Document yryJ7dJNp8DeR2mJQDzaL0qGX

Fa .. n8 a. :*-. ' Retrospective Mortality an Surveillance Studies of Workers in Benzene Areas of -Refine-ries -- 7 shan 9.Twi, Ph.D.;C. P. Wen, M.B.; N a y S. Weiu, M.P.H.; Otto Wong,Sc.D.; William A. McClellan, M.D.; and Roy L. Gibson, M.D. Thr mortality andheolth experfences o f reffnery worfi e n employed in benzene processes w opemtions are d e suibed. A remspectlve cohort mortality study of benzene w i t e n employed from 1952to 1978revealed no exces in ovemll general momlity or in comer mwtality compand e i t h v with the experienceo f rltc U.S. g e n d population or with that of an internal control p u p . Ascertainment of vital status wrts accomplished for 99%of the cohort. Recent industnal hygiene data that included 1,394 pusonal samples indicated that 84% of all benzene exposures were Im than 7 part per million (ppm), with a median exposure of 0.14 ppm for fhc refinery workers, and 0.53ppm for those in the benzenmloted mi& Among dwse w k e n , no deaths from leukemia wen! observed. A medical surveillance program for benzene worken is also described, with special emphasis on the effectiveness o f laboratory screening Evaluation of datu for a 21-year period showed no signiff- cant chunges in the blood indices of rhe workers as a group. Tbe limited value o f estobiishlng screenrng guidelines without the support o f epidemioiogicai studies IS discused. T h e potential effects on workers o f chronic benzene exposure is a subject of increasing attention.I2 Animal and human studies have focused on potential insults by benzene to the hematopoietic system. Numerous case reports and several epidemiological studies have attempted to establish the leukemogenicity of benzene.3d The results, however, have n o t been consistent, and data on human health effects at the exposure level of current operations of petrochemical industries arc limited. Due to the ubiquitous nature of benzene in many industries, the control of its hazards will continue to be a complex issue.g This study examiner the mortality and health experi- ences o f refinery workers employed in benzene processes or operations. Of all the employees at the refinery, the work- From the Gulf Oil Corporation (Mr Weiss and On. Tui. Wen. McUellm, a d Gibson); and Environmtntll Hell* Azsoclates, Berkeley, wif.(or. Wong). Addreu cofmwndence to Gulf Oil Corporaclan, P.O. Bax 2100, Houston. TX 77252 (Or. Wen). - exceeding seven hours; pcMial samples are those with sampexpowre and would be the ones most likely to dernonstnte advene mortality and health patterns. In addition to exposures to benzene, these workers have also had potential exposures to other chemicals found in a typical refinwy senink This report is divided into two parts. The first part describes a retrospectwe mortality study of benzene workers employed from 1952 to 1978 at a Texas refinery. The second p u r i s an evahation of a 21-year medical surveillance system for benzene workers in place since 1959 a t the same refinery. 1. RETROSPECTIVE MORTALITY STUDY OF BENZENE WORKERS -Materials and Methods Definition o f the Benzene Cohort The study includes all male workers ever employed a t a large Texas refinery from Sept. 15, 1952. to Jan. 1, 1978 (regardless of the length of their employment) who worked directly on the benzene, ethylene, aromatic distillate hydrogenation (AOH) or cumene units which are the principal petrochemical units in the refinery. Maintenance workers assigned on a regular basis to the benzenerelated units, utility men, and laborers assigned to these units w e n included in the cohort However, maintenance workers who were assigned to benzene units on an irregular basis were not included. The ethylene unit began operation of Sept 15, 1952; benzene on Sept 14, 1959; ADH on May 30. 1962; and -cumene on Jan.5,1970. Exposure Data Oata of exposure to benzene for the years 1973 through 1982 w e n examined. Data for the years prior to 1973 w e n inadequate for assessing the worke n ' exposure. The personal sampling exposure data for the 10-year period have been compiled for employees from the entire refinery and are shown in Table 1. Benzene was measured by the silica gel or midget im- pinger method at an earlier time period and by charcoal tubes later. Data were divided into two categoria according to the sampling time. Ful/ samples are those with sampling time greater than or equal to four hours, with the maiority Journal of Occupational MedicineNol. 