Document npw4J4YLyLbLOwbVBEkgE1J7m

3MCompany Page Iof 63 . Date: October II, 2001" FINAL REPORT Epidemiology, 220-3W-05 Medical Department 3M Company St. Paul, MN 55144 Title: A Longitudinal Analysis of Serum Perfluorooctanesulfonate (PFOS) and Perfluorooctanoate (PFOA) Levels in Relation to Lipid and Hepatic Clinical Chemistry Test Results from Male Employee Participants of the 1994/95, 1997 and 2000 Fluorochemical Medical Surveillance Program Study Start Date: $.uly 1, 2001 Protocol Number (not applicable) IRB Approval Exempt X E:'-_eAited IRB Approval Date: (not applicable as these data are from a medical surveillance program) Principal Investigator:. GearyW. Olsen, D.V.M.,Ph.D) Co-investigators: MicheleM. Burlew,M.S.t Jean M. Burris, R.N., M.P.HJ Jeffrey H. Mandel, M.D., M.P.HJ Study Director:. Jeffrey H. Mandel, M.D., M.P.H.' 1. 3M Medical Department, 220-3W-05, St. Paul, MN 55144-1000 "(Corrections made from previous version) 3M Company Page 2 of 63 ( ABSTRACT The 3M fluorochemical medical surveillance pro_a_arwn as conducted in 1994/95, 1997 and 2000 at the company's Antwerp (Belgium) and Decatur (Alabama) manufacturing plants. Although cross-sectional assessments of the data have been reported, the opportunity to conduct a longitudinal assessment became possible as a result of a large number of employee participants in the 2000 fluorochemical medical surveillance program. A total of 175 male employees voluntarily participated in the 2000 program and at least one of the two previous program years. A total of 106 (61 percent) of the 175 employees participated in the 1994/95 program and 110 (63 percent) of the 175 participated in the 1997 program. Of these 175 employees, a total of 41 (24 percent) participated in all three years (Antwerp = 21, Decatur = 20), 65 (37 percent) participated / in 1994/95 and 2000 (Antwerp = 45, Decatur = 20) and 69 (39 percent) participated in 1997 and 2000 (Antwerp -- 34, Decatur = 35). There were insufficient number of female employees to conduct any meanin_ul lon_tudinal assessment. Only 14 female employees participated in the 2000 fluorochemical medical surveillance program and at least one of the previous program years. Serum perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) were assayed in each surveillance program year although the method of analysis (high performance liquid chromatography mass spectrometry) differed slightly between years. A different research laboratory was used to assay PFOS and PFOA in each year. The same hospital laboratory analyzed the clinical chemistries for all three surveillance years. These included: cholesterol (m_dl). high density lipoproteins (I-IDL --,., 3MCompany Page 3of 63 m_dl) and triglycerides (m_dl); alkaline phosphatase (IU/I.,), gamma _utamyl ( transferase (GGT, IU/L), aspartate aminotransferase (AST, IU/L), alanine aminotransferase (ALT, I'U/L), total and direct bilirubin (m_dl). Most reference ranges remained relatively constant over time except for ALT. In each surveillance year, potential confounding factors were also determined. These covariates included age, body mass index, number of alcoholic drinks per day and cigarettes smoked per day. The continuous outcomes of lipid and hepatic clinical chemistry tests were evaluated as repeated measures incorporating the random subject effect fitted to a mixed model by the MIXED procedure in the SAS statistical package. Restricted maximum likelihood estimates of variance parameters were computed. Adjusted regression models were built by introducing all covariates and testing the covariance structure. Significant coefficients were defined when the p value was < .05. There was a positive association between PFOA and serum cholesterol and triglycerides over time but not with PFOS. This was association was limited to the Antwerp employees and, in particular, the 21 Antwerp employees who participated in all three surveillance years. This positive association between PFOA and serum lipids is opposite the inconsistent toxicological evidence that suggested a possible hypolipidemic effect of PFOA in rodents and no effect in primates. Adjusting for potential confounders, there were no temporal changes associated with the fluorochemical tests, PFOS, PFOA and TOF, and the hepatic clinical chemistry, tests. Limitations of this study included the number of employees with three years of surveillance data (only 24% of the 175 subjects), the inability to analyze temporal changes due to small numbers in female employees, the use of different laboratories and / 3M Company Page4 of63 the associated systematic (experimental error) with each fluorochemical assay for the three surveillance progTam years and the lower levels of serum PFOS and PFOA .. measured in each program year among these employees compared with those that cause effects in laboratory animals. i 3M Company Page 5 of 63 ENTRODUCTION i" The 3M fluorochemical medical surveillance program is conducted on a routine basis at the company's Antwerp (Bel_um) and Decatur (Alabama) manufacturing plants. Employee participation is voluntary. Priorto 1994, only total organic fluorine was measured and no specific fluorochemical analytes were measured. Serum perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) have been routinely assayed since 1994/95 rather than total organic fluorine as the analytical capabilities have improved. Cross-sectional analyses of the 1994/95 and 1997 medical surveillance program data and the 2000 data in relation to Antwerp and Decatur employees' serum PFOS levels have been reported elsewhere (Olsen et al, 1999a, 1999b, 2001). In the 1994/1995 medical surveillance program, a total of 178 male employees participated (Antwerp = 88; Decatur = 90) and 149 male employees participated in the 1997 program (Antwerp = 65; Decatur = 84). For these two program years, there were too few female -o -. participants to include in the data analysis (Olsen et al 1998). In the 2000 fluorochemical medical surveillance program, there were considerably more participants: 421 males (Antwerp = 206; Decatur - 215) and 97 females (Antwerp - 49; Decatur = 48). It was suspected that the increased voluntary participation in 2000 was due to increased employee awareness of the persistence and prevalence of PFOS in human tissue and the environment and the company's May 16, 2000 phase out announcement that it would cease the production o_,perfluorooctanyl chemistry in certain repellents and surfactants by the end of 2000. Regardless of the surveillance year, there have been several consistent differences between the Antwerp and Decatur male employee populations. The Antwerp male 3M Company Page 6 of 63 ,_" employee population has been si_ificantly younger than Decatur, has had lower Body Mass Indices (BMI) and hig.her self-reported daily consumption of alcohol. In addition, the Antwerp male employee population clinical chemistry profiles were different for several tests including lower mean alkaline phosphatase and triglyceride values and higher total bilirubin and I--IDLvalues than the Decatur male employee population. Analyses of workers' lipid and hepatic clinical chemistry results have not been associated with hypolipidemic effects and PFOS as reported in rodents (3M Company 2000; Haughom and Spydevold 1992; Ikeda et al 1987; Pastoor et al 1987; Seacat et al 2001a; Sohlenius t al 1993) and primates (Seacat et al 2001b). In the 2000 medical surveillance program, statistical analyses also examined the relation between PFOA and a calculated total organic fluorine index (TOF) to clinical chemistries, hematology, thyroid hormones and urinalyses (Olsen et al 2001). A positive association was observed between triglycerides and PFOA; however, this association was opposite the data that have inconsistently reported a hvpolipidemia effect of PFOA in rodents (I-Iaughom and Spydevold 1992; Pastoor et al 1987) and no effect in primates (Butenhoff et al 2001). Furthermore, this positive association between PFOA and triglycerides has not been observed at the 3M Cottage Grove manufacturing plant (Olsen et al 2000) where employees' serum levels have, historicaUy, been much higher than those measured among Antwerp and Decatur employees (Olsen et al 1999; 2001a; Olsen et al 2001b). The inability to assess temporal changes in cross-sectional studies is a wellknown limitation of this design. The large participation of employees in the 2000 fluorochemical medical surveillance who may have participated in the I994/95 and/or 1997 surveillance programs at these two manufacturing sites allowed for an opportunity  -, i 3M Company Page7 of 63 to conduct a longitudinal analysis among the male employee population. Altogether, a total of 175 employees (Antwerp = 100; Decatur = 75) who participated in 2000 had also participated in at least one previous fluorochemical medical surveillance exam since 1994/95. Therefore, the purpose of this analysis was to conduct a longitudinal assessment of this 6 year time period regarding the relationship of PFOS, PFOA and TOF to the medical