Document Lp2yJJV9Zavo2GGeB6kZDGBeb

(SlJPfc Haskell Laboratory for Health and Environmental Sciences 1090 Elkton Road Newark, DE 19714 AR226-1307-6 From: Robin C. Leonard Epidemiology Program Tel: (302) 366-6594 Fax: (302) 366-5207 Date: To: Subject: January 27,2003 Paul Bosssrt, Washington Works Dave Ramsey, Washington Works Anthony Playtis, Washington W orks Beckie Elder, La Porte Leigh Belcher, Haskell Lab Gary Jepson, Haskell Lab Rudy Valentine, Haskell Lab Gerry Kennedy, Haskell Lab Barbara Dawson, BMP Richard Wilder, M.D., IHS Epidemiology Files BY: Cancer Incidence Report 1959-2001 All-Cause Mortality Report 1957-2000 W ashington Works, Parkersburg, West Virginia After a hiatus of several years, I am pleased to announce the resumption of the Epidemiology Program's Standard Cancer Incidence and All-Cause Mortality Surveillance Program for U.S. DuPont sites. Attached please find the subject report from the Epidemiology Surveillance Program. If you have any questions, please do not hesitate to phone me. Robin C. Leonard, Ph.D. 000031 R L003145 EID645647 E pidem iology S urveillance Report Cancer Incidence fo r W ashington W orks Site 1959-2001 Interpretation of Surveillance Data This report consists of tables showing the numbers of cancer cases reported from 1959 through 2001 at the DuPont Washington Works site in Parkersburg, West Virginia. For each specific cause the number of observed cases is compared to the number expected based on the experience of the entire U.S. Company population by the ratio of the observed to the expected numbers of cases. This ratio is the Standardized Incidence Ratio (SIR). This ratio is calculated only for those cancers for which at least five cases are observed. This is because the statistical estimates for small numbers are very uncertain, and unlikely to give any useful information. In addition, numbers smaller than 5 may well be due to chance occurrence, and seldom represent significant population trends. An SIR value of 1.0 indicates that the observed number of cases is equal to the expected, and therefore no increased risk is indicated. Accompanying the tables is a descriptive text that summarizes the main points. Sources of Surveillance Data Cancer cases that occur among active employees are recorded in the U.S. Company-wide Cancer Registry that was started in 1956. Through 1988, cases were reported to the Registry primarily by diagnoses entered on Accident and Health Insurance (A&H) claims and by death certificates that accompany life insurance claims filed by beneficiaries of deceased employees. Beginning in 1977, these sources were supplemented by Cancer Registry Report forms submitted by Company Medical personnel. Beginning in 2000, ascertainment of cancer cases is accomplished by a combination of Cancer Registry Reports from the plant sites, a screening of health insurance claims data, and case capture from death certificates acquired for the Mortality Registry. Cancer cases are included in the observed numbers for the plant site if the person worked there at the time of diagnosis. For cases for which the date of diagnosis is unknown, the person is included in the observed numbers for the last known site at which he or she worked. Methods of Analysis To determine expected numbers of cases for the standardized analysis, cancer incidence rates for DuPont employees, specific for gender, 5-year time, and 5-year age categories are computed for each cancer diagnosis observed. Then, the Company-wide rates are multiplied by the person-years contributed to each of those categories by the site population. The sum of the products over all age groups is the expected number of cases. This approach constitutes an 1 000032 R L003146 EID645648 internally standardized analysis, which is generally preferred because it provides age-adjusted expected numbers and is based on actual plant populations. Tests