25, No. 9Beptsmber 1983 685 I <ai0 0.1-0.9s 1.0499 5.0.9.99 ia024.99 325.0 1 fotd 378 (5P 270 (371 "66 (9) 10 (1) 110) 2 (0 222 (33) 301 (488) 97 (14) - 1943, 12(2) 13 (3 e n in this operation havepmumably reuived the greatest ling time less than four hours or those recorded as partid samples whose data on sampling time were missing. No attempt has been made to adjust the exposure numbers to e. igThhtheosuarmtimplien-gwestigrahtteegdyauvesreadgebsy( WAS). industrial hygienists for determining personal exposure h u u s u d l ~been based on representative sampling of various job assignments or worst case sampling on spills or turnaround. Job classifica- tions selected for sampling M uwdly those in which the highest potential for exporun exisrs during routine opefa- tions. Replications o f these job classifications are then -sampled to characterize the exposum Data Collccfion Copies of all personnel records sin= the refinery operation was started in 1901 were made, and pertinent demographic and work history informatmn were screened according to the cohort definition previously described. Of the approximately 20,000employees whose records were collected and screened, 454 have been identlfied as benzene workers. Data coltected on each cohort member included name, date of birth, race, KX, social Kcunty num- ber, date of hire, employment status, date of last =pan- tion, i f any; vital status: date of death for known decedents, and cause of death; and work history. Each work history consisted o f a chronological listing o f all jobs ever held by the individual and included the dam of job assignment, department, unit and job utle. Standard epidemiologrc procedures for ascertaining the vital status of the cohort were followed. They included matching the cohort employees with rosters o f cumnt employees and annuitants, matching with death certificate fila of company records, obtaining information from tha Social Security Administnuon, and the Texas and Louisi- ana Departments of Motor Vehicles and a plant follow-up. Death certificates were collected from the vital statistics departments of individual stater for all known decedena The underlying caux of death was coded by a trained nosologist according to the rules of the International Classification of Diseases (ICD) revirion in effect a t the time of death and then recoded to the ICD adapted (ICDA) eighth revision.10 For analytical purposes, numben of penon-yean at risk wem calculated. The denominator data wen determined .. 686 from S e p t 15, 1952. or from the date o f first benzene exposure, whichever was later, to the end of the study period (jut. 1, 1978) or to the date o f death, which ever was eariier. Individuds with unknown v i t a l status were assumed lost to followup on the date of termination or on the last data of contact, whichever was later. The number of years contributed by each worker w u classified by a g , race, and cdcndju year. The U.S. nation4 age-xx-raceyear. causespecific mortality rates for five-year time intends during the study period were! applied to these numben o f person-yean to obtain the number of expected deaths fnwn a specific cause. Standardized mortality ratios (SMRs)wem computed and their significance testing is based on the assumption of a Poisson distribufim for the observed numben o f deaths, utilizing a two-sided test o f significance. Part of the analysis was performed using a computer program developed by MonsoA.ll An internal comparison group consisting o f a 10% sample of the employees frem the same ptmt who worked during the same peiiods in operations not related to ben- zene was selected. The selection guide was social security- number and the analysis was based on relative risks determined by usingthe Mantel-Haenszelprocedure.12 Results Table 2 shows the chiuacteristia of the benzene cob- Of the total cohort of 454 employm, 369 (81%) w e n white and contributed a combined total o f 5,900 