surveillance data collected on these 175 Antwerp and Decatur male employees. METHODS Dam Collection Data werecompiledfrom the1994/95,1997and2000 fluorochemicmaeldical surveillanpcreoem'amdatabasesA. totaolf175maleemployeesparticipatiendfl_2e000 programandat]eastoneofthetwo previousprogramyears.A totaolf106(61percent) -o" ofMe |75employeesparticipatiendthe1994495program and 1I0(63percento)fthe 175participatiendthe1997program.Of these175employees,a totaolf41 (24percent) participatiendalldt,'eyeears(Antwerp= 21,Decatur= 20),65 (37percentp)articipated in1994/95and2000 (Am_verp= 45,Decatur= 20)and69 (39percentp)m'ficipatiend 1997and 2000(Antwerp= 34,Decatur= 35).Forpurposesofbrevityt,hesethree subpopulations will hereafter be referred to as subcohorts A, B and C. Demom-aphic data (age, BMI, alcoholic drinks per day and cigarettes per day) were recorded for each employee in each surveillance year. A standard set of clinical chemistries and hematology, data was also obtained for each employee. Given results from previous toxicological studies, the longitudinal analyses focused on lipid ..... i ,., .o-, \ 3M Company Page 8 of 63 [cholesterol (m_dl), high density lipoproteins (HI)L, mffdl) and triglycerides (m_dl)] and hepatic [alkaline phosphatase (IU/L), gamma glutamyl transferase (GGT, IU/L), aspartate aminotransferase (AST, rU/L), alanine aminotransferase (ALT, IU/L), total and direct bilirubin (mg/dl)] clinical chemistries that were measured in each program year by the same laboratory (Allina Laboratories, St. Paul, MN). Reference ranges were relatively constant over time, although for ALT the range declined from 20-65 IU/L in 1994/95 to 1-40 IU/L in 1997 and 2000. Fluorochemical Analyses PFOS and PFOA were assayed in 1994/95, 1997 and 2000. However, the method of analysis differed slightly for each year. In 1994195,the method used tetrabutylammonium to ion-pair with PFOS and PFOA in the serum (Johnson et al 1996). The ion-pairs were then extracted with ethyl acetate. The abstraction product was then analyzed using high-performance liquid chromatograph-thermospray mass spectrometry. In 1997, the serum samples were analyzed by liquid chromatography/mass spectrometry, using selected ion monitoring in the negate-ion mode (Anderson and Mulvanna 1997a; 1997b). In 2000, sera samples were extracted using an ion-pairing extraction procedure (Hansen et al, 2001). Only in 2000 were the extracts quantitatively analyzed for PFOS and PFOA as well as the other analytes: PFHS_(perfluorohexanesulfonate), PFOSAA (Nethyl perfluorooctanes,ulfonamidoacetate), M570 (N-methyl perfluorooctanesulfonamidoacetate), PFOSA (perfluorooctanesulfonateamide) and M556 (perfluorooctanesulfonamidoacetate). High-performance liquid chromatom'aphy/electrospray tandem mass spectrometry (I-IPLC/ESMSMS) was the 3M Company Page 9 of63 f technique used in 2000. The samples were evaluated versus an extracted curve from a human serum matrix. Analyses were conducted at different laboratories in the three surveillance years. For purposes of this longitudinal analysis, a total organic fluorine index (TOF') was determined by calculating the percent of PFOS and PFOA that was attributed to organic fluorine (64.7 and 69.0 percent, respectively) multiplied by the ppm measured for each of these two fluorochemicals and then summed to produce the TOF. Data Analysis Briefly, mixed models can be used in the analysis of repeated measures data which are simply data sets with multiple measurements of a response variable on the same subject over time. Detailed explanation of these models is provided elsewhere (Littell 1996; 2000). Mixed models contain factor effects which are considered both <,r ( fixed and random. An effect is fixed if the level.s.-. in the studyrepresent all possible levels of the factor, or at least all levels about which inference is to be made. Factor effects are random if the levels of the factor that are used in the study represent only a random sample of a larger set of potential levels. The focus of the standard linear model is to model the mean of y by using the fixed.effects parameters B. That is, y=X#+e where y represents a vector of observed data, fl is an unknown vector of fixed effects parameters with known design matrix X, and is an unknown random error vector modeling the statistical 'noise' around Xfl. The residual errors are assumed to be _.... 3M Company Page l0 of 63 independent and identically distributed Gaussian random variables with mean 0 and Sj - variance o-9. A generalized standardlinear model is a mixed model which is: y=X/3+Zy+ where y is an unknown vector of random-effects parameters with known desi_ matrix Z, and is an unknown random error vector whose elements are no longer required to be independent and homogeneous. If y + e are assumed to be Gaussian random variables that are uncorrelated and have expectations 0 and variances G and R, respectively, then the variance ofy is: V = ZGZ' + R The varianceofthe data,y, can be modeled by specifying the structure of Z, G and R. The model matrix Z is designed in the same fashion as X, the model matrix for the fixedeffects parameters. For the matrices G and R, a covariance structuremust be selected in risingmixed models. Since observations on different subjects are assumed to be independent, the structurerefers to the covariance pattern of repeated measurements on the same subject. For most of these structures, the covariance between two observations on the same subject depends only on the length of the time interval between measurements and the variance is constant over time. Numerous covariance structuresexist. Common examples include the following. Simple covariance structure (SIM) specifies that the observations are independent, even on the same subject, and have homogeneous variance. It is usually not realistic for most repeated measures data because it specifies that observations on the same subject are.independent. Compound symmetric (CS, otherwise 3M Company PageIIof63 known asvariancecomponents)structursepecifietshatobservationosn thesame subJect havehomogeneouscovarianceandhomogeneous.variancCeo.rrelatiobnestweentwo observationasreequalforallpairsofobservationosn thesame subjectA.utoregTessive orderI (AR(1))covariancsetructursepecifiehsomogeneousvariancbeutthatcovariances betweenobservationosn thesamesubjectarenotequal,butdecreasteowardzerowith increasintgimeintervabletweenmeasurements(lag).Itslimitatiiosnthatobservations on thesame subjecftarapartintimewouldbeessentialilnydependentA.utoregressive withrandom effecftorsubject(AR + RE) covariancsetru_:tusrpeecifiehsomogeneous variancpelusthecovariancbeetweenobservationosn thesame subjecatrisefsromtwo sourcesI:)anytwo observationsshareacommon contributiboencausetheyareon the same subject; and 2) the covariance between observations decreases exponentially with lag but only to the common contribution (not to independence), Toeplitz (TOEP) r. ! structure specifies that covariance depends only on lag but not as a mathematical.., function with a small number of parameters. The 'unstructured' structure (UN) specifies no patterns in the covariance matrix and is therefore completely general. The above structures are appropriate if equal spacing (of data) is assumed in a time series analyses. In situations where unequally spaced longitudinal measurements exist, spatial covariance structures can be used. In the present analyses, equal spacing was assumed _ven there were approximately 3 years between each medical surveillance program examinations. Akaike's inform, ation criterion (AIC) and Schwarz's Bayesian criterion (SBC) are indices of relative goodness-of-fit that were used to compare models with the same fixed effects, but different covariance structures. SBC penalizes models more severely for the il <<_ ! 