral inteclions Neuroses Other diseases ot the respiratory system, except tonsillectomy Anal fissure and fistula Synovitis and bursitis Residual ' Claim Frequency Obs Exp 0/Ef 736 1,000.6 74 12 9.7 124 12 8.0 150 17 11.8 144 27 25.3 107 9 5.6 161 9 7.0 129 650 933.2 -- Disability of eight or more consecutive days tSee Appendix A for code from ICDA, Eighth Revsion'4 jo/E = (observed/expected) x 100 ' Causef Total Streptococcal sore throat Unspecified viral infections Other infective and parasitic diseases Malignant neoplasia Neuroses Other diseases of the nervous system Acute respiratory infection Influenza Hernia of the abdominal cavity Other diseases of the urinary system Arthritis and rheumatism Residual Claim Frequency Obs Exp 0/Ef 612 723.9 85 5 1.1 471 8 4.7 170 12 9.2 130 29 26.8 108 6 4.1 148 43 30.4 141 17 14.0 121 26 22.7 115 64 58.0 110 11 9.6 114 16 10.2 157 375 533.1 -- Disability of eight or more consecutive days TSee Appendix A for code from ICDA, Eighth Revsion" T0/E = (observed/isxpected) x 100 497 DBZ0200903 Table 5. - Observed and Expected Frequency of Disability,* for Selected Causes, Among Male Chemists Usually Employed In Petrochemicals, Aged 20-64 Years, 1964-1977, Based on the Male Nonchemist Cohort. Causet Tolal Pneumonia Duodenal ulcer Hernia of the abdominal cavity Residual Claim Frequency Obs Exp O/'Et 203 242.3 84 9 4.9 184 6 4.2 142 26 22.8 114 162 210.4 -- "Disability ol eight or more consecutive days tSee Appendix A for code from ICDA, Eighth Revsion14 iO/E = (observed/expected) x 100 group. Excesses are seen for several diagnostic categories and departments, but the ratios are based on small num bers of claims. For five departments, Tables 3 through 7 present illnesses whose O/Es are elevated, limiting men tion to findings based on five or more claims and differ ences of at least one between the observed and expected numbers of claims. No notable excesses were seen in elastomers or in administration. For all causes of disease, the nonalcoholic chemists filed 1,886 claims, while 2,831.6 were expected (O/E =67). Among the identified alcoholic chemists, 103 claims were filed, as compared to 96.6 expected on the basis of the ex perience of the identified alcoholic nonchemists (O/E = 107). Elevated ratios were seen among the alcoholic chemists in several diagnostic categories, but only a few were based on five or more claims: mental disorders (i.e., alcoholism, depression, or neuroses); cardiovascular disease excluding heart disease; influenza; duodenal ulcer; and synovitis and bursitis (Table 8). During the period 1964 to 1977, the female chemists accrued 769.2 person-years of observation and filed 54 claims (Table 9). On the basis of the experience of the female nonchemists, 73.4 claims were expected (O/E = 74). The average number of days per claim was 31.7, Table 6. -- Observed and Expected Frequency of Disability,* tor Selected Causes, Among Male Chemists Usually Employed In Research, Development, and Engineering, Aged 20-64 Years, 1964-1977, Based on the Male Nonchemist Cohort. Claim Frequency Causet___________________________ Obs Exp 0/St Tolal Neuroses Other mental disorders (alcoholism, depression) Other diseases of the nervous system Acute respiratory inlection Other diseases ol the respiratory system, except tonsillectomy Hernia ol Ihe abdominal cavity Other diseases ot the digestive tract Other diseases ol the genital organs Other diseases ol the bones and joints, except synovitis and bursitis Residual 164 192.8 85 5 1.7 302 7 2.1 327 10 8.3 121 5 3.9 127 7 5.5 126 19 17.1 111 13 5,4 239 6 4.4 137 7 4.3 164 85 140.1 -- Disability of eight or more consecutive days tSee Appendix A lor code from ICDA, Eighth Revsion14 pO/E = (observed/expected) x 100 498 Table 7. -- Observed and Expected Frequency of Disability,* for Selected Causes, Among Male Chemists Usually Employed in Atomic Energy, Aged 20-64 Years, 1964-1977, Based on the Male Nonchemist Cohort. Cause) ' Total Acute respiratory infection Pneumonia Ulcer of the duodenum Hernia of the abdominal cavity Gallbladder disease Hyperplasia of the prostate Accidents, poisonings, and violence Residual Claim Frequency Obs Exp 0/E| 133 154.1 86 6 3.5 170 6 4.6 130 7 3.6 195 15 13.6 110 5 2.4 209 5 1.0 483 11 7.0 156 78 35.7 -- "Disability of eight or more consecutive days tSee Appendix A for code from ICDA, Eighth Revsion14 40/E = (observed/expected) x 100 slightly less than