penon- yean. The average age of study subjects at enVy into the cohort was approximately 34 yean and the average length of followup 13 yean. The average number o f yean of benzene-related work (exposure) was 8.0 for the white employes and 4.5 for the nonwhite employees. The cohort is relatively young with only 7.5% (N=34) of the employees found to be deceased. All death certificates were identified and only three employees were lost to follow-up. Thus, the followup was successful for 99.3% o f the cohort Distributions of the benzene cohort by year of first exposure and duration of exposure are shown in Tables 3 and 4, respectively. One-half o f the cohort began work in benzene area before 1965 and almost 8096 had exposum lasting one year or longer. -Tabla 2 Benrrm Cohort StrtisciCr S ~ p 1t5, 1952 to Jaa. 1,1971 I Na studied (K) Penon-yean of observation (%I Aversgr age at entry A v e q follow-up, yr Average exporun, yr Ma a l i n No. d a d 011thurtificatm unmilrblr No. lost to followup ('K) I 369 (81) 5'000.7 (851 33.0 13.6 8.0 340 26 0 3 (aa 85 454 (19) 894.3 5.90) (151 37.3 33.8 10.5 13.0 4.5 7.4 77 417 8 34 00 ..0. 3 (an I Rufinery Workers and 9enzeneflsai et d -. . T* 3 O t a i k b . c l d C U d Fimt C r p . m Y r r of Ficrr E q p . r m 1952-1954 19581959 1960.1 964 196B1969 1970.1974 1975.1977 Tod - M a (%I 61 (13.4 im(tz5) I65 414.3) -73(16.1) (5~21 lo4 (2291 44 (10.8) ~ 451 (1OaO) a 7 Lmythr fExposuo,yr <1 14 5-9 1&14 15.19 >20 Tod we. (%I - 95(20.9t 145 (31.9) 77 (17.0) 52 ( 1 1 4 47 (10.9 38 (8.4) 4!X(lOQo) The total number of sampler o f personal exposure to benzene obtained ms 1,718. Averages were taken when there were two or mom samples from the same individuals on the same dav, resulting in 1,394 samples for analysis, as shown in Table 1. For 8996 of the full samples and 79% of the partial samples exposum were less than 1 part per million (ppm) benzene. For 10% of the full samples and 17% of the partial samples exposures were between 1 and 10 ppm. In addition, for 4% of the partial samples expo- sures exceeded 10 ppm. Table 5 shows benzene exposures by selected units. As expected, the median benzene exposure was highest among those assigned to the cumene and benzene and cyclohexane units. This was followed by amine, waste water treatment, pump house, laboratory, ethylene, A D H unit, and docks, in descending order based on full samples. For partial s ~ D I ~ s . the order was cumene, benzene and cyclohexane, ethyi& d o c 4 ADH unit, laboratory, amine, wuta water treat- ment, and pump house. Since the study design limited to those wortin# in ths banzcne and cyclohexane, cum-, ADH, and ethflene units, duse units we= grouped in Table 6 as " b a n d a t e d units" while the remaining unia listed in Tabk 5 were classified u "other benzene areas.'' All of the remaining samples taka in the refinery were clarified u "JI o t h m " The exposum from the benzenerelated units w r e the highest with a median of 0.33 ppm, follomd by those fmm ocher benzene mu with 0.24 ppm. Exposum from ail otherr wrte ctedy the lowest with a median of 0.07 ppm. T J l e 7 presents the-SMRs for selected causes of death for all males (white andnonwhite combined). The cohortas a whde s h o d a favorable overall mortality experience. . There wen a total of 34 deaths with 58.69 deaths ex- p.cttd. The expected numbers of deaths were calculated far whim and nonwhite mdesindependendy based on the US. white and nonwhite male mortality rates and then were added together. The SMR for all causes (0.58) is SiIgrificantIy decreased w i t h p < .01. The mortality for arteriosclerotic heart disease showed a similar significant deficit (SMRq.46). The number of cancer deaths (10) was slightly smaller than expected (11.53; SMRzO.87). There were no c a m of leukemia or of cancer of the lymphatic system. Exceuo were seen for canccn of the stomach, laqp intestine and b d n , each with two deaths. Cancer of the buccal cavity and pharynx and cancer of the larynx were also increased, each with one death. The numbers were too small to permit any conclusion and chance variation c w l d explain the excesses. However, this points out the need for continuation of the surnillance of benzene workers