3M Company Page12of 63 number of estimated parameters than AIC and thus the two criteria did not always agree on the choice of."oest' model. SBC was preferred. In the present study, the continuous outcomes of lipid and hepatic clinical chemistry tests were evaluated as repeated measures incorporating the random subject effect fitted to a mixed model by the MIXED procedure in the SAS statistical package (Littell et al 1996). Restricted maximum likelihood estimates (REML) of variance parameters were computed. Adjusted regression models were built by introducing all covariates (see below) and testing the covariance structure. Based on goodness-of-fit tests described above, AR+RE, was routinely considered the best covariance structure chosen for the mixed models. Covariates included PFOS (or PFOA or TOF), years of observation, the-interaction term of PFOS and years of observation, age, body mass index (BlVlD, cigarettes smoked per day, alcohol drinks per day, year at first entry and baseline (at first observation) years worked. For hepatic clinical chemistry tests, serum ... triglycerides was also considered a covariate (OIsen et al 2001a). RESULTS Provided in Table 1 are cross-sectional analyses of the study subjects who participated in each of the three years (1994/95, ].997 and 2000) stratified by location. As reported previously in the complete cross-sectional analyses of these programs (Olsen et al 1998: 1999; 2001), Antwerp employees in this longitudinal investigation were younger, had lower BMIs and drank more alcoholic beverages than Decatur employees. They also had consistently lower triglyceride and alkaline phosphatase levels and hi_et HDL and total bilirubin levels. Decatur employees, on average, had serum PFOS levels 3MCompany Page 13of 63 ,< that were higher by approximately 0.5 ppm in each cross-sectional analysis. Similar findings were observed for PFOA except with the 1997 data where the two populations had comparable mean PFOA levels. Provided in the following two tables am the cross-sectional analyses for the three subcohorts by location. Among Antwerp employees (Table 2), each of the three subcohorts had lower mean serum PFOS levels in 2000 than at their year of entry whereas there were no consistent changes across subcohorts with PFOA. Among the three Decatur subcohorts (Table 3) mean PFOS values declined over time but mean PFOA levels tended to increase. Provided in tables 4 through 30 are the mixed model coefficient estimates, standard errors, p-values and 95% confidence intervals from testing potential determinants of lipid and hepatic clinical chemistry change. The natural log was used for ', all depe ndent variables. oo.. Tables 4 through 6 contain the analyses for cholesterol. There was no change in cholesterol associated with PFOS (Table 4). Overall, PFOA was positively associated with cholesterol as the main effect coefficient was significantly positive but its interaction with time (years variable) was negative ('Table S). Provided in Tables 5A through SD are separate analyses for Antwerp for all subjects (Table 5A) and by each subcohort. The PFOA and cholesterol association appeared to primarily reside with the 21 Antwerp employees in subcohor_ A (Table SB). This finding can also be observed in Table 2 as the subcohort's mean PFOA levels went from 1.32 ppm. to 2.37 ppm and then declined to 2.06 ppm at the same time their cholesterol values rose from 208 m_dL to 226 m_dL to 229 m_dL. There were no associations between cholesterol and PFOA observed among \ 3M Company Page 14of 63 the Decatur employee population (Table 5E) nor were there significant associations ,,.' between TOF and Antwerp or Decatur but when the two sites were combined there was a significant positive association between TOF and cholesterol (Tables 6, 6A and 6B). There were no significant associations between PFOS, PFOA or TOF with I-IDL (Tables 7 through 9). BMI, alcoholic drinks per day and cigarettes smoked per day were the most significant associations with HDL. Tri_ycerides were not significantly associated with PFOS over time when both Antwerp and Decatur populations were examined together (Table 10). However, among the combined Antwerp and Decatur populations, PFOA was positively associated with triglycerides (Table 11) as seen with the significant positive coefficient for the main effect of PFOA and the nonsignificant positive main coefficient of years and the negative coefficient for their interaction (PFOA x years). The significant main effect of PFOA /e < was the consequence of the Antwerp population (Table 1IA) and primarily subcohort A (table 11B) and, to a lesser extent subcohort B (Table 11C), but not subeohort C (Table IID). Therefore, the association appeared to be related to the Antwerp workers who were enrolled in this lon_tudinal cohort be#nning in 1995, but not 1997. There was not a significant association between PFOA and triglycerides among Decatur workers (Table 11E). Among the Antwerp subcohort A, their mean triglyceride levels rose from 85 mg/dL to 115 m_dL to 123 mffdL at the same time their PFOA levels increased from 1.32 ppm to 2.3"/ppm ,and then declined to 2.06 ppm. Although the main effect for TOF was si_ificantly positive, the interaction term with time (years) was not significant (Table 12). Again. this association was more consistent for Antwerp employees (Table 12A) than Decatur employees (Table 12B). 3M Company Page15of63 ( Among the hepatic clinical chemistry tests that were adjusted for the various chan_ng demographic factors and triglyceride levels, there were no significant associations between PFOS, PFOA and TOF with changes in alkaline phosphatase (Tables 13 - 15), GGT (Tables 16-18), AST (Tables 19-21), ALT (Tables 22-24), total bilirubin (Tables 25 - 27) or direct bilirubin (Tables 28-30). Observations apparent in Tables 2 and 3 can also be seen in these mixed model analyses. For example, the two most significant predictors of alkaline phosphatase were time (years) and location (as seen with the lower values among Antwerp employees). For ALT, entry period was also significant as it reflected the higher reference range values for ALT that were used in 1994/95 than in subsequent years. ,. DISCUSSION : These analyses were the first lon_tudinal assessment of the fluor0chemical . medical surveillance program at 3M's Antwerp and Decatur manufacturing sites. Overall. we observed no associations that were consistent with the toxicological evidence" that PFOS produces a hypolipidemic effect at threshold dosages in rats and primates (3M Company 2000; Haughom and Spydevold 1992: Ikeda et al 1987; Pastoor et al 1987; Seacat et al 2001a: 2001b: Sohlenius et al 1993). Our results did suggest a positive association between temporalchanges in cholesterol and triglycerides and PFOA; however this is also inconsistent with the toxicolo_cal evidence that PFOA may result in a hypolipidemic effect in rats (Haughom and Spydevold 1992: Pastoor et al 1987) but produced no effect on blood lipide in primates (Butenhoff et al 2001). < 3MCompany Page16of 63 (. Even though we were able to perform a longitudinal assessment, there were \ several limitations to our analyses. We were limited to I75 employees of which only 41 (24 percent) participated in all three surveillance years. Although a geater absolute number of Decatur employees (but not percent-wise) have participated during each year, for this longitudinal assessment there were more Antwerp (57 percent) than Decatur (43 percent) employees. Antwerp employees have had lower serum PFOS level by approximately 0.5 ppm (Olsen et al 1.998;Olsen et al 1999a; 1999b; 2001a; 2001b; 2001c). There were insufficient numbers of female employees for any meaningful longitudinal analysis. Given the variability inherent in the analytical method (Hansen et al 2001) and the different laboratories used, serum PFOS and PFOA levels may have systematic error incorporated in each measurement that we were unable to assess as blood samples were analyzed only at the time of the surveillance program. This i .+ systematic error may have masked associations with lipid or hepaticclinical chemistries, .-, although the range of PFOS and PFOA measured was relatively consistent throughout the study time period. Because 3M has announced a phase-out of the production of perfluorooctanyl chemistry-related materials, we doubt that there will be many more subjects in the future that can be included in this lon#tudinal assessment. Also, the finding from this assessment would suggest that serum PFOS levels have either remained constant or declined slightly over time among these 175 employees. On the other hand, serum PFOA levels appeared to trend upwards, on average, by approximately 0.5 to 1.0 ppm for these employees. Another limitation is the fact that the serum PFOS and PFOA levels measured in these employees were lower than those that cause effects in laboratory, animals. 3M Company Page 17of 63 f In summary, a longitudinal analysis over a six year time period of 175 Antwerp and Decatur male employees did not show significant changes, consistent with toxicological data, of lipid or hepatic clinical chemistry values associated with either PFOS or PFOA. The PFOS and PFOA serum levels measured in these employees were lower than those that cause effects in laboratory animals. / / \ REFERENCES 3M Company Page 18 of 63 3M Company (2000). SIDS Initial assessment report: Perfluom_:tane sulfonic acid and -. its salts. St. Paul (_-_g):3M Company, (unpublished report). Minneapolis (MN):University of Minnesota, (unpublished report). AndersonDJ, Mulvana DE (1997a). Analytical Report for the Determination of Perfluorooctanoate and Perfluomoctanesulfonate in Human Serum by LC/MS. Ithaca NY:Advanced Bioanalytical Services Inc; August 1997. Anderson DJ, Mulvana DE (1997b). Analytical Report for the Determination of Perfluorooctanoate and Perfluorooctanesulfonate in Human Serum by LC/MS. Ithaca N'Y:Advanced Bioanalytical Services Inc; September 1997. Butenhoff JL, Costa G, Elcombe C, FarrarD, Hansen K, Iwai H, Jung R, Kennedy G, Lieder P, Olsen GW, Thomford P. Toxicity of ammonium perfluorooctanoate (APFO) in cynomol_s monkeys after 26 weeks of oral dosing. (unpublished report) Hansen K.I,Clemen LA, Ellefson ME, Johnson .IHO (2001). Compound-specific, quantitative characterization of organic fluorochemicals in biological matrices. Environ Sci Technol 35:766-770. Haughom B, Spydevold O (1992). The mechanism underlying the hypolipcmic effect of < perfluoroctanoic acid (PFOA), pcrfluoroctanesulphonic acid (PFOSA) and clofibric acid. ! Biochemica et Biophysica Acta 1128:65-72. Ikeda T. Fukuda K, Mori I, Enomoto M, Komai T, Suga T (1987). Induction of cytochrome P.450 and peroxisome proliferation in rat liver by perfluorinated octanesulfonic acid. In: Pert):osmes in Biology.and Medicine. (HI) Fahmi an(:/HSies, eds) New York:Springer Verlag, pp 304-308. Johnson JD, Wolter.lT, Colaizy GE, Rethwill PA, Nelson RM. Quantification of Perfluorooctanoate and Perfluorooctanesulfonate in Human Serum Using Ion-Pair Extraction and High Performance Liquid Chromatography-Thermospray Mass Spectrometry with Automated Sample Preparation. St. Paul, IVDr,:3M Env Lab, 1996. Litteil RC. Pendergast J, Natarajan R (2000). Modeling covariance structure in the analysis of'repeated measures data. Star Med 19:1793-1819. Littell RC. Milliken G_,, Stroup WW, Wolfinger RD. SAS System for ,.MixedModels. Car,/NC:SAS Institute Inc.. 1996, 633 pp. Olscn GW. Burris .I'M,Mandel JI--IZ. obel LR (1998). An epidemiolo_c investigation of clinical chemistries, hematology and hormones in relation to serum levels of perfluorooctane sulfonate in male fluorochemical production employees. St. Pauh3M Company (unpublished report). i 3MCompany Page19of 63 i ' Olsen GW, Bun-isJM, Mandel JH, Zobel LR (1999a). Serum perfluorooctane sulfonate and hepatic and lipid clinical chemistry tests in fluorochemical production employees. JOEM 41(9):799-806. Olsen GW, Logan PW, Simpson CA, Hansen KJ, Burris JM, Burlew MM, Schumpert IC, Mandel JI-I(1999b). Fluorochemical exposure assessment of Decatur chemical and film plant employees. St. Paul:3M Company (unpublished report). Olsen GW, Burris JM, Burlew MM, Mandel JH (2000). Plasma cholecystokinn and hepatic enzymes, cholesterol and lipoproteins in ammonium perfluorooctanoate production workers. Drug Chem Toxicol 23(4):603-620. Olsen GW, Burlew MM, Burris JM, Mandel J'H (2001a). A cross-sectional analysis of serum perfluorooctansulfonate (PFOS) and perlfuorooctanoate (PFOA) in relation to clinical chemistry, thyroid hormone, hematology and urinalysis results from male and female employee participants of the 2000 Antwerp and Decatur fluorochemical medical surveillance program. St" Paul (MN):3M Company (unpublished report). Olsen GW, Schmickler _. Tierens JM, Logan PW, Burris J'M, Burlew MM, Lundberg IK, Mandel JH (2001b). Descriptive summary of serum fluorochemical levels among employee participants of the year 2000 Antwerp fluorochemical medical surveillance program. St. Paul:3M Company (unpublished report). c i Olsen GW, Logan PW, Simpson CA, Bums J'M, Burlew 19_1L, undberg J'K,Mandal JH ..... -(2_00Ic).Descriptive summary of serum fluorochemical levels among employee participants of the year 2000 Decatur fluorochemical medical surveillance program. St Paul:3M Company (unpublished report). Pastoor TP, Lee KP, Peril _L_., Gillies PJ (1987). Biochemical and morpholo_cal studies of ammonium perfluorooctanoate-induced hepatomegaly and peroxisome proliferation. Exp Mol Pathol 47:98-109. Seacat AM. Thomford PI, Hansen IO, Clemen I.A, Case MT, Butenhoff JL (2001a). Sub-chronic dietary toxicity of potassium perfluorooctanesulfonic acid in rats. Toxicol Sci (submitted 2001a). Seacat AM. Thoford PJ, Hansen KI, Olsen GW, Case MT. Butenhoff JL. Subchronic toxicity studies on peffl,uorooctanesulfonate potassium salt in cynomolgus monkeys. Toxicol Sci (submitted, 2001b). Sohlenius AK, Eriksson AM, Hogstrom C, Kimland M, DcPierre JW (1993). Perfluorooctane sulfonic acid is a potent inducer of peroxisomal fatty acid B-oxidation and other activities known to be affected by peroxisome proliferators in mouse liver. Pharmacol Toxicoi 72:90-93. \ I'F()S 'l'ahlc I. ('lti.%s-_eetional Analysis ()1 Me;Ill and ,%.1....,ard I)eviali(m of _t:rtt111I)FOS, I'rOA, "r()F, DcnmBruphic Characlorislics and (linical ('helnislries of Anlwer I) and Deculur Male l".mldoyces Who ll:tl-licipaled hi Two or More Medical _Ul'Voill:ll|CerC_xllnlill;l|iOllSI}clwecn 1994195 and 2(XX) 199411995 1997 20(X) ill (=3 Ai!lwerp(N = 66) Mean SIt 1.87 1.96 ..D.e_c.a_It=l4r!_)()N._ . An!w!rp.(N- 5.5).....])t[a_l(.Nu=r _.L .... .A_)II_W_Pe.r_.(N-._I._.().}_.... ]_ec;.i(.NIL=_r75.).. Mean SIt Mean SIt Mean Sit Mean SIt Mean SIt 2.62' 1.78 1.42 1.26 1.85 1.64 I.16 1.07 1.67" 1.39 I'F()A 1.08 1.53 1.90h I.OX 1.54 1.61 1.4! !. 17 1.43 1.21 1.83" ! .53 T()F 1.06 1.77 3.(X)" 1.77 1.98 1.4X 2.17 1.76 1.74 -I.24 2.3,1h 1.72 ABe 3(_ 6.h 43d 6.0 32 6.8 43d 7.3 38 7.8 47d 7.0 IIMI 23:} 2.,I 28.0d 3.7 23.2 2.4 "...). 9Id 4.3 24.7 ")I_' 29.0d 4. I A Icoh(ll I. I'l I. I Baseline Years I 1.0 5.7 Worked ('igarelles 4 7 O.3 It.f} 20.4 a 7. I !()=' 15 ().9'l 1.0 7.0 5.2 5 7 O.I O.I 20.()'t 6. ! 5 !0 I. I d 1.0 9. I O. I 5 8 O.I 0.2 2().3' 8.5 5 II ('holesler(d 217 43 2!9 38 202 45 214 35 220 41 213 40 HDL 54" 13 43 13 49" Ii 43 10 53 '_ 12 44 10 Triglycerides III 79 204 d 122 108 53 181d 112 131 80 178_" 115 AIk Plms 72 18 103a 27 68 15 87 d 20 58 14 74 d 2I (i(.H" 37 25 47 24 23 II 33" 27 25 19 29 18 AST 25 II 3 !' 11 26 6 26 8 24 7 25 7 ALT 44 i7 49 25 30 12 32 15 23 i! 32 d !4 "rolal llilirubin (1.9a 0.4 0.5 11.2 0.8 i' 0.4 0.6 0.2 I .()" 0.3 0.8 0.2 Direct Riliruhi,_ 0.2 11.4 0.2 {I.(H 0. I h 0.(!7 0.1 0.04 0. I b 0.05 O. I 0.06 a. p < .(15 b. p < .(11 c. p < .(XII d. p < .O(X)I COmlmrcd In year-specific analyses of other manufacturing site. PF()S PF(IA ' "l'_d_Ic2. ('ioss-Seclioual Ail,ilysi._,d Meal! aild Standard l)cvi;ilioll of Serum I'F()S, I)FOA, T()I'. l )cnlogr;iphi(: ('hal';It'lel'iSlics al|d ('lillic;ll ('h11|islries (fl"Three _ui)gl'Olll'l,_ O1"Anlworp Male l],.l)h_yees (A, B alld (') Who l):li'licipated in Two or Mo,'e Medic.l Surveillance Ex:m)iimlioils Bclw_n 1995 and 2(XX) 1995 1997 2000 l-_ig_ ,i).ui). ul()3 .A..(_.N.=2.1) .... Mean SI) I1(N=45) Mean SD A(N=21) Mean SI) C(N = 34) Mean SD A(N=21) Mean SD B (N=45) Mean SD C(N .= 34) Mean SD 2.19 1.27 1.72 2.20 2.24' 1.32 0.91 0.93 1.53. 0.87 1.211 1.31 0.87 t' 11.7I 1.3"-) 1.28 11.96 1.64 "_).37 a "_).'_18 1.0,1 ; 11.61 2.06 a 1.04 1.17 1.37 1.38 I).9,1 T()F 2.33 1.211 1.78 1.98 3.08:' 1.59 1.31 11.91 2.41 a 0.81 1.58 1.51 1.52 0.89 Age 33 6 37" 7 35" 5 30 7 38" 6 42' 7 32:' 7 I Y IIMI 23.4 2.7 24. I 2.3 23.8 2.8 22.7 2.0 24.3 3.4 25.5" 2.7 23.9 2.4 I )rinks/tl_ly 1.3 1.4 1.0 0.9 1.2 1.2 0.7 0.8 1.7 e 1.3 1.2 i. I 11.8 0.6 ('igarelles/day 4 7 Baseline Years 9.9 4.3 Worked 4 8 11.6 63 5 7 9.9 4.3 5 7 5.3 4.9 6 8 4 8 5 8 9.9 4.3 11.6 6.3 5.3 4.9 Cholesterol 208 46 220 41 226' 50 187 34 229 46 233" 38 196 3I HDL 56 II 53 13 51 9 4g II 56 12 52 !1 52 13 Triglycerides 85 49 AIk I'lms 69 211 G( IT 30 9 123 " 88 73 16 41 29 115 69 67 19 25 10 104 41 69 12 22 II 123 65 154 95 105 55 55 17 (d) 14 58 12 23 13 3If 25 19 I1 AST 25 5 25 13 26 5 26 7 22 5 25 9 23 6 AI.T 42 8 46 20 31 12 30 13 22 II 25 13 21 7 Totnl Ililiruhi. I.(1 11.3 0.8 0.4 l)irecl lliliruhilt 11.2 0.1_I 0.2 0.04 0.8 0.3 0.8 ().4 0. I 0.05 0.2 0.08 1.0 0.3 1.0 0.3 0.9 0.3 O.I 0.03 0. I 0.06 O.I 0.06 .... / II_ ,h..., ,.,h.,,- ,.,,.,,,_.,,;,.,,,,e_ ;,, ,,.,.,r K n ._ IlK lily,ill c_s't_sssA_ r s__" iJ_ Ih:ln {_rnlsn lr_ /F_"_ L !' PF( ).'4 _ ,, 'l':,ble 3. ('ross-Scctio,aaAl nalysisof Mean and,Slanth,rdI)vialion of/_eruln l)l_O,";,I'FOA, T()i', I)emographicCharacteristicsandClinicalChenlistricsof Three Subgroupsof DecaturMale Emph)yees(A, B and C) Who Participatedin Twoor More Medical St,rveillanceExaminationsl]etween 1994and 2(IX) ............................ 1994 : 1997 i 2000 ---_ \ ,n,lj:m) i';,_ -2 ,_16"3 ___A(_._N.=_21...).).. Mean SI) 2.(17 I.(,7 II (N : 2111 Mean SD 3.17' 1.76 A (N = 2()) _ Mean SD 1.93 1.76 C (N = 35) Mean SD 1.80 1.59 A (N : 20) Mean SD 1.78 2.14 B (N = 20) Mean SD 1.84 1.03 C (N : 35) Mean SD 1.51 0.98 I'FOA 1.50 I).87 2.30= !. 14 1.41 I. 16 1.41 1.20 1.46 1.34 2.60" 2.02 1.60 I. 14 TOF 2.37 1.5-1 3.64" !.79 2.22 1.78 2.14 I."/8 2.16 2.16 2.98 1.76 2.118 1.32 Age 42 7 44 5 45 7 42 8 48 7 50 5 /-15 8 IIM! 28.2 4.11 27.7 3.3 28.3 4. I 29.6 4.5 28.9 4. I 28.7 3.3 29.3 4.5 I)rinks/day (1.I 11.3 11.4 (1.8 11I. I). I 1).I (1.2 0.(1 0. I (1.2 (1.3 11I. 0.2 ('igaretles/day 6 13 Ilaseline Years 19.6 8.6 Worked 14 15 21.3 5.2 ,1 10 19.6 8.6 5 II 211.2 I().0 3 8 10 15 4 9 19.6 8.6 21.3 5.2 211.2 10.0 Cholesterol 235= 34 2(14 36 223 31 2t18 37 228 39 199I' 33 212 41 HDL 47_ 16 39 9 45 I! 41 9 47 12 38a 6 45 9 Triglycerides 181) 88 229 148 188 ! 15 178 112 193 124 187 76 165 128 AIk I'hos 98 29 1117 24 85 22 87 18 70 27 82 17 72 20 (;(IT 41 25 52 22 35 36 31 21) 29 16 34 23 27 !7 AST 32 15 29 6 25 6 27 8 26 8 25 4 25 7 ALl" 52 33 47 11 30 15 34 - 16 30 18 32 !1 33 14 Total llilirt,bi,i 0.6 0.2 0.5 Direct I_ilirubin (I.2_ 0.115 ............................................. , n ( 11_ li,.,. ,ohl.," t',_mn_irlt'nnlc_ 0.2 ;n t..:,r 0.2 11.6 0.2 (1.6 0.2 0.8 0.04 0.I 0.04 0. I (1.05 0. I It n c" ll(; I|i:i,I orc_iirt A r. ,_ " (I_ lh:m ornltrt (_ 0.2 0.7 0.08 0. I 0.2c 0.8 ().'