the expected average of 32.7 (O/E = 97). Slight excesses were seen in the numbers of claims due to infective and parasitic diseases and to malignant neo plasia. Two claims were submitted for absences due to neuroses, 0.6 were expected. Too few claims were sub mitted by the women to be analyzed in detail. Myocardial Infarction Incidence. -- While actively em ployed at Du Pont and prior to diagnosis of a first myocar dial infarction, the 3,686 male chemists accrued 44,550.1 person-years of observation. Ninety men suffered their first myocardial infarction during the years 1964 to 1977. Based on the experience of the nonchemists, 149.8 myo cardial infarctions were expected, yielding an SIR of 60 (95% confidence interval = 48-74). No excesses or trends are seen in the anailysis of myocardial infarction, work history, and diagnosed alcoholism. To investigate the effects of potential risk factors, data are needed from all subjects, regardless of disease status. The myocardial infection file contains records of blood pressure and body weight for people with myocardial in farctions, but no comparable data are easily accessible for people without the disease. However, among men who developed a first myocardial infarction during the Table 8. -- Observeil and Expected Frequency of Disability,* for Selected Causes, Among Male Chemists Identified as Alcoholics, Aged 20-64 Years, 1964-1977, Based on the Male Nonchemist Cohort. Causet , Total Neuroses Other mental disorders (alcoholism, depression) Other diseases of cardiovascular system, excluding heart disease Influenza Ulcer of duodenum Synovitis and bursitis Residual Claim Frequency Obs Exp 0/Ef 103 96.6 107 11 2.6 421 19 16.5 115 7 3.7 188 5 2.8 178 5 2.5 198 5 0.1 -- 51 68.4 -- "Disability of eight or more consecutive days tSee Appendix A for code from ICDA, Eighth Revsion14 tO/E = (observed/expected) x 100 DBZ0200904 Table 9. -- Observed and Expected Frequency of Disability* by Cause Among Female Chemists, Aged 20-64 Years, 1964-1977, Based on the Female Nonchemist Cohort. . Causef Total Infective and parasitic diseases Malignant neoplasia Neoplasia, benign and nature N0S Neuroses Diseases of the nervous system and sense organs Diseases of the cardiovascular system, except heart disease Infections and other diseases of the respiratory system Diseases of the digestive system Diseases of the breast and genital organs Diseases of bones and joints Accidents, poisonings, and violence Residual Claim Frequency Obs. Exp. 0/Ef 54 73.4 74 5 1.5 333 6 3.1 192 9 9.1 99 2 0.6 -- 1 2.2 45 4 4.4 91 10 11.6 86 3 1.3 225 5 8.0 62 3 4.0 75 3 4.4 69 3 23.2 Average Days Per Claim Obs. Exp. 0/E 31.7 13.2 43.2 36.7 37.0 100.0 32.7 19.2 73.0 39.4 55.4 35.4 97 '69 59 93 -- 283 17.8 29.0 61 22.5 20.4 27.3 40.7 27.6 30.1 12.3 25.8 81.3 36.6 110 67 92 48 222 Disability ol eight or more consecutive days jSee Appendix A tor code from ICDA, Eighth Revsion" io/E = (observed/expected) x 100 N0S = Not otherwise specified years 1964 to 1977, the chemists had a smaller proportion of cases with hypertension or obesity than the nonchem ists. Smoking histories were not available for any study subjects, but the chemists' extremely low lung cancer in cidence and mortality suggest low prevalence of smok ing,' No myocardial infarctions were diagnosed among the female chemists. Discussion This paper describes the frequency of disability among actively employed chemists. Associations between illness and department are presented. Overall, the chemists ex perienced fewer absences of eight or more consecutive days than expected on the basis of the experience of the nonchemist comparison group. Fewer claims than ex pected were submitted for absences due to benign and unspecified neoplasia, heart disease, and peptic ulcer, and to diseases of the urinary system, bones and joints, and skin. Mental disorders were responsible for more dis ability among the chemists than expected. The rates of absences attributed to neuroses, depression, alcoholism, and other diseases of the nervous system were greater among the chemists than among the nonchemists usually employed in plastic resins and fibers, basic chemicals and finishes, research, development and engineering, and atomic energy. Some chemicals to which the Du Pont chemists may have been exposed have been associated with the devel opment of neuroses, other mental disorders, and other nervous system disorders. For