to verify the significance, or the lack of significance, o f thw observation% Table 8 shows SMRs for those who worked one year or mom in the benzene areas. The total number of the cohort decreased from 454 to 359. The mul0 an similar to those shown in Table 7. The total number of cancer deaths de- creased from 10 to four and the SMR for cancer changed from 0.87 to 0.49. -Table 5 Brnmne EXPOSUN Conamntioa, (ppm)by Selected Units* Full Benrrm and cyclohrxam 85 0.79 Cumine 13 1.18 ADH 18 0.06 Ethylene 32 0.09 Pump h o w No. 11 57 0.18 Docks 32 0.05 Amine 9 a75 laboratory W l l t m t a r mrmrnt 38 0.14 55 aw 1.34 t 1.39 6.10t 10.89 0.33i 1.01 0.11 t 0.07 0.55t 1.12 0.10 0.14 1 . 2 5 ~1.24 a m 3.92 i . n t 220 0 . 8 0 2 2.96 1.37: 6.58 0.04i 9.94 0.07 2 2.73 0.12t15.52 0.06 235 0.642 4.12 0.06: 14.10 a37t 5.04- 57 38 21 80 26 63 5 sa 4 1.25 1.60 0.59 0.85 0.0) a75 a41 aM ai5 5.56 t 8 7 4 5.96 t 14.42 0.74 t 0.70 0.27 t 0.38 7.45223.79 3.6s 1 1 6 0.49: 0.29 1 ~ 0 75-0~3 ailt 0.34 SO indiummndrrddoviation; GM, gaomrbic mean; GSO, gtomotric mnQrddnutim Journal of Occupationat Medicinr/Vol. 25, No. SBeptembrr 1983 .- . .._ 1.44t 6.48 1.85 t 3.29 0.08 i 16.75 0.11 t 3.75 0.12213.26 0.55 9.96 0.40: 2 1 1 0.49: 3.93 o a t zaa 687 _. Full p+thl <.I ppm 30 (2431 . l a 9 9 ppm 76 (51.4 1.0-4.99 ppm - 38625.7) 5.09.99 ppm 1 ~ ~ 2 4 . p9p9ma 2.(.1..4 - >25.opprr, 2 (1.4) 51 (28.a 12 (38.7) 51 (26.0) 7 (3.6) a (4.1) l(3.6) 264 (6aO) 113 (23.1) 9 (261 2..t..a..g 139 (44.1) 153 (48.6) 21 (6.7) 110.31 i.(.a.31 Tomi 148 1I 191 1% Medim Mur, so GM GSO a35 f.38 3.66 a35 -6.43 - 0.61 2.96 8.29 am - 9.21~- ai7 a76 2.22 ai4 8-52 a48 3.51 1-213 a40 1.78 I SO indicrtm standud drvution; GY, geometric m m ; GSD. gmmatric standad drvution 388 0.01 a19 - 0.57 0.06 - 4.19 - 315 - -. 0.1 1 - 0.35 1.08 . 0.10 4.79 Discussion Analysis of overall mortality for the benzene cohort revealed no significant excesses. In general, mortality from all causes and from diseases of the circulatory system was significantly lower than expected based on the experience o f a comparable group of U.S. mdes. Thus, benzene workers in this study demonstrated a hellthy worker effect13 similar to that seen among other employed groups As part of the original design, an internal comparison group was selected (using social security number as a XI=tion guide) for further validation of mul& This compari. son group included 10% of the worken who were employed in the same plant during the same time period in opetations not related to benzene. There were 823 people in the CMparison group, approximately twice as many as in the study cohort. Table 9 pments the relative risks for the study co. Caurr of Deatb (ICDA 8th Coda No.) I All caumt All canccn (140209) Cancer of buccal cavity and pharynx (140-149) Cancrr of dipatwe system (150.159) Cancer of stomach (151) Cancer of l a m intestina(153) Cancu of mprratory system (160.1633 Cancer of larynx 4 161) Cancer of lung (162-163) Canen and tumor of brain (191-192.225,238) All lymphopoietic u n w (200.209) Leukemiaand alwkemu (204207) Oiraaspr of circulatory system (390-458) Amrioabmtic heart disease (410413) Cembrovacular disaeitl(430.438) Nonmalignant respintory disease (460.5191 Pneumonia (480.4861 Diseases of digestive system (52&577) Accidents, poisoningsand v i o l e m (E800-E999) Accidrna (800.949) 0- Dntbr 34 10 1 4 2 2 3 1 2 2 0 -0 15 9 2 1 1 1 5 4 EX- Deaths ~ 58.69 11.53 a43 3.16 0.62 0.89 4.12 0.19 3.88 0.47 1.12 0.42 27.89 19.49 X70 3.14 1.26 3.37 7.52 4.72 SMR 0.58, 0.87 2.33 1.27 3.23 2.25 a73 5.26 0.52 4.26 0.00 0.00 0.54t 0.46t 0.54 0.32 0.79 0.30 0.66 0.85 95% Confideam Limits Low ~ ~~ 0.40 0.42 0.03 0.34 0.36 0.25 0.15 a07 0.06 aa 0.00 0.00 0.30 0.21 0.08 0.00 0.01 0.M 0.21 0.23 UPW 0.8 1 1.60 1294 3.24 11.65 8.1 1 213 29.28 1.86 15.36 3.28 8.73 am a88 1.95 1.77 4.42 1.65 1.55 217 688 Refinery Worksn ami Bentan~/T~ait at 2.2 a 6 8 as* a% u 2 4 8.18 I a19 ai3 1.25 1 0.30 3.33 am 1854 2 224 0 3 9 0.10 322 ~ 1 U43 232 1103 1294 1 a65 1.n 0.02 asti 1 295 1 P,.2.79 0.34 0.00 036 - - am 1-89 1.99 1 0.31 3.23 0.04 1735 0 4.78 0.00 = 0.06 4.70 4 029 aeo 0.00 1265 12 19.87 a60 -0.31 ~ 1.08 9 14.13 0.64 a29 1 3 2 255 0 3 8 0.09 283 0 224 0.00 a00 1.64 1 2.23 