_ 0.04 11.1 0.06 ICompany l'ag2e3of63 Table4. Mixed M(alelCoel'ficile.n,l.'slJmSatlaensd,ardErrors(SE),IS-Valueasnd95% ConfidenceLimits fiom"l'csliPnogtentiaDletenninanlosfCholesteroClhange IncludinPgFOS and theInteractiwointh Number ofYearsofObservatioonfAntwerpand DecaturMale Employees " 95% ConfidenceLimits .....................C.:._._.l_.-t.li..i..ac_m_e) Intercept 4.868 PFOS 0.(110 Years Observation O.(XX)9 PFOS x Years Obs - 0.0004 Age 0.007 BMI 0.1)06 Drinks/day Cigarettes/day l._ation* 0.014 - 0.0009 0.034 SI". O.127 0.()08 0.0(}5 0.01}2 0.01}3 0.()03 0.012 0.001 y (I.037 p-wtluc <.0001 .18 .84 .83 .01 .07 .27 .41 .36 Lower 4.618 - 0.005 - 0.008 - 0.004 0.002 -0.0005 - 0.011 - 0.003 - 0.039 Upper 5.I18 0.025 0.010 0.003 0.013 0.013 0.038 0.00 ! O. 108 Entry Perit)tl** Baseline Years Worked 0.064 - 0.004 0.028 0.003 . .02 .20 0.009 - 0.009 0. 119 0.002 #natural log *Antwerp vs Decatur ** 1994195 vs 1997 Co,npa,w Page24 of 63 Table 5. Mixed Model Coefficie,t I;'sthnales, Standard Errors (SE), P-Values and 95% Confidence Limils from Testing Potential Determinants of Cholesterol Change Including PFOA and the Interaction with Number of Years of Observalion of Antwerp and Decatur Male Employees 95% Confidence Limits ........................ lnlercepl PFOA YeaJ_ Oi)scrvalion PFOA x Years Obs C.',__ffic__i_e.t 4.812 11.(132 11.(X15 - 0.0115 SF. 0.127 O.IXF) : 0.005 11.0112 p-wdue <' .000 ! .01X)8 t .24 .005 Lower 4.56 ! 0.013 - 0.0(H - 0.009 Upper 5.063 0.05 I 0.014 - 0.002 Age BMI Drinks/day Cigareltes/day Location* Entry Peritxi** Baseline Years Worked 0.008 0.1107 0.014 - O.(X)I I).041 0.068 - 0.004 0.003 0.003 0.012 O.(X)I 11.037 0.027 0.003 .007 0.002 0.013 .049 0.00004 0.013 .263 - 0.010 0.037 .32 - 0.003 .27 - 0.032 0.001 O.114 .01 0.015 .15 - 0.009 0.122 0.001 #natural log *Antwerp vs Decatur *'!994/95 vs 1997 r-,. Table 5A. Mixed Model Coefficient Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits from "l'csting Potential Dctermina,_ts of Cholesterol mChange Including PFOA and Iho Interaction with Number of Years of Observation of Antwerp Male Employ_s 95% Confidence Limits Compa,ly Page25 of 63 Intercept PFOA Ct_flicient 4.710 0.029 S1 O.149 0.012 p-value < .0001 .01 Lower 4.414 0.006 Upper 5.005 0.053 Years Observation PFOA x Yeal's Obs 0.(X)5 - 0.(X)3 0.006 0.(X)3 .36 - 0.006 .20 - 0.009 0.017 0.002 Age 0.009 0.003 .008 0.003, 0.016 BM[ 0.008 O.(XI5 .14 - 0.003 0.019 Drinks/day 0.022 0.013 .09 - 0.004 0.047 Cigarettes/day Entry Peri_al** Baseline Years Worked 0.{XX)7 0.0.79 - O.{X)2 0.(X}2 02038 0.004 .70 - 0.003 .04 0.004 .67 - 0.010 0.004 0.1"53 0.007 Wnaturallog **1994195 vs 1997 Table 5B. Mixed Model Coerlicient Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits fi'om Testing PotentialDeterminants or CholesterolmChange Including PFOA and the Interaction with Number of Years of Observation of Anlwcrp Subgroup A (I 995, 1997 and 2000) Male Employees : 95% Confidence Limits ? .. Company Page:26of 63 Coefficient SE p-value Lower Upper lntcrcq)t PFOA Years Ohservation 5.059 0.1144 0.1)37 ().4I,I 0.1)2(| 0.015 < .(RIOI .03 .02 4.194 0.0{14 0.007 5.925 0.084 .. 0.067 PFOA x Years Obs - 0.013 11.(105 .02 - 0.023 - 0.002 Age - 0.0007 0.012 .95 - 0.025 0.024 BM! 1).004 0.012 .77 - 0.020 0.028 Drinks/day - 0.016 0.019 i .41 - 0.054 0.023 Cigarettes/day (i.(103 (}.(Xl5 .5 i - 0.007 0.013 Baseline Years Worked 0.015 0.0 ! 6 .36 - 0.018 0.047 _naturallog :" : ..... 4, "_'-"x Table 5(;. Mixed Model Coefficient I_,titnates,Spmdard EtTors (SEt, P-Values and 95% Confidence Limits from Testing Potential Determinants of Cholesterol t Change Including PFOA and the Interaction with Number of Years of Observation of Antwerp Subgroup B (1995 and 2000) Male Employees 95% Confidence Limits . .4 Company Page27 of 63 Intercept PFOA Years Observation PFOA x Years Obs Age BMI Coefficient 4.838 0.018 0.005 - 0.002 0.004 0.014 SE 0.244 0.017 0.008 0.004 0.(X)5 O.(X)9 p-value < 0.001 .30 .55 .58 .40 .13 Lower 4.346 - 0.016 - 0.0 ! I - 0.009 - 0.006 -0.004 Upper 5.330 0.052 0.021 0.005 0.015 0.031 Drinks/day Cigarettc_day Baseline Ye;trs Worked 'natural log 0.017 0.002 - 0.0001 0.022 0.003 0.006 .45 -0.028 0.061 .44 - 0.004 0.009 .99 - 0.011 0.01 I Table 5D. Mixed Model Coefficient l-:slimates, Standard Errors (SE), P-Values and 95% Confidence Limits from r,iestm g | Potential Determinants of Cholesterol' Change Including PPOA and the Interaction with Number of Years of Observalion of Antwerp Subgroup C (1997 and 2000) Male Employees 95% Confidence Limits .,,C_ompa,ty Page28 of 63 Intercept PFOA Years Observation PFOA x Years Obs C,_cl'l'icient 4.528 0.004 - 0.010 0.010 SE 0.281 0.062 !. 0.026 0.022 p-valtte < .000 ! .95 .70 .65 Lower 3.955 - 0.125 - 0.065 - 0.035 Upper 5.100 0.132 0.044 0.055 Age 0.016 0.006 .01 0.004 0.028 BMI 0.009 0.012 .43 - 0.015 0.033 Drinks/day Cigarelles/day Baseline Years Work mnaturalog 0.033 - 0.0001 - 0.005 0.032 0.(X}3 0.(108 .31 .97 .49 a - 0.033 - 0.007 - 0.021 0.100 0.006 0.011 MCompany Page29 of b3 Table 5E. Mixed Model Coefficient Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits from Testing Potential Dcterminams of Cholesterol mChange Including PFOA and the Interaction with Number of Ye,us of Observation of Decatur Male Employe,cs 95% Confidence Limits Interccpl PFOA Years Ohscrvalit)n PFOA x Years Obs Age BMI Drinks/day Cigarettes/day Entry Period** Baseline Years Worked C_cl'l]cienl 5.220 " 0.016 - 0.002 - 0.003 0.002 0.003 - 0.088 - 0.002 0.047 - 0.002 S[': 0.206 0.016 0.(X)8 0.003 0.005 0.004 0.035 0.001 0.040 0.004 p-value < .0001 .34 i .77 .22 .70 .55 .01 .i5 .24 .62 Lower 4.810 - 0.017 - 0.017 - 0.009 - 0.008 - 0.006 - 0.158 - 0.004 - 0.032 - 0.010 Upper 5.630 0.048 0.013 0.002 0.012 0.011 - 0.0 !8 0.0007 0.125 0.006 Snatural log **1994195 vs 1997 .. Company Page 30of 63 Table 6. Mixed Model Coefficien! Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits from Testing Potential Determinants of Cholesterol j Change Including TOP and the Interaction with Number of Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits Intercept TOP Ctx;fficient 4.828 0.021 SI.;. O. 127 O.(X)8 p-value < .0001 .(X)7 Lower 4.577 0.006 Upper 5.079 0.035 Years Ohservalion TOP x Years Obs O.(g)4 # - 0.003 O.(X;5 .37 0.001 : .07 - 0.005 - 0.005 0.014 0.0003 Age BMI 0.007 0.006 0.003 0.003 .01 0.002 0.013 .05 - 0.0001 0.013 Drinks/day Cigarettes/day Localion* 0.012 - 0.001 0.042 0.012 0.001 0.037 .32 - 0.012 .37 - 0.003 .26 - 0.032 0.036 0.001 0.115 Entry Pli{Ml** Baseline Years Worked 0.063 - 0.004 0.027 0.(X_3 .02 0.010 .17 - 0.009 O. ! 17 0.002 tnatural log *Antwerp vs Decatur .. **1994195 vs 1997 "l'ablc 6A. Mixed Model Coefficient I.'_timatcs,StandardErrors (SE), P-Values and 95% Confidence Limits from Tcsling Potential Determinants of CholesterolmChange Including TOF and the Interaction with Numberof Years of Observationof Antwerp Male Employees 95% Confidence Limits ...... I ..... J _c 31 ul"63 Intercept TOF Years Observation TOF x YearsObs Coefficient 4.721 0.017 0.1)04 - 0.0008 SIS 0.150 0.011 0.007 0.002 p-value < .4)001 .12 .55 .70 Lower 4.424 - 0.004 - 0.009 - 0.005 Upper 5.018 0.038 0.017 0.003 Age 0.009 0.003 .01 0.002 0.016 BMI 0.009 0.006 .12 - 0.002 0.020 Drinks/day Cigarettes/day Entry Period** Baseline Years Worked 0.019 0.0008 0.070 - 0.0008 0.013 .14 0.002 ' .69 0.038 .07 0.004 .85 - 0.006 - 0.003 - 0.006 - 0.009 0.045 0.005 0.145 0.008 #natul_l log **1994195 vs 1997 /, _.,I Coml)any Page32of 63 " Table 61). Mixed Model Coefficient l_limales, Standard Errors(SE), P-Values and 95% Confidence Limits from TestingPotenlial Determinants of CholesterolmChange Including TOP and the Interactionwith Number of Years of Observationof Decatur Male Employees 95% Confidence Limits Intercept TOP Years Observation TOIz x Years Obs Age BM! C'.oel'l]cient 5.214 0.014 - 0.002 - 0.002 0.002 0.002 SI_ 0.202 0.011 0.008 0.002 0.005 0.004 : p-value < .0001 .19 39 .22 .69 .55 Lower 4.81 ! - 0.007 - 0.018 - 0.006 - 0.008 - 0.006 UDDer 5.616 0.035 0.013 0.001 0.012 0.01 I Drinks/day Cigarettes/day Baseline Years Worked - 0.090 - 0.002 - 0.002 0.035 0.001 0.004 .012 - 0.160 - 0.020 .16 ! - 0.004 .60 - 0.010 0.0007 0.006 mnaturallog *'1994/95 vs 1997 t Table 7. Mixed Model Coefficient Estimates, Standard EiTors (SE), P-Values and 95% Confidence Limits from Testing Potential Determinants of I-IDLmChange Including PFOS and the Interaction with Nuudlcr of Y;u's of Obscrvalitm of Antwerp and Decatur Mal Employees 95% Confidence Limits ,, Cuml_aay Page33 of 63 Intercept PFOS Years Observalioq PFOS x Yca,'s Obs Age BM ! Coefficient 4.212 " - O.I)10 0,002 - 0.001 0.004 - 0.0 i 7 SI'_ : 0.144 0.IX)8 O.(X)5 0.002 0.003 0.(XH p-wdue g <'.0001 .24 .71 .52 .22 < .0001 Lower 3.930 - 0.026 - 0.008 - 0.005 - 0.002 - 0.025 Upper 4.495 0.007 0.012 0.002 0.011 0.010 Drinks/day Cigarettes/day Location* 0.064 - 0.