I lllllll IllU 1111 11111 111 IllU 11111 Ilm 1111 111 - PBS I lllllll IllU 1111 11111 111 IllU 11111...

F I lllllll IllU 1111 11111 111 IllU 11111 Ilm 1111 111 BAOO 0 0 6 0

o s S s z - i - ~ eD

H 43 c.

W W

, -

Y

Epidemio logica I In ves tiga tio n of the Polyvinyl Chloride Industr-y . - in Reference to Occupational Acroosteolysis

by The Institute of Industrial Health The University of Michigan Ann Arbor, Michigan

February 1969

4 , ‘.$:

*I,

Confidential Report to the d ” Medical Advisory Committee Manufacturing Chemists Association

BAO 0 0 0 61

L

I

The Institut.e of Industrial HeeIth The University of Michigan

Ann Arbor, Michigan

Confidential report. to the Medic~I Advisory Committee

Manufactufing Chemists' Association

I

BAO 0 0 0 62

21 w

Q a u a c h, '0 W 0

PREFACE-LIMITATIONS OF TED3 STUDY

.. .!Chis study was undertaken to determine the prevalence o f acroostedysfs in

In interprethg the data that follows we would urge the P.V.C. Industry and to IdentUgr those occupational exposures presenting a Mgh risk of this disease. careful consideration of these cautions:

1. The fact that a campany or plant does not seem t o have any cases of the disease at the t ime of the survey does not r u l e out the poss ib i l i ty that cases may develop subsequent to the study.

2. Some of the early manifestations of the disease, e.&, Raynaud's phenomenon, and some nonspeciffc x-ray changes, m y occur in diseases other than acroosteolysis. nonspecific x-ray changes presents particular difficulty in determining the presence or absence of the disease Fn a particular plant, cf. Table XXII.

Thus, the interpretation of this symptom complex or o f

3. The over-all prevalence rate of acroosteolysis determined in the present study is sufficiently low that the apparent absence of the disease i n certain s m a l l plants may not reflect a true absence o f the disease or hazard, but simply reflects insufficient experience.

Therefore, we strongly recommend that: (1) continuing medfcal surveillance and, (2) measures directed toward minirdzation of hand cleaning of reactors bo undertaken Sn a l l plants regardless of presence or the apparent: absence of cases of acroosteolysis. . .

I

BAOO 0 063

! r

I

.

I

.. .

TABLE OF COmTENTs

ZLST OF TABLB

IZST OF FIGURES

Section

I. A. Background of Pro%lem B. Medical Bibliographic R e v i e w -

II. EPIDEMfOLOGIC DESIGN CONSIDERATIOk

A. Phase I-Preplanning 1. Questionna5rz 2. X-ray exandnations 3 Work his tory evaluation 4. Bibstrial hygiene evahation

1. Clinical history 2. X-ray examination 3. Work history questionnaire 4. I n d u s t r i d hygiene survey Survey Problems A r i s i n g In the Course of study 1. Clinical h i s t o r i e s 2. X-raxs 3. Work his tory 4. I n d u s t r i a l hygiene survey

B. Phase II-Pilot Run

C.

- -- lp. coMpuTATIoMALMEmms .

A. General Outline 1. fdentificaffon of s S y respondents 2. QuaLLty control of coding 3. C a r d PWchfW

B. Data Conversion Techniques

2, X-ray examinations 3. Work history 4. Materials and exposure data

I.. 2. CoxQutation manipulation

1. Clinical h f s t q

C. Computer Storage Techniques D e c k "clean-up" and disk storage

I

iii

page vii

X

1

1 2

3 3 3 - 6 8 1 2 ' 15 15 15 17 17 20 20 20 20 20

22

22 22 22 22 22 22 23 23 23 25 25 26

. :

TABLE OF CONTENTS (Continued)

Section

D. Statistical Methods of Collation

rv. RESULTS

A. Population Characteristics B. Signs and Symptoms in the PVC Study Population

1. Definitions 2. Results-Table I V 3. Factor analysis 4. Comments concerning nonspecific nature of symptoms

studied C. Hand X-Rays Results

Total. defects in PVC and control populations Agreement of radiological interpretation, wc population Relative number o f fingers involved among 595 x- ray abnonaalitfes in PVC workers Occurrence of abnormal x-ray by defect and plant Occurrence of x-ray defects by corporation Comparison of x-ray defect occurrence between Tecum- seh control population and plants performing only compounding

1. 2.

3.

4. 4. 6.

D. Occupational Experience of PVC Study Populatfon 1. General characteristics 2. Work experience by plant and job man-months 3. Materials and processes

E. Diagnostic Categorization - 1. Description

2. Discussion of diagn6s.t;i.c category distribution -

Yo m C T I o N S OF SIWS AND SYMPTOMS, X-RAY Fp(IDLmGs, JOB HISTORIES, AM) DIAGNOSTIC CATEGORIWITION

A. -

Distribution of Symptoms Between Study Group, its Sub- Groups, and Contro l Population I. Occurrence o f diagnostic category 2. Occurrence of Signs and Symptoms 3. Discriminant ability to differentiate plants with AOL

from plants probably not having AOL Job History and its Relationshfp to Possible Distribution of AOL Among Various Plants

3.

page

26

28

28 29 29 36 37

38 39 39

40

h2 43 43

43 46 46 46

' % 63 63 65

69

69 69 71

79

79

.-

BAO 0 0 0 6 5

' TABU OF CONTIXTS (Continued)

SecMon

1.

2.

3.

4.

5.

Relationship of respondent's job his tory and man- months exposure t o diagnostic categories Clustering of high associational. relationships between jobs and AOL Discussion of clustering o f jobs with hiagnostic categories 5, 6 Dif feren t ia l distribution of duration of work experience i n s p e t i f i c jobs Job experience o f individuals with Rsynaud's phenomenon on the baais of diagnostic category 3 among var ious plants

Distribution and groupings o f t he x-ray defects i n this group Comparison of x-ray defects in diagnostic categories 5, 6 and Tecumseh controlgopulation Distribution of x-ray defects i n diagnostic category 4 Comparison of x-rays i n control population vs. p l a t s beUeved not t o have AOL

C. Analy8i.s o f X-Rays vs. Diagnostic Categories 5, 6 1.

2.

3.

4.

Differences in Exposures in Plants With and Without AOL 1. Materlals used 2. Reactor cleaning 3.

D.

Procedures following completion of polymerization - VI. SUMMARY AND cmcLusxoNs

A. The D e f M t i o n o f Acroosteolysis B. Course of 3fsease .

1. Early 2. Active disease 3. Recovery phase Possible Variations in t he Nature and Course of AOL 1. 2.

Relationship Between Job Assignments and t h e Occurrence of AOL 1. Reactor s z a p i n g 2. Compounding operations

C. Raynaud's phenomenon Hthout bone change Bone changes without Raynaud's phenomenon

D. Prevalence of AOL E.

* TABLE OF COImEN?S (Concluded)

Section

F. Material Exposure and AOL G. Work Practices and AOL

1. Recovery phase 2. Air fluah of reactor before entry of cleaner 3. Reactor cleaning procedures Control Measures In Prevention of Acroosteolysis H.

VII- mOMMWDATI0IPS

I

.. . .

. .

vi

4 a - B 6)

N m - N

Page -.)

101 101 101 102 102 102

104

105

BAO 0 0 067

L

.

1

.. J

.

Table

I.

II.

III.

m.

VI.

VIL

LIST OF TABLES

. .

Stundard 1 I E Job TLtles

Magnitude of Dose Expressed 8s S w e Concept

Plant Papulation

Occurrence of symptom by =ant

Standard Deviation, and Standard Error of the Mean Per- centage o f Symptcm Occurrence in All Plants

Rank Ordering of Frequency of Positive Symptoms in Wc Study Population

Number o f X-Ray Defects and 'Percent of Plant Populatfon WLth Sucb Abnormalities i n Control and PVC Populations

Agreement and mumber of Fingers Sharing X-Ray Defects . in 1362 PVC X-Rays

Total m e r of Fingers Involved, by Percentage

Specific X-Ray Defect by Plant

Occurrence of X-Ray Defects by Corporation

Occurrence of Specific X-Ray Defects in Control PopuLatfon' ana fn Workers at P l a n t s Where* only Compormding Opera- t ions are Performed - Job Experfence by Plant i n Tenns of Mtnaber of Respondents Reporting Specific Job Experience, Total and Mean Man-Moneha in Specific Job

Functional Grouping of Jobs, Men, Man-Months, and Kern Man-Mmths

Extene of. Exposures to Materials in Vinyl Chloride Monomer. Production

Extent; of Exposure t o Catalysts

BAO 0 0 0 68

- pw3e

Id

26

29

32

36

38

39

41

42

44

45

4s

47

51-

58

2

TaLz m r . XVIII.

X I X .

xx.

fMI.

m1.

mIr . a m.

t ., 'LIST OF TABLES (Continued)

Reactor Cleaning Methods

Absolute Pressure-Time Factor

Roentgenographic end Symptomatic C r i t e r i a for Diagnostic Categories 0-6

. .

Distribution and Frequency of Diagnostic Categories by CorporaZions

Distribution and Frequency of Diagnostic Categories by Plants

Distribution of AOL in PVC Study G r o u p

Distribution of Dfagnostic Categories Among the Three Sub- Populations within PVC Study Group

Distrfbutfon of Raynaud's Phenomenon Between Plants with Probable, Possible, and No Cases of AOL, Corrected for Male Population

. .

Occurrence of Signs and Symptoms i n Plants with Probable Cases of AOL

xxvf. Occurrence of Signs and Symptoms in Plants Carrying Out Conpounding m'erations Only

-

XXVII. Symptom Dlstrtbution Among Plants Excepting Those Engaged in - - . Compounding Only (KO Cases in Diagnostic Catagories 3, 6)

=I. Distr€bution of signs aGd SymptOms in Plants Having mssible Cases of AOL (i-e-, G, BB, CC)

X X Z Summary of Symptoms, Distribution, and Mean Occurrence in Plants Known to Have Diagnostic Categories.5, 6, Possibly Having, or Not Believed to EIave AOL. Conpunding Only

Similar Data for Plants

X X X . Sunanary of Comparison of Symptom Occurrences

3wcT. Ability o f Symptams in Populations Believed to e v e AOL to Discriminate f r o m Populations not Ee%Leved to Have AOL

. . I .

v i i i

- .. .. - .

Page

61.

62

.64 '

65

66

67

60

70

72

73

74

75

76

T7

78

Table

XXKII . -11.

-0

XXXVII .

XXXVIII.

XLI.

XLII.

- - LIST OF TABLES (Concluded)

Cases per Man-Month Exposure in Specific Jobs

Cases per Exposed Population

Cases with Expertence in Each Job per Total mumber of Cases i n Each Dlagnosttc Category

Clustering of Jobs with Diagnostic Criteria

Number of Men aidburation of Work as Reactor Cleaner ia Plants with AOL vs. Plants wlthaut AOL

Possible Work Experience in Job 17 Among I n d i v i U with Raynaud's Phenomenon Only (Dx Cat. 3) i n Ugh, Possible, and Law RZsk Plants

The Occurrence of Specific X-Ray Defects in the Tecumseh Control Population aad Population in Plants Believed t o Eave Cases of AOL

Comparison of X-Ray Defect6 in Tecumseh Populatioq with Plants Believed n-&t t o Have AOL Cases

Reactor Cleaning Procedures

X - R a y Defects Associated with IndirLdusLs i n Dfagnostic Categories 3 tbraugh 6 . .

Prevalence of AOL by Diagnostfc Category and Reactor Cleaning Exposure (Job 17)

, ix

BAO 0 0 0 70

LIST OF FIGURES

. . . . ngure

CJ a c . !

1. Medfcal hfstory questionnaire. ' 4

2. Hand x-r& survey (Instttute of Industrial Health). 7

3.

4.

Job h i s to ry form (p r io r t o p i l o t s t u d y ) .

Exaqle of job history card entries.

ll

12

5. Plant sheet. 13

6. Job sheet. 14

16 . . 7. Materid exposure sheet.

8.

9.

Job history fo ra (as revised d t e r p i l o t study).

Job and material .exposure sheet.

18

19

10. Job and material exposure sheet. (8s revised). 24

ll. 25

12. ~ ~ d i c d history questionnaire (as revised after pilot study). 30

Example of coding emosure data for a given material.

13. Hand x-ray s m y (Institute of In&strial Health). ' 40

14. Standard Im job t i t l e s . 56

15. Absolute pressure-time values. - 9 5

. . I

x r:

BAO 0 0 071

CI I .)

I i

SECTION I

lloTRODUCTION

I

I J . -t

L

t J

'5

d

f 1

-.

*

7

A. BACKGROUND OF PROBLEM - Between 1961 and 1963, a bizarre, previously undescribed syndrome p a d -

ally emerged within a polyvinyl chloride synthesis plant i n the United States. This complex of symptoms was characterized, i n part, by Raynaud's phenomenon and associated with finger t i p tenderness and pain. While the Raynaud's phenomenon-a temporaxy small vessel constriction involving t h e fingers-could normally be expected t o occur within any large population of males, the l o c a l tenderness differed from tha t expected w i t h Raynaud's. hrents L e d t o x-ray examination of the hands o f several workers in the PVC synthesis operation with r a t h e r unexpected findings. vealed w h a t appeared t o be mlt iple f racture Lines with associated i r r egu la r i ty of the margins of the phalangeal tufts.

A t the d i s t a l phalanges of the hand the x-ray re-

The x-ray defects resembled those of the rare famiLfal condition, Familial Acroosteolysis. Bowever, sweral aspects of the current groblem were incon- s i s t e n t with t h e famil ia l disease. Only 47 cases of t he la t ter had been reported i n t h e world's U t e r a t u r e whereas several Cases were seen i n the relatively small population In a PVC synthesis plant. associated abnormality was limited t o the hands whereas t he familial. condition involved both upper and lower extremities. In View of these differences, aug- mented by others a t a later date, the new abnormality was referred t o as occupational ecroosteowsis (AOL).

In addition, the occupationally

Simultaneously, D r . Lef evre,' of Solvay Corporation in Belgium, reported similar lesions of t h e d i s t a l phalanges i n same WC synthesis workers. of the apparent 8ssocia+&on between this unusual clinical. condition ana some phase of vinyl chloride polymerization, the Medical Advisory Committee of the Manufacturing Chemists' Association recommended further investigation of this

Because

phenomenon i n order to: _.

1.

2.

Protect the heal th of workers engaged i n vinyl chloride polymetizatfon.

Determfoo t h e par t icular phase of polyvinyl chlortde production associated with tMs condition.

3. Learn whether any workers involved i n the compounding and fabrication of products from tfie polprlnyl chloride polymer w e r e affected by this condition.

1 !

4 I .

1

c

BAO 0 0 0 7 2

-e

u em

m

4. D m 8 t measures t o protect potent ia l ly exposed workers a t such phases of the operation as might be indicated by investigation. N

m - N

e The research program caLled for both epidemiologic studies in the industry y)

as wellas laboratopy studies i n animals. This.report is confined t o the former.

. _

B. MEDICAL BIBLZOGRAFSIC REVlEW

Rgview was undertaken of t he toxicological and medical Literature per t inent t o t h e vinyl chloride and polyvinyl chlorlde industry, u t i l i z i n g the resources of several U. S. manufacturers of vinyl chloride. This disclosed a report by Suc$u published fn Roumania i n 19632 which described t h e occurrence in some workers of physioLogica1 manifestations consistent w i t h Raynaud’s phenomenon. Associated s e n changes involving the hands and wrists and i n several cases the neck and face were characterized by gecul lar shiny, tight, somewhat thfckened lesions in addition t o small plaque Lesions of a similar nature involving t h e wrists and fingers. While c l in i ca l ly these lesions were referred t o as sclero- de.=, biopsy of these lesions indicated t h a t they were consistent with derma- tomyositis. In addition, there were gastrointest inal disorders which, while not hepatotoxic i n origin, were suggestfve of some l i v e r involvement. Central nervous system manifestations were noted which could be associated with over- doses of vinyl chloride monomer. No mention was made of hand x-rays.

In 1967, an a r t i c l e From Franc2 described lesions more representative of those seen by t he Belglum and U. S . manufacturers. report an involvement of t he s tyloid processes of the ulna and radius a t t he wrist. and Adam5 in Great Britain, described more cases of occupational acroosteoly- s i s . a combination of the skin lesions, Raynaud’s phenomenon, and bony lesions were again described. a spondylit ic lesions involving the sacroilAac jo in t . ing number of reports of this unique condition, end for the reasons noted above, the urgency of this study became more apparent.

This paper was the first to

Following this report, Wilson, e t a ~ , ~ i n the United States, and Harris

In the l a t t e r paper, and i n another subsequently published i n France,6

Furthermore, the -British publication5 noted t h e presence of In view of t he munt-

- -

2

BA000073

SECTION II

E2IDEMIOLCGIC DESIGN CONSIDEXATIONS

A. PHASEI--PREPLAN”G

As a result of conferences between representatives of the Manufacturing Chemists’Association and the I n s t i t u t e of Xndustrial H e a l t h l a te i n the fall of 1966 and January 1967, all 23 manufacturers of polyvinyl chloride in the U, S. were invi ted t o par t ic ipate i n the epidedologic study. A l l but four agreed t o do so. duction of polyvinyl chlorfde as of January 1, 1967, would be examined. plants had a l l employees i n the plant e&ned, regardless of t h e employees‘ association with t h e polymerization process. ployees engaged i n the fabrication of finished resins were Included. the results t h a t are here reported a re based on examination of & of the workers d i rec t ly engaged in the production of the polyvinyl chloride polymers, but the workers i n other categories represent but a sample of those employed i n the 19 partfcipat ing companies,

It w a s agreed tha t all workers directly involved in the pro- some

In others, a portion of the em- Thus,

Job histories, process, and relafed chemical information w a s t o be obtained for t h e five years preceding t h e study, even though the employees might have worked in the plant for a longer period of time. company during the past f i ve years, job histor ies were obtatned only f o r their experience w i t h the present company.

For employees who entered the

O n the basis of a Visit t o one corporation’s production Operation, dis- cussions with technical personnel Fn the plaxt, and conferences with those of- ficials responsible. f o r corporate indus t r i a l hygiene and medical affairs, the mode of epidemiological approach was formulated. In addition, following a re- v i e w of’ the literature and other information made available t o t h e Institute of Industrial Health through t h e cooperation of concerned corporations, it became apparent t h a t answers t o questions of a c l i n i c a l nature as w e l l . as x-ray studies and indus t r i a l hygiene evalwt;tolrr were required. tially designed ccnsidering four phases o f investigation by means of‘ a m e d i c a l questionnaire, x-ray examinations, job history, and indus t r i a l hygiene survey.

A program was ini-

1. Questionnaire

A self-administered type questionnaire w a s constructed t o include those sfgns or symptoms associated with AOL (Fig. 1). €n words clear ly ur?derstood by t h e several levels of understanding t h a t m i g b t be encauntered i n the plant population. to be expllcit but not leading and of such nature as not t o cause undue con- cern among the employees.

The questions had t o be asked

Furthermore, the questions w o u l d have

Fhlally, sincelargenuinbers ofquest ionnairesueretobe

3

BAO 0 0 0 74

L

To ARSWEX THE FOLLUdIMG QUESTIONS, FIE A CGLCLE AROUND ZSJ-BOX WITR YOUR

RIGHT NAD F&quent tenderness, soreness, or

pain at fingertips LEFTHAM)

Fig. 1. Medical hisbry questionnsire.

t

4

BAO 0 0 0 7 5

r ' 't

T .I 1.:

I : 'I :

Have ytni noticed 'by of the follawing changes of the skin of your hands, wrists, or arms:

- *Ifyes, i s it 03 JEG5Z-j JWRIsTsJ )-

White, raisedpatches of skin *= Whitish pimples -+= lZ5-J

Shiny skin patches *lyEs pEJ

\

+If yes, is it on ]HANDsI ~WRISTS~ JGEI .

*If yes, is it on -1 JWRISTSJ IEEJ

Thickening of the skin .

35

36

*-

I 34

, - +IP yes, i ~ i t On JEiiEl J~RISTS~ j m q J Small , raised areas, ar lumps less than 1/2 inch around (NOT including *= w a r t s 1

3 1

38

Pits or scars of the fingernails m m - -

- Ridges or cracking of the fingernails

Have you used hand held power tools or air-driven tools at work in the past six years.

*If yes, i s it on l j j IWRrsTs/ )E] '-

Have

39 40

you been bothered by these ftngemail troubles: I

Have you in jured any fingers in the past month? lyEsl @ * 42

Which finger was in.Jured-- -1 JLnaexJ

IRinfff E\

43 f Middle I - -

Fig. 1. (Concluded)

5

BAO 0 0 0 7 6

L answered, it was necessary to design a self-completion type examLnatton f o m which lent itself to rapid handling wfth a minimum of supervision.

The manner in which the questionnaire would be administered was l e f t t o It was emphasized t h a t management

A t others, the dis t r ibut ion was

the discretion of local plant management. s h e d not Influence employees‘ answers. hrnRpd aut without any further explanation, preceded by an explanation that a study w a s being made of the health of workers ia the PVC industry. Is the completion of the form.