of Significance Beginning with the 2002 Surveillance Reports, no symbols will represent statistically significant increases or decreases. However, the 95% confidence interval around the SIR will be presented. There are two primary reasons for this change. First, the choice of significance level of p=0.05, while customary, is also somewhat arbitrary. We have decided to emphasize the size of the SIR (the magnitude of the difference), along with the stability of the estimate (the width of the confidence interval), as the indicators of possible need for further investigation. However, it remains that if the 95% confidence Interval includes 1.0, then that finding is not statistically significant for p = .05. It may be that the observed number for a particular cause is greater than the expected number, but the 95% confidence interval may still include 1.0. In this instance, it does not necessarily follow that there can be no occupational risk factors associated with this moderate excess of cases. If the number of persons at the plant is small, excess cancer morbidity would be difficult to detect because of dilution by data from the rest of the plant in addition, the duration of exposure may be too short for effects to be manifested by excess cancer cases. To provide additional information in those situations in which there appears to be an excess of cases, we have incorporated in this report a table that lists (without personal identification) individual cancer cases if that type shows an SIR greater than 2.5. This will enable us to examine such things as age at diagnosis, duration of employment, and the length of time between hire and diagnosis. These data often give us a good indication as to whether or not the pattern presented is indicative of occupational risk factors being involved. It is very important to understand that excess risk may occur because of other factors, such as smoking, diet, alcohol use, or family history. This type of information is not accumulated and analyzed in the routine Registry surveillance analyses. 2 000033 RI<003147 EID645649 Plant-Specific Summary of Findings-- Washington Works Table 1 is the-cancer incidence surveillance report for Washington Works: 1959-2001. Of the 42 types of cancer reported to have occurred in employees a t Washington Works, only 14 types had at least 5 cases observed. These types of cancer were colorectal (32 cases), pancreas (9 cases), larynx (6 cases), lung (61 cases in males), malignant melanoma (14 cases in males), female breast (8 cases), prostate (19 cases), kidney (18 cases), bladder (18 cases), brain (8 cases), lymphoid and histiocytic tissue (9 cases), multiple myeloma (7 cases in males), myeloid leukemia (8 cases), and unspecified sites (14 cases in males). Except for bladder and kidney, none of the confidence intervals around the SIRs excluded 1.0. The SIR for bladder cancer in males is 1.94, with the 95% confidence interval ranging from 1.15 to 3.07. These numbers indicate an increased risk for bladder cancer in males. All cases were male. The SIR for kidney cancer is 2.30, with the 95% confidence interval ranging from 1.36 to 3.65. These numbers indicate an increased risk for kidney cancer in males. All cases were male. Figures 1 and 2 compare the incidence rates of bladder cancer and kidney cancer for W est Virginia females and males to the rates for the U.S. and Ohio, based on data from the respective State Health Departments for the year 1998. West Virginia men tend to have higher rates of both kidney and bladder cancer. West Virginia women tend to have higher rates of bladder, but not kidney cancer. 