0.44 0.01 243 an4 4.63 486 2.21 4 290 1.38 0.37 3.53 hort and the comparison group for di causes and for cancer of ail sites. The relative risks are based on the adapted Mantel-Haenszel procedure for a cohort study,12 and are adiustcd for age and race. The relative risk for all causes was decreased (0.90) while the increase in relative risk for all cancer (1.37) was not stqnificant a t the -05level. This gives further support to the finding that the study cohort did not show any increased risk, compamd with either the US. general population or the general refinery employes. The levels of benzene exposun monitored between 1973 and 1982 from workers in this refinery were quite variable, although most of them were low and the medianexposure was 0.14 pprn, for the refinery workers, and 0.53 ppm for those in the benzene-Elated units. More than 84% o f the samples showed a n exposure of less than 1 ppm benzene. Of those samples showing exposum greater than 1 ppm benzene, two thirds, 144/220, were partial samples. Only three full samples from a total of 727 samples exceeded the 10-ppm limit. It should be pointed out again t h a t full samples hada sampling time of fourhours up to eight houn. Of particular importance i s that no leukemia deaths were obserwd for those with greatest exposure to benzene in this refinery. Even i f we were to assume that one leukemia death was observtd, the SMR for all lymphopoietic cancer would be qual to 0.89 with a 95% confidence interval of 0.01 to 4.497. Since no leukemia deaths were observed, an SMR greater than 3.30 is highly unlikely despite the rela- tively small sample size of this snrdy. This result is not compatible ,with the report by the National Institute for Occupational Safety and Health (NIOSH) o f an SMR for leukemia of 5.60 among workeo manufacturing rubber hydrochloride.' The two studies w e n similar in aspects such as methodology and time periods of observation. In the NIOSH study the sample sizes were 748 for the toul cohort and 311 for those ex- posed one year or more, compared with 454 for the t o u l cohort w d 359 for those exposed one year or in this study. The benzene exposure was interpretedin the NlOSH s t u d y as falling mostly "within the limits considered (legally) permissible at the time of exposure" and likewise in this study. 8enzcne exposure of the rubber worken appears to have been much higher with median exposures in eight different units between 3.4 and 9.4 ppm from 1973 through 1976, a period for which comparable exporurr . - ~~ ~ Tablr 9 R d a t i n R i s k for Death From A l l Causa and From C a w of All Sit- for the Smdy Cobort and tha Comparison Groap Caws of Death Study (N4W Cornparboo ani Rdrth Rat X2* P vdw All causes CUKlr of dl rim Llukmir md lymphopoirtk Umr 34 10 0 t 60 30 S * Mant&Hmmzd x2 with 1df Journal of Occupational MedicineNol. 25, No. SBeptember 1983 0.30 0.34 1.31 a s ... ... .. . 028 0..2.3. 689- e. dam wore a v l i h b b from refinery workers exposed to ben- zme. It should be noted that the majority (5896) o f the NIOSH study subjects had been exposed for less thur one y w . Exposure data in both--studies were extremely limited for the earlier ye- Through analyst, of these exposum data, fira additimal uniu or areas in the refinery have ban identified as yielding the next highest I m l o f expawre. They am the pump house, waste water treatment unit, docks, laboratory and amine u n i t Studies in the fuhrrs would be more completc if the benzene cdXort also included these areas or units, even though benzene exposum in these areas were very much lower. In summary, examination of the mortality exprieacs o f benzene workers (1952-1978) showed a fav- ovenll mortality (SMR = Q.58). The mortality risk from all causes from all cancer or from lymphopoietic cancer is not in- creased. II. MEDICAL SURVEILLANCE PROGRAM OF BENZENE WORKERS Concomitant with the e f f o r t to minimize warttm' benzene exposunl their health is closely monitored. The importance of a medical surveillance progtam i s generally recognized. However, screening procedures v u y and results an difficult to interpret Of particular controversy i s the utility of blood scmning (hematology and chemistry) for