A t some plants, the questionnaire was

I n some cases, supenrisory personnel were present t o aid

To assure uniformity i n x-ray interpretations, standardized techni&es f o r obtaining hand x-rays were recammended by D r . Walter M. Whitehouse, mo- feeqor and C h a i r m a n , Department of Radiology, and Dr. Theodore Ditchek, former Asgistant ProfessQp, Department of Radiology, The University of Michigan Medical School. The requirements were:

FXLM: Non-screen, Kodak Standard Ready-Pak, 10 x l2 inches (or 10 x 12 Blue-brand Kodak film i n cardboard caset tes)

EXPOSURE FACTORS: 45-55 Kvp (depending on thickness af hands) loo ma 1/2 sec 36-40 inches, distance -

!I!BXNIQm: Single exposure. Both hands t o appear on sFngle 10 x 12 f i l m , ProxindL phalanges a t middle of fila. not be close t o edge of film (.@low at least one inch margtn). Fingers must be flat against film holder.

“ips of f inge r s should

IDENTIFICATION: By Social Security Number of the employee.

- D K ’ C a r o U . F. Tatum, Radiology. Consultant t o The B. F. Goodrich Company, reviewed wtth D r . Whitehouse a n b t h e r s o f the Institute of f n d u s t ~ a l Health’s staff, a INmber of hand x-rays thought to show variolls stages of acroosteolysis. Since the radiologic manifestations of the early disease d g h t be equivocal, uur [email protected] decided to report t h e i r findings i n terms of morphologic defects rather than as a diagnosis of acroosteolysis or no acraosteolysis. The decision as t o whether the disease wa6 present was arrived a t only after data frum both the control and i ndus t r i a l populatiop was obtained.

fn order to expedite t h e reading of large numbers of x-rays and record f i n d h g s in a unffonn manner, an x-ray reporting form was designed (Fig. 2). This conststad of a white sheet ( a s i l l u s t r a t ed ) accompanfed by three rep l i ca t e pages of carbon pager-Like fabrication. The r a t i n g of the blanb sheets was as follms:

I . 6

BA000077

*

I.

SOCIAL SECURITY NO. I I

- NO positive findings j I

PLANT NO. COMPANY NO. I 1 1

Badiologfst 's initials: - i Fig. 2. Hand x-ray survey ( Ins t i tu te of Industrfal Health).

1, Receive x-rays, Room L5l8, School of Public Heal th

2. Secretary, by typewriter, enters Social Security No., Plant no., and coipany NO. on a set of x-ray report forms. -

3. X-ray film plus white and yellow copy is sent to:

fnst?tute of Industrial H e a l t h Radiology OffLce ~ 6 4 0 University Hospital

I .

-? I

BAO 0 0 0 78

I

L 4. F i r s t Radiologist reads x-ray and checks his interpretatto= oq w h i t e copy of report form, i n f t l a l s the copy, and returns it and the x-rey to the Secretary, Room ~5630, University Hospital

5 . Second Radiologlst reads x-ray and checks hFs interpretat ion on yellar copy, i n i t i a l s the copy, and returns f t and the x-ray to the Secretary, Room w5630, University Hospital

6 . Secretary, Room ~5630, University Hospital, sends reports and x-rays t o Secretary, Room 1518, School. of Public Xealth

7. Secretary, by typewriter, consolidates the two reports.

Where identical, check X on report Where different, single diagonal slash

8. Secretary sends pMc cupy back t o the plant with x-rays

9. Blue copy is retsfned far, our f l les

Thus, each hand x-ray was read independently by Doctors Whitehouse and Ditchek, whose findings were brought together on a green sheet indicating whether there was concurrence or discordance between the two readers f o r each defect involving each finger.

To serve as a control population, Dr . Horace J. Dodge, Professor of Epj.~miology, School of Public Realtb, The University of Michigan, made available t o t h e investigators 2407 hand x-rays previously obtained as p a r t of continuing epidemiologic studies in Tecmeh, Michigan, a town near Ann Arbor. Since the PVC plant papulation was essentially a male papulation, the x-rays sanrpled from the Tecumseh Study p a p were U e s over the age of 18. proach made avaiJ.able insfghts in to the normal e s t r i b u t i o n of l e s ions that might be faund in an unexposed population.

T h i s ap-

Inquiring i n t o the work h i s t o r y of personnel i n the WC industry soon revealed t h a t c0nsiderabI.e job mobility was the general rule. duction practices, economfc conditions, plant leyaut, the existence of other product l ines, as w e l l a s normal turnover markedly influenced the job experience of each worker in t h i s .industry.

Seniority and pro-

Concerning the f i rs t of these, job seniortty, usually hourly employees enter at the lmest job ra t ings requiring l o w SUS and yielding relatively l o w pay, such as a reactor cleanfng Job o r bagging and shipping. able periods, as job openings developed, he may have the opportunity t o pro- gress; t o higher paying, mort skiUXd jobs.

. After vari-

0 The normal tendency f o r such in=-

8 ! '

BAO 0 0 079

Vidual is t o move i n t o one of several jobs x i t h i n the polymerizatlon area. For example, there are duties associated with reactor operation, viz, addition of catalysts and additives t o each batch, recovery of unreacted monomer and other unreactect ingredients, movement of charges to blend tanks , operations of dryers, e%. The period of t h e for advance from the lowest grade t o these higher grade jobs w o u l d depend upon e c o n d c conditions, piant turnover, and other factors determining job mobility within the plant.

To fb ther complicate t h e matter, practices varied from plant to plant as t o job assignments within the polymerization area SO t h a t there were varying combbations of duties assigned t o specific job t i t les . Some plant6 period- ically rotated employees through the various. j o b s i n the polymerization area. In mamy plants, t h i s rotat ion extended beyond the immediate polymerization area i n t o the drying operation. . .

I n v i e w of' the multitudinous combinations and divisions of duties that were fuund i n the 32 plants of the 19 corporations, each with i t s own l ace l job t i t le , a standard l i s t of 26 job t i t l e s - w a s devtsed for the purposes of this study. This l i s t took i n t o account s imilar i ty and dispar i ty of exposure t o materials i n combining various un i t operations under a single job t i t l e o r i n s e t t i n g up separate job t i t l e s . The list of these standard job t t t l e s is given i n Table I. _ .

C e r t a i n job categories require fu r the r comment. Additive Makeup (Job #1) and Catalyst Makeup (Job #3) duties w e r e combined as a full-time job in some plants,while i n other plants these specif ic tasks w e r e the responsibi l i ty of

-- react or operat o r s . _. ~ . . .

The bagger and the EhipFer usually worked in t h e same area; though performing different types of work they both were exposed t o similar concentrations of the finished polymer.

Workers engaged in compuunding and fabricating were involved in a wide Since no cases of variety of jobs, many with different types of exposures.

acroosteolysis have been found among t h i s group, information obtained on these operations during the surveys has not been included In the present analysis.

The pipe f i t t e r s , welders, mechanics, and others of t he maintenance department employees who must sgend a portion of t h e i r t i m e i n all areas of the plant, w e r e grouped as d i s t i n c t from electr ic ians and instrument technicians who were considered t o be less exposed.

Several local. job classif icat ions were used in most plants f o r persons i n the mOnOmer production areas. t he same materials i n similar concentrations' a t a given plant, the one job t i t l e of monomer operator w a s pruvided in t he standard IIH classif icat ion. me monomer maintenance operator was l i s t e d separately as his exposure was believed t o be more intermittent with ?art of h i s time being spent i n the malntenance shop.

Since a l l hp these were exposed t o e s sen t i a l ly

9

I IllRl 111 I! 1111 I1 11111 1111 111 Ill BA000080

c

-= z3 B

(9

TABU I N

STANDARD I M JOB TITLES Y)

m * N

m CI code - NO.

01 02 03 04 05 06 07 08 09 10 l l 12 13 14 15 16 17 18 19 20 21 22

(Water purification, treatment, etc.) 23 24 25 30

Additive (other than catalyst) Makeup Bagging and Shipping Operator Catalyst Makeup Man Clerical, Other Office, Staff Companding and Fabricating Dryer Operator General Laborer, C u s t o d f a l Laboratory Supervisor Laboratory Technician Maintenance, Electr ical MaLntenance, Mechanical Monomer Maintenance Operator Monomer Operator Monomer Recovery Operator Monomer Supervisor Plant Guard, Security Officer Reactor Cleaner Reactor Cleaner, Solvent Reactor Operator Slurqf mender Supervisor, Polymer Tank Farm Operator U t i l i t i e s Area Operator and Supervisor

Warehouse Operator pansf er Operat o r Other than WC

I

- The mozIQmer recovery operation re fers t o recovery from the palymerization

In some plants react ion and does not constl tute a part o f monomer production.

operator.

various products from Me area to another, using power trucks, fork mts, and other equipment for t h i s purpose. these workers for use where they were not located in any one area with i t s own classiffcat ion.

- it was a full-time! job, i n others it is a p a r t i a l responsibi l i ty of the reactor

In a number of plants, workers were engaged pr incipal ly i n t ransfer of

A separate c lass i f ica t ion was se t up for

;

I

10

d

- _

BA000081

MBny polyvinyl chloride workers. +n multlproduct plants spent months or W W h3

y e a r s during the perlad of the study i n areas where there were no operations

personnel, often spent a portion of t h e i r weekly activity in bulldings In a involving vinyl. chloride or its polymers. Other workers, notably madntenance ' ,"

large plant t h a t were not involved with these materials. u H t i t le , "other than.l?VC," was set up to have a classification f o r such activi ty.

u W

The Standard IIR job

Some difficulties were experiencedin classiFyfng a number of the workers aCC0rdjsg to t& Standard In job t i t l e clsssificatlons but it sellred its purpose satisfactorily i n most instances and fitted i n t o the occupational pa t te rn w e l l .

The Job History Form (Fig, 3) vas designed t o be filLed Ln by t h e personnel department, or the employee himself. department number, t he Standard IIB Job tiCle could be ident i f ied and the corresponding code number entered by the coder.

From the local job t i t l e and the

*

Jantrary 2961 t o present

1, Job T i t l e from to

f ram to

2. El@. No. fr am to

from to

from to

f r o m to

from - t o

Mg. 3. Job h i s t o r y form (prior to pi lot study).

Where a given local 3ob t t t l e at a plant involved operations covered by tvo or more of the Standard IIH job t i t l e s , an estimate was made of the frac- t i o a of t3.me spent in each. each month tbat .a worker w a s engaged full time in various Jobs during the year,

Since the fractian Of one-twelfth WRS used for

I

ll

BAOOO 0 8 2 -- I I - - -

It was convenient to u s e t h i s same fraation of twelfths wherc h i s daily 'or weekly work was divided between bro or more jobs.

An example is given in Fig. 4 of the manner i n which the Job History Form w8s coded for a worker who had t he job t i t l e of Helper-Bagger €n Depart- ment 5661 from November 1964 through December 1966, with his duties including reeactor cleaning for two-twelfths of h f s tkne md bagging f o r ten-twelfths.

From To Job The Fraction -- 'Job T-Ltle Dept. Mo Yr Mo Yr Code No. Assimed

Helper-Bagger 5661 1l 64 12 66 3-7 02 02 10

Ng. 4. E x w e of job history card entries.

Thus the employee had on ly to f U aut the job title, department designa- tion and dates worked i n tha t job *om the starting t o completion i n terms of the month and year during the s t u d y period 1961-1966. The coders would be prawided wSth a table converting local job tf t les 8t each p h l t t o the proper I131 job code ( i n the above exanq?lc of Fig. 4, Jobs 17 and 02) and t he unit t i m e interval w a d be apportfoned to the time spent in carrying aut the dutfes of tke standard job t i t l e s on that par t icu lar job i n t ha t particular plmt.

4. In&striaL Hyglene Evaluation . ..

The indus t r ia l hygiene phase of the study was designed to obtain b f o n n a - t i a n on equipment used, procedures followed, JntLterials involved and names and descriptions of jobs in order to compare practices i n t he plants where cases have occurred with plants withcut cases of AOL,

. .

To obtain t h i s information withtn a reasonable t h e l i m i t and expendfture *of man-dqys required certain comprdrmises w i t h . i d e a l i n d u s t r i a l hygiene practices. measurement of exposure to tech material ic terms of time and amount with de- termination of the time-weighted average exposure was not feasLble.

A l i s t of materials used was obtained at each plent but a precise

Conduct of the u s u a l fndustr la l hygiene methods of evaluation of the exposure by analysis of the air l o r materials absorbed through the respiratory tract or the analysis of biological. samples such as the urine, blood, or breath to include a B J S S S ~ ~ ~ of materials a lso absorbed through the skLn w a s obviously impraceicable. basis af general observation, a knowledge of the physical, chemical, and physiological properties of the material, and discussion with technically in-

It was necessary t o estimate the extent of exposures on the

12 ! a

c

i.

I I I I I

1 ib

famed persons a t the plants. m b e r of haws per week and of the relative severity of the exposure to both the respFratory tract and the skin, it was believed that useful semiquantita- tive estimates could be obtained.

By setting up but three broad categories of the

Pr io r to u n d e M r @ a formal pi lo t study, a preliminary survey was msde . at a large vLnyl chloride polymer production plant t o become familiar with plane layuut, equipment, and operations.

Fol3.awlag this survey, a number of forms were prepared t o record the de- s ired information for transfer to 80-corumn punch cards.

A "Plant Sheet" was provided for the narrat-lve report based on a description of' plant processes and a surrey of the operations. coatactsd at 'tge-plant were placed on this form. s h m in F€g. 5 .

Names of persons The head of this sheet is

Name of' Plane D a t e of Survey

Location

Contacts at plant: Management -

Medical -Industrial Hygiene

Technical.

Safety

Operations described by

Description af plant processes, operatLons, general c h d s t r y involved, and maintenance procedures, according to year If changes since Jamzary 1961.

and materZals

F€g. 5. P l a n t sheet (1 o f ) .

A "Sob Sheet," Fig . 6, was set up to record the name of the local job title, the name of' the plant and its location, and the materials involved In this job. -6 for-notation of the years during the s t u d y geriod that the job t ik l e had been in effect. scription of Job, " In the event th i s description applied f o r only certain years rlurLng the study period. spondfng to the l o c a l job t i t l e 8s descrlbed was then entered. proTlided for individual materials, repeated six tines on this Job Gheet. each material to wh€ch the worker on th i s job was exposed, a number was entered and the name of the material. concentration of the material used in order t o &ve an idea of whether the

Following the locfd. job t i t le were the years 1961, -2, -3, -4, -5,

This series of years W a s a l s o placed after the heading, "De-

The canparable Standard I M stuw job t i t l e come- Headings were

For

Space was provided for the amount or

13 t .

BAO 0 0 0 84

.. Name of Plant Location L o c a l Job Title (1961, -2, -3, -4, -5, -6) Description of Job (1961,.-2, -3, -4, -5, -6)

Comparable to IIH Job T i t l e Fiumber of Materials

Material No. Amount or concentration Y e a r s i n comparable use (1961, -2, -33 -4, -5, -6)

._

Material No. Amount af concentratlon Y e a r s in comparable use (1961, -2, -3, -4, - 5 9 -6)

Matierial No. Amount or concentration Years in comparable use (1961, -2, -3, -4, -5, -6)

Material NO. Amount or concentration Years in comparable use (1961, -2, -3, -4, -5, -6)

Material No. . - - Amount or concentration years in canparable use (1961, -2, -3, -4, -5, -6)

Material No. Amount or concentration

- - -

%ears in comparable use (1961, -2, -3, -4, -5, -61

Fig. 6. Job sheet.

Along the right side of the sheet were! placed spaces far the code number and t h e punch card column numbers f o r the keypunch operator. sheet would use 39 columns of the Bo-~olumn card if six materials were listed. Addi t ional cards would be used for a greater number o f materials under the

The one job

@

BA000085

em CE)

same job t i t l e . A space was provided for t h e number of materials t o gtve an indication of whether there was mare than the one card f o r the local. job t i t l e . ”,

Y

N W m m

Information on the extent of t h e exposure t o each material was entered One sheet was used for each

material on each job f o r each plant. The comparable standard I I H Job ! N t l e w a s also entered on this sheet. at the plant, it was necessary merely to place a check mark i n the proper spgce a f t e r the notations on ekosu re as being negligible, moderate, o r appreciable by inhalation and s i m i l m degrees of exposure by the skin with the check mark befng placed under the column indicating the duration of exposure as being less than 1 hour, 1-8 hours, or more than 8 hours a week. Since the exposure might vary f r o m year t o year, headings were provided for three s e t s of such information and for the years dur-g which It applied. O n this sheet also, spaces for the code number and notations for the keypunch operator were provided along the right: side of the sheet. used f o r each Material Exposure Sheet.

on a Material Exposure Sheet, shown as FLg. 7.

For rapid notation whFle recordtng the data

A separate punch card was t o be

B, PHASE 11-PILOT RUN

The adequacy of the forms as designed was t es ted i n a pilot run of a plant c l i n f c a l and industrial. hygiene survey. of the study are described below:

Modtficatians i n the various aspects

The questions asked in the CL in fca l history appeared t o be clear to the Minor revision was made in Question 42 (Fig. 1) t o make it a more qloyee .

open ended question as it became apparent that old fractures might cozplicate the in te rpre ta t ion of the x-rays.

The recomnendod x-ray technique provlded high resolution, and good quality: pictures. luwing alterative x-ray procedure was permitted.

However, since same plants utilized portable x-ray units, the fol-

- ”For maximum bony detail and for optimum standardization of ’

f W fo r the survey, it would be help- if, as many ins t i tu t ions as possible w o u l d use non-screen f i l m and K o d a k Standard Ready-Pak as origbally recommended. St has been cal led t o uur a t ten t ion t h a t x- ray facZlf t ies available i n some plants are not suf f ic ien t ly pawer- fW. t o u t i l i z e this tecbdque.

0

“Cassette techniques may be used. Standard 10 x 12 casset tes with high-speed screens F d Blue-brand’film usually require 45-55 roTp and

15

BAOO 0 0 8 6

W A P r o j e c t

Mater ia l Ecposure Card ,

6-7 (. )

1-3 ( 4-5 ( 1 8-10 ( ) n:=( 1

H a w , of P l a n t Location W d Job Ti t l e Ccqarable t o IIH Job n t l e Material Years AppUcable (1961, -2, -3, -4, -5, -6)

a p O S U X = (per veek) Less than 1 hr 1-8'hr More than 8 hr

Ne@igLble by in- halati on -

Moderate by inha- la t ion -

Appreciable by Inhalatian -

Negligible by skin -

Modcrate by skin - Appreciable by

sktn - Years appllcable (1963, -2, -3,

negligible by in-

Moderate by Lnha-

Appreciable by

Negligible by

b - halation - la t ion - inhalation - skin - - -

- - - -

Moderate by skin - - - Appreciable by

I9egligfble by iw

Moderate by inha-

Appreciable by

Negligtble by

halatim - - la t ion - iahalation - - skin - -

Moderate by skin - - Appreciable by

Years applicable (1961, -2, -3, -4, -5, -6)

- -

- - S k k

. .. - .-

34(, I 35-36 ( 1

Fig 7. M a t e r i a l exposare sheet.

I .

16

BA000087

1.25 milliampere seconds. ampere seconds. The t h e trill thus vaqr depending on the available mflliamperage. Because of the marked variation i n accuracy of set- t ings i n smaller and older equipment, technical variations may be necessary. Before a large number of pat ients are processed i n the m y , €t is suggested t h a t a saapl.e f l l m b e submitted t o t h e local radiologic consultant t o e n w e that; adequate bony detail is being demonst rated.

Pa r speed screens w i l l r e q d r e 2.5 m i n i -

"Positioning of the hand should be done 60 that the axes of the fingers 1€e in the l O - l n c h dimension of the 10- x =-inch film holder, with palms and fingers flat, and with a i least a 1-inch margin between the edge of the f i lm and the fingertips. &re included i n this single exposure, with the thumbs ne8d.y approximated i n the cen t r a l area of the fllm."

Both hands

3. Work H€story Questionnaire

The structurfng and arrangement o f the proposed form proved 'too open ended and unstructured, ab IIH job ti.t;le and proportion of t i m e on that job, a new job history form was designed (Fig. 8).

Since insuff ic ient space had been provided f o r the coding

4, Industrial Eygiene Survey

Use of these farms in the piLot; stuw showed t h a t some duplication could be avotded by combining on the one sheet the information on jobs and materials prevlausly entered separately on the Job Sheet (Fig . 6 ) and the Material Sheet (Fig. 7). As shown in Fig. 9, the new Job and Material Sheet provided, on a single computer card, rrU. the b i t s of information f o r the one job t i t l e a t a given plant and up t o five materials with the extenk of the exposures t o them. The revised Job and Material Exposure Gheet w a s identi& to the pr-lor Job Sheet through the headings concerning the job. The new format then permitted entry of all of the desired inf&tion on each o f ftve materials. As the .Lnfomnat;ion was obtained at the plant, the name o f the material w a s entered, s-e d i g i t s indicating the two inclusive y e a r s were placed after the card column numbers 14-15. 1966. Xn spaces under headings representing the duration of inhalation and skin exposure were entered the numbers 1, 2, or 3, t o indicate whether the exposure was considered negligible, moderate, o r appreciable, A separate card, using up t o 2 l columns was t o be punched for each material a t each job.