example, exposure to organic solvents can result in depression of the central nervous system, fatigue, mental depression, anorexia, irri tability, inattentiveness, and irritation of the respiratory tract," Exposure to carbon disulfide has been associated with psychological and neuropathological effects; how ever, myocardial infarction incidence, also associated with carbon disulfide exposure,20 was not elevated in the cohort, Tetraethyl lead, produced by Du Pont and in cluded in basic chemicals and finishes, is associated with neuroses, impaired psychological performance, and cen tral nervous system damage.2122 The excess of mental disorders is not consistent with Pell's earlier study23 of Du Pont research personnel. The research group had lower incidence rates of illness due to mental disorders and neurological disorders than the other salaried personnel. The other previous studies of chemists were mortality studies; therefore it is not possible to compare results with respect to nonfatal mental illness, but deaths due to suicide can be examined. The mortality study of this cohort revealed more deaths due to suicide than ex pected among male chemists no longer actively em ployed by Du Pont' Li et a!124 reported a higher propor tion of deaths due to suicide among American Chemical Society members dying before age 65. The chemists who died after 64 years of age were as likely to die as a result of suicide as were members of the general population. Chemists who died in Washington state between 1950 and 1971 had approximately twice the proportion of deaths due to suicides expected.2 California chemists who died during the years 1956 to 1961 had a PMR of 248 for suicide (18 observed, 7 expected).3 The Registrar General's report6 on deaths in England and Wales during 1970 to 1972 also showed excess suicides among chemists. The Pell study23 of research personnel showed fewer suicides than expected on the basis of U.S. general population mortality rates. Suicide is combined with accidents as a cause of death in the published reports on the Swedish chemistry graduates and the category had fewer deaths than expected.5 23 Chemists who are identified alcoholics have higher dis ability rates than alcoholic nonchemists for mental dis orders, cardiovascular diseases excluding heart disease, influenza, duodenal ulcer, and synovitis and bursitis. Diagnosed alcoholics appear to lose the health advan tage, relative to the nonchemists, of being chemists. There may be interactions between alcohol and chemical expo sures, as described in the toxicological literature.26 Many 499 DBZ0200905 chemicals induce hepatitis which in turn can alter alcohol metabolism. Conversely, alcoholic liver damage can alter metabolism of chemicals. The diagnoses mentioned earlier were also found to be excessive in a case-control study of absenteeism among alcoholics.27 The combination of alcohol and occupational chemi cal exposures does not appear to adversely effect the mortality experience of the chemists. In this cohort, cir rhosis deaths were not more frequent than expected, compared to the experience of the nonchemists (5 ob served, 5.7 expected).' Deaths due to cirrhosis were less common than expected in American Chemical Society chemists dying under age 65.' More deaths than expected occurred in members over age 64 from cirrhosis of the liver. Milham2 reported no excess in cirrhosis deaths among Washington chemists. None of the other pub lished reports on chemists contained data on cirrhosis. Chemists had a lower incidence rate for first myocar dial infarction than nonchemists. Deficits in O/Es were seen for absences due to acute myocardial infarctions and other ischemic heart disease, but more standardized diagnostic criteria are needed to allow internal compari sons over time and for comparisons with other incidence studies. Deficits in myocardial infarction mortality were ob served in most of the previous chemist studies. Swedish chemistry graduates,5 chemists dying in Washington state,2 professional and technical staff at Dow Chemical Company,2* and research personnel at Du Pont23 experi enced fewer deaths due to cardiovascular disease than expected. However, the research personnel had slightly higher rates of absences due to heart diseases and other cardiovascular disease than expected. Li et.al' reported a higher proportion of deaths due to diseases of the heart than expected. Chemists