3 000034 R Z.00314S EID645650 Figure 1. Comparison of Kidney and Bladder Cancer Incidence Across WVA, Ohio, and total U.S. (1998 rates) 01 eo0u0a, 14 12 10 oQaQo~ 8 6 u0a 4 M9 2 m0 Kidney iBWVA Women OHIO Women ;OUS Women Urinary bladder W est Virginia women have higher bladder cancer than the comparison groups, but their kidney cancer rate is the lowest o f the female comparison groups. Figure 2. Comparison of Kidney and Bladder Cancer Incidence Across WVA, Ohio, and total U.S. (1998 rates) " 50 0 i 40 1 30 20 %10 s 5o Figures 1 and 2 indicate that West Virginia men have higher rates o f kidney and bladder cancer than Ohio men and higher than the general U.S. male population. 4 TJ00035 R L003149 EID645651 A dditional Inform ation S pecific to th is Site Report Bladder Cancer Bladder cancer is the sixth most common cancer in the United States, excluding non-melanoma skin cancers. The American Cancer Society estimates that in 2003 there will be about 57,400 new cases of bladder cancer diagnosed in the United States (about 42,200 men and 15,200 women). In 2003, there will also be about 12,500 deaths from bladder cancer in the United States (about 8,600 men and 3,900 women). The following risk factors have been linked to bladder cancer Sm oking: Smoking is the most important risk factor for bladder cancer. Cancer-causing chemicals in tobacco smoke are absorbed from the lungs and get into the blood. From the blood, they are filtered by the kidneys and collect in the urine. These chemicals in the urine damage the cells that line the inside of the bladder and increase the chance of cancer developing. W ork exposure: Certain chemicals used in the dye industry have beon linked to bladder cancer. Other types of industries use chemicals that may put workers at risk if good safety practices are not followed. Smokers who work with cancer-causing chemicals have an especially high risk of developing bladder cancer. Race: Whites are two times more likely to develop bladder cancer than are African Americans. Age: The risk of bladder cancer goes up with age. > Chronic bladder inflammation: While chronic bladder irritations such as urinary infections and kidney and bladder stones don't cause bladder cancer, they have been associated with it in some studies. Personal history of bladder cancer: People who have had bladder cancer have a higher risk of forming another tumor. Birth defects o f the bladder: Very rarely a connection between the belly button and the bladder fails to disappear as it should before birth and can become cancerous. Use o f the herb, Aristocholia Fangchi: This Chinese herb, taken by some people to help them lose weight, has been linked to bladder cancer. Kidney Cancer The American C ancer Society estimates that there will be about 30,800 new cases of kidney cancer (18,700 in men and 12,100 in women) in the United States in the year 2001, and about 12,100 people (7,500 men and 4,600 women) will die from this disease. These statistics include both adults and children and 5 00036 R L003150 EID645652 include renal cell carcinomas as well as transitional cell carcinomas of the renal pelvis. Renal cell carcinoma is the most common type of kidney cancer in adults. In about 50% of cases, the renal cell carcinoma has not spread outside ' the kidney when it is discovered. In another 25% of people the cancer will be found to have grown locally outside the kidney, and in the remaining 25% it will have metastasized (spread farther away) to other parts of the body such as the lungs or bones. The following risk factors have been linked to kidney cancer Sm oking: Smoking doubles the risk of getting kidney cancer. O veruse o f certain pa in kille rs: Pain killers containing phenacetin were once popular non-prescription medications, but they have not been available in the United States for over 20 years. - A sbestos: Some studies show a link between exposure to asbestos in the workplace and kidney cancer. - Cadm ium : There may be a link between cadmium exposure and kidney cancer. Also, cadmium may increase the cancer-causing effect of smoking. Workers can be exposed to cadmium in the air