early detection of blood dyscrasia. In addition to leukemia, blood disorders such as pancytopenia 01 aplastic anemia have been associated with benzene exposure. Greenburg e t ail4 in 1939 and Aksoy e t a l l s in 1971 found significant abnormalities in hematological results o f workers with high levels of exposure to benzene (recorded exporum up to 210 ppm). Townsend e t alls studied the health examination findings of 282 men occupationally exposed to benzene and found a slight decrease in mean total bilirubin and red blood ceil (RBC) values as compared with a control group, but these changes were riot regarded as clinically significant. Fishbeck e t a l l 7 found no persistent hematological effects except for a slight rrduction in hemoglobin and an increase in the mean corpuscular volume (MCV)of 10 employees with chronic benzene exposure for several years exceeding the 25 ppm eight-hour TWA. The remainder o f this report presents data from a group of refinery worken who are employed in benzene areas and who have been under medical surveillance since 1959. Blood screening has been conducted on there workers for a period of 21 years, from 1959 through 1979. Subjects and Methods of Study As part of the medical surveillance program o f morbidity and mortality at the refinery, beginning in 1959, blood tests were performed on each benzene worker up to four times a year. These tesll included hemoglobin, hematocrit, white blood cell count with differential, platelet count, bleeding and dotting times Frequency of examination was later reduced to once a year in the early 1960s. W i t h the advent of the autoanalyzer and the Coulter counter, a 12-panel automated blood chemistry (SMA 12) or SMAC and a complete blood count w e n added in the early 1 9 7 0 ~ E x h year benzme workus wn defined by the super- visor for each o f the unils and induded workers assigned for a certain period of time to one of the following units: benzene, ADH, ethylene, pdymCriUeim, cumene, pump houses, loading shop and d o c k ad barges o f the bulk oil department The benzene w a k e n included worken from units other than those manciocrtd in part I; time p e r i d overlapped (from 1952 thrar& 1978 for p u t 1 and from 1959 thm@1979 fw p u t 2). Outcome of Workers in the Medical Surveillance Program - - -- - Results of the hematology and the blood chemistry p r a f i l n m pnrented in Tables 10 and 11, respectively, in t e r n s o f the mean, ths standard deviation, the value, at 5th and 95th percentiles, and the number of examinations recorded and analyzed, Appmimately 1,460 tests of hematology and 900 tests of blood chemistry were con- ducted. As a group,_ths workers showed valum within normal limits; no tendency of pancytopenia(aplastic anemia), bleeding or clotting deficiency, judlptd from the mean and the 5th and 95th percentile values in the hematology data profile (Table lo), was sari. Results of the blood chemistry profile were also gcnedly within nwmd limits, including bilirubin, which should haw increased if hemolytic anemia occurred or liver function was impaired (Table 11). A small portion o f the value) falling within the lower or upper 5% might have been considered clinically abnormal, but, when tesB were repeated, the results usually returned to normal and were thenfort judged to be clinical- ly normal. Most of these dixregancks may have molted from individual and laboratory variations. Only in one of 303 workers did repeated blood tests lead to an early diagnosis of multiple myeloma Among these 303 employees who had received medical surveillance a total o f 11 deaths had occurred as of the end of 1979, whereas 21.46 deaths were expected (SMR = 0.51). Causes of these 1 1 deaths included: coronKy heart disease (seven observed; 9.68 expected); cirrhosis o f the liver (one observed; 0.90 expected); "natural causes" (one observed); and cancer (two observed; 4.42 expected). The cancer deaths were from malignant melanoma and multiple myeloma. As of 1979, none of the reported deaths had been ascribed to leukemia nor h d any living employee developed leukemia Discussion The medic4 surveillance system i s important for multi- ple reasons. h o n g them are creation of baseline