. For example, 36 for exposure over the years 1963 through

m

Q)

rJ Q c 6 W

La

Ccoapsny

Plant Location Y)

--- Name of Employee Sac. Sec. KO,

Dates Inclusive Jan. 1961-Dec. 196t

From To Mo Year Mo Year

Do not Job Title

- -

On completion o f form, return t o Medical D&p%t&nt.

Flg. 8. Job history form (as 'revised after pilot study).

I I

18

I

BA000089

I

. . . - .

Nslne of Company P l s n t Location Local Job Tltle (1961, -2, -3, -4, -3, -6) Description of Job '

Ccmparable to I M Job T L t l e mmber of Mater ia ls

Meterial =-13 ( 1

Brposure 14-15 ( ) 16 ( ) 17 ( ) 18 ( ) 19( ) 2 0 ( 1 a ( ) u-13 ( 1

Esrposure 14-15 ( ) 16 ( ) 17 ( ) 18( 1 19 ( ) 20 ( ) 21 ( )

RMIunt Cancentrat ion

Ccncentrat ion Material Amount

Exposure on inhalatfon and to skfn: ciable (3).

Negligible (I), Moderate (2), Appre-

Fig. 9. Job and materid. exposure sheet ( of ).

BAO 0 0 0 90

C, SURVEY PROBLEMS ARISING IN PIE COURSE OF STUDY Lb

, - N m c 1. Cl in ica l Histories w

- 0, Y

These did not appear to pose any signif icant problems regardless of levels of l-lteracy.

2. X-Rays

There w e r e occasional films fn w h i c h variations in technique occurred. R a r e l y were repeat x-rays necessary. or failure t o include a l l fingers, required some repeat x-rays.

Failure t o ident i fy l e f t and r igh t hand,

.In sane cases, the radiologfsts c d d not determine f romthe i n i t f a l x-

Accordingly, €n 1- such cases the plants w e r e requested to obtain re-

Plants were

rays whether devfattons were due to a fracf5ure of e i the r recent or remote origin. peat x-rays, as w e l l a~ t o review the campany medical records and question the warker regard5ng any hand trauma history, remote o r recent. cautioned not t o give any indication to the employee as t o the f lnger or hand invciLved in the qpestionwg.

Through t h i s approach, trauma histories that had been overlooked and nedlcal .records that had not been consulted were found to confFrm that many of these x-ray d.eviations represented trauma. part the respondents i n diagnositc category 0-2 (see Sectfon N, El).

These x-rays made up i n large

3. Work History

Althaugh job t i t les peculiar t o each plant were provided by the management of that plant, employee respondents Prequent3y recorded job t i t l e s which were e i t h e r inconsistent with, or w e r e combinations of, t i t l e s reported.by supervision. vfduals and t h e i r work experience, To o b d a t e this, medical investigators asked that the job history forms be completed prior t o the plant visit, titles glven by the employee were then reviewed with the management and in- consistencies eliminated. t ions became a responsibility of the Lnvestigators.

This resulted i n considerable delay in c l a s s i f m g these ind€-

The

Codlng of job titles and proportioning of 'time frac-

The investigators end t h e plant representatives usually agreed on the

For example, at many of the plants the statement estfmete of the severity and duration of exposures t o materials, falls had to be avofded. was made that med were not allawed t o enter the reactor until vent i la t ion had

Certain p i t -

20

BAO 0 0 0 91

,b removed flammable vapor t es te r . The first graduation on the scale of the usual inst ru- ment o f this type indicates a concentration of 2$ of t h e lower explosive I fmi t . Since the LEI, of -1 chlor€de i s k$, a reading at the half point of the first graduation indicates a c9ncentration less than or equal t o 400 p e s per mil- Uon of Vi-l chloride, somewhat less than the current threshold limit value o? 500 parts per million. However, One of the companies believes that toacicolcgic evidence w a r r a n t s reducing the TLV t o as l i t t l e as 50 parts per miUion.9

of the V I W ~ cfilpride vapor as Shawn by a negative reading on the

I n the production of vinyl chloride and polypfnyl chloride and related

These were l i s t e d alphabetically and material code numbers copolymers, it was expected that many compounds w o u l d be used. number was 227. assigned. used in compounding and fabricating operations. Since it was f e l t t h a t few, if any, cases would be found i n these operations, the problem was handled by setting rxp 16 categories of compounding materfals, such as plastfcizers and pigments. qyiry i n t o the ccnnponents of the impl-icated category could be undertaken.

Actually, the

A greater problem was poaed by the many more hundreds of materials

i i S h d d cases be discovered i n these operations, then a *her in-

In additLon to materials used i n the prodnctlan o f t h e polymers and t o the fFnished polymers themselves, t h e reactor cleaners were exposed t o cer ta in intermediate materi82.s together w i t h unreacted momers and other campounds in the residue. pletely polymerized resh but azSo sQme partially polymerized product of varying: molecular weight. Atlly known, these materials were designated as p o l y d n y l chloride s~rapfngs, or scrspings o f the pertinent copo2ymer, and given a code number.

I I Xt is understood that; these scrapings include not only the con- @

As the actual carupasition o f taese screpings is not

I

BAO 0 0 0 9 2

.. .

A. GENERAL, O V T m

We a r e indebted t the I n s t i t u t e f o r Social Research for t h coding of t h e primarg data in 80-column form, and the. punching o f the majority of ~O-CO~LURI cards, as w e l l as providing t h e number sequence f o r respondent identity. wish t o express our appreciation t o Miss Joan Scheffler, I n s t i t u t e for Socfal Research, who provided r e l i ab le help and a id i n the design of coding procedures and was responsible f o r t h e i r execution.

we

SECTION I11

COMFuTAnONAL METHODS -

I.. Ident i f icat ion of Survey Respondents

While posi t ive respondent ident i f icat ion could be achieved by use of t he nine-digit Social Security number of each employee, the use of nine colums was unnecessary ror t h e present purposes of the study since cer ta inly there were fewer than 9999 respondents i n t he survey, including control papulation. Accordingly, when c x n i c a l history, t h e c o q l e t e d x-ray report, and t h e completed job h i s to ry form were assembled, they were transmitted t o t h e coders for numbering and coding. number and t h e plants were ident i f ied by t h e i r code l e t t e r s .

A t t h a l t i m e , each respondent was given a sequential

2. Quality Control of Coding

F o l k r i n g the assembly o f the code sheets, a one-in-f%ur sample was taken

By t h i s technique, it w a s found that there was an error of O.l$with- of the coded sheets and compared with t h e primary data coding. in this 25% sample.

forms transmitted t o

3. Card Punching

Fol&&g quality control, t h e code sheets were transmitted, t o t he key- . p u n a n g section forpunching and veriffcation.

B. DATA COE3vERsION TECHFIIQUES

1, Cl in ica l History

Each respondent's c r n i c a l his tory was contained on one 80-coiumn card, the 5047 cards' consti tuting Deck 1. Removal of individuals who had been em- p3.oyed af ter December 31, 1966 reduced Deck 1 t o 5OU. respondents as a study population.

1

22

.. .

BAO 0 0 0 93

2. X-Ray Examinations L If the x-ray was posit ive f o r an indiviclual, t he l a s t two columns of t he

preceding c l l n f c a l history indiceted 8 Zead-in t o a new card according t o x- ray findings. In this way, the number of these new cards of x-ray findings, Deck 2, was kept t o a I n I ~ i m a L size. h i s t o r i e s indicating finger involvement end 11 possible defects each required up t o three 8O-coL~mn cards for recording. tion, cards were punched only i n the case of abnormal x-rays.

This w 8 S especially important a s x-ray

For t he Tecumseh control popula-

3. Work History

The job h i s to ry i n accordance with the standard IIH Job code was recorded This large number of cards was required where on up to three 80-~01umn cards.

there were multiple job changes over the sti tuted Deck 3. dates, t he problem of time overlap had t o be resolved. lows: The f i r s t job i n any sequence for one respondent was given t h e -11 period of months from beginning through the end of that period. sequence f o r t h a t respondent had i t s beginning date advanced one month so t h a t it was no longer coincident with t h e ending date of the previous job. sequent job received t h e full span period, and so on. ner, it was believed tha t since t h e sequence of worker forms were in random order, this program would randomly dis t r ibute throuefiout this popuLation thfs

sfx-year study period. These con-

This w a s done as fol-

The next job

Because the job history fo rm (Fig. 8 ) requested inclusive

The sub- In this a l t e rna t ing man- e

matter of job period overlep. -

I. Materials and Exposure Data

It had been intended t o use a separate card f o r each materLal on each job as coded on t h e Job and Materiel Ekposure Sheet (Fig. 9). information had been obtained and recorded on these sheets, It was noted t h a t the 80-column card w a u l d permit punching of the codes on a job t i t l e and on sFX materials with extent o f exposures t o each on the one card. A new Job Material Exposure Sheet (Fig. 10) was prepared omitting t h e deSC~ptlOn of the l c c e l - job, as this was Fmpllcit i n the cornperable IJH job t i t le , and renumbering the information bi ts for t h e keypunch operator. This transfer o f the data f r o m each i n i t i a l Job and Material. m-osure Sheet for al.1 32 plants t o tbe final form fo r coding was done by t h e investigator (WAC) who had visited the greatest’ number of Qlants. of thg estimates of exposures i n the Ught o f accumulated experience.

A f t e r all of t h i s -

This operation also served a s au opportunity t o review a l l

From this Job and Materials EQbSuiSheet , code conversion was performed, followed by qual i ty control, punching, and verification. The date i n this set of cards became Deck 4.

The exposure data was coded f o r punching f n six columns for each material, %bee columns for the three t i m e in tervals of l e s s than 1 hour p e r week, 1 t o 8 hours, and more than 8 hours per week for exposure through t h e respiratory

BAO 0 0 0 94

... .

-.

Job and Material Ekposure Sheet { of )

Name of Company W

5-6 ( 1

p~an t ( b c a tion )

Local Job Tit le (1961,-2,-3,-4,-5,-6)

7 0

Comparable t o IIEI Job Tit le 8-9 ( 1

No. of Materials 10-11 ( )

* Years Inha l a t i on Bkin

- I. hr 1-8 hr 8 hr + -1 hr 1-8 h r 8 hr +

Fig. 10. Job and material exposure sheet.

tract and three columns for these same time intervals for exposure through the skin. moderate, and appreciable, coded as 1, 2 and 3, respectively.

For each of these there were three degrees *of severity: negligible,

As an exampLe, an expasure of l ess than I hour a week of appreciable degree through the respiratory tract and o f moderate degree through the sk in plus a negligible exposure th- both for more than 8 hours would be coded as

I ,

24

sham I U F i g . 11.

.. . -. Column Numbers

Severity of E x p O 6 U r e 1- 2 3 4 5 6 Appreciable X Moderate X Ne@ g ib l e X X

Fig. 11. Example of coding exposure data for a given material.

It-was desired t o combine duration and severity i n t o a single expression f o r dose. Obviously a dose of mre than 8 hours a week of appreciable severi ty would-be considered a major dose and a dose o f less than an hour a week of negligible seve r i ty would be a sl.i&t dose. duration and severity that a r e considered major, intermediate, and slight doses.

Table If shows the Carnbinatfons of

The dose is considered as of major import when there has been a severe arposure for a period of 1 t o 8 hours o r more per week and a l so when there has been a severe exposure for less than I hour per week combined with a moderate exposure for a period of 1 t o 8 hours or more per week. The dose i s considered as of i n t e r m 6 a t e import when a moderate exposure has csntinued over the ge- r i o d of l t o 8 hours or more pe r 'week o r when the eqosu re for less than l. hour per week is severe. negUgibLe only f o r any duration or moderate f o r less than one hour.

The dose was consfderea as slight when the exposure was

Obviously the relat ion between duration and severity 'h~ producing a phys- i o log ica l e f f ec t depends on 8 number of factors such as the physical and chem- i c a l properties of the material, i t s physiological action and the e f fec t under consideration. The conibinations of, duration and severity of exposures listed i n Table I1 as being considered "Major, It "Intermediate, I' o r I'Slight" appeared t o f i t most of the materials u t i l i z e d i n t h e industry.

d I.4 .

C. CO& STORAGE TECHNIQUES

1. Deck "Clean-Wp" and Disk Stcrrage

Prior t o computer storage of data, each Deck was run through t o check for punching errors. w e r e i den t i f i ed by computer print-out and checked against primary data sheets and repunched as necessary.

Those cards which were inconsistent in t h e i r punched format

In t h i s mrltler, clean decks were obtained for sub- 0

BAO 0 0 0 96

l- kL .

I I I

I I

I

I I@

I I

I

I I I

I 18 I r

TABLE XI. m m m E OF DOSE EXPXSSED AS SINGLE c m c m

(Duration of expormre i n hours per week)

. Major Dose Intermediate, Dose S l i g h t Dose More Less More

.than hr than than hr than than hr than Less More Less

I hr 8 hr I hr 8 h r 1hr 8 hr 0 0 2 2 0 1

1 0 0 0 3

2 0 2 1 2 0 2 0 0

0 3 I 0 2

0 1 0 3

0 3 0 1 0 0 0 1 0

0 3 0 2 3 1

3 3 2 0 0 0 1 0 0 3

1-8

note: 1 = negligible exposure; 2 = moderate exposure; 3 3 appreciable exposure. the computer as having been recorded are included i n this table .

Only those combLnations printed out by

sequent manipulation and data therein transferred t o computer f i l e storage. This was accompLfshed by disk storage on the IBbf 360-70, following which each card was filad i n a card image format for recall and m?UIfpUlRtiOn a s necessary.

2. Computation Manipula+ion

I n programming collations o r computations, data were recalled from f i l e The program was tes ted an8 NU for grint-mt. storage t o tape.

operations were carried out a t The University of Michigan Computing Center. Upon completion of the runs, data was returned t o f i l e storage as acquired or new card decks were punched utilizing collated matrices.

Most of these

D. STATISTICAL -OB OF COLLATIOI'?

C o m p l t e r f a c i l i t i e s were largely u t i l i zed to echieve col la t ion of data a s wel l ' as reduction to tabular form f o r further s t a t i s t i c a l analysSs.

Because this population appeared t o seearate itseLf i n t o groups t h a t f o r

A considerable reliance was placed upon ch i square various purposes could be considered as discontinuous, the simplest s t a t i s t i c a l techniques were ut i l ized. analysis between such groups. In a f e w places, standard measures of deviation from the means of distributions were employed. means of dis t r ibut ion were examined, there was a tendency t o d i s t r ibu te i tself iu a platykurt ic distribution w i t h varylng degrees of skewness aggesting at least two d i f f e ren t populations.

However, i n several cases where

.

26

. . I

..

BAOO 0 0 97

In order to determine whether or not the symptoms e l i d t e d tended to group 'c-

themselves i n clusters, a Factor Analysis was carried out.

The investigators w i s h to acknowledge the aid and guidance of K. E. Patno, ph.D., Professor of Biostat i s t ics , The University of Michigan, School of public Eealth, i n developing the s t a t i s t i c a l treatment of the data, and Miss B e v e r l y Beers f o r computer prognzmming data storage, collation, and manipulation in addition to program t e s t ing and deck clean-up.

_ .

. . - - . . . . .. -

- --

! '

I

BAO 0 0 0 98

I I I I

I

I I

I

I

I

I

I I

I

SECTIOR IV

RESULTS

A. F’OPUUTION CHARACTERISTICS

1, Toted. Population Studied in Polyvinyl Chloride

Industry

Tota l Man-Months Represented by This

papulation

No. of $ of T o t a l Respondent 8

Years

< 20 * 20-24

25-29 30-34 35-39 40-44 45- 49 50-54 -55-59

60 or more NO response

-. - = 9 = 330 = 720 = 725 = 804 = 793 613

= 477 = 266 = I27 = 87

= 0.2 = 6.7 = 14.6 = 14.7 = 16.3

= 13.7 = 16.1

= 9.7 = 5.4 = 2.6

excluded

~edian age range = 35-39 years Modal age range = 35-39 years Mean age = 35.8

\

3. Sex Distribution

Males = 4751 = 96.4s Females = lfl = 3.64 lo response = 82 = excluded

I

28

1

4. Race Distribution

P l a n t Population Plant Population Plant Population

34 S 21 T 220 U

V 255 tl 28

59 X* 141 Y* 198 49

100 Z

36 97 349

3 K L

N P e

164 A 472 B

. 135 C 0 74

643 E 46 F

G 522

M* 183 308 9

5. Distr ibut ton of Study PopuAatfon Among Plant-s. The subjects of t h i s study were employed i n 19 corporations, TEBI~~ of which w e r e clult iplant opera- t ions. Thus, 32 p l a n t s i n the United States and Canada and t h e i r employees were the subject of t h i s study. The dis t r ibu t ion Of the 50U. employees among t h e 32 plants are shown i n Table 111.

Plant Population

139 AA 275 BB 83 cc

DD 50 Fz 80

26 FF GG 32

HH* 9

pLAmT POpULATION

, E 69 R sL5

B. SIGNS AND SYMPTOMS IN TBE WC STUDY PO-TION

1. Definit ions

a. Symptam Numbers. Reference t o the self-c0nqleti.m questionnaire il- lustrated i n Fig. 12 U u s t r a t e s the quesaon asked and +he number associated with tht symptom at the right-hand edge of the questionnaire.

b. The number of persons responding t o any one question was not neces- s a r i l y consisten% with the number i n the plant population since some questions may not have been answered. I n all cases, however, these nonrespoases amounted to considerably less than 1% of the plant population studied.

was repl ied to in the affirmative. . Accordingly, t h i s accounts f o r the widely varyFng percentage (Table N) answering appropriately t o this question, i. e., noting p e s t i o n #25 affirmatively. However, €-h should be noted that the t o t a l

c. Qpestion 27. This question was to be answered only i f question #25

’ -

I 29

L

, . - - BAOOOlOO

.. .

Doc. Sec. Ra. EhJa ( 5-13 1

ll *Int' (3-41

To MmJpl TEE PO- QUESTIONS, P'W A CIRCLE AROUND TIfE Box WITH POUR

-. FCIR MAMPLE: HOfIMY IS THE DAY

R a n you noticed SWDER vttlteness, blanching, or paleness ai' the fingers

&your RICK!! XAXD

RIGBP HAND

fuvE YO7Rl m S OR FINGERS BEEN BOPEWED WITH ANY OP DE FOLI#XQlC:

Tingling or plas-and-needles feel- RIGW HAND

W a n n feeling i n bands or fingers RICm HAMI lVEsr 29 LEPTRAmD JYEs((N0I RIm HAM)

LEFT HAND C a l d feeling fa bmda or ffmgers

RIGHT HAND Frequent tenderness, soreness, or pain at fingertips m HAND

Fig. 2. Medical history questionnaire.

I

7

L A

A

.

H ~ * you noticed any of the fallawing changes of the skin of your hands, mists, or arms: . Thickening of the sktn *= @

*IO yes, _ - is it on JHANDsl GI 1 3 White, raised patches of skin .-]nsl

Whitish pimples *JyEsJ

Shiny skin patches *@ @

than.l/2 inch around (NOT including *m @

*If yes, is it on j7iEEf JEiii?‘~

*E yes, ts it on jwRIsTs-1

*If yes, i s it on -1 pRIsTs) rziq Small, raised areas, ar lumps less

warts) *X yes, is it on FJ j”F) lAREAsl

t

3b

-35

36

37

38

I

‘ 0 Ridges or cracking of the fingernails

Have you used hand held power t oo l s or air-dr€ven t o o l s at work Fn the

I 40

I

Eave yau every broken (or fractured) or crushed a finger? Ip yea, was .it tbe r s 1 or ]EJ hand?