from working with products such as batteries, paints, or welding materials. Gene changes (m utations): Genes are made up of DNA and are the basic units of heredity. They are the reason we resemble our parents. Changes or mutations in certain genes can increase the risk of developing kidney tumors. Some of these changes are inherited (people with a family history of renal cell cancer have an increased risk) and some can be caused by later damage, fo r example, by cigarette smoke. von Hippei-Lindau syndrome: This disease, caused by an inherited gene mutation (change), increases the chances of renal cell cancer and other types of cancer. Tuberous sclerosis: Patients who have this disease often have cysts in the kidneys, liver, and pancreas and are more likely to get renal cell cancer. Diet and w eight Some studies show a link between being overweight, a diet high in fat, and renal ceil cancer. Long-term d ia lysis: People who have been on dialysis fo r a long time may develop cysts in their kidneys that can give rise to renal cell cancer. Age: RCC is rare bn children and young adults; it is found mostly in adults between the ages of 50-70 years. " G ender Men are twice as likely to get renal cell cancer as are women. Not enough is known about the causes of renal cell cancer to say for sure how to prevent it Since smoking is linked to this cancer (as well as to other cancers), if you smoke, you should q u it Also, if you work with asbestos or cadmium, be sure to follow good safety practices. 6 000037 R I.0 0 3 1 5 1 EID645653 Recommendations for Follow-up We recommend that complete work histories on tine cases of bladder and kidney cancer be examined for any commonalities of occupational exposure, and the medical records be reviewed for the presence of other risk factors for kidney and bladder cancer. We also recommend that consideration be given to determining the feasibility of conducting a case-cohort study. This approach would provide an assessment of exposure potential and enable analyzing for associations with the health outcomes. The design of such a study would provide for one series of controls to be used for all the cancer cases (bladder and kidney.) It is important to remember that the ongoing TFE epidemiology study being conducted by the APME should provide important information about cancer outcome in at least part of the Washington Works cohort In addition, the work that will be done to categorize exposures for different jobs/tasks over time will be useful for a case-cohort study of the entire plant workforce. There is potential for leveraging these efforts to productive use in the surveillance program. I f ! can answer any questions, please do not hesitate to call. Sincerely, Robin C. Leonard, Ph.D. Principal Epidemiologist, E. 1. du Pont de Nemours, Inc. 7 000038 R X .003152 EID645654 TABLE 1. - CONFIDENTIAL - o Boo 0 O' 1H11 U1 01 CO MALES Cancer Type Observed Expected NEOPLASMS- H P , ORAL CAVITY, PHARYNX UR TONGUE MAJORSALIVARY GLANDS OTHER & UNSPECIFIED PARTS OROPHARYNX OTHER A ILL-DEFINED SITES 4 2 1 1 1 2 0.63 N/A. 1.36 N/A 0.78 'N/A ' 0.67 N/A ' 0.7 N/A 0.6 N/A NEOPLASMS - DIGESTIVE ORGANS & PERITONEUM ESOPHAGUS STOMACH 3 2 SMALLINTESTINE, INCLUDING DUODENUM COLORECTAL 3 32 Over a intrahepatic bile ducts GALLBLADDER a EXTRAHEPTIC BILE DUCTS 3 2 PANCREAS RETROPERITONEUMa PERITONEUM 9 1 OTHER a ILL-DEFINED SITES ' 4.82 N/A S43 N/A 0.67 N/A 308 1.04 2.98 ' N/A 1.35 N/A 9.13 . 088 0.29 N/A 0.23 N /f NEOPLASMS-RESPIRATORY A INTRATHORACIC ORGANS nasal cavities, mI ddle ear, a SINUSES 2 0.S6 LARYNX 6 3.39 N/A 1.77 * ObsaYeil/ltxptcictl Ratios art nut calculated wliai less than 5 cases art observed. 1/22/20(13 10:59:42 AM Page 1of 3 95% Confidence Interval FEMALES Observed Expected Ratio*" 95% Confidence Interval N/A - N/A N/A - N/A N/A - N/A N/A - N/A N/A N/A N/A - N/A WA - .M/A N/A - N/A N/A - N/A 0 71 - 1.46 N/A . ,M'A N/A - N/A 0.44- 187 N/A - N/A N/A - M/A . -------------- -------- ..