in the Guralnick report* had ap proximately the same number of deaths due to coronary heart disease as expected. The deficits in disability and myocardial infarction in cidence may be attributable to confounding or to expo sure misclassification. Potential confounding factors are socioeconomic status and smoking habits. Chemists are on the average more educated than nonchemists. They may have altered their dietary habits, which is a possible explanation for the decrease in the U.S. heart disease mor tality.2' 30 The Du Pont chemists may smoke less than the nonchemists,' thereby reducing their risk for myocardial infarctions and many other causes of disability. Exposure misclassification may be occurring. Some of the nonchemists at Du Pont sustain exposure to hazard ous chemicals which may be obscuring risks associated with the occupation of chemist. Summary This paper describes disability among actively em ployed chemists. Associations between cause of illness and department group are presented. Overall, the chem ist experienced fewer absences of eight or more consecu tive days than expected on the basis of comparison with a nonchemist group. Fewer claims than expected were ob served for absences due to benign and unspecified neo plasia, heart disease, and peptic ulcer, and to diseases of the urinary system, bones and joints, and skin. Also, the 500 chemists had lower than expected incidence rates of first myocardial infarction. A slight excess was seen among all chemists for absences due to mental disorders. The alco holic chemists experienced higher disability rates than al coholic nonchemists for absences due to all causes com bined, and to mental disorders, cardiovascular disease ex cluding heart disease, influenza, duodenal ulcer, and synovitis and bursitis. Stricter diagnostic criteria for myocardial infarction, data on potential confounding factors, and more detailed exposure histories are needed for further research on ill ness among chemists. The authors wish to thank Dr. Bruce W. Karrh, Corporate Medical Director of Du Pont for his support of the study, Drs. Brian MacMahon, Richard R. Monson, and David H. Weginan foi epidemiologic advice, Dr. Alan S. Morrison for his review of the manuscript. Dr. Vann C. Brewster for his administrative assistance, and Audrey E. L, Cidney for her programming assistance. William E. Faverweather, Brian H. Wright, and George W. Zervas also provided programming assistance. Ann D. Monson was the nosologist Clerical duties were performed by Mary Donahue, Louise J-towbray, Cerri A. Nardo, Lisa P. Ranken, and Delores Saunders. Kathryn C. Buchanan, Doris Carmo, Gertrude M. Crowley, Martha DiDomenicis, and Mary O Loughlin provided secretarial services. References 1. Li FP, Fraumeni JF Jr, Mantel N, et al: Cancer mortality among chemists. / Natl Cancer Inst 43:1159-1164,1969. 2. Milham S Jr: Occupational Mortality in Washington State. 1950-1971. U.S. DHEW, Public Health Service Publ. No. 76-175. Washington, D.C.: U.S. Govt. Print. Office, 1976. 3. Petersen GR and Milham S Jr: Occupational Mortality in the State of California, 1959-1961. U.S. DHEW, Public Health Service Publ. No. 80-104. Washington, D.C.: U.S. Covt. Print. Office, 1980. 4. Olin CR: Leukemia and Hodgkin's disease among Swedish chem istry graduates. Lancet 2:916,1976. 5. OJin GR: The hazards of a chemical laboratory environment -- A study of the mortality in two cohorts of Swedish chemists. Am Ind Hyg Assoc I 39:557-562,1978. 6. Occupational Mortality: The Registrar General's Decennial Sup plement for England cind Wales, 1970-1972. London: Her Majesty's Stationary Office, Office of Population Censuses and Surveys, 1978. 7. Study unveils British scientists' way of death. New Scientist 80:93, 1978. 8. Guralnick L: Mortality by occupation level and cause of death among men 20 to 64 years of age, United States, 1950. Vital Statistics, Special Reports 53:141, 1963. 9. Hoar SK: A retrospective cohort study of mortality and cancer in cidence among chemists. Thesis submitted to Harvard School of Public Health, Department of Epidemiology, Boston, Mass., 1980. 10. Rosenman KD: Cardiovascular disease and environmental expo sure. Br l Ind Med 36:85-97, 1979. - 11. Hoar SK and Pell S: A retrospective cohort study of mortality and cancer incidence among chemists. I Occup Med 23:485-494,1981. 12. Pell S, O'Berg MT, and Karrh BW: Cancer epidemiological sur veillance in the Du Pont Company. I Occup Med 20:725-740,1978. 