42

Fig. l.2. (Concluded)

BAO 0 0102

0 0 0 P 0 W

I C 3 133 2.3 h 133 3*0 2 ll 18.2 4 133 3.0

47 6.4 4 5 80.0 12 47 25.3 -- 2 163 1.2 P 5 47 10.6 3 A J 163 0,6 0 163 -- 0 1

80 10.0 Es 6 79 7.6 3 79 3 4 8 9 88.9 8 BB 7 n l 2.6 , 7 272 2.8 2 g 22.2 36 269 13.4 cc 9 83 10.8 6 83 9.8 7 11 63.6 23 83 27-7 H 1 69 i , t I 69 1.4 2 3 6647 11 69 15.9

19 47.4 36 217 16.6 3 32 9.4

U 9 219 4.1 13 219 5.9 9 PQ 0 32 -- 0 32 -- 0 0 J 1 33 2.9 1 36 2.8 0 1 -- 2 36 5.6 Q 3 48 6.3 1 48 2.1 5 20 25.0 13 49 26.3 K 5 96 5.2 1 96 1.0 4 4 100.0 18 96 16.8

w P 3 57 5.3 3 59 5.1 2 11 18.2 7 % 12.1 n, B 37 w3 8.0 27 468 3.8 26 62 41.9 102 469 21.7

B 21' 643 3.3 u1 643 1.9 13 21 7L4 75 641 11.7 HH. 0. 9 -- 0 9 -- 0 0 -- 1 9 u.1 PF 2 23 8.0 1 26 3.8 0 3 -- 6 26 23.1 2 1 loo 1.0 0 99 -- 1 11 100.0 3 LOO 3.0

28 7.1. N 1 29 3.4 1 29 3.4 2 & 100.0 2 AA 8 136 5.9 7 139 5.0 3 13 23.1 25 139 18.0 . L 6 339 1.8 11 341 3.2 12 20 60.0 44 343 12.8 G 11 518 2.1 8 519 1.5 10 15 66.7 26 3.18 5.0 T 0 21 -- 0 , R 4 114 345 3 114 2.6 3 5 60,O 16 114 14.0 DD 0 50 -- 0 % -- 0 0 -- 2 50 4.0

....

D 2 74 2.7 0 74 -- 1 6 16.7 6 74 8.1

20 -- 21 -- 0 0 -- Y a8 lg 4.1 b 197 2.0 1 10 10.0 37 1Yl 18.8 X 1 141 0.7 2 141 1.4 1 1 100.0 18 140 12.9 W 3 254 1.2 6 294 2.4 6 7 85.7 36 254 14.2 H 3 181 1.7 3 182 1.6 1 3 33.3 13 180 7.2

308 1.0 4 6 66.7 30 300 9.7 -- 3 33 9.1 V 3 308 1.0 3

620 4968 12.5 TOTAL 168 kgi 3.3 131 4985 2.6 131 279 47.0 -- 12.4 -e -- -- -- 3.4 -- -I 42.7 -- 3*3 -- -- -- 2.1 -- 36.3 I- -- 7.4 3.0

- g + BD 6E -- -- 0 . 9 -- 0.37 -- -- *Indlcrter o n 4 those responding appropriately t o quaation. 6 1 6 1 1 0 1 0 6 0 UdA

' 6 0 34 -- 0 31 -- 0 0

-.. 6-41 -- -- 1.3

I

m P 0 0 0 P 0 I@

w u

W IV (Continued)

C P A EE BB cc

U

J 0 K D P B E HH FB z N M L Q T R DD Y X w M V

n

aa

1 3

4 12 6 2

19

--

-- -_ 1 1 1 4 32 7 1 2 1

. 2 5

13 4

10

l o 6 5 5 14

-- --

133 0.8 46 10.9

163 -- 80 8.1

rll 3.3 83 13.3 69 11.6

220 10.0 32 -- 35 -- IrO 8-3 96 9.4 74 1.1) 59 6.0

W .9*8 643 5.0

g u.1 26 11.5 100 -- 29 --

137 0.8 344 7.6 319 2*3 21 --

113 9.7 50 2.0 ig 6.6 ill0 3.6 231) 6.7 180 5.0 307 5.2

10 4 6

44 23 . 14 35 2 1

J.2 19 10 7

101 68 1 6 3 2 24 42 29

18 3 48 20 32 13 34

--

47 16-69 80

270 82 69

220 32 36 49 96 74 59

471 643 9 26

100 29

138 344 519 20

114 YJ Irn 139 254 180 3al

4.5 10 21.3 12 2.3 8 7.3 9

16.3 49 28.0 17 20.3 17 15.0 49 6.3 2 2.8 2

2b.5 9 19.8 9 133.5 11 u. 9 10 21.4 119. 10.6 87 11.1 1 23.1 7 3.0 2 6-9 2

17.4 33 l2-2 . 51 5.6 33 ..- 1

15.8 18 6.0 9

24.4 19 14.5 24 U.6 32 7 *2 11

11.1 2.7

133 7.5 47 25.5

163 4.9 80 11.2

271 18.1 83 20.3 69 24.8

220 22.3 32 6.3 35 5.7 49 18.4 96 9.1, 74 1b.9 59 16.9

k'0 25.3 642 13.6

g 11.1 26 26.9

100 2.0 29 6.9 137 24.1 345 14.8 519 6.4 21 4.8

114 15.8 N 18.0

1% 9.6 141 17.0 254 l2.8 179 6.1 306 8.8

I

i

m P 0 0 0 I-J 0 cn

c 5 P 6 A PIC 6 RE 13 cc 5 H 2 U 13 w J 1

2 Q IS D 2 P 1 B 31- E 23 mi -- PP ..- z 1 n Ah 8 L 16 0 4 T -- B 6 DD 2 Y 9 x 6 n 5 I( 6 V 10

--

--

--

c,.

133 3.8 47 12.8 163 -- eo 7.5

g 0 8.8 83 6.0 69 2.9

220 6.0 32 -- 36 2.0 49 4*1 96 -- 74 2.7 59 1.7 469 6.6 642 3.6 9 -- 26 -- 100 1.0 29 -- 138 518 343 4.7 519 0.6 21 -- 114 3.3

!jO 4.0 1!+3 4.5 140 4.3 254 2.0 179 3.4 306 3.3

134 2.2 2 47 14.9 3 163 -- -- 77 3.9 b q o 4.1 . 13 82 3-7 2 69 7.2 3 220 9.1 lo

32 3.1 1 3S 2.9 49 -- 2 m 1.0 a-

'14 2.7 3 59 1.7 3 470 5.1 31 613 197 9 g 11.1 c-

26 -- 1 100 -- 2 29 -- -- 138 2.9 8 342 ,2*9 18 516 1.4 9

113 8.0 - 6 9 4.0 1 1g 1.0 4 140 -- 1 2% 1.6 3 180 0.6 3 308 1.6 6

-- 21 --

132 I7 163 79

rll e2 69

219 32 35 49 ¶ 7', 59

468 643 9 26 1Gt 29 138 341 514 21

113 3 196 140 254 180 308

3

7 22 8 3 19 1 1 3

4 4

68 22 2 1

1

--

--

-- 7 23 13 1 11 1 3 1 8 4 6

47 6.4 163 -- 78 9.0

q 3 8*1 4 9.9 69 4.3

220 8.6 312 3.1 39 2.9 49 6.1 m -- 74 5.4 39 6.8 467 14.6 643 3.4

9 22.2 26 3.8 loo -- 29 3.4 137 5.1 342 6.7 515 2.5 21 4.8 114 9.6

50 2.0 197 1.5 141 0.7 2% 3.1 180 2.2 308 1.9

I

l I

TABLE IV (Concluded)

I Bymptorn Number 37 TotLl Tokl Total Total

40 3839 $ + $ + Total + $ + Total + Re',

-- 44 -- 4 46 8.7

76 13.2

Total i R6(Ip, $ + Total t 130 5.6 1 132 0.8 6 134 4.5 133 0.8 3

47 2.1 6 163 -- -- 163 -- -- 163 -- 78 694 5 79 6.3 4 76 2.3

46 13.0 -- 163 -- 10

C P A Ex BB cc H U 0 J Q K

P B E nH gP z H AA L

T R DD Y X W

Y

n

a

n

PI1 82 69 220 32 35 49

I97 74 !%J

470 642 9 26 100 20

138 342 5.16 21

115 50 196 141 2% 1-77 30s

2.6 25 3.7 l2 2.9 6 2.7 18

2.9 1 2.0 5

3 2.1 7 3-4 . 7 5.7 37 0-5 30

3 -- 1

-- ..-

--

-- -- -- -- -" 2.9 9 2.0 23 0.8 18

4.4 la 2.0 3

9 7

0.4 15 1.1 3 1.6 13

-- -..

-- ..I

218 9.2

69 8*7 219 8.2 31 --

' 34 2.9 49 10.2 97 3*1 74 9.3 59 11.9 467 7.9 643 4.7

9 33*3 26 3-8 99 -- 29 -- 138 6.5 341. 7.3 914 3-5 21 -- 114 8.8

50 6.0 188 4.8 141 5.0 2% 5.9 177 1.7 303 4.3

d 11.6 2% 79 67

209 30 35 48 96 74 54 456 638

9 26 99 26

123 324 506 21 108 48 192 140 254 172 2%

c

29 11 2 40 1 1 7 4 s 6 41 45 2 1 1 1 26 36 15

20 4 9 11 23 11 22

--

. 265 10.9 77 14.3 6'7 3.0 218 -18.3 30 3-3 35 2.9 46 15.2 . g6 4.2 73 6.8 59 10.2 459 8.9 640 7.0

9 22.2 I 26 3.8 I

gg 1.0 28 3-5 134 19.4 331 10.9 !W 3.0 21 -- 110 18.2 50 8.0 192 4.1 141 7.6 2% 9.1 173 6.4 302 7.3

-= ZA. en B

mrmber of indiHduals ansvsring this question appropriately is very low.

category 3 in the absence of s ignif icant x-ray abnormalities.

mosZ C D . c H

4 +

answering questions #25 and #!27 i n the affirmative constitute diagnostic

2. Results-Table IV-

fie r e l a t ive ly large values f o r Stan- Deviation of the Me- (Table V) seen for each symptom suggests t h a t w e are probably dealing with a non- homogeneous population whose distflbution of characteristics do not f i t a Poisson Distribution. That t h i s i b l reasonable stems from the observation t h a t all plants, ere not affecte’d by t h e presence of AOL. when t h e mean percentage of posi t ive answers in the study poptitation and. t h Stan- Deviations and Standard Errors oi the @ans are tabulated, Table V.

This becomes apparent

Symptaa Mean Percentage +.SIC- NO. of Occurrence

25 3 -4 3.1 0.54 26 2.4 2.1 0.37 27 42.7 36.3 6.42

29 4.0 3-2 0.57 30 5.6 4.3 0.76 31 12.6 7.5 1-33

34 . 3.2 3.5 0.62 35 2.9 2.2 0.33 36 * 5.2 4.6 0.81

s 6.6 6.3 1.10

40 8.3 5.8 1.02

20 12.4 7.4 1.30

32 13.6 7.2 1.28 33 3.5 2.9 0.51

37 1-7 1.8 0.31

39 2.9 2.9 0.52

Accordingly, statistical analysis of these data is being perfonned uti- l izing a negative binomial. However, C h i Square Analysis can be used to study symptan occurrence a t such a point in t h i s study as the two-or possibly three-different populations can be defined on the basis cf plants with AOL. plants without AOL, end plants with possible AOL (see Section V on Synthesis af Symptm8, Job .Histom, and X-R8.y Findings).

I 36

3. Factor Analysis

To determine possible clusters of symptoms recorded .Ln the medical his- tary, Ye used a Factor A n a l y s i s Program from the Bio-Medical Series (BMD O3M). on the basis of correlations between pairs of' variables, a Factor Analysis separates the variables, if possible, i n t o grcryps o f variables whfch tend t o be answered consistently by the respondents. Thus, a Factor Analysis can be used as 8 check on our p r i o r expectations about which symptoms occur together, or as an indication of s y m p t a groups we might be UllarJare of. Variables are associated w i t h factors by scores called 'loadings," which mfght be used as weights in a future analysis t o construct variables which represent factors, L.e., r a the r than using a l l of the original variables. m y , questions could be designed using the Factors as a guide. really a group of variables w h i c h seem t o be associated wi th a par t icu lar dimension of the problem so that perhaps one question could be asked Frz regard to that dimension instead of several questions.

S h U a r l y , i n a future A Factor i s

Results of Factor Analysis:

VarFables not loading significaxtly on any Faceors:

questions #u, 13, 34, 15, 16, 41. Factor l-Unusual sensations, hands or fingers

&uest€ons #$43, 44, 46

Factor 3-Particular color changes of f ingers

symptom #25, 26, 27

Factor b-Physicd chan'ges in skin o f hands, arms, wrists

Factor 5-Abnormalities of the f ingerna i l s

The most s ignif icant of these clusterfngs was determined t o be Factor 3 following the Facto? Analysis. A l l of these syraptams in Factor 3 were required f o r c la s s i f i ca t ion as a possible case of AOL on the basis of symptoms alone, on the expectation t h a t respondents would report them as occurring together.

I . . .

37

BAOO 010 8

i i I

I

4. Ccmnnents Concerning Nonspecific Nature o f Symptoms Studied

-c

U

Q

a hL (D

...

Y

N

4

It can be seen f rom Table V that several of the symptoms occur i n . a - W

relatively high percentage of the study group. 'bins and needies" feellng) , #31 (numbness, fingers fa l l asleep), and #32 (stiffness, Fullness, or tightness of fingers) have a relatively high percentage of pqsitlve responses. ccnmcm occurrence that they are probably nonspecific. Accordingly, use of these specific questions to screen .for AOL would probably be &appropriate. !J!he same impression probably applies to use of question #40 (rldges o r cracking of the fingernails) though to a somewhat lesser degree.

6ymptoms #28 (tingling,

This suggests that these symptcons are of such

By contrast, Symptoms 825 (sudden whiteness or blanching of the fingers), #26 (sudden redness ar blueness of the fingers), #33 (frequent soreness, f e d m e s s , or pain at fingertips), #34 (thickening of the skin), #35 (whtte, raised patches), #37 (shiny s k h patches), and #39 (pits or scars of t h e fingernails) occur with such law frequency that their occurrence may be of some significance.

This rank-ordering of the occurrence of these symptom suggests three classes of symptoms as possible indicators of the presence of AOL in a plant pqpuletian.

TABLE V I

RANK ORDEBlNG OF FREQUENCY OF POSITIVE SYMPTOMS IN Pvc STUDY POPULATION

I

I.

I

a I

+sD ,m Frequency i n Symptom Mean of. Occurrence - PcrpuLatfon No.

LOW 37 26 35 39 34 25 33

1.7 1.8 2.4 2.1 2.9 2.2 2-9 2.9 3.2 3.5 3.4 2.1 3.5 2.9

4.0 3.2 5.2 4.6 3.6 4.3 6.6 6-3

. 40 . 0.3 5.8 12.4 7.4 28

31 12.6 7.5

K i g h

0.31

0.38

0.62

0.50

0-57 0 .a1 0.76 1.10

1-02 1.30 1-33

0.37

0.52

0-3'7

13.6 7.2 1.28 32 I .

38

__ . __ . . - , - - . . . . .

_ . -

- i - ,- - - .

c. FmD X-RAYS RESULTS

I. Total. Defects In PVC and Control Populations - TABLE

OF X-RAY DEFECTS AM) PERCENT OF PLANT FOETEATIOR W I T H SUCH MORMALITIES I N C O W O L AND W C POPULATIONS

Control PVC Population

No.

Defect No.* N = 2407 N = 50U 4

44 1.83 321 6. hl

6 0.25 40 0.80

1 0.70

0 -- 21 0.42 2 0.08 9 0.18

2 5 0.21 35 3 4 5

1 0,004 15 6 7 8 ' 4 0.17 43 9 10

0.30 1.26 0.86

2 0.09 40 0.80

~1 0.46 63

1 0.004 7 0.14

Total defects 76 594

defect3 In

*See Fig. 13. papulstion 3.16 . 11-85

- Comment: Although each defect occurs significantly more frequently i n the M o s e d PVC population than i n . the control ( p < O.Ol), it should be pointed out that these differences must-be t akenwi th reservation. ologists reading these films made their interpretat ions separately and w t t h - aut ident i fyhg the source of these filmsy the difference i n markings on t he Tecuarseh C o n t r o l Population ccruld have resulted .Ln a subtle bias. The dip- . .

ferences are particularly noteworthy i n the case of the xulnimum defect ( L e . , defect #ly Marginal Defect) where appraximatelY faur times as mazly of these defects were seen i n the study papulation than Ln the control papulation (corrected for sample size d€fference). Since there w e r e no cases of AOL in the control m, this difference i n defect #1. Occurrence can only be accounted for Ln part by def in i te or suspicious cases of AOL i n the PVC worker popula-

Although the radi-

tion.

I . . 39

BAOOOllO

SOCIALSECURITY No, . PLANT No. COMPANY NO. - n u NO positive findings I 1

Defect How 1.

2.

3.

4.

5.

6.

7 -

8.

9-

10 . ll..

Badlalogist 's initials:

Pfg. 13. Hand x-ray 8urvey (Institute of Industrial Health).

2. Agreement of Radiological Interpretation, PVC Populatton

Table VI11 demonstrates the extent of M8cordance between x-ray interpretations as w e l l as indicatiag the extent of the abnormalitfes seen in the W C worktng population, found to have an x-ray abnormality. specific defects (listed as 'bther osseous disease, l c rv are concerned only with defects numbers 1 t o 10 involving 595 x-ray abnor- melities.

Out of this total population af 5Ol.l workers, 1302 were Wh5J.e this latter number includes non-

#U) the anlrryses that fol-

BAOOOlll

TAm

?

1

AM) NUMBEX OF FINGERS SR~NIZG X-RAY DEFECTS IS 1302= wc X-RAYS

3 30.0

b

Agree 3 0 - 0 o c 3.0 isa agree & Total 2 9 1 -

1 0 - 1 - 5

33 -3

1287 13 *3

1262 12.5

B Totals A 57 g 2 4 0 2 0 1 2 I 78

I) ~ ~ ~ 8 . 1 ~ ~ ~ 2 A x

vIlhe difference between Eese two f i g u r e s end f r o m those in the Defect all Tota l k80 17 12 7 11 3 5 Ir 2 595+

' 0 ategory is not shown here.

.

41 1 . D

BAO 0 0 112

-. e n N

c1 N W

Comments: Noteworthy i s the grea tes t degree o f agreement as regards defect #3, i . e . , "transverse defect," while defects #2, i.e., ''Umrginal de- feet w i t h residual fragment," and #7, i.e., "shortening" produces S o I l l e W h a t

less agreement. However, t h e differences in percentage agreement between these three defects is not s t a t f s t i c a l l y significant. that meanlagArl comparfson between readers cannot be made.

Defect #5 occurred so r a re ly

A t the other extreme, the radiologis ts f a d agreement most difficult i n attempting t o di+gnose phalangeal tuft widening and soft tissue flubbing. (There are t o o f e w defects in #b Category t o comment.) relatively SU number of these defects, these differences are not statisti- cally valid.

However, because of the

While there is a large number of defect j#l readings, Le., marginal defect, comparison with the agreement ratio between readers fo r defect #3 Fndi- cates a s t a t i s t f c d l y Significant greater concurrence between readers for $he l a t te r defect,

3. R e l a t i v e Kumber of Fingers Involved Among 595 X-Ray Abnormalities i n pvc Workers

Extent of lesions in F'VC workers, Table VIfI above, Fndicates the number. @ of fingers involved by x-ray changes which might be associated with AOL. Study of the t o t a l s permits constraction of Table M, percentage of f ingers involved among Pm: watkers .

.. . _ .

TABLE Ix

TOTAL NUMBSI OF FINGERS INVOLVED, BY PERCENTAGE

No. of $ of T o t a l F'inners IhvoLved Finger Involvement

1 8007 2 9.0 3 2-9 4 . 2.1 5 1.2 6 1.8 7 0-5

0.8 a 9 0.7 10 or3

loo. 0

I

. .

42

BAO 0 0113

W

Comment: It should be noted t h a t item 3 considers o n l y defects #l thraugh 2 #lo, w h i l e the 1302 x-ray abnormalities is a t o t d of a l l defects. Thua w W e t h e t o t a l number of defects #l through #lO equals 595, the balance, or 707 defects, are the miscellaneous group classified under "others" (see the "x-ray

one person with multiple defects, the numbers c i ted here do not represent individuals, rather the t o t a l abnormal x-ray count, i.e., 1302 abnormal x-rays,

L Y

I4

method" section). Since.many of the defects l i s t e d here represent more than e

4. Occurrence of Abnormal X-Ray by Defect; and Plant

See Table X-Specific X-Ray Defect by Plant. Except for defect #l, there is relatively small nunibem of x-ray abnormalities. Accordingly, deterndna- t ion of Standard Dedations and Stan-d Errors are of l i t t l e u t i l i ty . Also, w h i l e nrany of t he plants have smal l s t u d y populafions (3.. e., less than 100 workers) t h e percentage occurrence for defect #l at such locations has l i t t l e meaning. occurrence are rclatively insignificant . Likewise, since defects fi and #9 occur infrequently, percentages of

Analysis suggests that where defect #3 occurs a t the l$ or more level of

This, however, i s o n l y relevant when plant popu la t ion frequency, or defect #4 a t the 0.85 level of frequency, that suspfcion might be cast oa such p h t s . is greater than 100 workers, e 5. Occurrence of X-Ray Defects by Corporation

See Table XI.

- 6. Ccmrparison of X-Ray Defect Occurrence Between Tecumseh Control PopUation and Plants Performing M y Compmang

See Table X I L

Comments: - -

Defect #l occurs mme frequently i n the compounding plants (p < 0.025) than i n the Tecumseh control. (Table VIII) that the degree of agreement f m d f o r the marginal defect x-ray interpretation was cons-iderably l ess than for others. quent G c m e n c e o? defects #2 through #9, s t a t i s t i c a l tests cannot demonstrate meadngful differences between these two populations.