- .................... ... 1 0.16 N/A N/A - N/A N/A - N/A 0 64 - 7.85 ..... -- --- ----------------- -- ....................... , WASHINGTON WHS PAREERSRL'RG IN EID645655 Cancer Tvtie TRACHEA, BRONCHUS, &UUNQ PLEURA MALES Obs/Exp Observed Expected R a 0 i- 61 60.2 101 . 4 Oi)S n/a : 95% Confidence Interval 0.77 1.30 N/A - vl/A FEMALES Observed Expected 3 1.32 Obs/Exp JRq li* N/A '' 95% Confidence In te rv a l N/A - N/A NEOPLASMS- BONE, CONNECTIVE TISSUE, SKIN, A BREAST BONE a ARTICULAHCARTILAGE 2 0.83 CONNECTIVE &OTHERSOFT TISSUE 3 2.62 MALIGNANTMELANOMAOF SKIN 14 10.6 FEMALE BREAST N/A - N/A N/A - N/A 0.72 - 221 " . J, !\ 1 021 < N/A - N/A 3 0.68 N/A. N/A - N/A 8 5.42 1.47 ' 0.63 - 2.90 NEOPLASMS - GENITOURINARY ORGANS CERVIX UTERI PROSTATE TESTIS BLADDER KIDNEY a URINARY ORGANS 18 22.1 ..os 5 3.41 . 1.46 16 8.26 18 7.78 0.61 - 1.34 0.47 . 3.42 1.15 - 3.07 1.36 3.65 1 1.25 N/A _ N/A - N/A NEOPLASMS-OTHER A UNSPECIFIED SITES BHAIN OTHER 8 UNSPECIFIED PARTSOF NERVOUS SYST THYROID GLAND OTHER ENDOCRINE GLANDS a RELATED STRUCT OTHER* ILL-DEFINED SITES UNSPECIFIED SITE 8 1 2 4 2 14 6.63 1.20 0.62 N/A 2.02 N/A 0.51 'n/a : 1.27 N/A 10.8 129 0.61 - 237 N/A - N/A N/A - N/A N/A - N/A N/A - N/A 0.70 - 2 16 1 U35 N/A 1 0.27 N/A N/A N/A N/A - N/A NEOPLASMS - L YMPHATIC A H E M A TOPOIETIC TISSUE LYMPHOSARCOMA* RETICULOSARCOMA 2 3.18 N/A N/A - N/A HODGKIN'S DISEASE 3 324 N/A N/A - N/A 1 0 18 N/A N/A - N/A OTHEn LYMPHOID * HISTIOCYTIC TISSUE 9 7.62 1.18 0 53 - 2 24 MULTIPLE WELOMA* IMMUNOPROUFERATIVE N 7 4.06 1.72 0 69 - 3 55 2 0.14 N/A N/A N/A M LYMPHOID LEUKEMIA 1 263 N/A N/A . N/A 1 0.11 N/A N/A - N/A 000040 * Obitrvui/Ex|>ecled Ratios are not calculated when less Ilian S cases arc otiscrvul. 1/22/21)03 10:39:42 AM Page 2 of 3 - f'ltb lP ID R N T tA I - WASHINGTON HVi.V PARKERSBURG H I Cancer Type MYELOID LEUKEMIA MONOCYTIC LEUKEMIA LEUKEMIAOF UNSPECIFIED CELLTYPE MALES Observed E^ ecud a " asa $ $ $ $ ? : 1 022 " 4 2J P S 5' : 9SH Confidence Interval 0.86 3.97 NIA - Is/A NIA . NIA FEMALES Observed - ---- ------ ------- - 1 Qbs/Bxp Expected '-".Ki{tie^ -" 'r-./V.. .'`fa.' i' 0.-0-7---- -"ti NiitI.An-".'-ii; ': 95% Confidence Interval N/A - NIA Location | SiteCodo PARKERSBURG WV 12560 WASHINGTONWKS PARKERSBURGWV 2661 WASHINGTON RES LABPARKSBGWV 2667 WASHINGTON WORKS WV 2569 Reference Population Total 282026 M |l97729 Dale 1/20/2003 1/20/2003 1/20/2003 1/20/2003 Sita Population Total 6523 F 1034 M 14409 Date 1/22/2003 1/22/2003 1/22/2003 1/22/2003 "0D41 M * Obstxved/Expecled R ubs are noi calculated wliui less titan 5 cusca arc observed. 1/22/2001 10:59:42 AM Patte 3 of 3 _ CONFIDENTIAL - WASHINGTON H US TA RKF.RSBI'RG H I ' E pidem iology S urveillance Report A ll-C ause M ortality fo r the W ashington W orks Site 1957-2000 Interpretation of Surveillance Data This report consists of tables showing the numbers of deaths from all causes reported from 1957 through 2000 at the DuPont Washington Works site in Parkersburg, W est Virginia. For each specific cause the number of observed deaths is compared to the number expected based on the experience of the entire U.S. Company population by the ratio of the observed to the expected numbers of deaths. This ratio is the Standardized Mortality Ratio (SMR). This ratio is calculated only for those causes of death for which at least five deaths are observed. This is because the statistical estimates for small numbers are very uncertain, and unlikely to give any useful information. In addition, numbers smaller than 5 may well be due to chance occurrence, and seldom represent significant population trends. An SMR value of 1.0 indicates that the observed number of deaths is equal to the expected, and therefore no increased risk is indicated. Accompanying the tables is a descriptive text that summarizes the main points. Sources