Document kDb6yzzjRX02RDjLYzq10B4jE

WVt 1*9"VI V| The search for a pattern lr||//| P in deaths among I WilihB Texas chemical workers he chemical worker calling the TClear Lake City, Texas, office of the Occupational Safety and Health Administration (OSHA) late in 1978 wanted some answers. He had a brain tumor, he told Richard Miller, an OSHA industrial hygienist. He knew of two fellow workers who had died of brain tumors, and he wanted somebody to investigate the Union Carbide plant (pictured at left) where they had all been employed. Spurred into action by that call, federal investigators by last November had found 18 confirmed deaths from brain cancer among people who had worked at the plant over the last 40 years. Their studies seem to show that longtime employees of the plant are Victor Alexander, an occiyttonal health faced with an elevated risk of dying of doctor, helped apply the science of brain cancer. epidemiology to brain cancer deaths In Texas The Union Carbide plant in Texas City is just one of several factories be ing investigated along the Texas gulf and plant and state officials, and came coast, where dozens of plants refine up with a total of four brain cancer oil and produce a wide variety of pe deaths among former employees at the troleum-based chemicals. At a fall con Union Carbide plant. The worker who ference sponsored by the New York made the original complaint was not Academy of Sciences, government re among them. His cancer turned out to searchers reported higher than normal have begun elsewhere than in his brain rates of brain cancer in certain groups (and he was still alive late in 1980). of workers at several of those plants. In January 1979 OSHA's Clear Lake Oneofthe grimmest pictures was paint City office sent a memo on the Union ed by William Nicholson, an epidemi Carbide findings to the regional office ologist at Manhattan's Mount Sinai in Dallas, which passed it on to OSHA School of Medicine. He found in a study headquarters in Washington. There it of members of the operating engineers' immediately aroused the suspicion of union that workers in chemical plants Victor Alexander, an occupational like Union Carbide's, compared with health physician, and his colleagues. those in oil fields or in refineries, suf "Someone else could have said the four fered an increased risk not only of brain cases were within the limits of sta cancer but of lung cancer and other tistical variation," said Alexander. kinds of malignancies as well. Some "But our prior knowledge influenced scientists at the meeting, particularly us. We knew that a number of potent those from industry, disagreed with the carcinogens are used in the chemical validity ofthese findings. industry." How the brain cancer cases came to Alexander decided to visit OSHA's re light is a dramatic story of epidemi gional office. While in Texas he went to ology, the science of tracing the course Union Carbide to talk to the plant phy of a disease in a population. After sician, David Glenn. Aware that Miller | he received the worker's complaint, had been looking into the brain cancer | OSHA's Miller began his own inves- problem, Glenn had already review ed | tigation. He called local doctors, union a collection of death certificates of 455 I officials, workers, families of workers. employees, primarily those who over ucc 053602 EPIDEMIOLOGY Richard Waxweiler and his colleague Sanford Laffingwall comb through filos in the NIOSH office in Cincinnati the years had died on the job or had be come eligible for pension benefits. He had found ten brain cancer deaths, in cluding the four that OSHA had uncov ered. Thus the toll was more than dou bled. "After we talked a bit, he showed me the list," Alexander recalls. "I must say that I was almost speechless." That number of deaths was enough to persuade OSHA to ask a sister agen cy, the National Institute for Occupa tional Safety and Health (NIOSH), for assistance in a full-fledged epidemiolog ical study. In February 1979 officials from both agencies met company rep resentatives and union leaders, and agreed on a plan of attack. Their first step was to try to find out whether more employees had died of brain cancer. Matching state of Texas death records against the names of all the employees who had worked at the Union Carbide plant since it opened in 1941, the re searchers had found three more brain cancer deaths by June. Union Carbide, which cooperated ful ly with the government investigators, went to work on its own files, combing its employee records for medical and work-history information. That search turned up two more brain cancer deaths. In the meantime, three more employees had developed brain tumors. All have since died, bringing the total to 18. No one contests the number of deaths. The task that now faces gov ucc 053603 ernment and industry is to find out whether the victims got cancer from their jobs, and if so, how. For answers, researchers are tracing the fate of 8,850 past and present Union Carbide work ers. This search, which Alexander calls "the real scut work of epidemiology," is necessary to establish whether the 18 deaths really represent a brain cancer rate higher than that ofthe general pop ulation. To determine what the expect ed number of deaths from brain cancer would' be for the Union Carbide work ers, scientists need to know not only the number of those employed and their ages but also the length of time they worked at the plant. The death rates are baaed on person-years (every year a worker lives after starting at Union Carbide counts as one person-year). To determine the actual number ofdeaths, the epidemiologists are trying to follow at least 95 per cent of the workers through an arbitrary cut-off point, the end of 1977. Last summer, when the study was about 80 per cent complete, Richard Waxweiler of NIOSH reported some preliminary results: for the entire Union Carbide population at the Texas plant, the risk of brain cancer was sbout 50 per cent higher than could be expect ed on the basis of the employees' ages and total person-years. For those who hsd worked less than ten years, there was no excess risk; for those who had worked ten to 20 years, some excess risk was evident; and for thoee who had been there 20 years or more, the risk was three times what it should have been. Waxweiler says these figures are just what one would expect with cancer, which can take 20 or more years to de velop. "The weight of the evidence points to an occupational risk factor," he reports. Finding a substance in the Union Car bide plant that might be responsible for the increased risk of brain cancer will be an even harder job. With this task in mind, the government scientists set up a second study, directed by NIOSH's Sanford Leffingwell. For each case of brain cancer, re searchers have chosen six other people, called controls, who closely match the victim in age, sex. race, date of hiring, and time worked at Union Carbide. Now they are studying the jobs of the controls and the jobs of the brain can cer victims, paying particular attention to each worker's exposure to chemicals, to see if any pattern sets the two groups apart. Because the plant's operation has changed dramatically over the years, chemical engineers must be brought in to tell the researchers about the com ponents and by-products of out-of-date processes. That investigation is more than half done, and so far there is no sign of any correlation between a specific chemical exposure and brain cancer. Though Lef fingwell thinks something at the plant caused the deaths, he says that there is to date no proof that the apparent ex cess risk at Union Carbide is not the re sult of chance. H. Michael Utidjian, Union Carbide's associate corporate medical director for toxicology and epidemiology, also makes this point. Utidjian emphasizes the failure so far to demonstrate a cause ofthe brain cancers. He is not convinced that they are the result of the work en vironment at Union Carbide. The cause ofthe deaths, he says, could be some fac tor in the community at large, rather than in the plant. Whatever the outcome, the Union Carbide case has changed the epide miologists' attitude toward brain can cer. Once so rare that it was ignored in surveys, the disease hss become so notorious, says Alexander, that it would almost be epidemiological malpractice to ignore it. With this kind of atten tion, and time, the answers may emerge. --James Gorman OnCOVBt / JANUARY 1981