Hkever, it should be recalled

Because of the infre-

43

BAO 0 0 114

i

24BU X t . BPECIPXC X-RAY DFPECT BY PUNT

47.2 135 74 643 46 522 69 36 97

349 183 28 59 49 115 34 21 220 3m 253 141 19.98 100 133 275 83 50 80 26 32 9

16 9.8 '41 12 2 48 '1 65

5 1 9 19 3 2 6 1 12 2

LO 7 5 4 9 3 2 17 4 5 8

-.

-- e-

l -

8.7 8.8 2 -7 '1.5 e.1

12.5 5.8 2.6 9 a 3

5.4 2 -7 6.9 10.2 2 .o 10.5 3.9

4.5 2 *3 2 .o

4.5 3 .0 1.4 6.2 4.8 10.0 10.0

--

2.8

-- -- ll.1 -

1 0.6

Total 5011 321 '15 k0 21 9 15 65 43 7 *Plants known tp have cases o f AOLb **Percentage o f plant population. (C) Compounding operatlono only.

TABU X I

OCCURRENCE OF X-RAY DETECTS BY CORPORATION

X-ray Defect No.

Cow. Population 1 2 3 4 5 6 7 8 9 10 Total lncidecce A l l Defects

100 Workers 2 2 2 13 *5

02 Is93 169 12 16 6 2 LI 26 31 3 17 293 15.5 01 a 4 1 6 - 1 - - - 3 - -

03 105 5 2 3 I - - 1 1 - - 13 32.4 04 97 9 3 3 1 - 2 6 1 2 2 29 30.0

23 4 .,3, 1 I - 3 - - - 35 6.6 4 13.8

05 532 06 29 2 1 - 1 - - - ' - - - 0 8 . 59 6 - - - 1 - - - - - 7 11.9 Q9 49 1 2.0 11 u 4 l 2 2 2 - - - 4 - - 1 23. 18.4 12 34 2 1 1 - - - - - - - 4 11.8 13 23. - - - - - - - - - e - - I4 220 1 0 1 2 3 1 1 5 - - 4 27 3.2 03 15 3- 7 1 - - - - - - - 1 9 2 *9 17 593 1.8 4 5 2 - - 3 4 - 7 43 7-3 l a 100 3 1 1 1 - - 4 2 1 3 16 l-6 .O rg 413 19 3 2 3 - - 3 3 - 1 3 4 8.2 20 83 4 - - 1 - 1 1 - - - 7 8.4 21 130 1 3 - I 1 3 - 1 I - 2 22 6.9 23 67 1 - - - 3 . - 2 - 1 - 5 7.5

Totals 50U 320 35 40 21 9 15 62 43 7 40 592 u.3 SD = k6.0

1 . - - - - - -

TABLE XII

OCCURRarCE OF SPECDIC X-RAY DEFECTS IN CONTROL POPULATION AND IFI AT PLAMTS wREI(EILp ONLY COMRXNDING OPEZtATIONS ARE FZRFORMED

X-ray Defect No. 1 2 3 4 3 6 7 ' 8 9 Population

Plants Compounding ~

Pvc only rgw 5 3 1 1 0 '1 2 1 556

Control Population 44 5 6 0 2 1 1 1 1 r l 2407 9 < 0.025

BAO 0 0 117

I.. .-1

c' - = W

Q

D. &CUPA'I!IONAL EXPEXIEIJCE: OF PVC STUDY POPULATION N

m

h,

00 4

L

1. General Characterist ics '9

AB expected, workers usually had had job experience within t h e pa r t i c i - 1) The normaL job and. pay progression with pa t ing companies as a result of:

increasing seniority, and 2) changing production needs Over a t ime! course.

Job progresston i n WC synthesis operations starts a worker a t the lowest paying job levels, f .e- , bagging and shipping and/or reactor cleanfng. senfar i ty is accumulated, higher paying, r e l a t ive ly m r e s k i l l e d jobs became available. t h e s i s areatae.g., i n t o a craft ar t rade category, as maintenance man, labora- t o r y assis tant , etc., or promotion may lead t o more s k i l l e d dut ies i n the sya- thesis area of the operation, e.g., reactor operator, d r i e r operator, monomer recovery area, etc. To fur-ther complicate analysis, no% isfrequent ly a worker may perform several of these tasks at any one time, div5ding h i s time between several such tasks in the course of the day's duties.

As

These job advancements may also take the employee out of the syn-

The other source of job experience variation, ie., changing production needs, s t e m s from the economics of the market place. If production requirements temporarily diminish, t he employee may be sh i r t ed t o another job, "bumped" t o a lower paying job, or changed t o an equal ly w e l l paying job elsewhere h the plant. temporary "layoffs" may supervene.

At; the lowest job grade, during such slack periods,

Given these sburces of job variation, no consistent job t i t l e can be at tached t o any one subject in order t o develop the usual prevalence data. Accordingly, the approaches u t i l i z e --months as the pertinent denominator obtained In t h i s s t u d y . The number of jobs represented by t h i s population was 10,843. _ _

2. Work Experience by Plant and Job --Months

See Table XIII. For job code numbers see Fig. 14, p. 56.

Comments: Those jobs having the longest mean man-months, i.e., c l e r i c a l or staff' (4), securi ty of f icer (16), laboratory worker (8,9), compounding or fabricating ( 5 ) represent d i f fe ren t populations from the r e s t of the production workers. little mobility w i t h i n the plant, or else-as in the case of t he last two categories-represent special technical skills. In addition, t he crnnpounding job category contains large increments of man-month experience contributed t o by those plants in which t t i s type of operation is the sale manufacturing process.

The first two of these four groups represent jobs which have

46 I

e

0-

Q)

w m N m c. Iu '0 QI Q)

a ...

TABLE XU:

JrlB EXPERIENCE BY PLANT It? Ti2 OF "MBE3 OF RESPOND~TS REPORTING S P E I E C JOB EXPZUENCE, TOTAL AND MEAN MAR-MOEITXS IN SPSCITIC JOB

Job NO. 1 Job No. 2 Job No. 3 P l m t Total Man Mean Man T o t a l Man Mean M a n T o t a l Man M e a n an

Resp. Months Months Pesp. Months Months Resp. Months Months

A B C D E F a G . H J IC L M N P Q R S

U V W

2 ,

- x Y Z AA BB cc DD EE FJ? GG

-

3.8 2.3 2.0

12.0 3-7

4.0 1.0 1;7 6.7

-- --

-- 1.0 2-9 1.6 3-9 4*3 4.7 - 6.2 4.4 3.6

6-7 9.3

' 6.6 2-8

-- --

40 212 47 31 37 16

2 5 18 4

7 31 7 38 4 3

32 112 63

--

--

-- -- 15 2l 56 L5

- 8 34 -- 2

30 6 1

2 12

14 7 6

25

7 7

16 25 14 7

46 93 40

--

--

--

-- -- 13 20 42 14 la 2

lo --

101 35 7

24 42

--

-- 56 a 10 104

7 20 26 98 59 33 243 346 74

--

-- -- 125 97 261 53 18 3 -- lo

3.4

7.0

L2.0 3-5

4.0 1.1 1.7 I

4.2

1.0

5.8

--

--

-- 2- 9 1.6 3.9 ._ . . 4.2 4.7 5.3 3 *? 1.9

9.6 4.9 6.2 3.0 1.0 1-5

1.0

-- --

-- --

4. $I 14.77 3.90 M e a n Man-Months of All Plants

. . ! *

47

BAO 0 0 119

Q CF)

Q N m

Job No. 5 Nob No. 6 * Job No- I( L

Plant Total Man Mean M a n Total Man M e a n M a n Total' Man Me= Ma% Resp. Months Months Resp. Months Months Resp. Months Months-

A 7 B 31 C 3 D 5 E 40 F 12 G 85 H 22 J 5 IC 6 L' 39 M --. m I P 3 Q 7 R 5 S 1 T 1 u 23 V 34 W 21 X 1 Y -- 2 -1

AA 8 RE 8 cc 9 DD 7 EE 3 FF -- GG 8

20.7 16.8 U. 6 u. 1 42 -7 10.0

10.5 6.4 1.7 6.4

2.0 20.4 15*3 399 4.8 12.1 27.3 8-9 16.6

--

--

-- -e

19.4 12.9 20.4 9.3- 3*0 16.2

1-9 --

EH -- -- -- 9 56.2 . * -I -- -- 393 16291 ' 1257 Z6X 820 13373

41.45 45-81 * 16.31 Mean Man-Months of All mants

. IL

I . 4a

BAO 00 12 0

TABLE XIU: (Continued) em - Job NO. 7 Job NO. 8 Job NO. 9 N

Plant Total Man Mean Man Tots1 M a n Mea Man Total. Man Mean Man ," Resp. Months Months Resp. Months Months Resp. Months Months

A 3 C D E F G €I J

* K L M H P

B S I U V w

' X Y z AA BB cc DD EE

Q

f 1

u. 90 33-60 39.60 Mean Man-Months of All Plants

! ' 49

BA00012 1

.- TABLE XIII (Continued) m - N .- - c Job NO. 10 Job No. 11 Jab No. 12 N

Plant T o t a l Man Mean Man Total Man Mean Man T o t a l M a n M e a n MG - - , . Resp. Months Months Resp. Months Months Resp. ' Months Months"

A B C D E F G H

K L. M N P Q R S T U v w X. Y z AA BB cc m FP GG

3 .

14 605 78 3140 26 824

3 92 3-35 5847 1 . 67 1 2 -- .I-

5 39 4 115

106 2206 .I" -- .I- d-

6 220 2 308 43 767 14 395 2 85 45 Z 8 T . 45 2423 67 -1587 4- *-

-.-

s 2 17 829 56 2698

5 76 284 9

2 16

4- -- a- . -0

-" --

3 5 28 359 88 6.12 Mean Man-Montbs o f All Plants

! 50

BAO 0 0 122

TAm X I 1 1 (Continued)

Job NO. 13 Job No. 14 Job NO. 15 Plant Total Man, Mean Man T o t a l M a n Mean Man Total Man Mean Man

Resp. Months . Months Resp. Months Months Resp. Months Months

A

1 28.97 26.87 Mean Man-Months

o f All Plants

1

1 .I

1 i

51 . .

L

BAO 0 0 12 3

TABLE X U J (Continued) 43

N

cp

u Job No. 16 Job No. 17 Job No. 18 L

Plant T o t a l M a n Mean Man Total Man Mean Maa Total Man Mean % Resp. Months Months Resp. Months Months ' Resp. Months Months- _.

A B C

E F G H 5 f f; M I? P Q B S T U V W X Y Z AA EB cc DD EE FF GG

n

466 u o 5 229 163

3871 246 -- 275 24

3-86 251

3 121 235 260 210 90

289 1694 839

--

-- -- 83

347

375 84 246

6

a79

--

44.46 12.40 14.33 Mean Man-Months of . A l l Plants

. . I .

52

I I 118 Ill Ill 1111 II 1 Ill 1111 IIP BAO 0 0 124

c a tr'

cp QJ

W TABU =I (Continued) N

L

Job Nu. 19 . Job NO- 20 Job No. 21 Plant T o t a l Man Mesn Man Total Men Mean Man Total Man Mean M a n 5

Resp. Months Months Resp. Months Months Rem. Months Months

A B C D E F G H J K L M N P Q R S T W f W X Y z AA BE cc DD E FF GG

28.4

7.0 23.0

1.4

16.7 42-7

-- 42.5 5.1 7-3 13.2

b.1 33.9 23-8 18.1 6-9

J.3-7 35.9 25.7 12.3

21.9

30.1 19- 4 4.9 11.7

4.7

--

-- -- 24.8

--

23 -71 21.69 34.12 Mean Man-Months of All Plants

! - 53

TABLE XIII (Continued) m . A

Job NO. 22 Job No. 23 Job No. 24 43

P l a n t Total Man M e a n M a n T o t a l Man Mean Man Total Maa Mean Ma& R e s p . Months Months Resp. Months Months Resp. Months Months-

ul -- A B 3 C 2 D E 5 F 4 G 1 II 2 J. 7 K 6 L 5 M -- m 2 2 Q 1 R S 9 T W V 3 W 9 X Y -- 2 AA 1 BB 22 cc 3.7 DD 4 EE 3 FF -- GG -&

--

-_ -- -- --

* - 5 --

12.35 30.23 27-89 Mean Man-Month6 of A l l Plants

, 54

BAO 0 012 6

-= z=? u - 4 '. & i s 43 Q

= N W

OI

T A ~ X I f I (Concluded) w

L-L

Job No. 25 Job No. 30 Total Man M e a n Man T o t a l M S n MemMan - ResQ 9 Months Months Resp. Months Months

Plant

A B C D E F G H J K L M 1p

P e R S T - U V W X Y 2 AA BB cc DD EE FF GG

39 104

. 33

59 1'

363 1 13

- 34 l8l

--

--

587 1450 ,

1291

1609 48

UJ.85 10

480 1514 7091

-- .

--

15; 1 13.9 39.1

27.3 48.1 30.8 10.0 36.9 44.5 39-2

I-

--

27-09 27-89 Mean Man-Months of All plants

. . -

I

55

, . *

BAO 0 0127

AddLtive (other than catalyst) Makeup Bagging and Shipping Operatar Catalyst Makeup Man Clerical, Other Office Staff Canp&ding and Fabricating Dryer Operator General Laborer, Custodial Laboratory Supervisor Laboratory Technician Maintenance, Electrical Maintenance, Mechanical Monomer Maintenance Operator Monomer Operator Monomer Recovery Operator

Plant Guard, Security Wficer Reactor Cleaner Reactor Cleaner, Solvent Reactor Operator Slurry Blender fltpanrlsdr, P d p r Tank Farm Operator Utilities Area Operator and srtperrlsor

Warehouse 'Operator Pansf er operat or Other than WC

. M o n o m e r SupenrLsor

(Water purification, treatment, etc.)

Code No.

23 24 25 30

m m m N m c. H

. w W 4

Fig. 14. Standard IIH job titXes.

This tendency to remain in the & w e noted job are&*, becomes mare apparent when one groups the jobs by their functional relationships within the plant. Table XTV prwides such grouping.'

3. Materials and Processes

a. Monomer Production,

In the nine plants which include vinyl chloride montrmer production per-

In one process, hydrogen chloride is added t o acet- T'. a second process,

In same plants

sonnel anrong the study population, either or both of the two published methods of production are used. ylene i n the gresence of a mercuric chloride catalyst. bydrogen chloride fs removed from ethylene dichloride ( lY2-dicbloroethaae) by pyroly8iq. Chlorine is added as a cstalyst i n this prOcc!ss.

I .

56

BAO 0 012 8

I i C

I

' FUNCTIOmAL GROUF'IIJG QE' JOBS, &: MAN-MONTHS, AND MEAN MAN-MONTHS

Job Grouping Men Man-Months Mean-Months

Other Marmfacturing (02, 16, 14, 20)

Clerical Staff (04)

Support of Manufacturing (04, 16, 21, 23, 24, 25)

Other than WC

624

2927

2287

784

1257

393

474

880

1.1- - I I (30)

12324

38461

32794

28079

57657

16291

18 444

21247

32826

19 -75

13 -14

14.34'

35-82

45.86

41.45

38.91

24.18

21-89

the hydrogen chlorlde p r a c e d I s ,used as a raw material in the former process.

In some of these plants, personnel engaged in monomer production rotated through the polymerization reaction area. separate mintenace personnel assigned exclusively t o the monomer production mea. curic chloride change-over. visory personnel indicated.

In only two or the plants were

In one other plant a maintenance worker handled the operation of mer- In only four of the plants were separate super-

Only one plant prdduced VLnyl chlofide monamer - t h a t also producing the polymer. con&cted i n a central room. in a gaseous or vapor state. polymer, exposure t o gascars materials was for the most part also found to be

This plant was of modern designwith most of the operation being Ligh t exposure was estimated f o r all materials A t the eight plants making both monomer and

t . , . 57

slight with the exception .of that t o vlayl chloride. was considered t o be major in one plant and intermediate in three others. fntermediate exposures to ethylene dfchloride and to nlercuric chloride were estimated in one plant and t o hydrogen cuoride in two plants.

Exposure to this gas

Tfie estimates of extent of exposure at the nine plants to the several materials used in monomer prodsrction are presented i n Table XV.

TABLE xv

EXTENT OF EXPOSURES TO MATERIALS I T VINn CHLORTBE MONOMER PRODUCTION

Exposure Estimates to Designated Material

'Iant Job vinyl EDC E l Clz .Hg21= HCCI.3 DCE* CP)+ No. No. Chloride

E

F

(3

H

R

U

2

AA

13 S

13 . :: I:

3.2 S 13 S 15 S

- 1 6

13 I 15 8

13 s 15 s

12 13 M 1s S

12 S 13 S 15 s

1 6 s s s S

S S

s S s s S 6

S S

I S

I S

S S s

6 S S s S S

S S

I * s

S

S 6 S s S S s S S

Ea 1-j S S S S Note: The job number 12 refers to monomer m s l n t m c e personnel, ntnn-

ber 13 to rrll monomer operators, and nunher 15 to monomer super- visors. The egosure estimates are indicated by the le t ters M (Major), I: (Intermediate), and S (Slight) as defined by Table If.

*DEE refers to l,l-dichloroethane. +*cP refers. to chloroprene.

58

BAO 0 0 13 0 . - I

b. V i n y l Chloride Polymer Production

Over 227 materials were used or produced in the plnnts producing polyvinyl chloride, Vinyl chloride-*nyl acetate copolymer and other co- and

.tcr-poly-mrs. The severi ty and duration of exposure t o each of these was estimated fo r each Job. catalysts or inLt€ators, dispersants, inorganic compounds for coalescence r e m a t i o n , enniisifying agents ( including soaps and surfactants) , organic solvents (including chlorinated hydrocarbons), and a rrumber of trade named materials tha t were not identified'.

These materials included monomers and polymers,

The ca ta lys t s are all act ive chemical. compounds and as such were especially suspect as a poss-Lble fac tor in t h e causation of AOL. exposure t o these is shown In Table XVL for 24 p lan ts producing vinyl chloride polymers. the t a b l e t o protect confidentiali ty. degrees of extent; of exposure. include both sevef i ty and duration.

The extent of

Catalysts which are unique are i den t i f i ed as "A," nB, " etc., i n As discussed i n Section 111, the three

EXTENT OF IXPOSURE TO CATALYSTS

Extent of Exposure No. of -_ C a r b o n KO. of Major Lntermediate Slight

plants A t o m 8 t 2 Job No, Job No. Job No. Catalyst t - ~ n ,~,

Inorganic peroddes 12

Ammonium persulfate Potassium persulfate Sodium perborate mewen peroxide ( 3 W

Organic peraxides

Peroxide A '1 2 1 1 Dt isopropylperoxydt carb mat e - 12 4 4 1 7 1 4 Peroxide B 1 5 1

Cayrylyl percnrLde 3 8 1' 2 Peroxide C

Pelargonyl. peroxide - 2 9 2 1 Percprfde I) 1 10 1 1 Ter-butyl peraocypivalate 7 ll 1 3 1 3 Lauroyl peraxide 23 12 9 6 10 5 Benzoyl peroxlde 2 7 2 Azobisi sobutsrronitrFle 2 3. 2

1 6 1

I .

59

BAO 0013 1

a Another suspect material is t h a t scraped from the reactors. This is be-

lleved t o include some p a r t i a l l y polymerized resin, the h igh molecular weight polymer, plus some unreacted ca ta lys t together with additives. information i s available. vinyl chloride is present. Also, it has been reported t h a t use of lauroyl percA.de is associated w i t h a residual of t h i s cata lyst in PVC par t ic les , whereas l i t t l e of the diisopropyl peroxydicarbonate relaains unreacted i n the product.7 Even though Job 17, Reactor Cleaning, includes no specif ic notation of exposure t o the ca ta lys t s or additives, It i s t o be understoog tha t the m a t e r i d designated as scragings ( W C scrapfngs, copolymer VC-VA scrapings, etc.) probably inclucks an indeterminate amount of the materials introduced ln t h e batch.

L i t t l e specif ic c 43 - It is known t h a t approximately 1s unpolymerized Y

c. Cleaning Procedures

The usual method for cleaning reactors before s t a r t i ng a new batch requires hand scraping with putty knives or nylon scrapers f o r the s ide w a l l s , and ham- mer and ch i se l for surfaces where heavier deposit accunulates. Occasionally when the batch polymerizes t o a solid mass inside the reactor, the r e s in has to be chopped out with axes. af'ter the charge has been drawn off, the usual pract ice i s t o break any vacuum with air, then t o remwe the manhole cover and wash out any loose residue with water . A f l e x i b l e hose is dropped t o the bottcm of the reactor and air blown i n by a portable blower. out t he reac tor charge f loo r and ictroduces 811 exposure t o the reaztor operations in t h e area. A t p lants in the warmer climstes, recrztors are usually oiit 3.n t h e open. A i r analyses have shown vinyl chloride concentratfans i n the reactor p r i o r t o r inslng and venting t o be in the order of 3000 par t s per m i l - l ion .