of Surveillance Data Deaths that occur among active and pensioned employees are recorded in the U.S. Company-wide M ortality Registry that was started in 1957. Deaths are reported to the Registry by the corporate Benefits division through death certificates that accompany life insurance claims filed by beneficiaries of deceased employees and pensioners. Deaths are ascribed to the observed numbers for the plant site at which the employee worked at the time of death, or the site at which the pensioner worked at the time of retirem ent. Methods of Analysis To determine expected numbers of deaths for the standardized analysis, m ortality rates for DuPont employees and pensioners, specific for gender, 5-year time, and 5-year age categories are computed for each cause of death observed. Then, the Company-wide rates re multiplied by the person-years contributed to each of those categories by the site population. The sum of the products over all age groups is the expected number of deaths. This approach constitutes an internally standardized analysis, which is generally preferred because it provides age-adjusted expected numbers and is based on actual plant populations. 1 000042 R X .003156 EID645658 Tests of Significance Beginning with the 2002 Surveillance Reports, no symbols will represent statistically significant increases or decreases. However, the 95% confidence interval around the SMR will be presented. There are two primary reasons for this change. First, the choice of significance level of p=0.05, while customary, is also somewhat arbitrary. We have decided to emphasize the size of the SMR (the magnitude of the difference), along with the stability of the estimate (the width of the confidence interval), as the indicators of possible need for further investigation. However, it remains that if the 95% confidence interval includes 1.0, then that finding is not statistically significant for p = .05. - It may be that the observed number for a particular cause is greater than the expected number, but the 95% confidence interval may still Include 1.0. In this Instance, it does not necessarily follow that there can be no occupational risk factors associated with this moderate excess of deaths. If the number of persons at the piant is small, or the plant is recently built or acquired, excess mortality would be difficult to detect because of the small probability of this population having any deaths. To provide additional information in those situations in which there appears to be an excess of deaths, we have incorporated In this report a table that lists (without personal identification) individual deaths If that cause shows an SMR greater than 2.5. This will enable us to examine such things as age at death, duration of employment, and the length of time between hire and death. These data often give us a good indication as to whether or not the pattern presented is indicative of occupational risk factors being involved: It is very Important to understand that excess risk may occur because of other factors, such as smoking, diet, alcohol use, or fam ily history. This type of information is not accumulated and analyzed in the routine Registry surveillance analyses. 2 000043 IU L 0 0 3 1 5 7 EID645659 Plant-Soecific Summary of Findings--Washington Works Table 1 is the all-cause mortality surveillance report for Washington Works: - 1957-2000. The only causes of death for which the SMR was greater than 2.0 were diseases of blood and blood-forming organs in males (SMR = 2.97; 95% Cl = 0.95-6.94); and rheumatic heart disease in males (SMR = 3.55; 95% Cl = (1.14-8.30). Note the exclusion of 1.0 in the confidence interval around the SMR for rheumatic heart disease in males. Two other categories of circulatory system diseases were significantly elevated. These were acute myocardial Infarction (SMR = 1.38; 95% Cl = 1.151.64); and atherosclerosis and aneurysm (SMR = 1.98; 95% Cl = 1.17-3.14). Recommendations An increased risk for mortality due to heart disease Is not a new finding at W ashington Works. An earlier study on heart disease at this site did not identify any occupational risk factors. We recommend the following: 1. Consider undertaking a feasibility assessment for a case-cohort study for heart disease, with emphasis on detailed exposure assessment and identification of other risk factors for heart disease. 