data for future comparisons; provision of trend analysis of groups or subgroups; early detection of adverre health effects; provision o f medical benefits to the employee; and increased awareness emphasizing health maintenance and promotion. From the clinicd standpoint, e u l y detection of Ever- sible health effectl i s probably the mort important justification for a medical surveillance program. However, the experience of the program described in this report suggests the need for establishing different interpretation criteria of scmning for adverse effects of benzeneexposures. As in every b i d o g i d phenomtnon, the weening results varied within expected rmgn from individualto individual - Refinery Wodcenand BsmendTsiet al WBC, lOW blood Polymaphonudrr Lymph#ym Homoglobin, dml blood Hmatanf 3 RBC, l@/dof blood R i a l 1~p/rlblood Blndinp timr,rein cfottingt i m a 6,952 4.177 2.406 14.8 44.4 5.0 293.0 1.64 coo 1,891 1,458 782 1.1 3.7 Q4 97.0 U.80 1.oa - 42.112 1,350 13.0 39.0 - 4.4 t60.0 0.62 3.38 10.300 6.588 ` 3.860 ` 16.7 5a4 _- 5.6 459.0 3.0 6.58- 1,101 1,396 T,3M 1.w 1,399 602 1,085 1401 - 1,002 i Tab& 11 -Somnuy of B l d CbWProfW Glucotr, m d d BUN. mgldl Uric acid, mddl Calcium. mudl Photphorw rnddl Cholrmrol. mddl Triplycrridr. mg/dl SGOT. units SGPT,units Bilirubintotal, mddl Bilirubind i m mddl Bilirubinindirect. rng/dl TotJ Protein. ddl Globulin. g/dl Albumin. p/dl 94.1 15.1 6.3 9.9 3.3 207 134 30.6 31.1 0.65 0.15 0.56 7.2 2.7 4.4 17.4 4.2 1.3 0.5 a5 41 a4 15.2 23.1 027 0.13 0.32 0.5 0.5 0.4 73 10 4.5 9.1 24 144 48 14 9 0.2 ai 0.1 6.4 2.0 3.8 118 21 8.4 10.5 4.1 275 287 54 56 1.1 0.2 1.o 7.9 3.2 4.9 903 902 902 895 895 897 266 902 272 902 245 266 904 270 903 Data haw b m collKtDdb r t w m 1970 and 1979: the numbrr of examinations on individuilsvrrks and within individuds from year to year. These variations have been intensely confounded by different laboratory standard practices and advancements throughout the years. I n the past, such variations have been handled on an indi- vidual basis by the judgment of clinicians. With the advent of computer storage which makes statistical analysis of large sets of biological data possible, the lack o f a "healthy population" standard has emefgcd as an obstacle to an objective analysis. A study by Frederik'* of approximately 20,000 employees indicated that the usual hematologic normal ranges derived largely from hospital experiences are not applicable to an employed population. Thus, even though in several instances the nsults may be judged to be marginal or even "abnormal" by hospital standards, they could well be within the "normal" variations. For regulators, this would certainly p m t difficulties when vying to establish screeningguidelines and referral criteria. It should be realized that the predictive valueLs of each test or of a battery of. terts is far from established for muginally abnormal test results.20 Only when values are overtly abnormal can one conclude that health status may be either compromised or in true jeopardy. In this study, as the result of hematological screening one patient was sus- pected of and latcr proved to be suffering from multiple myeloma. Other less serious conditions have also been diagnosed. However, not all early interventions proved to be early enough, including the case of the patient with multiple myeloma The occupational medical surveillance program should undergo periodic evaluation to define priorities among competing goals and to improve effectiveness and effici- ency. This could include activities to determine the predic- tive value o f each screening procedun. Findings should be translated into beneficial actions for the employees, whether such actions are health education for life-style changn, proccu engineering changes to reduce ex~osued, or clinical referral and intervention to prevent illneu and disability. Screening per sa i s just one of the many pnven- Journal of Occupationat MedicineNol. 25, No. S/Saptembw 1983 691 ti- m e ~ mthat constitub the surveillance program, but guidelines as to its fmqumcy and interpretation should be based on cumulativs (long-tern) epidemiological results from diffemt screening pfZtiCn Acknowledgment l d n g Takohaw, M.0, and h b zuociatm conducted th.Infell phasa of pu( Iof the smdy. c References -- 1. McMtchrl AG. Spirtas R, Kup& LL, Gal: Solvan! ex- sun and leukemia amon& rubber workers: An rpidemblogk%tudy. J Occup M i d 17:234-239,1975. 