Before the reactor cleaner en ters the vessel

This method disperses t h i s vinyl chloride Through-

After venting f o r varicns periods, usually 15 t o 20 mirmtes, the presence of residual -1 chloride i s tested, often by sr,iffFng and 5~ some plants 3y mews of a PLamable vapor indicator. million as the lowest concertration detectable by odor. ' A s previously noted, it requires 400 p a r t s per m i l l i o n f o r the usual type of flammable vapor detector to gtve an Indication half way up t o the f i rs t graduation. analyses have been mde with more sensi t ive instrumexts, such as the gas chromatograph, €t; has been found tha t the vinyl chlorede concentratian teads t o be below 100 par t s per million and generally about 50 ppm. containing i n the order of' 1% of absorbed wpolymerized of this is released as s c r a p i x is car-ried on. cantimed during t h i s operation.

The literature reports 4000 p a r t s per

Where a i r

W i t h t h e r e s i h e vinyl chloride, some

Aeration of the reactor is

As shown i n Table XVII, a t most plants residue is scraped out a f t e r every batch though at sane p la r t s t h i s i s done -here from every one t o three cycles up t o every 10 cycles or every two weeks.

! ' 60

BAO 0 0 13 2

http://percA.de

v

a 3 i 1 1 ..

1 i I .

4 .

1 I

1 1 I

7 A

7

9 1.

\II

TABLE XVII

REACTOR CLEXKIXG METHODS

No. of P l a n t s Method

Mamtal scraping af te r every reactor cycle 17 Manual scraping af te r 2 t o 3 cycles 2 Manual scraping after 4 to 12 cycles 4 Water j e t cleaning 3 O r m i c solvent cleaning 3

. A small number of plants use high pressure water j e t s up t o 4000 psig to remove t h e residue from reactor surfaces. 2500 psig might cause injury t o the glass linings of the reactors proved un- founded.

Early fears that pressures over

T h r e e plants-use organic solvents f o r reactor cleaning. The original design of one plant provided for t h i s method of' cleaning.

d. Blowdawn or Stripper Tanks

I n 1 4 of the 26 polymer plants, the completed polymer charge was dropped _ - -

to a blowdawn or stripper tank where the unreacted viny3. chloride was removed, usually by application of vacuum, though volati l ization at atmospheric pressure and elevated temperature was u t u i z e d in other plants.

-

In 15 of these plants, the Vrayl chloride i s stripped directly from the batch i n the reactor. In three of the 26 plants, both methods are used.

e. Procedure Follwing Completion of Polymerbation

On review of the data collected during tke surveys, the question arose a s to whether differences amone; procedures follawing completfan of polymerization might produce significant qualitative differences i n the exposures o f t h e reac- t o r cleaners. AccordFngly, information w a s requested of each plant conducting palymerization operations concerning the f o l . l & ~ three elements of the procedure:

(1) Time from end of polymerization phase t o opening of the reactor;

(2) extent of vacuum, if any, during this phase and hm long applied;

(3) rate of temperature change from te-mitation of polymerization to opening of react or

I .

61

.. .

BAO 0 0 13 3

The infonaation on decrease in temperature frm the time of completion of polymerization t o the opening of the reactor w a s so variable that comparison between p lan ts was not feasible.

_ .

I

62

i

It w a s thaught t ha t exminat ion of the responses might sha; t ha t d i f fe r - tnccs i n the extent of t h e vacuum o r t h e time m e r which it was qpl ied .might

2 43 b.4

a l t e r the amount of res idua l vinyl chloride in the residue o r the amount re- W Q 0 W

leased i n t o the space within the reactor.

i n the residue and the scrapings to which the cleaners were exposed as wel l as

The temperature change procedure might very w e l l have an- e f fec t on the extent o f part ia l ly polymerized material

the anmaant of Vinyl chloride monomer rematning.

- Nineteen of the 26 polymer plants presented the data on vacuum and t h e t i m e over which .Lt was applied in such a maaner t h a t conlparhons could be made. The amaunt of vinyl chloride rernaintng tn t h e residue wcluld be generally i n proportion t o t h e vacuum within the reactor (expressed as the number of inches of mercury absolute pressure) and inversely to the t i m e over which the reduced pressure w a s appliea. The ionger t h e reduced pressure is applied, the smaller the amaunt of vinyl chloride remabing i n the residue. a t which the remaining vlnyl chloride would be drawn f'rorn the residue w o u l d decrease. approximated by applying the logarithm of' the time factor. A single value representing the r e l a t tve e f f ec t of these two values was obtained by d i d d i n g the yacuum as absolute pressure in number of inches of m m x r y by the logarithm t o €he base 10 of the time (t) in k n u t e s . pressure-time factor.

Accordingly the rate

!l!hus the effect of time would not be l i n e a r but would be more c losely

This value i s termed the absolute

Since the perfod of time t h a t t he vacuum was applied varied i n cer ta in of the plants, two values are given f o r such plants i n Table XVfII, bution o f these values f o r the "positive" and "negative" plants is presented i n Section V, p. 95."

The distri-

ABSOLUTE PRESSURE-TIME: FACTOR (Between Completion of Polymerfzation end Opening of Reactor)

Inches €&/log t: 0.5 0.7. 4 5 6 9 10 U. 14. 16 17 ZO 21

NumberofPlants: 1 1 1 1 2 2 5 1 4 4 2' 1 1

BAO 0 0134

E. DIAGNOSTIC CATEGORIZATfON

I. Description

a. Diagnostic Category 0

Subjects w e r e placed i n t h i s category if they showed no Raynaud's phenomenon and t h e i r x-rays were e i the r totally normal or the x-ray abnormality w a s l imi ted t o defect #Ue

b. Diagnostic Category 1

Subjects were placed tn t h i s category if they reported a his tory of hand trauma iavolvixlg a spec i f ic flnger and the hand x-ray w a s consistent wtth th€s history, and if they did not have posi t ive symptoms 25 and 27. I n those cEtses where there was a question, individuals were placed i n t h i s category when repeat x-rays showed sequential changes. consistent w i t h trauma.

c. Diagnostic Category 2

These were asyzptomtic siibjects with x-ray changes consistent with hand In many cases, his tory of hand trauma could not 'be recalled though trauma.

the f inger les ions were clearly consistent w i t h traumatic inJury.

d. Diagnostic Category 3

Subjects who had negative or equivocal x-rays i n t h e presence of the mi- indicator of Raynaud's phenomenon, Lee., "white f hgers" prec ip i ta ted by cqld exposure were placed i n t h i s category. - gory were taken to mean:

Esu ivocal x-rays i n this cate-

(1) nonconcurrence betweenroentgenologists in re: a single marg ina l l es ion (defect #l), and absence of x-ray defects #2 through e, o r nonspecific x-py defects susceptible t o in te rpre ta t ion o f a subjective nature, i.e., x-ray defects #8 t o #u.

- (2)

e. Diagnostic Category 4

me major criteria for t h i s category, l i s t e d as "possible AOL," are t h e absence of the cardinal sfgns and synrptams of Raynaud's phenomenon and x-ray changes not c lear ly compatible vLth Am. t h i s category were some which w e r e nei ther inconsistent with AOL or old trauma,

Among the x-rays of. those placed in

! * 63 _. ..

BA000135

e- - = ---

2 ==. Q

as w e l l as other x-rays with none of the more clearly definable defects asso- c ia ted with AOL (i.e., defects #2yf:5). N

0 c W

se, f. MagnostLC Category 5 c r , -

This consisted of individuals i n whom the cardinal signs and symptoms of Raynaud' s phenomenon were present. fects without changes compatible with trauma by appearance and/or history. A t the one extreme there were a few individuals having a single marginal defect with concordance between x-ray readers through those having multiple severe f inger changes. Accordingly, t h i s category represents active, symptomatic cases of AOL.

In addition, they had e i ther multiple de-

- .

g. Diagnostic..Categary 6

These cases were similar t o category 5 except thut they did not have the character is t ic Raynaud's phenomenon but did have severe multiple f inger x-ray change. However, i n all cases but one, the changes 81.e believed t o represent old disease which was becoming, or had become, quiescent. Accordbgly, this group p n e r a l l x could be considered as displaying inactive, "burnt out" cases Of AOL.

The diagnostic c r i t e r i a referred t o i n this study- may be summarized as i n Table XM.

TABLE XIX .. .

R0EN"aFUPHIC AND SYMPTOMATIC CRITEKU FOR DIAGNOSTIC CATEGQRIES 0-6

H a n d X-Ray Specific X-Ray Defect Dtagnostic Raynaud's Abnonnal Consistent w i t h AOL

- Category Phenomenon Consist. 1 2 3 4 5 6 7 8 9 10

6 - + + + + + + + + + + + *Indicates only those lesions that may be seen i n AOL (e-=) though not diagnostic of AOL.

*Single finger involved, no concurrence between readers. -No concurrence be-heen readers.

64 I .

BAO 0 0 13 6

s 1 1 1 1 1 1 1 .

.*

1 1 1 I 1 1

1 i

- e

-=. a I-. c,

en 43 Q N m

cy

TABLE xx DISTRIZAPllION AND FREeuENcy OF DIAGNOSTIC CATEGORIES BY CORPORATIONS L

Dia o s t f c Cate fories cozy. Population n_2m ,., ,, 1 2 3 4 5 6 8 9 u. 12 13 14 15 17 Is 19 20 21 23

164 1892 105 97 532 28 59 49

u 5 34 21 220 308 594 100 414 83 E 9 67 -

34 1 2 4 1 1 3 3

4 3 4

2 6 6 -

7 14 2 1 1 2 1

1 .2

3

2 2 1

1 1

.- - Totals 50n 4896 74 16 22 3

2. Dbcussion of' Dfagnostic Category Distributfon

a. High Probability Locations for AOL

The highly characteristic combination of Raynaud' 6 phenomenon and/or mltiple bony phalangeal fnvnlvement which characterizes diagnostic categories 5 and 6 should be indices to the location of this disease, AOL; it would appear that plants B, E, F, K, R, TJ, and EE probably have cases of t h i s accupatianal disease (cf. Table =I).

Accordingly,

This probably being the case, further analysis will accordingly proceed f r o m this assumption (see Section V).

b. Plants Demonstrating Possible Cases of AOL

While the cha2ycterlstic course of AOL begins with symptoms of Raynaud's phenomenm, one case has been seen to have multiple active bony lesions in

-. 65 ! *

.. .

BAO 0 013 7

TABLE )3u:

DISTRIBUTTOW AND FR?QUENCY OF DIAGNOS!CIC CATECQRIES BY PLAmTs

Diaqnostic Cateaories 0-2 5 3 4 5 6

=ant Population

A

c I] E F G H J K L M a P Q R S T U v W X

. Y 2

AA EB cc DD EE FF GG HA

a 164 472 135 74 643 46

522 69 36 97 349 a 3 28 59 49

=5 34 21 220 308 255 141

U)O 139 275 83 50 79 26 32

_9

17

1 7 2 7 1

2 3 1 1 1 3 3

4 3 3

1*

1 - 1 2 6 1

6 1

3

3 1

1

I

1

2

6

6 2 3 -c

2 1

2

2

1

Totals 5011 48% 74 16 22 3 *Female.

BAO 0 013 8

.4

the absence of symptoms of the vascular disorder usually associated wit& AOL. Since so l i t t l e is known of all possible variations in AOL, and because of the nonspecific nature of .the m i n i m a l bony lesions of minims1 or early AaL (l.e., x-ray defec ts #l, e), several plants la which either questionable x-rays or a possibly high incidence of Raynaud's phenomenon occur must be considered as poseible cases of A05. In this category are plants G, BB, and CC (cf. Table =)

C. Plant Locations Probably Not mrlng Cases of' AOL

By exklusion, aU plants not noted above in the precedent probabXy-do not have a problem of AOL occurrence. - .

These first assumptions that AOL occurs in the stu& papulatim as ihdi- cated above are summartzed in Table XXII .

TABLE XXII

.. DISTRIBUITON OF AOL IN WC STUDY GROUP

Corpaiatian: - - Plants with probable Case8 Of AOL - By E, F, K, R, U, EX 2, 4, Jl, 14, 2l.

Plants with possible * C8Se8 Of AOL - G, BBy CC 2, 19, 20

Plants Probably without A, Cy D, H, J, L, M,

x, Y, z, M, DD, FF, 23

1, 2, 3, 5, 6, 8, 9, cases of AOL N y P, % S y T, V, W y 12, 13, 15, 17, 21,

HR NB: It shoula be noted that plants which only perform campounding

operations (cf. Table m) are in the last of' the three above categories but can be separated for analytical purposes.

.d. Population Distribution Among the Three Groups of Plants, L e . , Plants with Robable Cases of AOL, Possible Cases of AOL, and Those Probably Free of AOL (see Table X I I I )

I .

BA000139

I

m3m XXIII W Y

W m cu. YL

DXSTRIBWION OF DIAGNOSTIC CATEGORIES AMONG THE SUB-POPULATfOI?S WIZHIX WC STUDY GROUP

.. Tota l Diamostic Categories Papulation 0-2 4 5-6 3

Plants with cases of AOL 1673 1598 41 9 25

plants with ?

cases 910 893 14 3

Plants probably without cases

Totals - 2428 s g - - 4

4896 74 16 25 SO11

Cmsfdering plants which carry aut campounding operations only, the dis- t ~ h t i o n of diagnostic categories in this p a a t i o n ‘is as follows:

554 Dx 3 2;; Dx 4 1 Dx 3,6

DX 0-2

0 Total Papulation = 557 *Includes one woman

-

It should be maerrstood that those indivfduals with Dx 3 and D x 4 do not necessarily indicate the presence of AOL (cf. pg. 101, paragraph 2). represen* eltber persons With Raynazd’ s phenomenon withme significant bone changes in the case of Dx 3, or fndiviattals without Raynaud’s phenomenon but w i t h bone changes not necessarily characteristic of AOL (cf. pg. 65).

They

-

60

_ -

BAO 0 014 0

_ - -

-=I,

'e '.XI

(r, em Q N

c W SECTION V L

DITERACTIONS OF SIGNS AND SMPTOMSt X-RAY FINDINGS, w JOB ElISTqRfES, A W DIAGNOSTIC CATMfORIZATION

The previous section described the c r i te r ia by which it was concluded

For analyt.l+ purposes, we have divided the t o t a l PVC tbat certain plants and corporations have active cases of acroosteolysis (AOL) and others do not. s t u d y populetlon into four groups: synthesis plants divided a8 follows:

The flrst three p o u p s are employees of

. 1. P h t s with AOL

2. Plants w9thaut AOL

3. Plants where AOL is questLonable

The fourth group derives from plants where compounding canstitutes the only manufacturing process.

A. DIS1IRXBUTION OF mMpIlOMS BFTWEEX STCIDY GROUP, ITS SUB-GROUPS, AND CONTROL POPULATION

1. Occurrence of Diagnostic Category 3

This diagnostic category is based upon the presence at? Raynaud's phenomenon i n the absence of significant x-ray findings. this diagnostic category represent incidental cases of Raynaud's unrelated t o .AOL, while others may have U d or early AOL. naud's phenomenon in a Canpar8ble population is not known and could not be obtained from the Tecumseh data, it becomes necessary t o tittempt estimate of the frequency of Raynaud's phenomenon by i n t e r n d c o m p ~ i t ; ~ s . Disease occurs more frequently i n females, and since there are a few females i n the PW: Study group (including one i n diagnostic category 3), t h e coxnpari- sons that f U c w are based on males Oris. (see Table XXIV).

Probably some individuals in

Since the prevalence of Ray-

Since Rayn~ud's

m e r rpodng all females from the study populaticn, and those Fndivld- uals who have Raynaud's phenomenon associated with AOL, camparison w a s made of the incidence of Raynaud's among t h e four groups described above (Table XXIV).

I

69

I lllrm 118 Ill 1111111111 1111 1111 811 111 BAO 0 0 14 1

TABLE XXN

I

I

I

I I

c

II I

b

m DISTRIBUTION OF RAYNAUD'S PfIENOMENON BETWEEN PLANTS W I T H PROBABIZ, POSSI=E,N,

c w

c1

AND NO CASES OF AOL, CORRECTED FOR MALE! POWLATION Q

Dx C a t . 3* c

$ Total Dx Cat, 0-2* Total* MalePop.

1582 1623 Plants d t h AOL 41 - 2.5 Plants without AOL 1 4 0.60 2294 2308

Plants

*Includes males only.

Plants with ? AOL 19 2-3 817 830

Ccmrpmndinp; only 1 0.23 438 439

Conclus i ons :

(1) Table XXN shows t h a t plants known t o have AOL have a considerably higher number of Raynaud's compaxed w i t h t h e "control" population made up of plants where there axe probably a0 cases of AOL,

(2) A significantly higher incidence of Raynaud's ( p e 0.01) occurs i n plants where AOL may possibly ex i s t than i n p lan ts believed not t o have AOL.

(3) Workers i n the compounding operation show the lowest incidence of Ragnaud's phenomnon, only 0.23%. i can t ly d i f fe ren t from the population i n which no cases of' AOL occur (€.e., fhe putat ive "control" group).

This is lower but not s t a t i s t i c a l l y signlf-

(4) IncLuded among the plants where the presence of AOL is questionable Although no posi- is one engaged only fn the synthesis of monomer ( p h t G).

tive or questionable x-ray cases occurred here, the presence of Raynaud's i n 1.355 of a population of' over 500 men, raises in te res t ing questions as regards the et iology of AOL. I n another. of the p lan ts i n t h i s category of possible casesI whFle having a smal l population, 7.2% of t h e 23 men eq loyees reported m a u d ' s spptuns, a questionable x-ray (i.e., defect #11).

.

In addition, one emplayee i n thLs group appears to have

( 5 ) Possible cases of AOL in diagnostic category #3: Among the 74 md.es in this category, there are 8 workers whose constel la t ion of Reynsud's phenomenon plus nondefinitive x-ray changes suggest that they may heve AOL, Whether these a r e cases, and if they represent m i n i m a l developing, stationary, or de- crudescent cases, are questions which cannot be answered by single examination.

Also among those within diagnostic category #3, there are eleven who have such a mult ip l ic i ty of associated hand symptoms as t o arouse suspicion. W e

! * 70

4

J

t.

d

c

c

1-

i

c

f

-m

I J

know tha t one man among these eleven workers has developed a fully developed

played in plwts where t h e existence of cases is known t o t h e employees, the question of the power of suggestion must be considered.

m AOL subsequent t o this study. On t h e other hand, since all of these a re em-

r 4 = c w Q L

N

2. Occurrence of Stms and Symptoms

The distr-ibution of signs and symptoms i n the populations of the four categories of p lan ts described above i s shown €n Table X X V , plants believed to have AOL cases; Table X X V I , plants engaged only i n compounding; Table XXVII, plants without AOL; and Table XXVIII, plants with possible cases of AOL. above de ta i l ed tabLes are sunmaxized i n Table XXIX. ous graups were t e s t ed by the ch i square technique with the results of these comparisons shown i n Table XXX.

The Comparisons of the vari-

a. Discussion

Comgzarison of plants believed t o have AOL t o those believed not t o have AOL (Table XXX, (1) vs. (4)). It is quite clear t h a t t he occwrence of symp- - .

toms i n the p l a t s believed t o have AOL is s igni f icant ly more frequent than i n those believed not t o have t h i s disease (with the exception of symptoms 35 and 38). of the assuxptions involved i n SubdiVidLng the study population, one can order t he symptoms by their order of p values t o construct a ranking for possible d i f f e ren t i a t ion between papctlations on t h e basis of syrqpl;ams (see Table Xxm) .

A d m i t t h g t o the gross nature of tbfs analysis, and with a rea l iza t ion

b. Comparison of Symptoms Fn W C Synthesis and Compounding Plants Believed Free of AOL Compared w i t h Plants O n l y Compmdirig PVC (Table XXX, (2) vs= (4))

Comparing the dif'fererices between these two populations suggests quite - st rongly that there is l i t t l e consistent difference between persons i n the except f o r symptoms ccunpoundtng plants anB others. begeved t o be free of AOL,

#3l, l.e., "numbness, f ingers fall asleep." difference between the two populations, the absolute number of symptoms w a s

However, i n those four instances of si&flcant differences between "plants without AOL" vs. "plants compounding only" as regards symptans @8, 32, 34, 37, such differences a re not striking i n corkrast to the differences between other gruupings, e.g., (1) vs. (2), (1) vs. (4)

In those cases where there was a

W great.er" in the plants believed f r e e of AOL.