2. Provide additional communications to workers concerning the known risk factors for heart disease, and consider on-site preventive programs. If you have any further questions, please do not hesitate to call me. Robin C. Leonard, Ph.D. Principal Epidemiologist E.l. du Pont de Nemours, Inc. 3 000044 R L003158 EID645660 - CONFIDENTIAL - Cause o f Death Observed Expected MALES Oh&Bxp R a ti** 95% Confidence In te rv a l INFECTIOUS A PARASITIC DISEASES INFECTIOUS AM) PARASITIC DISEASES 12 11.7 1.0? 0.62 - 1.78 N EO PLASM S-U P, ORAL CAVITY, A PHARYNX 1UP, ORALCAVITY, ANOPHARYNX 4 3.01 N/A N/A - NI,A NHUPLASMS - DIGESTIVE ORGANS A PERITONEUM ESOPHAGUS 3 4.31 N/A STOMACH 2 4SI N/A SMALL INTESTINE. INCLUDING DUODENUM 2 0.40 N/A COLORECTAL 19 20.3 093 LIVERA 1NTRAHEPATIC BILE DUCTS 3 2.83 N/A GALLBLADDERS EXTRAHEPTIC BILE DUCTS 2 1.19 N/A PANCREAS 7 8.68 OSO N/A - N/A N/A N/A N/A - N/A 0.66 - 1.46 N/A N/A 0.32 - 1.66 11 1f N EO PU S M S - RESPIRATORY A INTRATHORACIC ORGANS NASALCAVITIES. MIDDLE EAR, &SINUSES 2 OS LARYNX 2 1.31 TRACHEA. BRONCHUS.8 LUNG 64 66.1 PLEURA ! 0.79 N/A N/A 097 N/A N/A - N/A 0.73 - 1.27 N/A N/A NEOPLASMS- HONE. CONNECTIVE TISSUE. SKIN. A BREAST BONE AARTICULARCARTILAGE 2 0.42 CONNECTIVE AOTHERSOFT TISSUE 2 1.24 MALIGNANT MELANOMAOF SKIN 2 3S3 N/A N/A N/A N/A N/A N/A - N/A N/A - N/A * Oteavul/lixpotlotl Ratios arc nol culcnlaial when less Ilian 5 eases arc obvxvcd. 1/22/2003 10:58; 18 AM Page 1of 5 _ CONFIDENTIAL - FEMALES QbsfB*p Observed Expected R a th * 9S% Confidence Interval 1 0.11 N/A N/A - N/A 1 1.06 N/A N/A - N/A 1 012 N/A ` N/A - N/A WASHINGTON WHS PARKERSBURG WV 000045 Cause o f Death OTHER SKIN FEMALE BREAST Observed 1 Expected 0.33 MALES Obs/Exp B atid11 m 95% Confidence In te rv a l N/A - N/A Observed FEMALES Obs/Exp Expected Bada* 95% Confidence Interval 2 1.79 N/A N/A - N/A NEOPLASMS - GENITOURINARY ORGANS PROSTATE TESTS BLADDER KIDNEYA URINARYORGANS 9 14.2 : 0.63 0J28- 1.20 1 0.38 ' m ' N/A N/A 7 4.34 1.61 0.64 - 3.32 8 &20 1 0.68 - 3.02 ` ... NEOPLASMS - OTHER A UNSPECIFIED SITES BRAIN OTHER ENDOCRINE GLANDS&RELATED STRUCTUR OTHER & ILL-DEFINEDSITES SECONDARYMALIGNANT NEOPLASMOF RESPIRATO UNSPECIFIED SITE 8 3 1 1 13 5.49 1.09 032 N/A 0.75 N/A 0.1 N/A 9.85 1.32 039 - 237 N/A N/A N/A - N/A N/A - NIA 0.70 - 2.25 1 0.27 N/A N/A - N/A NEOPLASMS - LYM P H A TIC A HEM ATOPOIETIC TISSUE LYMPHOSARCOMA A RETICULOSARCOMA 1 1.61 N/A N/A - NIA HODGKIN'S DISEASE 2 1.49 N/A N/A NIA OTHER LYMPHOIDA HISTIOCYTIC TISSUE 3 4.99 N/A MA - NIA MULTIPLE MYELOMAA IMMUNOPROUFERATIVE NEO 5 3.44 1.45 0.46 - 3.39 2 0.11 N/A N/A - N/A LYMPHOID LEUKEMIA 1 0.06 N/A N/A - N/A MYEL0I0 LEUKOMA 8 3.49 1.71 0.62 - 3.74 MONOCYTIC LEUKOMA 1 0.12 N/A NIA NIA LEUKEMIAOF UNSPECIFIEDCELL TYPE 3 1.83 N/A N/A - N/A NEOPLASMS O F UNCERTAIN BEUA VIOR * 1NEOPLASMSOF UNSPECIFIED NATURE 1 0.98 N/A NIA NIA M H ENDOCRINE, NUTRITIONAL. & M ETABO LIC DISEASES, A IM M U N ITY DISORDERS ...... I 1 \ II ii__ [diabetes melutus 12 7.60 1.67 0.81 - 2.75 1 0.21 N/A 000046 Ot*>crvul/t:x|Hx;lul Ramiti arc mil calculated whui less ilion 5 casts arc lit. Tvul. 1/22/2003 10:58: IKAM Pute 2 of S - CONFIDENTIAL - WASHINGTON WHS PARKERSBURG WV Cause o f Death DISORDERS OF UP0ID METABOLISM Observed 1 OTHER AND UNSPECIFIED DISORDERS OF METABOU 1 OBESITY ANOOTHER HYPERALIMENTATION 1 Expected 0.68 0.48 0.33 MALES Oks/Exp R atio* MA MA M'A 95% Confidence In te rv a l MA - MA MA MA MA - MA Observed FEMALES Ohx/Exp Expected R a th * 95% Confidence In te rv a l DISEASES O F BLOOD A BLOOD-FORMING ORGANS DISEASES OF BLOODAND BLOOD-FORMINGORGANS 6 1.68 '2S7 0S5 - 6.94 *1 il ! t1 MENTAL DISORDERS [senile AND PRESENILE ORGANIC PSYCHOTIC COM3 2 1.43 m DISEASES O F NER VOUS SYSTEM A SENSE ORGANS OTHER CEREBRAL DEGENERATIONS 1 PARKINSONS DISEASE 2 OTHER CONDITIONS OF BRAIN MONONEURITIS OF LOWERLIMB ! 