(XH 0.006 0.013 0.001 , 0 0.0,13 < .0001 .0(KI5 .89 0.037 - 0.007 - 0.079 0.090 - 0.002 0.091 Entry Period** Baseline Years Worked 0.025 - 0.007 0.032 0.003 , .44 .03 - 0.038 - 0.013 0.088 - 0.0005 'natural log *Antweq) vs Decatur ' ** 1994195 vs 1997 Company Page 34of 03 Table 8. Mixed Model Coefficienl Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits from Testing Potential Determinants of HDL j Change Including PFOA and the Interaction with Number of Years of Ohservalion of Antwerp and Decatur Male Employees 95% Confidence Limits ...............Coefficient Inlercepl 4.216 PI:OA "-().{WK| Years Ohscrv|ttion 0.(X)5 PFOA x Years Ohs - 0.iX)2 Age 0.004 BMI -0.017 SI_. 0.1,15 tl.I(IJ 0.(X)5 0.(X)2 0.003 0.004 p-wilue < .0001 .50 I .35 .40 .28 < .0001 Lower 3.929 -0.027 - 0.005 - 0.006 -0.003 -0.024 Upper 4.503 0.015 0.014 0.002 0.0I0 -0.010 Drinks/d:ty Cigarcltc.s/day Location* O.(X_2 - 0.(XH 0.(X)8 ILl)13 (l.(X)l !. 0.043 < .0001 .(XXH .85 0.036 - 0.007 - 0.076 0.088 - 0.(X)2 0.093 Entry Pe|'itxl** Baseline Years Worked 0.020 - 0.007 0.032 0.003 .53 - 0.042 0.082 .04 - 0.013 - 0.0004 Wnaturallog *Antwerp vs l:)ecatur ** 1994/95 vs 1997 Table 9. Mixed Model Cocfficienl Estimates.StandardErrors (SE), P-Values and 95% Confidence Limits from Testing PotentialDeterminantsof HDL' Change Including TOF and the Interaction with Numher of Years of Ohservationof Antwerp and Decatur Male Employees 9.5% ConfidenceLimits ..Company Pag3:.5of63 Intercept TOP Years Ohservalion TOP x Years Obs Age BMI Ctxd'ficienl 4.217 - 0.006 0.004 - 0.(}0(F) (}.(}04 - 0.017 SL 0.145 0.(}08 : ! 0.(105 O.(X)l 0.0(}3 0.0(H p-value < _0001 .44 I .47 .57 .25 < .0001 Lower 3.931 - 0.023 - 0.007 - 0.004 - 0.003 - 0.025 Uplgr 4.502 0.0 i 0 0.014 0.002 0.010 - 0.010 Drinks/day Cigarelles/day Location* Entry Period** Baseline Years Worked 0.063 - 0.(}04 0.007 0.021 - 0.007 0.013 0.0(}1 0.043 0.032 0.003 < .0001 .0005 .88 .50 .04 0.037 - 0.007 - 0.078 - 0.041 - 0.013 0.089 - 0.002 0.0'92 0.084 - 0.0004 'natural log *Antwerp vs Decatur .' *'1994/95 vs 1997 Table 10. Mixed Model Coefficient Eslimatcs, Standard Errors (SE), P-Values and 95% Confidence Limits fi-om Testing Potential Determinantsof'rriglyceride mChange Including PFOS and the Interactionwith Numl)cr of Ycars of OI}scrvalionof Antwerp and DecaturMale Employees 95% Confidence Limits Company Page 36of 63 Intercept PFOS Years Observation PFOS x YearsObs (:ocfficicnl 2.730 0.025 - 0.(XH 0.006 SI'." 0.335 0.020 0.013 0.005 p-vah,c < .0(XII .22 .73 .22 Lower 2.068 - 0.015 - 0.029 -0.004 Upper 3.392 0.065 0.02 I 0.015 Age 0.003 0.008 .67 -0.011 0.018 BMI 0.006 0.009 < .0001 0.048 0.083 Drinks/day Cigarettes/day Location* -0.l)29 0.(I I ! 0.052 0.0.13 0.003 0.099 .37 .0002 .60 -0.094 0.005 - O.143 0.035 0.017 0.247 Entry Period** Baseline Ye;u's Worked 0.089 0.(X)5 0.()73 .22 0.007 !. .50 - 0.055 - 0.010 0.234 0.019 #natural log *Antwerp vs Decatur ** 1994195 vs 1997 Table I I. Mixed Model Coefficient [L,;tim'des, Standard Errors (SE), P-Values and 95% Confidence Limits from Tesling Potential Determinants of Triglyceride' Change Including PFOA and the Interaction with Nmnhcr of Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits LUllil)ally .-age 37 of 63 Intercept PFOA Years Observalioil PFOA x Years Obs Ct_l'ficienl 2.539 0.(FM 0.(X)7 - 0.008 St, 0.332 0.025 0.012 0.005 p-value <!.0001 .0002 .57 .12 Lower 1.883 0.045 - 0.017 - 0.018 Up_r 3.196 0.144 0.03 I 0.002 Age BMI Drinks/day Cigaretles/day Location* Entry Period** Baseline Years Worked 0.006 0.[)66 - 0.027 0.011 0.072 0.098 0.003 ,0.007 : 0.009 0.032 0.003 " 0.096 0.(170 0.007 .42 < .0001 .40 .0002 .46 .17 .67 - 0.009 0.049 - 0.090 0.005 - 0.118 - 0.04 1 - 0.011 0.021 0.083 0.037 0.017 0.262 0.236 0.017 inlilttrltl log *AIII,'PIII I It:,:lllllr tllJt)4/Osw lY Jl . Table I I A. Mixed Model CocMcien! l;.slimales,!.Standard Errors (SE), P-Values and 95% Confidence Limits from Testing Polential Determinanls of'l'riglyceride mChange Including PFOA and the Interaction with Number of Ye.'a_ of Observation of Antwerp Male Employees 95% Confidence Limits _,_ Company Page38 of 63 Intercept ._ PPOA Years Ohservalion PFOA x Years Obs Age BM! Drinks/day Cigarette:,dday Entry Period** Baseline Years Worked tnatural log *'1994/95 vs 1997 Coefficient 3.(X)7 0.089 0.024 - 0.0 !0 0.1)12 ,0.039 - 0.026 0.018 0.013 0.003 SI'.' 0.394 0.030 0.015 0.007 0.(X)9 0.014 0.033 0.005 0.100 0.011 p-value <.0001 .005 .12 .15 .22 .007 .44 .0004 .90 .76 Lower 2.286 0.028 - 0.006 - 0.023 - 0.007 0.011 - 0.092 0.008 - 0.212 - 0.019 Upper 3.848 0.149 0.053 0.004 0.030 0.068 0.040 0.028 0.186 0.026 .,MCompany : Page39of63 TableIIB. Mixed Mc_JelCoeMcienlF.stimaleSst,andardEn'or,(qSE),P-Valuesand95% ConfidenceLimits fromTestingPolentiall_terminanlosf'l'dglyceliCdhea'nge IncludinPgFOA andtheInteractiwointh Number ofYearsofObservaliooafAntwclpSubgroupA (1995,1997and2000)Male Employees 95% ConfidenceLimits ..............._._.f._...'.cl._..i_.n..t_.e(I_ Intcrcepl PFOA 3.104 0.182 YearsObserwilio,1 0.167 PFOA x YearsObs -0.061 SF. 1.063 0.051 0.037 0.011 p-value .009 .001 < .0001 < .0001 Lower 0.g80 0.079 0.091 -0.084 Upper 5.328 0.286 0.241 -0.039 Age BMI Drinks/day Cigareltedday Baseline Years Worked mnaturallog - {).01)7 0.039 - 0.082 0.008 0.036 {}.031 (}.030 0.051 0.0 i 2 0.04 1 .83 .20 .i2 .52 .40 o - 0.069 - 0.021 - 0.186 - 0.017 - 0.049 0.056 O.100 0.023 0.033 0.120 / , ,,, Tal)lc I I C. Mixed Model Coefficient I_,stimate_S, tandardErrors (SE). P-Values and 95% Confidence Limits fronl Tcsling Polcnlial Dcterminamsof "rriglyceride j Change Including PFOA and the Interaction with Number of Yearsof Observationof Antwerp Subgroup B 0995 and 2000) Male Employees 95% Confidence Limits .! Compa,y P-'.g4c0of b3 .............................. (_pcl'ficicnt Intercepl 2.775 PFOA 0,(_)7 Years OI)scrvalion PFOA x Years Obs 0.005 0.0008 Age 0.019 BM! O.(H2 Drinks/day Cigmcllc_day Baseline Years Worked 0.038 0.022 - 0.002 "natural log SE 0.523 0.038 J! 1).018 0.(}09 0.011 0.019 y 0.(}._0 O.(X)7 0.012 p-value < .0001 .02 i .80 .93 10 .04 .45 .(X)3 .83 Lower 1.720 0.019 - 0.033 - 0.017 - 0.004 0.003 - 0.062 0.008 - 0.026 Upper 3.830 0.174 0.042 0.018 0.04 1 0.081 0.138 0.036 0.02 I , 'rablc IID. Mixed Model Coefficicn! l,_timatesS, tandardEn'ors(SE), P-Values and 95% Confidence Limils from 'TestingPotentialDetenninants of Triglyceride' Change Including PFOA and the |ntcraction with Numl_r of Yearsof Observationof Antwerp SubgroupC (1997 and 2000) Male Employees 95% Confidence Limits M Company Page41 of 63 Intercept PFOA ........... Years Ohscrvation PFOA x Years Ohs .Ctmfficient 3.285 0.032 - 0.072 0.020 Age - 0.(M)5 BMI 0.057 Drinks/day Cigarclles/day Baseline Years Worked - 0.025 0.019 0.010 _natural log __.. Si] 0.744 (}.149 0.063 0.051 0.016 0.030 0.081 0.(}08 (}.021 p-value .0001 .83 .26 .70 .77 .08 .76 .032 .65 Lower i .769 - 0.280 - 0.203 - 0.087 - 0.038 - 0.006 0.195 0.002 -0.034 Upper 4.80 I 0.344 0.058 0.128 0,028 0,120 0,145 0,036 0.053 ? r ,, .d Company Page42 of 63 "Fable I IE. Mixed Mtulel Coefficient Eslimates, Standard Errors (SE), P-Values and 95% Confidence Limits fi'om Testing Potential Determinants of Triglyceride mChange Including PFOA and the Interaction with Number of Years of Observation of Decalur Male Employees 95% Confidence Limits ............ lnlercepl PFOA Years Observation PFOA x Years Obs Age BM ! Drinks/day Cigarettes/day Entry Period** Baseline Years Worked Ct_el'l]cienl 2.581 0.054 - 0,028 0.002 - 0.0008 0.073 - 0.038 0.t)05 0.274 0.009 Si_, 0.567 0.(H6 0.022 O.(KI8 0.013 0.012 : 0.099 0.004 0.109 0.011 p-vahte < .0001 .24 .20 .85 .95 < .0001 .70 . !6 .01 .44 Lower 1.450 - 0.037 - 0.071 - 0.014 - 0.028 0.050 - 0.235 - 0.002 0.058 - 0.013 Upper 3.712 0; i 45 0.015 0.017 0.026 0.096 0.158 0.012 0.49 I 0.030 mnatural log **1994195 vs 1997 Table 12. Mixed Model Cocfficienl [:.,_limates,Standard Errors (SE), P-Values and 95% Confidence Limits fi'om "l'estiqgPolenlial Determinants of Triglyceride _ Change Including TOF and the Interaction with Number of Years of Obserwdion of Antwerp and Decatur Male Employees 95% Confidence Limits A Company Page43of 63 ........ Intercept TOF Years Obsm'valion TOF x Years Obs Coefl]cienl 2.612 (I.053 - 0.0(X)5 - 0.0005 SI'.' _ 0.334 0.020 0.013 0.004 p-value < .0001 .(X)8 .97 .91 Lower 1.953 0.014 - 0.027 - 0.008 Upper 3.272 0.093 0.026 0.007 : Age BM! Drinkdday Cigarettes/day Location* Entry Pt:,iod** Baseline Years Worked mnaluralh)g *Antwerp vs D_atur **1994195 vs 1997 0.004 0.066 - 0.031 0.01 ! 0.074 0.085 0.004 0.007 0.009 0.032 0.003 0.098 0.071 0.007 .57 < .0001 .34 .0002 .45 .23 .58 - 0.010 0.049 - 0.095 0.005 - 0.119 - 0.055 - 0.010 0.019 0.084 0.033 0.017 0.266 0.226 0.018 / Table 12A. Mixed Model Coel'ficicnl Eslimales, Slandard En'ors(SE), P-Values and 95% Confidence Limils fi'om 'l'csling Potential Delerminams of TriglycerideuChange Including TOF and the [nteraction with Number of Years of Observation for Antwerp Male Employees 95% Confidence Limits A Compa,w Paso44of 6.1 Intercept TOP Years Observation TOF x Years Obs Coefficient 3.078 " 0.053 0.016 - 0.0007 SE 0.395 0.028 0.017 0.006 p-value < .0001 .06 .35 .90 Lower 2.294 - 0.002 - 0.018 - 0.012 Upper 3.862 O. 107 0.049 0.010 Age 0.009 0.009 .32 - 0.009 0.028 BM! 0.042 0.014 0.004 0.014 0.071 Drinks/day Cigarelles/day Entry Period** Baseline Years Worked Jnatm_l log ** 1994/95 vs 1997 - 0.032 0.018 - 0.045 0.007 0.034 0.