2 Bmwning E: Toxiclty and Metabolism of tnduuld Sdvma New York: Elswiur~PullhlngCo,1965. 3. Vigianl EC: Leukemia associated w i t h banzsk exposure. AnuNY AcadSci 271:143-151,1976. 4. A ksoy M: Leukemia in shoeworkers exposedchronkAly to benzene, Blood 44:837-841, 1974. 5. Infante PF, Rinrky RA. Wagoner J K. e t d: Leukemu amon# workenexposed fo knzena. Lmcet 2: 76-78, 1977. 6. Thorp. I J : Epidemiologic survey of leukemia in pemonr potentially exposed to benzene. J Ocwp Mcd 16:375-382. 1974. 7. O n MG, Townsend IC,F h h k c k WA, e t al: Mortality among Indlvidudr occupatlonllly exxpowd to benzene. A& Environ H d r h U:3-10, 1978. 8. Rinrky RA, Y w n g RI. Smith AB. Leukemia in bemeno workers. A m J Ind Mid 2:217-245, 1981. 9. Whim MC, Infanta PF. Walker 6. Jr:OccupstionJ exporum to bentem: A review of clrsfnogenic and related health effects fdlow~th.u~~bunbcBkh/IndMJ1:233) 243.1980. 1Q Eighth R a r b k n Intunattonal Cladflcalkn of Ohasas, *ad far use &I the United States DHEW, Public Health Serrk. publkarlon 67-1693. W & i n g m , 0.C: US. Gm't P r i n t l a O m , 1967. 11. M o n r n RR: A d v s b of rrlulrr sunid and prooonknrl m a u U t y . Comput BIamrd R a 7:325-332,197+ 1 2 Mmul N, H w m u l W: S t r r b t l d O f rh. analy~hof d l t l from rsuap.c- 1ardb of d h r v } N U 32:719-748. 1959. 13. W.cr CP,T d S f , Cltwa RL: Anatomy of the k u l t h y w o r b Q affwt: A dtk.lh o w . J Ocwp Med 25:283-289.1983. 14. Greenburg L, Ma- MR, Goldwater L. e t al: Benzcm (b.rUol) poisoning In th.mtognwn prlntir indusw In New Y w k Ciw. / Ind Hyg To*d 21:2%420,1939. 15. A b y M, Dbicd K, Akgun T. e t d: Hematologicaleffects of chronic knrnn pobonir In 217 wverken. & J Jnd Med W.296- 302. 197r. r6. Townsod IC,On MC, Fhhbeck WAC Health exam flndings unong indivldvlh ocurpadonal~yexpored to k n z e n t [Occup Med 20543-548.1978. 17. Flshhbock WA. T o w n m d IC. Swank MG: Effects of chronic oceupatlonal exporum to-measumd consantntiowpf benzcnr / Ocwp M d 20539442,1978. 18. Frederit WS: Am the present IIm& of normality in hem, td- acceptable far multiphaic health xrwning? J Occup Med 16:3 13-323.1974. 19. V a c h k TI: Th. prodlcttnvalue of a p o l i t h and a negatln tatNEngl'JMed274:1171-117%1%h 20. US. 0- of Techndogy Assumene The Implkatlom of Cost4ffecttvrmr Analysis of Mkrl Technology: Background Paper No. 2. C a s Studla of Medical Technologies. C w Study No. & Codaffecttvcmr of Automated Multfchannel Chemirvy Andy- rem, May, 1981. Fleeting Beauty Gptured Throughout the English-speaking wodd the art of pwtraiture h a been haunted for close to 350 years by the impossible perfection of Sir Anthony VUI Dyck, who W ~ bSorn in Antwerp in i599 and died in London in 1641. Van Dyck was M ornament to whatever society had the good sense to make him welcome. A great master of the public portrait, he had also a sense of human nature as something that needed all the order and the dignity that could legitimately be granted. How could he not have been welcome wherever he went? When in l u l y in his eariy 30s. he took it for gra~~tethdat he would be accepted as a great gentleman. He looked the part, he dressed the put, he had a whde train of senants. he was in his own right an outstandingcollector (he owned 19 Titians, for instance), and he felt entitled to negotiate with his sitters as prince with prince. All this would have been ridiculous if the painting had been leu good than they are. Van Dyck might have been one of the hundreds o f ponr;lit painterr who tell white lies about their sitten and dress them up beyond their deserls. 8ut that he never did. He saw men and women alike with the e y a of an elegiac poet for whom beauty and riches and power were ephemeral attributes and all that mattered in the end was the beauty of the lived moment that would never recur. - From " Art View: Van Dyck Remains the Quintessential Permit Painter"by lohn Russell in The New York Tima, December 19,1982 . ,. 692 RefineryWorkers and BenzeMsai e t d