2.32 N/A 1.89 MA 0.02 MA MA MA MA - MA MA - MA 1 0.05 MA 1 0.05 ' m a MA - MA MA - MA DISEASES O F CIRCULATORY SYSTEM RHEUMATIC HEART DISEASE HYPERTENSIVE DISEASE ACUTE MYOCARDIAL INFARCTION OTHER ACUTE ANOSUBACUTE FORMS OF ISCHEMIC OTHER FORMSOF CHRONIC ISCHEMIC HEART DISEA ACUTE PULMONARY DISEASE OTHER CARDIOPATHY CEREBROVASCULAR DISEASES ATHEROSCLEROSIS AM) ANEURYSM OTHER VASCULARDISEASE S 6 128 1 71 3 48 27 18 4 1.41 3.99 6.23 o s e 92.3 199 2.0S m a ' ' 71.5 0.99 7 9 M A 39.3 1S2 28.1 096 9.06 1.98 2.96 M A 1.14 - 8.30 0.39 - 2S9 1.15- 1S4 M A - MA 0.77- 125 M A - MA OSO - 1.62 0.63 - 1.39 1.17 - 3.14 MA - MA 1 0.89 M A 1 0.67 MA 1 0.60 MA 1 0.79 M A MA - MA MA - MA MA - MA MA - MA __- J M DISEASES O F RESPIRA TOR 1' SYSTEM OTHER BACTEFUALPNEUMONIA __ ..... 0.67 MA MA - MA PNEUMONIA. ORGANISMUNSPECIFIED 8 ......... ............... ......... ......... ---- --------- --- ------------- ------------------- .... 8.63 0.92 0.39- 1.82 * Obscrval/lixjkJtial Raiins arc wit calculated when less than 5 eases arc oIjmyvuI. 1/22/200.1 10:58:IK AM Page 2 of 5 .. CONFIDENTIAL - -- -- WASHINGTON WAS PARKERSBURG WV 000047 i ! i. 000048 Cause o f Death Observed EMPHYSEMA 0 CHRONIC AIRWAY OBSTRUCTION. NOT ELSEWHERE 12 ASBESTOSIS 1 PULMONARY CONGESTION ANO HYPOSTASIS 1 POSTINFLAMMATORY PULMONARYFIBROSIS 2 OTHER DISEASES OF LUNG 2 Expected 8.11 10.7 0.77 0.33 1.51 1.67 MALES Qbs/Bxp R ati* |.17 1.12 WA m' m W ' 9S% Confidence In te rv a l 0.42- 235 0.57 - 135 NIA NIA NIA - NIA NIA - NIA NIA NIA DISEASES O F DIGESTIVE SYSTEM CHRONIC UVER DISEASE ANDCIRRHOSIS OTHER DISORDERS OF GALLBLADOER GASTROINTESTINAL HEMORRHAGE 8 IL21 037 1 0.26 NIA 1 036 NIA 0.41 - 132 NIA NIA NIA NIA DISEASES O F GENITOURINARY SYSTEM [DISEASES OF KIDNEYAND URINARYTRACT 5 6.34 0.78 0.25 - 184 DISEASES O F THE MUSCULOSKELETAL SYSTEM A CONNECTIVE TISSUE [MUSCULOSKELETALAND CONNECTIVE TISSUE DISE 2 1.36 NIA NIA - NIA CONGENITAL ANOMALIES OTHERCONGENITAL ANOMALIESOF NERVOLf SYST CONGENITALANOMALIESOF URINARYSYSTEM 1 1 0.11 NIA 0.18 NIA NIA - NIA SYMPTOMS, SIGNS, A ILL-D EFIN ED CONDITIONS [SYMPTOMS. SIGNS, AND ILL-OEFINEOCONDITIONS 3 7.39 NIA NIA NIA EXTERNAL CAUSES OF INJURY A POISONING OTHER TRANSPORT ACCIDENT MOTOR VEHICLEACCIDENT ACCIDENTAL FALL FIRE OR EXPLOSION OTHER EXTERNALCAUSES OF INJURYAND POiSONI SUICIDE 2 10 1 1 4 7 2.40 .NIA 18.7 1.01 2.31 NIA 1.96 NIA 6.02 NIA 13.0 0.53 NIA NIA 0.61 1.58 NIA - NIA WA WA WA - WA 0.21 - 1-10 * Otacrvod/Expeclut Ratios arc not calculauxl when less Ilian 5 cases arc nlr a vud. I/22/20U3 IO:58:IKAM Puge4olS - CONFIDENTIAL - Observed 1 FEMALES Obs/Exp Expected R atio* 0.06 WA 9S% Confidence In te rv a l WA - WA 1 0.13 WA WA- WA ] ' ....' ........ " 1 1 0.19 WA WA- WA 1 2 086 WA WA - WA 1 0.06 WA WA - WA WASHINGTON WHS FARRERSBURU WV Cause o f Death ASSAULT INJURY DUE TO WARFARE MALES ObsfExp Observed Expected 1 2.80 m ' ' 1 0.02 " T P f 95% Confidence In te rv a l NIA NIA FEMALES Obs/Exp Observed Expected R ptfaf 95% Confidence In te rv a l 3"" Location 1SMeode 1 ReferencePopulation Date Site Population Date PARKERSBURG WV 2560 frfcoi 1262026 1/2012003 Total 5523 1/22/2003 i WASHINGTON WKS PARKERSBURGWV 2661 F 164275 1/2012003 F 1034 1/22/2003 WASHINGTON RES LABPARKSBGWV 2567 N 197729 1/2012003 M >4489 1/22/2003 WASHINGTONWORKS WV 25 1g 1/2012003 1/22/2003 00004 n Obscrvul/Expoclul Ratios arc not calculated wluai less lbau 5 cases are observed. 1/22/2003 10:58:18 AM Page 5 of 5 . CONFIDENTIAL - WASHINGTON WKS PARKERSBURG WV