(X)5 0.101 0.011 t .34 .0005 .66 .56 - 0.099 0.008 - 0.245 - 0.016 0.035 0.029 0.155 0.029 Table 12B. Mixed Model Coefficient Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits from Testing Potential Determinants of Triglyceride mChange Including TOF and the Interaction with Number of Years. of Observation for Decatur Male Employees 95% Confidence Limits A Company Page45 of 63 Intercept TOF Years Observation TOP x YearsObs Age BM! Drinks/day Cigarettes/day Entry Period** ? Bascli,tc Yca,'s Worked 'natural log *'1994/95 vs 1997 Coefficient 2.644 0.031 - 0.035 0.004 - 0.002 {).073 - 0.030 0.005 0.275 O.(X)9 S F. 0.563 0.1130 0.022 0.005 t 0.014 0.012 0.099 0.004 0. I i I 0.011 p-value < .0001 .29 t .12.45 .90 < .0001 .76 .20 .01 .44 Lowcr 1.521 - 0.028 - 0.079 - 0.007 - 0.029 0.050 - 0.227 - 0.003 0.055 - 0.013 Upper 3.766 0.090 0.009 0.015 0.025 0.010 0.166 0.012 0.494 0.030 / _1Company age 46 of 63 Table 13. Mixed Model Coefficient Estimates, Standard Errol_ (SE), P-Values and 95% Confidence Limits from Testing Potential Determinants of Alkaline Phosphatase t Change Including PFOS and the Interaction with Number of Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits ......................... lnlerccpl PFOS Years Observation PFOS x Years Obs Coefficient 3.789 " 0.002 - 0.051 0.002 SFO. 176 O.(X)9 0.006 0.002 p-value ' < .000I .87 < .O(X)I .47 Lower 3.442 - 0.017 - 0.062 - 0.003 Upper 4.137 0.020 - 0.040 0.1)06 Age 0.005 0.004 .20 - 0.003 0.012 BMI - 0.0007 0.004 .88 - 0.009 0.008 Drinks/day Cigarettes/day Location* - 0.001 0.003 - 0.242 0.014 0.001 0,049 .94 .03 < .0001 - 0.030 0.0004 - 0.338 0.028 0,006 - 0.145 Entry Period** Baseline Years Worked 0.098 - 0.006 0.037 0.004 .008 0.025 .10 - 0.013 0.171 0.001 Triglycerides m 0.113 0.024 : < .0001 0.067 O.160 mnaturallog *Antwerp vs Decatur **1994/95 vs 1997 _i / _'- mpany Pagt_47 of 63 Table 14. Mixed Model Coefficient Estimates, StandardErrors(SE), P-Values and 95% Confidence Limits fi'om Testing Potential Determinantsof Alkaline PhosphatasemChange Including PFOA and the Interaction with Number of Years of Observation of Antwerp and Decatur Male Employees ' 95% Confidence Limits ....................(.:.l._f._.l._..ei.n_._tc.i.... Intercept 3.785 PFOA 0.005 YearsObservation - 0.047 PFOA x Years Ohs - 0.001 Age 0.005 BMI - 0.0009 Drinks/day - 0.002 Cigarettes/day 0.003 Location* Entry Period** Baseline Years Worked - 0.243 O.!00 - 0.006 SE 0.176 0.012 0.006 0.002 0.004 0.004 0.015 0.001 0.049 : 0.036 0.004 p-value < .000 ! .69 < .0001 .62 .19 .85 .89 .03 < .0001 .007 .09 Lower 3.437 - 0.019 - 0.058 - 0.005 - 0.002 - 0.010 - 0.031 0.0004 - 0.340 0.028 - 0.014 Upper 4.133 0.028 - 0.036 0.003 0.013 0.008 0.027 0.006 - 0.147 O.172 0.001 Triglycerides _ . 0.114 0.024 < .0001 0.066 0.161 #natural log ' *Antwerp vs Decatur *1994/95 vs 1997 t 'l'ablc 15. Mixed M_el CoelTicicn! l,.'slinmles,StandardErrors (SE), P-Values and 95% Confidence Limits from t'1l cslln'g II Potential Determinantsof Alkaline PhosphalaseI Change Including TOF and the Interaction witil Number of Vem_ of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits JCoral)any vagc48of b3 ...................... l.ntcrcepl TOF Years Observation TOF x Years Obs Age BM! Drinks/day Cigaretles/day Location s Entry Period** Baseline Years Worked (:'t_ffieienl 3.789 " - O.(XXXH - 0.049 - 0.00006 0.005 - 0.001 - 0.002 {1.003 - 0.245 0.100 - 0.006 SR 0. 177 0.009 0.006 0.0017 0.004 (l.(_14 0.015 0.001 0.049 0.036 0.004 p-value < .0001 .99 < .0001 .97 .19 .82 .9 i .03 < .0001 .007 .10 Lower 3.44 i - 0.019 - 0.060 - 0.003 - 0.003 - 0.010 - 0.031 0.0004 - 0.342 0.028 - 0.013 Upper 4.138 0.018 - 0.037 0.003 0.012 0.008 0.028 0.006 - 0.148 O.172 0.001 Triglycerides m O. 115 0.024 < .0001 0.068 O.162 mnatural log " *Anlwe;TJ vs Decatur ** 1004/95 vs 1997 | Company rage 49 of 63 Table 16. Mixed Model Coel'ficienl I_;limales, Standard Errors (SE), P-Values and 95% Confidence Limils front Testing Potential Determinants of GG'I_ Change Including PFOS and the Interaction with Number of Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits Intercep! PFOS Years Obserw_tion PFOS x Years Obs Age BM 1 Drinks/day Cigarettes/day Location* Coefficient 1.883 -0.004 - 0.075 0.004 - 0.0{)3 0.005 0.042 - 0.0009 - 0.096 SE 0.373 0.020 0.012 0.004 U.(X)8 0.009 0.031 0.003 01104 p-value < .0001 .84 < .0001 .42 .74 .59 . i8 .77 .36 Lower 1.146 -0.043 - 0.098 - 0.005 - 0.019 - 0.013 - 0.020 - 0.007 - 0.030 Upper 2.620 0.035 - 0.051 0.012 0.013 0.024 0.104 0_005 0. ! 10 Entry Period** Baseline Years Worked Triglycerides m 0.358 0.008 0.251 0.078 0.007 ! 0.050 < .0001 .30 < .0001 0.204 - 0.007 0.152 0.512 0.024 0.350 mnatural log *Antwerp vs Decatur **1994195 vs 1997 1-- Table 17. Mixed M(_lel Coefficient Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits fRmt Testing Potential Determinants of GOT' Change Including PFOA and the Interaction with Number of Years of Observationof Antwerp and Decatur Male Employees 95% Confidence Limits M COral)any Page50of 63 .............. Intercept PFOA Years OI)scrvalion PFOA x Years Ohs C(_fficient 1.876 " - 0.1X)9 - 0.077 0.005 SE 0.374 11.1125 0.012 0.005 p-value < .0001 .72 s < .0001 .29 Lower 1.138 - 0.058 - 0. 100 - 0.004 Upper 2.613 0.040 - 0.054 0.014 Age BMI DrinkrJday Cigarcite,'dday Location* Entry I)eri(_l** Baseline Years Worked - 0.002 0.004 0.043 - 0.0007 - 0.097 0.355 0.008 0.008 0.009 0.031 0.003 0.104 0.077 0.008 .76 .64 .17 ,82 .35 < .0001 .30 - 0.018 - 0.014 - 0.019 - 0.007 - 0.301 0.203 - 0.007 0.013 0.023 0.105 0.005 0.108 0.507 0.024 Triglyceridcs w 0.256 0.051 < .0001 0.156 0.356 mnatural log *Antwerp vs Decatur **1994/95 vs 1997 Tahle 18. Mixed Model Coefficient Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits from Testi,lg Potential Determinants of GGTmChange Including TOF and the Interaction with Number of Years of Ohservatitm of Antwerp and Decatur Male Employees 95% Confidence Limits M Company Page51of63 Intercept TOF Coefficient 1.870 " 0.002 SE 0.374 11.1120 p-valne < .0001 .93 Lower 1.132 - 0.037 Upper 2.608 0.041 Years Observation TOF x Years Ohs - (I.079 0.004 0.013 0.003 < .0001 .25 - 0.104 - 0.003 - 0.054 0.011 Age - 0.003 0.(KI8 .75 - 0.018 0.013 BMI 0.005 0.009 .57 - 0.013 0.024 Drinks/day Cigarettes/day Location* 0.043 - 0.0009 - 0.088 0.031 0.003 0.104 .17 - 0.019 0.104 .77 - 0.007 0.005 .40 - 0.294 0.117 I_.nll'y Period** Baseline Years Worked 0.352 0.008 11.078 0.008 < .01101 .31 0.200 - 0.007 0.505 0.024 Triglycerides m 0.249 0.050 < .000 ! 0.149 0:348 mnaturallog *Antwerp vs Decatur **1994/95 vs 1997 Table 19. Mixed Model Coefl'icicn! Eslimates, _;landard Errors (SE), P-Values and 95% Confidence Limits I'romTesting Potential Dclcrminants of AST' Change Including PFOS and the Interaction with Numl_r of Years of Observalion of Antwerp and Decatur Male Employees 95% Confidence Limits _.MCompany Page52of63 .................C.o.c.ff.icicnl Intercept 3.080 PFOS " 0.010 YearsObservation -0.009 PFOS x Years Obs 0.0007 SI': 0.198 0.()II 0.007 0.003 p-value < .0001 .39 l .19 .79 Lower 2.689 -0.013 -0.024 - 0.005 UpDcr 3.471 0.032 0.005 0.006 Age BM! - 0.008 0.004 0.004 0.005 .05 - 0.016 - 0.00005 .47 - 0.007 0.014 Drinks/day 0.()30 0.018 .II -0.007 0.066 Cigarcltcs/day Localio,l* Entry Period** Baseline Years Worked -0.003 0.102 0.039 0.005 0.002 0.053 0.039 0.004 .07 -0.006 0.0002 .06 - 0.206 0.003 .32 - 0.038 .22 - 0.003 0.116 0.012 Triglycerid_es 0.063 0.029 .03 0.005 0.121 t'natural log *Antwerp vs Decatur **1994195 vs 1997 JM Company ' P,',ge53of 63 Table 20. Mixed Model Coefficient Eslinulles, $landard Errors (SE). P-Values and 95% Confidence Limits from Tesling Polenlial Delermin:mls of A-S'I_ Change Including PFOA and Ihe Interaclion wilh Numl_r of Years of Ohservalion of Anlwerp and Decalur Male Employees 95% Confidence Limits Intercept PFOA __Coefficient 3.053 0.027 Yem:,_Ohscrvalion PFOA x Yeal_ Obs - 0.(_)8 - 0.002 Age - 0.007 BM! 0.004 Drinks/day 0.030 Cigarelles/day -0.003 Local ion* Enlry l)crio(l ** Baseline Years Worked - 0.097 0.04,1 0.004 ,_I': 0.199 0.015 0.007 0.003 0.004 0.005 0.018 0.(}02 0.053 0.039 0.004 l)-value < .0(_) I .06 .28 .41 .08 .39 .10 .07 .07 .26 .27 Lower 2.661 - 0.002 - 0.022 - 0.008 - 0.015 - 0.006 - 0.006 -0.006 - 0.201 - 0.032 - 0.003 Upper 3.445 0.056 0.006 0.003 0.0008 0.0 !5 0.066 0.0002 0.008 0.120 0.012 Triglyceridcs' 0.054 0.030 .07 - 0.004 0.113 "natural log Antwerp vs Decatur * 1994195 vs 1997 Table 21. Mixed Model Cocfficien! Eslimales, Standard Errors (SE), P-Values and 95% Confidence Limits fl'om Testing Potential Determinants of AST a Change Including TOF and the Interaction wilh Nu,nber of Years of Ohservation of Anlw_rp and Decatur Male Employees 95% Confidence Limits _1Conqmny i,ag .54of 63 Interccpl TOF Years Observalion TOF x Yem:s Obs Age BM! Coefficicm 3.058 0.018 - 0.009 - 0.006 - 0.