71 I .

BAO 0 0 14 3

I i

I

i

TABLE xxy . OCCURREME OF SIGN3 A I D 8YbW"W Xl PXAlFB WITH PRaBABLE CASE9 OP AOL

(Plentr with dieaortic categories 5, 6)

T o t a l $ Occurrence, B . E F I[ R U EE Reamnse hesponse Remonse Response Renwnee Rcsponrre Response pos. tote1 Symptom

NO .pas. t o t a l pm; t o t a l pos. total pos. t o t a l pos. t o t a l 'po8. total pos. t o t a l reap. reap. bM

- . 25 37 465 21 613 5 47 5 96 1( 114 g 219 6 79 87 1663 I3.2 26 27 4&3 l2 ab3 3 4? 1 96 3 114 13 219 3 79 62 1666 3.7 27 26 6 2 1 5 21 4 5 k 4 3 5 9 19 8 9 6 9 1 2 5 55.2

29 32 470 7 641 3 47 1 95 10 113 19 219 b 80 76 1665 4.6

31 101 471 68 643. 10 43 Lg 96 18 114 35 220 6 Bo 257 1671 14.5

28 103 469 75 6kl 12 h7 ~8 96 16 114 36 217 8 80 267 1664 16

30 46 471 32 643 5 h6 9 96 11 113 22 220 7 80 132 1669 7.9

32 119 470 @ 642 12 47 9 96 l8 114 49 220 9 80 303 1669 18.2 33 31 469 23 642 6 47 00 96 6 111 ~5 220 6 80 87 1668 5.6 34 24 470 11 643 7 47 L 97 9 l l 3 20 $20 3 77 75 1667 4 .s 33 31 . 468 9 643 3 47 0 97 6 113 i o 219 4 79 63 1666 3.8 36 68 467 22 6h3 3 4'7 0 97 1L I l k 19 220 7 78 130 1666 7.8

. 37 q 470 3 642 L 47 0 97 5 114 6 220 5 78 47 1668 2.8 38 37 467 30 643 6 46 97 LO 114 18 219 9 79 1666 6.6 39

~ 40 41 439 35 6401 4 46 4 96 20 110 40 218 10 76 164 1646 10.1

4 N

td P 0

OCCURREKCE OF 8IW AHD 8YMPTCW IN PLANT8 CARRYING OW CWPOWINU OPEEUTIONS ONLY (NO casea i n diagnostic categorielr 5, 6)

Symptom No. 26 27 28 . 29 30 31 32

Reaponae - only

. M 3 181 3 182 I. 3 13 180 5 178 9 180 13 180' 11 179 * x 1 141 2 141 1 1 18 140 0 141 5 140 20 139 24 1 4 1

Y 8 im 4 ig 1 10 37 197 10 198 13 197 40 1%' 19 1%

25 Plants

Reeponee Reaponae Raaponsa Reeponee Reeponee Response Response hpaiw. 90s. t o t h l p a . t o t a l pos. t o t a l poe. t o t a l poe. t o t a l pos. t o t a l poe. to ta l poe. to ta l

FF 2 25 1 26 0 3 6 26 2 26 3 26 6 26 7 26 HH 0 9 0 9 0 0 1 9 1 9 1 9 1 9 1 9 TOTAL 14 353 10 555 3 17 73 5% 24 552 31 552 88 . 551 62 553

?;I poa. reap. 2.5 1.8 17.7 13.6 4.3 5-6 15.9 11.2

SmtMP NO. 33 34 35 36 37 38 39 40 P l a l t e

Reeponae Reeponae Rcsponae Response Response Reaponae R06pOneO Response Cmpdlng . M 6 179 . 1 180 3 180 4 180 2 177 3 177 11 173

On'' son. t o t a l poa. total p a . t o t a l ws. t o t a l Foe. t o t a l pea.. t o t a l nos. total Foe. t o t a l

X 6 140 0 140 1 140 1 141 0 141 7 141 11 1 4 1 Y 9 198 2 197 4 196 3 197 0 196 9 188 9 1% FF o 26 o 26 1 26 1 26. o 26 1 26 1 2 6 HH 0 9 1 9 0 9 2 9 0 9 3 9 2 9

$ ROB. rem. 3.8 0.7 1.6 2.0 0.4 4.3 2.0 6.3 TOTAL 21 552 4 532 9 551 11 552 2 % 23 9 0 11 539 34 340

. .

I

I

U XJ 0 0 0

ar ‘mri

~ -- SY(pPta No.

23 26 a 28 29 30 31 12 Reiwnae Rearmnee- Acrpoare Asspoaae Rerwnrr Reawnae Rerwnie Miwnrr

pJI. totd pi. tota pol. tow POL. t o m . poi. total Do.. total poi. total IIII. totrl A 1 163 0 163 0 1 2 163 o 163 0 163 4 163 8 163 C 3 133 0 l33 2 11 4 133 1 133 2 l33 6 133 10 l33 D e 74 0 74 1 6 6 74 1 74 1 7L 10 74 11 74 w 2 io5 2 io5 2 L 13 iq e 103 8 10s 13 iol 19 iol L 6 339 11 341 Iz 20 44 343 15 341 26 344 42 3b4 51 343 I 1 2 9 1 2 9 2 2 i? 28 2 29 0 29 2 29 P 29 P 3 51 3 59 2 11 7 % 4 59 4 59 7 5 9 1 0 39 0 3 48 1 48 7 20 13 49 i 48 3 48 12 49 9 49 8 0 3 4 0 34 0 0 3 33 3 3 4 0 34 2 34 2 34 i 0 2 1 . 0 21 0 0 0 20 0 21 0 21 0 20 1 2 l Y 3 3% 3 300 I 6 30 3- 14 307 ui 307 34 3w n job Y 3 2% 6 231, 6 7 36 2% ‘ 5 2% 17 2% 32 2% 32 254 M 8 a36 7 139 3 13 25 u9 3 138 12 137 24 138 33 157 DD. 0 ~ 0 5 0 0 0 2 W O % l 5 0 3 % 9 % J 00 0 32 0 32 0 0 3 32 0 32 0 32 2 32 2 52 A 1 1 0 0 yg 1 1 3 1 0 0 1 1 0 0 0 1 0 0 3 1 0 0 e 1 0 0 WCAL 36 l882 38 U-39 bo 103 193 1881 54 1886 91 l889 198 l 8 p 228 1890 5 WI. -mu. 1.9 2.0 39.0 10.2 Q.0 h.8 10.4 12.1

1

8 ~ t - XO. 33 34 ?5 36 37 38 39 40

R t l p o n m Rerponir RarpMlae Reiwnee Rcrponia Reaponre Reaponma Raaponre poi. total ma. totml pi. to ta l mi. tota l pol. total POI. total wi. total PI. total

A 0 163 0 163 0 163 0 163 0 163 0 163 0 163 0 163 C 3 133 3 134 2 1% 3 133 1 153 3 130 I 132 6 134

2 74 6 .,.a .uJ 10 3‘12 0 29 1 5 9 0 49 1 34 0 21 5 308 0 2% 4 I38 2 5 0 1 32 0 100

39 1890 2.1

4 I4 4 Iih b 342 1 2 9 4 5 9 3 49 2 35 1 21 6 308 8 231, 7 137 1 5 0 1 32 0 106 8 W

3.6

3 75 2 102

19 324 0 2 8 1 % 4 4 8 0 33 0 2 1 5 298

22 2% 9 123 1 48 0 30 0 9 9

67 a31 3.7

* o b - -

S ~ Y OF a m o m , DISTRIEWION, AND MEAN OCCU~~RENCE IN PLANTS miom TO HAVE DIAONOSTIC CATEUORIES 5, 6, POSSIBLY MVTNO, OR NOT BEUEVED TO HAVE AOL. SIMIUR DATA FOR PLAloTS COMPOUNDING ONLY

Sympton No. 25 26 rl 28

Response Reaponea Respon s e Response pos, t o t a l 4 poe. pas. t o t a l $ pos. p a . t o t a l $ po S I pos. t o t a l g p os.

3 (1) Dx cat. 5, 6 6'7 1663 , 5.2 62 1666 3.7 69 123 55.2 267 1664 16.0 (2) Cmpding. only 14 553 2.5 10 555 1.8 3 17 17-7 75 552 13.6 (3) 4 AOL 872 3 - 1 21 874 2.1) 19 35 54.3 85 870 9.8 (4) No AOL 35 1882 1.9 38 1889 2.0 40 103 39.0 193 1887 10.2

Plant Population

-4 cn

29 30 31 32 Response Response Response Plant Population . Response

pas. t o t a l $ po s. pos. t o t a l 4 poa. poa. t o t a l % PO a. pos. to ta l 4 pos. (1) DX Cat. 5, 6 76 1669 4.6 132 1669 7.9 2S7 1671 14*5 303 1669 18.2 (2) Cmpding. only 24 552 k.3 31 552 5.6 88 551 15-9 62 553 11.2 (3) ? AOL 22 872 2.5 32 873 3.7 96 871 11.0 99 873 11.3 (4) No AOL A 1886 2.9 91 1889 4.8 1% 1890 10.4 228 189 l2.1

33 34 35 36 Reeponee Response Response Reeponae Plant Population

s. t o t a l e. p a . total $ poa. we. t o t a l a. pos. to ta l $ pos. (1) Dx cat. 5, 6 1668 ' E 6 75 1667 4.5 63 1666 ' ;!3 130 1655 7.8 (2) Cmpding. only 21 552 3.8 4 552 0.7 9 551 1.6 11 532 2.0 (3) 1 AOL 22 319 4.2 21 870 2.4 21) 867 2.8 43 869 4.9 (4) No AOL $I 1890 3-0 39 1890 2.8 53 1887 2.8 68 1888 3*6

37 38 39 40 Response Response Responee Res pons e Plant Population

pou. t o t a l 4 po s e we. t o t a l $ PO E. p a . t o t a l 4 po 8 . pas. t o t a l $ p os. (1) Dx cat. 5, 6 47 1668 2.8 109 1666 6.6 53 1627 3.3 164 1646 10.1 (2) Cmpding. only 2 9 8 0.4 23 540 4.3 11 539 2.0 34 540 6.3 (3) 0 AOL 14 869 1.6 55 868 6.3 17 843 2.0 55 849 6.5

-(h) No AOL 21 1888 1.4 97 1882 5.2 67 1831 3.7 144 1862 7.7

23

26

28

29

30

51

32

33

34

35

36

37

s

39

40

Relrti omhi? PopulstFan

PapdLati6n

PopulatiM!

Popuhtian

Population

Fwpu3.ation

Relatlonshlp

Rehticmship

Rclatfanship

Relationship

Relatianship

Relrtian8hip Population

Relationship Populatloa

BehtlOnShip Popubticp1

Rthtiansbip P o p u Z a t f O U

Rehtioaship

Rclatiauahip PopulatLon

Rehtianahip PopuzstiOa

RdlatiCUWhiP Papulation

Eelatloneaip

1 > 2 < .01

1 > 2 < .#

1 > 2 NSD

2 1 2 m 1 > 2 A9)

1 > 2 rn

1 > 2 < .ool

1 > 2 HSD

1 > 2 c .a 1 > 2

< -025

1 > 2 < .001

1 > 2 < .oQ1

l > 2 m . 1 > 2 lpsb

1 > 2

2 2 4 HSD

2 > 4 < -05

2 > 4 K60

2 2 4 rn

2 c L - Population - -01 < .m5 < .oos t?sD m

+(1) plants believed Eo have AOL; (2) compaunding plants; (3) plants wfth possible A m , (4) plants believed to be f i e e ai AOL,

WHSJ = nb significant biffe*nca between poprilatlan.

Rate: Probabflities obtained by,use of Flaher'a Table of Square andlyeir, d.f. = 1.

I TI

BAO 0 0 14 8

TABLE XXXI

!

ABILITY OF SYMPTOMS IN POPULATIONS BELIEVED TO HAVE AOL To DISCRIIEXATE FROM POPULATIONS NOT BELIEVED TO HAVE AOL

tion on Asked

Sudden whiteness, paleness, or blanch-

Thickening of skin of hands, wrists

Question

25

34 36 Whitish pimples 37 Shiny skin patches

Dfscriminance

ing of fingers

Moderate

Slight

Sudden redness or blueness of fingers Stiffness, fullness, or tightness of

Ridges or cracking of fhgernails

26 32

40 fingers

28 of fingers

29 Warm fee l ing i n fingers 30 Cold fee l ing in fingers 33. Rmibness, fingers fall asleep 33

‘Plngling, pins and needles sensation

Frequent soreness of f inger t i p s

None 38 Smell raised areas or lumps l /2 in . round

39 pits or scars of Pingernails

C. plants Believed t o Have Cases of AOL vs. Those P l a n t s Compounding onfy (Table XIM, (1) vs. (2))

In all categories, the occurrence of symptoms appears greater in the plants rbrth AOL than in the compounding plants. significant only in the case of symptoms 825, 26, 32, 34-37, and 40. t h i s variation I s a function of the lou discriminant puwer of the symptams (Table xlacr) and the relatively smaller group in the compounding popula+ion.

Rawever, this difference fs Same o f

d. Plants With Possible AOL Compared With Plant$ Believed Free of AOL (Table XXX, (3) vs. (4))

The distribution of differences in symptom occurrence between the two populations does not permit one to discriminate between them on this basis, Again, the problemsein relatively s m a l l pogulatfon size in this group af questionable plants ( L e * , 555 subjects) compared to the relatively large

!

78

BAO 0 0 14 9

I .

79

Q .- population i n p lan ts believed f r e e of AOL may account f o r t h i s i n a b i l i t y t o a

Q N m C

W

discriminate between t h e two populations on the bash of symptoms.

Q)

m - - e. Plants With Possible AOL Campared’WFth Plants Believed t o Have AOL (Table XMI, (l)-vs. (3))

By contrast t o the above group, the population appears to be di f fe ren t Fn the p lan ts w i t h possible AOL from that i n the plants with AOL on the basis of symptoms, This suggests t h a t our diagnostic c r i t e r i a a re rigid and err on the side of conservatism (see p. 63). .

3. Discriminant Not Having AOL

Abili ty t o Different ia te P l a t s With AOL from Plants Probably

Comparing the s t a t i s t i c a l l eve ls of significance of symptom occurrence differences between plants H t h AOL, plants without AOL and plants compounding only, a hierarchy of discriminant capabilfty f o r each syqtom can be constructed (Table Xxm)

Discussion: ‘7t should be emphasized that the foregoing analysis appl ies o a l y t o cases ofpAa have soreness of their finger tips (question #33>, the other smtoms i n t he law discrfmirrance grmp are probably so common -h a normal populatton as t o render them weak d.%scriminators fn R papulation poten t ia l ly

apul a t ions ra ther than individuals, While .it is quite true that most

8t d s k .

Finally, 6f signs assocfateclwith AOL, Raynaud’r; phenomenon i s s t i l l the most powerful screeninq tool of any of the questions asked i n detectfng AOL in a population.

B. JOB HTSTORY AND ITS RELATIONSHIP TO POSSIBLE DISIT(1BUTION OF AOL AMONG - .- VARIOUS PLAms

1, Relationshlp of Respondents Job History and Man-Months Drposure t o Diag- nos t ic Categories .

a. Cases per man-month exposure i n specific j c b . (Table =I)

b. Cases per exposed population. (Table -11)

C. Cases with experience i n each Job per t o t a l number of cases in each diagnostic category. (Table XXXIV)

BAO 0 015 0

CASES PER MAN-MONTH EXF'OSURE IFI SmClFIC JOBS

Job 3 Total response per job: Tota l man-months per job: 2274 32332 1860

Ratio, case/mos., DX 5,6

Sob 1 Job 2

498 860 477

10 1 5 ' 1 - = - - = - -E- 5 1860 372 2274 455 12532 =YZs

Ratio, case/mos., Dx 4

Ratib, case/mOs., DX 3

2 1 2 1 2274 1137 3.232 1392 1860 930

- = - 1 - - x -

10 1 40 1 11 1 - c - - = - -= -

169 Ratio, cmse/mos., Dx 3-6 2274 227 12 53 2 313 1860

Job 4 Job 5 Job 6

820 Total man-months per job: 16291 5'76TT 13373 Total response per job: 393 125'7

mtio, case/mOs., DX 5,6 0 0

mtio, case/mos., Dx 4 0

1 8 = - 13373 1672

5765' 19219 13373 4458 1 A=:& A=-

Ratio, case/ms., DX 3

Job 7 Job 8 Job 9

418 Total man-months per job: 2452 1893 16551 % Total. response per job: 206

Ratio, case/mos., Dx 5,6 1 2452 - 0

-= 4 - 1 16551 4138


mtio, case/mos., DX 3

0 0 0

1 3= - 2452 817

1 1893 - 1 7 = -

16551 2364 4 1 ll 1

-= - 3. - - = - 16551 1505 Ratio, case/mos., Dx 3-6 2452 613 1893

I .

'80

BAO 0 0 15 1

TABLE: ZXII (Continued)

ia 43 m u .- a

W

N h)

L Job 10 Job 11 Job 12 0

697 169 61 Total response per job: Total man-months per job: 3069 25010 103 4

1 Ratio, case/mos., DX 3,6 3 6 9 25010 8333 103 4

0 Ratio, case/mos., Dx 4

1 - 1 A= - -

0 2 1 25010 12505 -e-

Batlo, case/mos., DX 3 1 -

3069 I - 17 1 -= - 1 345

- mtio, case/mos., Dx .. 3-6 15? 5 25010 1471 1034

Job 3.3 Job 14 Job 15 . .- - -

61 Tota l man-months per job: 83 42 4782 1633 T o t a l response per job: 288 441

4 1 7 1 - = - 0 8342 7086 4782 683

- = - mtio, case/mos., DX 5,6

Ratio, case/mos., Ijx 4


1 - 1639

0

Job 18 -1

Job 16 Job 17

12 Total response per job: 24 1.047 . Total man-months per job: 1067 3-2984 172

1 7

18 1 12984 721 1’72

A= - 129% 1623

-= - Ratio, case/mos., Dx 5,6 0

0 1 0 -.-Ratio, case/mos., Dx 4 - -

Ratio, case/mos., Dx 3 0 0

1 - 1 0 49= - 172 I2984 265 Ratio, case/mos. , JJX 3-6

BAO 0 0 152

TABLE XXXLI (Concluded)

Job 21 Job 20 - Job 19

Total response per job: 893 166 234 T o t e l --months per job: 2U71 a07 790s

A= - 1 Ratio, case/mos., Dx 5,6 2u72 162

1 2107 - 1 -

7985

-0 0 4 1 21171 5293 -L: - Ratio, case/mos., Dx 4

Rat lo , case/mos., DX 3 2 - z - 1 7985 2662

0

1 - - - - -1 4 :?lo7 7985 I996 - 1 A= 7

Ratio, case/mos., Dx 3-6 2U71 605

Job 22 . Job 23 Jab 24 Total response per job: 106 108 253 Total man-months per job: 1309 3625 4988

Ratio, case/mos., Dx 5,6 0 - 4988

mtio, case/mos., Dx 4 0 0 1 - 4988

Ratio, case/mos., Dx '3

1 5= - 1 1.709 262 653 4980 998

- 1 ---s= - Ratio, case/mos., I>x 3-6

Job 30 - Job 25

Totsl response per job: 55 11-77 T o t a l man-months per job : 1490 32826

Ratio, case/mos-, Dx 4,6 0 I - 32326

Ratio, case/mos., Dx 4 .