13. International Classification of Diseases, Adapted for Indexing Hospital Records by Disease and Operations. U.S. DHEW, Public Health Service Publ. No. 62-719. Washington, D.C.: U.S. Govt. Print. Office, 1962. 14. Eighth Revision International Classification of Diseases, Adapted for Use in the United States. U.S. DHEW, Publ. No. 67-1693. Washington, D.C.: U.S. Govt. Print. Office, 1967. 15. Pell S and D'Alonzo CA: A five-year mortality study of alcohol ics. I Occup Med 15:120-125,1973. 16. Pell S and D'Alonzo CA: Acute myocardial infarction in a large industrial population. JAMA 185:831-838, 1963. , 17. Monson RR: Analysis of relative survival and proportional mor tality. Comput Biomed Res 7:325-332, 1974. 18. Miettinen OS: Comment. I Am Stat Assoc 69:380-382, 1974. 19. Casarett MG: Social poisons, in Toxicology: The basic science of poisons, L. J. Casarett and J. Douil (Eds.). New York: MacMillan Pub lishing Co. Inc., 1975 20. Davidson M and Feinleib M: Carbon disulfide poisoning: A review. Am Heart I 83:100-114, 1972. 21. Haenninen H, Hernberg S, Mantere P, et al: Psychological per- - DBZ0200906 formance of subjects with low exposure to lead, j Occup Med 20 683-689,1978 22 Bellies RP Metals, in Toxicology: The basic science of poisons, l I Casarett and J DoulKEds.). New York: MacMillan Publishing Co., Inc, 1975 23. Pell 5 and Fleming A): The health of research personnel. Research Management 4:235-24-4,1961. 24, Li FP: Suicide among chemists. Arch Environ Hetlth 19:518-520, 1969. 25 Olin CR and Ahlbom A: The cancer mortality among Swedish chemists graduated during three decades. Environ Res 22:154-161, 1960. 26 Plaa CL: Toxicology of the liver, in Toxicology. The basic science of poisons, L, |, Casarett and J. Doull (Eds,). New York: Mac Millan Publishing Co,, Inc., 1975. 27, Pell S and D'Alonzo CA: Sickness absenteeism of alcoholics. I Occup Med 12:198-210.1970. 28 Ott MC, Holder BB, and Langner RR: Determinants of mortality In an Industrial population, I Occup Med 18:171-177,1976. 29. National Heart Lung, and Blood Institute: Report, Working Croup on Heart Disease Epidemiology. U.S. Public Health Service Publ. No 79-1667, Washington, D.C.: U.S. Covt. Print Office, 1979. 30 Walker W); Changing United States life-style and declining vas cular mortality: Cause or coincidence? N Engl I Med 297:163-165, 1977 Appendix A Causes of Disability and Code from the international Classification of Diseases, Adapted, Eighth Revision.14 Cause* Infective and parasitic diseases Streptococcal sore throat Mumps Unspecified viral Infections Other Malignant,neoplasia Neoplasia; benign and nature NOSf Mental Disorders Neuroses Other Diseases ol the nervous system and sense organs Neuralgia and neuritis Other Diseases of the heart Acute myocardial Infarction Other Ischemic heart disease Other ICDA Code 000-136 034 072 079 000-136, except above 140-209 210-239 290-315 300 290-315, except 300 320-389 350-355 320-389, except above 390-429 410 411-414 390-429, except above Causa* ICDA Code Diseases of cardiovascular system, except heart disease 430-458 Cerebrovascular disease 430-438 Hemorrhoids 455 Other 440-458, Infections of the respiratory system 460-491 Acute respiratory infections 460-465 Influenza 470-474 Pneumonia 480-486 Bronchitis 490-491 Other diseases of the respiratory system Tonsillectomy 492-519 500 Other 492-519, Peptic ulcer Ulcer of stomach 531-533 531 Ulcer of duodenum 532 Ulcer, NOSt 533 Appendicitis 540-542 Hernia of abdominal cavity 550-553 Other diseases of the digestive system 520-577 Gastritis and duodenitis 535 Gastroenteritis and colitis 561 Anal fissure and fistula 565 Gall bladder disease 574-576 Other 520-577, Diseases of the urinary system 580-599 Calculus of kidney and ureter 592 Other 580-599, Diseases of male genital organs 600-607 Hyperplasia of prostate 600 Prostatitis Other Diseases of breast and 601 602-607 female genital organs 610-678 Diseases of the skin 680-709 Pilonidal cyst 685 Other 680-709, i Arthritis and rheumatism 710-718 Diseases of bones and joints 720-738 Displacement of intervertebral disc 725 Affectation of sacroiliac joint 726 Synovitis and bursitis 731 Other 720-738, i Accidents, poisonings, and violence E800-E999 Diagnostic categories were grouped by E. I. Du Pont Epidemiology Section tor use in epidemiologic surveillance system fNOS = Not otherwise specified 501 DBZ0200907