(}08 0.004 SI. 0.198 : ! 0.011 0.iX)8 0.002 0.004 0.005 p-value < .0001 .12 .24 .79 .06 .40 Lower 2.666 - 0.005 - 0.024 - 0.005 - 0.016 - 0.006 Upper 3.449 0.04 0.006 0.004 0.0002 0.015 Drinks/day Cigaretlc "s/day Location* 0.029 - 0.003 - 0.095 0.018 0.002 0.053 .12 - 0.007 .07 - 0.006 .08 - 0.199 0.065 0.0002 0.010 Enlry Period** Baseline Years Worked 0.039 0.005 0.039 0.004 .31 - 0.037 .24 - 0.003 0.115 0.012 Triglycerides w 0.058 0.029 .05 0.00003 0.1 !6 #natural log *Antwerp vs Decatur **1994195 vs 1997 :" Table 22. Mixed Model Coefficient l_slhnales, Slandald Errors (SE), P-Values:and 95% Confidence Limils from Tcsting Potential Determi,_ants of ALl"_ Change Including PFOS and the Interaction with Number of Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits , (.;Oral)JUly . ._c 55of63 ...................... Intercept PFOS Yeai_ Ohsurvaiion PFOS x Years Obs Age BM! Drinks/day Cigarcueshlay Location* Entry Period** Baseline Years Worked Triglycerides' ('ocliicicnl 2..501 " 0.010 - 0.095 - 0.00003 0.0009 0.010 - 0.012 - 0.008 - 0.088 0.329 0.001 0.159 ,";I_ ..........._, 0.273 0.016 O.OlO 0.004 0.006 0.007 0.025 0.002 0.073 0.054 0.005 0.040 p-value < .0001 .54 < .0001 .99 .88 . !7 .63 .001 .23 < .0001 .84 .0001 Lower 1.961 - 0.021 - O.l 15 - 0.008 - 0.012 - 0.004 - 0.062 - 0.012 - 0.233 0.222 - 0.010 0.079 Upper 3.04 I 0.04 1 - 0.075 0.008 0.010 0.024 0.038 - 0.003 0.056 0.436 0.012 0.238 *natural log *Antwerp vs Decatur ** 1994195 vs 1997 f---- -. Tahk_23. Mixed M_xlclCoeflicicnt Estimales, Standard En'ors (SE), P-Values and 95% Confidence Limits from Testing PotentialDeterminants of AL'!"mChange Including PFOA and the Interaction with Numberof"Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits C_m)pa.y Page56of 63 ................ ( :_._.fficji.e.!!! Intercepl PFOA 2.479 0.015 Years Observation - O.107 PFOA x YearsObs 0.005 Age - ().1)001 BM! 0.011 Drinks/day - 0.009 CigareUcs/day Location* - 0.007 - 0.079 SE 0.272 0.02(I 0.010 0.(g)4 (}.()()6 0.007 : 0.025 0.(X)2 0.072 p-value J < .0(X)! .46 < .0001 .19 .98 .13 .72 .001 .28 Lower 1.942 - 0.02.5 - O.126 - 0.003 - 0.011 - 0.003 - 0.058 - 0.012 - 0.222 Upper 3.016 0.(i54 - 0.087 0.013 0.011 0.025 0.040 - 0.003 0.064 Entry Period** Baseline Years Worked 0.330 0.0008 0.053 0.005 . < .0001 .89 0.225 - 0.010 0.434 0.011 Triglycerides _ 0.151 0.041 .0003 0.071 0.231 #naturallog *Antwerp vs Decatur ** 1994D5 vs 1997 / .f-- ,,.. .. _Company _.agc57 of 63 Table 24. Mixed Model Coefficient Estimates Standard Errors (SE), P-Values and 95% Confidence Limits IYom"l_sting Potential Determinants of ALT mChange Including TOF and the Interaction with Number of Years of Observation of Antwerp and D_atur Male Employe_s 95% Confidence Limits lntercep! TOF Years Observalion TOP x Years Obs C,xdTicicnl 2.476 0.020 - 0.102 0.002 SI,.' :. 0.273 0.016 : 0.011 0.003 p-value < .0001 .21 i < .000l .50 Lower 1.936 - 0.011 - 0.123 - 0.004 Upper 3.0 i 5 0.051 - 0.082 0.008 , Age BMI Drinks/day Cigarettes/day Libation* - 0.0005 0.0 ! I - 0.013 - 0.008 - 0.073 O.(X)6 0.007 0.025 0.002 0.073 .03 - 0.011 0.010 .11 - 0.003 0.026 .62 - 0.062 .001 - 0.012 .32 - 0.217 0.037 - 0.003 0.072 Entry Period** Baseline Years Worked 0.325 0.0008 0.053 (}.005 < .0001 .88 0.220 - 0.Ol0 0.430 0.011 Triglycerides= 0.148 0.040 .0003 0.069 0.228 =natural log *Antwerp vs Decatur ** 1994195 vs 1997 i ', ? Table 25. Mixed Model Coefficient Estimates. Standard Errors (SE). P-Values and 95% Confidence Limits from "Vesting Potential Determinants of Total Bilirubin e Change Including PFOS and the Interaction with Number of Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits Cuml}mly rage 58 of 63 Intercept PFOS Years Ohservation PFOS x Years Obs Age BMI Drinks/day Cigarettes/day l.._alion* Entry Periotl *_ Baseline Years Worked Coefficient - (I.334 "- I).1)18 0.(133 - 0.002 0.0 ! I - 0.003 0.016 - 0.008 0.315 - 0.114 - (I.002 SE 0.2,11 0.1)14 (1.1109 0.004 0.005 0.007 0.(124 0.(102 0.064 0.047 0.005 : p-value '. 17 .22 .(1005 .94 .02 .67 .50 < .0001 < .0001 .02 .64 Lower - 0.gl I - 0.046 0.0148 - 0.007 0.001 - 0.016 - 0.030 - 0.012 0.189 - 0.9_06 - 0.011 Upper 0.142 0.01 I 0.052 0.007 0.020 0.010 0.062 - 0.004 0.441 - 0.022 0.007 Triglyceridest - 0.088 0.037 .02 - 0.161 - 0.015 "natural log *Antwel9 vs Decatur **1994/95 vs 1997 _ICompany Page59of 63 "l'ab2l6c. Mixed Model CoefficienEtslimalesS,tandardErrors(SE),P-Valuesand95% ConfidenceLimits fromTcsdngPotentiaDleterminantosfTotalBilirubmiCnhange IncludinPgFOA andtheInteractiwointh NumScr ofYem,'osfObservatioonfAntwerpand DecaturMale Employees 95% ConfidenceLimits intercept PFOA ..........C.'._.l'l'icicnt - 0.315 " - 0.030 YearsObservation PFOA x Yeat_Obs 0.028 0.005 Age 0.010 BMI - 0.003 Drink's/day Cigarette "s/day Location* 0.014 - 0.008 0.318 Entry I)crio(I** Baseline Years Worked - 0.124 - 0.002 Triglycerides j - 0.082 SF. 0.242 0.019 0.009 0.004 t 0.005 "!. 0.007 0.023 0.002 0.004 0.046 0.005 0.037 p-value .19 . !6 .003 .18 .04 .61 .55 < .0001 < .0001 .007 .71 .03 Lower - 0.793 0.066 0.001 -0.002 0.0007 - 0.016 - 0.032 - 0.013 0.193 - 0.215 - 0.011 - 0.156 Upper 0.162 0.007 0.046 0.013 0.020 0.010 0.060 - 0.004 0.444 - 0.034 0.007 - 0.008 mnatural log *Antwerp vs Decatur **1994/95 vs 1997 "rab| 27. Mixed Model Coefficicnl ILstimates, Standard Errors (SE), P-Values and 95% Confidence Limits from "l'esling Polenlial Determinanls of Total Bilimbin I Change Including TOF and the Interaction with Number of Years of Ohscrvation of Antwerp and Decatur Male Employees 95% Confidence Limits ,Company t'age60 of 63 ........ lntercci)l 1"OF Years Ohse,'vation TOF x Years Ohs Age BMI Drinks/day Cigarcttcs/day Location* 'Entry Pcriod** Baseline Years Worked Triglycerides' Cocfficienl - 0.318 " - 0.020 0.029 0.003 (}.011 - 0.003 0.016 - 0.1X)8 0.316 (I. 120 - 0.002 - 0.085 SI,_ 0.242 0.014 0.010 0.003 (}.IX}5 0.007 0.023 1}.002 0.064 0.046 0.(R)5 0.037 P-Value . I9 .I0 .(X)5 .36 .03 .65 .5 i < .0001 < .0001 .01 .68 .02 Lower - 0.796 - 0.049 0.009 - 0.003 0.001 - 0.016 - 0.031 - 0.012 0.190 - 0.210 - 0.011 - 0.158 Uplx:r 0. !60 0.008 0.049 0.009 0.020 0.010 0.062 - 0.004 0.442 - 0.029 0.007 - 0.012 mnaturallog *Antweq_ vs Decatur ** 1994195 vs 1997 l Compaaly Isagc61 of 63 Table 28. Mixed Model Coefficient Estimates, StandardEnTors(SE), P-Values and 9595Confidence Limits from Tesling Polential Determinanls of Direct BilirubinmChangeIncluding PFOS and the Interactionwith Number of Years of Ol_sen'valionof Anlwerp and Decatur Male Employees 9595 Confidence Limits Interccpl PFOS Years Observation PFOS x Years Obs Coefficient - 1.756 - 0.013 - 0.097 - 0.006 SE 0.2(H 0.013 0.009 0.()04 p-value ; < .0001 .29 < .0001 .18 Lower - 2.160 - 0.039 - O.116 - 0.014 Upper - 1.355 0.012 - 0.079 0.003 Age BMi Drinks/day Cigarettes/day Location* 0.007 - 0.003 0.018 - 0.0005 0.076 0.004 .06 0.006 .58 0.020 .37 0.002 : .79 0.053 .15 - 0.0003 - 0.015 - 0.022 - 0.004 - 0.028 0.015 0"008 0.058 0.003 0.180 Entry Petite** Baseline Ycans Worked 0.353 - 0.001 0.038 : / 0.004 . < .0001 .74 0.277 - 0.009 0.428 0.006 Triglycerides w - 0.090 0.033 .006 - 0.155 - 0.026 natural log *Anlweq) vs Decalur **1994/95 vs 1997 Table 29. Mixed Model Cocfl'icic,+t Estimates, Standard Errors (SB), P-Values and 95% Confidence Limits from Testing Potential Determinants of Direct Bilimbin u Change Including PFOA and the Interaction with Number of Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits .............. Intercept PFOA Years Obsc,'vation PFOA x Years Obs Age BMI C(l'ficicnt - 1.753 - 0.0 !2 " - 11.095 - 0.004 0.006 - 0.003 SE 0.207 0.017 0.1X)9 0.(XH O.(XH 0.006 p-value < .0001 .47 < .0001 | .27 .! I .66 Lower -2.161 - 0.045 - 0.113 - 0.012 - 0.001 - 0.014 Upper - 1.345 0.02 I - 0.077 0.003 0.014 0.009 Drinks/day Cigarettes/day Location* Entry Peri(ul** Baseline Years Worked Triglycerides _ 0.014 - 0.0005 0.083 0.345 - 0.0008 - 0.089 0.020 0.002 0.053 0.038 0.004 0.033 .50 .80 .12 < .0001 .83 .009 - 0.026 - 0.004 - 0.022 0.271 - 0.008 - 0.155 0.053 0.003 0.188 0.420 0.007 - 0.023 IIItatura[ log *Antwerp vs Decatur ** 1994195 vs 1997 { Table 30. Mixed Model Coefficient Estimates, Standard Errors (SE), P-Values and 95% Confidence Limits from Testing Potential Determinants of Direct Bilirubin' Change Including TOff and the Interaction with Number of Years of Observation of Antwerp and Decatur Male Employees 95% Confidence Limits .: .A Company Page63 of 63 Intercept TOF Years Observation TOF x Years Obs Age BMI Drinks/day Cigar, ties/day Location* Entry Period** Baseline Years Worked Coefficient - 1.748 - 0.013 - 0.093 - 0.004 0.007 - 0.003 0.017 - 0.0004 0.074 0.349 - 0.001 SE 0.205 0.013 0.010 0.003 0.004 0.006 0.020 0.002 0.053 0.038 0.004 , p-value < .0001 .33 < .0001 .21 .08 .57 .39 .84 .16 < .0001 .77 Lower - 2.154 - 0.038 - 0.113 - 0.011 - 0.0008 - 0.015 - 0.022 - 0.004 - 0.031 0.274 - 0.008 Upper - 1.343 0.013 - 0.073 0.002 0.014 0.008 0.057 0.003 0.179 0.424 0.006 Triglycerides # - 0.087 0.033 .009 - 0.152 - 0.022 'natural log *Antwerp vs Decatur *'1994/95 vs 1997 !.