Ratio, case/mos. , Dx 3

0

0 14 1 32826 2345 -q - 18 1

32826 1824 - = - Ratio, case/mas., Dx 3-6 0

a2 I

BAO 0 0153

TABLE XXXIII

CASES FER ExTy)SED POPULATION

~~~~ __ -- Job 1 Job 2 Job 3 a Jobs Job 6

population at risk: 498 860 477 393 UT 820 Cases/acposed pop. (Dx 5,6) l/lOO 1/06 r/s 0 0 1/102 Cases/exposed pop. (Dx 4) 1/249 1/96 l /U9 0 0 Vn3 CMas/aKgoscd pop. (Dx 3) . i/r66 1/41 1/ll9 1/98 1/140 1/41 Casca/exposed pop. (Dx 3-6) 1/50 1/22 1/43 1/98 l/rc6 1/26

Population at risk: 206 % 418 87 697 169

Csae8/exposcd pap. (DX 4) 0 0 0 0 v349 0 Cases/exposed pop. (Dx 3) 1/69 1/56 1/60 1/87 1/63 1/85 Cases/eKpo sed pop. (Dx 3-6) 1/52 1/56 1/38 1/44 1/41 us6

Papulation at risk: 280 441 61 24 1047 22

caScS/eqO8ed pOP. (m 3) 1/56 v55 0 0 1/44 6 Casea/exposed pop. (Ilx 3-6) 1/22 1/25 1/61 0 1/21 1/12

Job 7 Job 8 Job 9 Job 10 Job 11 Job 12

Caseu/exponed pop. (Dx 5,6) 1/206 0 1/105 1/87 1/232 l/l69

Job 13 Job 14 Job 15 Job 16 Job 17 Job 3

~sses/urposed pop. (Dx 5,6) 1/42 VJ63 0 0 1/61 1/12 Cesea/expored pop. (Dx 4) 1/96 1/= 1/61 0 1/131 . 0

Job 19 Job 20 Job 21 Job 22 Job 23 Job 24 papulation at risk: 893 166 234 1 6 108 23 Caueh/exposed pop. (a 3,6) 1/69 1/166 US4 1/16 0 1/23 cnacs/urpoaed pop. ( n ~ 4) 0 0 0 0 1/2% Cases/ucpascd POP. (& 3) 1/50 0 1/78 1/27 1/22 L/&

1/21 1/22 l/5l Caaes/expOsed pop. (DX 3-6) 1/26 1/166 u59

Job 25 Job 30 population at risk: 55 u77

* ~ases/exposed pop. (Dx %6) 0 l/W? CIISe8/urpOted POP. (m 4) 0 1/39 cSse8/urposed pop- (DX 3) 0 1/84 Q s c s / ~ o e e d ~ - (Dx 3-61 o 1/65

BAO 0 0 154

I) t -

CASES UITB EXPERIEI9CE IM EACH JOB PER TGTAL NUMEER OF CASES IN ZXcH DLAGNOSTIC CATEGORY

- Job 1 Job 2 Job 3 Job 4 Job 5 Cases ~ L t h q e r l e n c e in Dx 5,6 5/25 10/25 (4q) 5/25 O/ o/ job per total number of Dr 4 2/16 9/16 (56,s) 2/16 O/ 3/16 cases in each diagnostic Dx 3 3/75 21/75 (2&) 4/73 4/75 9/95 cateaory Dx 3-6 lO/ll5 40/115 (35.7%) 1-1/115 4/u5 l2/11q

job per total number of Dx 4 3/16 O/ o/ o/ o/

- Job 6 Job 7 Job 8 Job 9 Job 10 h e s with experience in DX 5,6 8/25 1/25 o/ 4/25 1/25

cases in each diagnostic Dx 3 20/75 3/75 1/75 7/75 1/75

Job I1 .-. Job 12

Job per total number of 4 2/l6 O/ 3/16 2/16 1/16

. catepry Dx 3-6 31/115 4/11? i/us 11/u5 2/n5 - J= Job 14 Job 15 -

Cases w i t h experience in Dx 5,6 3/25 1/25 4/25 7/25 o/

cases in each diagnostic Dx 3 11/75 2/75 5/75 8/75 o/ cateuory DX 3-6 17/115 >/I15 19/U5 19/U5 1/u5

Sob 16 Job 17 -- Job 18 Job 19 Job 20 Cases with experience In Dx 5,6 O/ 3-8/25 (7%) 1/24 13/25 ( M ) 1/25 job per t o t e l number of 4 O/ 8/16 (5%) o/ 4/16 (2%) o/ cases in each diagnostic Dx 3 O/ 24/75 (3%) O/ 18/75 (244) o/ category Dx3-6 o/ 49/ll5 (43%) 1- 1 24

Cases with experience in Dx 5,6 1/25 1/25 1/25 o/ job per total number of DX 4 O/ o/ 1/16 o/

o/ O/

Job 21 Job 22 Job 23 Sob 24 Job 25

$75 3/75 m e a i n each diagnostic Dx 3 3/75 4/75 ca%egom DX 3-6 4/115 5/ 11.5 15/115 5h.15

Job 30 Cases with experience in Dx 5,6 1/25 job per total humber of Dx 4 3/16 cases in each d i w o s t i c Dx 3 14/75 category Dx 3-6 18/115

I .

84

BAO 0 0 15 5

c

2. Cluetedng of H l g h Associational Relationships Between Jobs and AOL

It is apparent that AOlL has a strong assaciatianal relationship to job 17, i.e., reactor cleaning. Using the ratios for job 17 8s a range finding de-dce, the ratics which approached those of job 17 were noted. clustered as indicated by Table XXW.

These

C L U S T " G OF JOBS WITH DIAGNOSTIC CRITERIA

Cases/Total Cases with This Job Ekperience.

Diagnostic Cases/--Mas. Cases/-. POP. Category Inciaence > l / & O Incidence > 1/75 Inc.tdence , 4&

I. 3 14 17

13 14

2

17

Cases/Total Case6 with Cases'm* pqp. Job Experience. Diagnostic Cases/Man-Mos.

Categary Incidence > 1/600 Incidence > 1/50 Incidence > 23$

2 2 6

2 6 8

3 14 17 9 22 13

14 17 19 22 23

3

17 19

L-

o f. . W

4 Population too snmsll for antilysis

I .

. --

BAO 0 015 6

. ..

I

T; P .-. . 3 ' 0

B m N

3. Discussion of Clustering of Jobs with Diagnostic Categories 5 , 6

a. Job 17 ( reac tor cleaning). This Job appears unquestionably t o be as- - I n fact, Table XXXV indicates t h a t a l l but 7 of t h e 25 cases w i t h de f in i t e

W e c3

soclated both with AOL and Raynaud's phenonenon without bone changes (Dx Cat. 3). AOL had his tor ies of w a r k i n g i n job 17. individuals reveals that 6 of these worked a t polycleaning during the decade preceding t h i s study period, €.e., p r io r t o 1961-66. apparently has AOL and w h o never worked a s a reactor cleaner, worked as a laboratory assistant. scale reactors. Accordingly, the association of reactor cleaning w i t h AOL appears highly signif icant .

Further inquiry concerning these 7

The other worker who

Despite some question, he reports having cleaned bench

b. Job 19. Since t h i s is i n l i n e of the usual upgrading route for many employees who t e w l n in WC production, it follows t h a t t h i s job his tory will be associated with that of having previously worked in job 17, reactor clean- Ixlg.

c . Job 1 ( additive makeup) and Job 3 ( ca ta lys t makeup}. jobs a re low pay grade duties. of the same individual who p e r f p m s the equally low-grade job of reactor cleaning.

Both of these Frequently these tasks a re the respons ib i l i ty

d. Job 2 (bageng end packing). The same cond.ltions apply to t h i s job as n i t h job 19 and/or jobs 1 and 3.

e.

The two who didn't give a history of t h i s job w e r e found on subsequent

Job 14 (monomer recovery operator). Once more, of t he seven workers with AOL having history of work on t h i s job, f i v e of these a l s o worked on job 17. inquiry t o have worked as reactor cleaners i n the 19ISO's.

f. Job 13 (monomer operator). The fuur persons with AOL had all worked at job 17 during the study or prevlausly i n the 1950's.

6. Jobs 22 and 23. In d e w aP the smal l t o t a l size of this population in terms of m@n responding an& man-months exposure, in te rpre ta t ion of this data is not feasible,

4. Different ia l Distributfon of Duration of Work Experience in Specific Jobs

A distribution of' the number of men who worked i n each job category, 1-30, for periods of 1-6, 7-12, 13-24, 25-36, 36-k8, 49-60, and 61-72 months was obtained. It w a s QUT i n t en t t o b v e s t i g a t e the poss ib i l i t y t h a t men i n plants where there w e r e cases of AOL remained i n high risk jobs for l o w e r periods of t i m e than men in the same jobs at p l a n t s .without cases of AOL.

BAO 00157

The r e su l t s of these computations faf led t o indicate that men i n plants ~

with AOL remained i n t h e h igh risk jobs longer than In plants without AOL. For t he higher risk job, i.e., job 17, reactor cletming, the number of men on, rc.

job 17 i n each time interval , t h e i r t o t a l man-months on job 17, and the cunnila- C.

6,

m

tfve percentage o f t o t a l workers reporting job 17 experience i s given in.Table Xxxvx. worked 36 or m o r e months i n reactor cleaning than bras the case 5n plants without AOL (p < 0.02). 24 or more months or 48 or m a r e months f a i l e d to s h w any significant difference i n the number of men engaged in reac tor c l e a n h g between eit.heP group of plants (i.e., with or without cases of AOL) suggests that plants with cases do not have the pr&lem of AOL because of .longer tours of duty or more men on such t o u r s Fn this job. it appeers that the period of exposure, i n most cases, ts less than two years. This observation tends t o reinforce the above conclusion. .

h)

CD Chi square analysis indicated t h a t in plants with cases, more men

However, the fact that similar analysis on the basis of .

Indeed, on the basis of the clinical history of AOL,

5. Job Expertence of Individuals w i t h Raynaud's Phenomenon on the Basis of Diagnostic Category 3 Among Various Plants

Considering the individuals placed i n t h f s diagnost ic category and t h e i r experience in reactor cleaning, t h e i r work his tor ies =e summarized i n Table XXXVII. -

R e d e w of t h i s data suggests the poss ib i l i t y that induction of Raynaud's phenomenon among PVC workers may not' require reactor cleaning exposure. That is, while AOL (bony changes plus Raynaud's phenomenon) i s associated with reactor cleaning, indtkt ion of Raynaud's component of AOL !hay be produced because of absorption of one of several components t h a t mLght also be found i n other locat ions in t h e PVC industry, e.&, workers potentially exposed t o Vrnyl chloride,8 as in Plant C where o n l y monomer i s produced.

However, as seen previously, because job histories' are available t o this survey only for the period l96l-66, w e are unable t o determtne whether any of those Fn diagnostic category 3 may have had reactor cleaning experience pr ior- . __ t o the survey period. -

- C. ANALYSIS UF X-RAYS VS. DIAGNOSTIC CATEGORTES 5 , 6

.. ._ . - 1. Distr ibut ion and graupings of the x-ray defects i n this group is as

follows:

BAO 0 0 15 8

I

4

- T O t d 8

Wontba on Job 17 1-6 7-l2. 15-24 2536 37-48 49-60 61-72

uo. w. uo. no. xo. uo. No. no Men k8. Won. Her* Hen Men kr* Men Ilo8* Hen wn. xen &I. *ka

cLamlllatiw 71 9> 99 100

TABLE XXXVII

possms WORK EXPERIENCE IN JOB 1 p AMONG INDNIDUAUI WITH RAYNAUD'S PHENOMEZON ONLY (Dx Cat. 3)

IN HIGA, WSSIBU, AND I x l W RISK PIAEETS

JOII 17 Worked Did not Work

Plsnts with AOL 18 27 Plante with Questionable AOL 0 8

5 15 Plants without AOL

*Male responilents only.

Cases Defect No. * Cases

Defect No.+

1 d o n e 5 1, '2 1 I, 2, 3, 10, ll 1, 2, 4, 8, LO, U. 1, 3, 4, 6, 7

1 1 1

1, 6 , 7, 10, IJ. 1: 1, 7 1 1, 7, 8, 9, 10, IJ 1 I, 7, 10, = 1 1, 2

2 1

1 1

1, 3, 7 1, 4, 7t 10

3, 7 1 5 alone 1

~mong those indtviduals i n diagnostic categories '5, 6, the prevalence of these x-ray defects, alone and/or in ccmibination is as follows:

Defect 1, o r in combination w i t h others 'Defect 2, or in combination with others Defect 3, or in conbinatkm k t h others Defect 4, or 5.n combination w i t h others Defect 5, only Defect 6, or i n combination with others DeOect 7, or in combinatLon with others Defect 8, or i n combination with others Defect 9, of: in cpmbination with others Defect 10, or in combination wlth others Defect ll, or i n combination with others

20125 = &$ 5/25 = 20$ 7/25 = 2 w 4/25 = 1 6 1/25 = 44 5/25 = 2 6 ll/25 = 44% 2/25 = 8$ 3/25 = mJ 9/25 = 36% 8/25 = 324

Further review of t h f s group of 25 probable cases of AOL (diagnostic categories 5, 6) revealed only four individuals who had o n l y one finger involved. finger, another two had defect fects l, 3, and 7 affect ing only one finger.

Of these four, one demonstrated x-ray defect 1 only on a single 2 on one finger, w h i l e t he last case lzad de-

I ss

BAO 0 0 16 0

J . :

5 1 Discussion:

must be recal led (c f . Table ViII) t h a t t h i s defect occasioned considerable discordance between x-ray readers. af defect 7 is high, it should be noted that it was rarely seen alone i n this group of individuals. Since t h i s group did have m u l t i p l e defects in the main (16/25, or a&), the presence of shortened f ingers should be considered as an accompaniment rather than a hallmark of acroosteolysis.

While d e f e c t 1 is seen most frequently among these casest it ZI), (t,

While the degree of agreement and frequency CI

z -

2. Comparison of X-Ray Defects i n Diagnostic Categories 5, 6, and Tecumseh Control Population (cf. Tables VII, X)

The plant populations in By E, F, 9, R, U, and along with t he fie- quencg of defects i n these p lea ts is compared with similar data frcun the control populaeion i n Table XXXVIII.

TABLE XXXVIII

PIE OCCURRENCE OF SPEclFIC X-RAY DEFECTS I N THE TECUMSEH CONTROL POPULATION AND POPUMTION IN PLANTS BnIEvED M HAVE CASES OF AOL

(i.e., Ihc Categories 5 , 6)

Occurrence i n Control

Population,

Occurrence .

P(df = 1) X-ray DfaeT.lostic C a t . 5, 6, Defect .. ._ - N = 2407 Plant Population = 1673

1 . 139 44 < ,001 2. 16 5 e 0 0 0 5 3 16 6 -005 4 9 0 m

5 3 2 - 6 10 1 < -001 7 10 ll. NSD 8 22 4 < .001 9 4' 1 -

Conclusions: Differences i n x-ray findings between the Tecumseh Control Papulation and the PVC population are subject t o possible bias for the reasons discussed previously.

From the x-rays of cases c lear ly consistent with AOL, several conclusions can be drawn:

a. This disease more frequently tends t o involve multiple phalanges.

b. The defects 2, 3, 5, while more consistently interpreted i n x-ray

. . I 90

reading, a re defects t h a t occur i n a normal population.

c. of AOL.

Defect 7, i.e., "shortenings'; alone i s a tenuous basis f o r suspicfon

d. Defect 1, i. e. , 'barginel defect ", while undoubtedly associated with AOL, is susceptible t o considerable subjectivity i n x-ray interpretat ion.

e, Defect 9, foe., "total. resorption of distal, portion phalanx," might be readily conFused i n surveys with old amputations.

f. Given the subjective considerations invalved i n d e t e d n i n g the presence o f defect 1, t h a t defects 2, 3, 4, are probably most def inf t ive in

'mdsing the diagnosis of AOL,

3.

involving multiple fingers) and who also had no symptoms of Raynaud's t h e d i s t r ibu t ion of defects w a s a s follows:

Distr ibut ion of X-Ray Defects i n Diagnostic Category 4

a. Among those workers whose x-rays were not def in i t ive (and usually net

Defect No.

1

No. of Occurrent e s

3 2 I 4 2 1 1 1 1

.-

b. The prevalence af these x-ray defects, alone or in combiaatim for diagnostic category is gs follaus:

Defect I, or in combination with others 6/16 = 38% Defect 2, or .Ln combination with others o h 6 = C% Defect 3, alone 4/16 = 256 Defect 7, alone or i n combination wlth

others 9/16 = 56$ Defect 8, i n combination with others 3/16 =.19$ Defect 9, tn cambination wlth others 3/16 = 19$ Defect 10, i n conibination with others 3/16 = 1%

BAO 00162

Discussion: It can be seen that defect 7, i.e., "short fingers," poses a severe problem in v i e w of a surprising high relative incidence (cf. Table XXXVIII) i n an ostensibly normal control population. While it may be argued that the individual manifesting th i s alterat'ion may have had asymptomatic AOL, given the plant constellation of cases ( o r the i r absence) one should be able t o Interpret the significance of such findings in a pupulation under investigation. As previously noted, defect 1 was rather coxuaonly found. Problems i n interpretation of the defect's presence or absence have been discussed,

e m c1> rc)

m c w c e

W W

4, Comparison of X-Rays i n Control Population vs. Plants Believed Not t o #ave AOL

The occurrence of the x-ray defects i n the Tecumseh population a s compared t o the plants believed not to have AOL cases is shown in Table )(xxM,

It fs apparent that while defects 1, 2, 3, and 8 appear mora frequently i n the PVC plant population believed not t o have cases o f AOL, t h a t t h e levels of significance for defects 2, 3, and 8 are not impressive when compared t o &he difference -in occ1uTences of defects indicated i n Table -1. Further- mnre, the difference in occurrence of defect L suffers from the lack of objec- t ivity in x-ray interpretation since we were unable t o mask the identifying marks on the control population films,

-

TABLE XXXIX

COMPARIBON OF X-RAY DEFECTS IN TECUMSXH POPULATION WITH PLANTS BELIEVED NOT TO IiAW, AOL CASES

Occurrence of Occwrence of X- X-Ray Defects . ray Defects i n

Defect i n Control B Plants Believed 4 P(d9 = 1) N o t t o pave AOL,

N = 2428 Population, N = 2407

I. 44 1-83 106 4.4 < -001 2 5 - 0.21 14 0.57 c -05

6 0.25 16 0.66 c -05 4 0 - 8 0.33 00 3

5 2 0.08 4 0.16 too feu 6 1 0.04 2 0,082 too few 7 li 0.046 18 0.74 ITS a 4 0.017 13 0.54 .m 9 I 0.04 1 0. ob too few

BAO 0 0163

D. DIFFERENCES IN EXPOSURES IN PLANTS WITH AND W 1 T K C ) U T AOL

31 c z c a Q m N 8

u m

. . 1. Materiels Used CL

' A comparison of those msterials used exclusively i n each p lan t and of t h e materials used i n plants with and w i t h o u t cases f a i l e d t o d isc lose any mater ia l or combination of materials present i n the "positive" p l an t s and absent Fn t h e n

I negative '' plants

I

On comparing the exposure to ca ta lys t s in the "positfve" p lan ts with those

This one p lan t experienced in the "negative" plants , a number of s i g n i r k a n t observations may be noted. All of t he p lan ts except one used lauroyl peroxide. cases of AOL. of t h e study. These results w o u l d fndicate t h a t no one spec i f ic catalyst is implicated.

Only one catalyst .was used by this p l sb t throughout the period FLve of the seven "positive" plants dLd not use t h i s catalyst.

2. Reactor Cleaning

Procedure followed i n the several plants in the cleaning of reac tors is : 1 presented in Table XL. ' 'Although all except one of the seven "posit iveh p lan ts cleaned the reactors by hand scraping d t e r every batch and only 10 of the 17 n negative" plants did so, it is to be noted t h a t there are three lfmes as many employees in t he "positive" plants than i n the "negative" plants . s i t na t ion tha t i n t e r f e re s with t h e va l id i ty of a d i r e c t comparison between these bdo groups of plants is t h a t the former group of seven p l an t s include t h e e with more than two hundred eq l3yees whereas t he la t ter group of 17 plants include only two plants w i t h more than th i s rxnber.

Another

Prior t o the end of the study period, w a t e r j e t cleaning of the r e a c t m s was introduced Tb two of the "positive" plants and i n one of the "negative" plants .

Of the three plants which cieaned reactors by mans of solvents, t h e one i n which possible AOL cases occurred cleaned by hand scraping except i n one process until 1966. That year solvent cleaning was applied in te rmi t ten t ly t o

December 1966, was the solvent method used en t i r e ly Tor the glass-lined reactors. I n one of' the "negative" plants, solvent cleaning was the exclusive method since the be&nn€ng of operation i n 1966; i n the other, solvent cleaning was not begun u n t i l September 1964.

the glass-l ined reactors but not the s t a in l e s s steel reactors. Not un t i l --

Inquiry was made concerning w e d n g gloves durfng hand scraping of reac-

Cases tors, but the prac t ice was so haphazard t h a t no conclusions may be drawn. Only i n two of the plants was there a requirement t h a t gloves be worn. were found in one of these and no cases in the other. gloves was found tin use: rubber and neoprene costed cotton being the most u s u a l .

A grea t assortment of canvas to avoid abrasion, PVC fnqregnated cotton,

I 93

BAO 0 0 164

-=A = CY-’

TABLE x1;

REACTOR CLEANING PROCEDURISS 6, Q)

N e h- W

UJ Hand Scraping en

Plant After Every L e s s Than Cleaning by Cleaning by Code Batch E v e r y Cy cle Wa.ter Jet Solvent

Plants with AOL Cases B E F IC

R W

Ex

B3

cc

A

C D H J L N P

Q S T

V 2

AA

DD GG

From 1965

Every 5-12 ErOlU 1962 cycles (Bldg. A) (Bldg. B)

X

X

Plants with Possible AOL Cases Every “several “ On glass-lined, cycles intermittently

in 1s6, entirely by Dec. 1966.

Plants with no AOL Cases X

(to Sept. 1964) X’

X Every 2-112 cycles Every 4 to 8 days

X .. Weekly tbru 1964 every 2 weeks w35-66

X X X

(Eimlsion) Every 1 to 3 cycles (Suspension) Every 10 cycles

X Every 4- or more cycles

X From start of operations

I

BAO 0 0 165

4

. -

. a

3. Procedures Following Completion of Polymerization

The dis t r ibu t ion of the values of the absolute pressure-time f ac to r as discussed i n Section IV.D.4 for the "positive" plants as compared witb those for the "negati.ve" p lan ts i s presented i n Fig. 15. Although there were more of the "negative" plants w i t h values of 16 or more than was the case with the "positive" plante, the values for t h e "negative" plants extended d m through the "positive" plant values. It Was considered tha t no significance could be ascribed t o these differences.

In Fig. 15, each X represents a plant excepting where two X's are joined, in which case they represent the high and l o w values of the absolute pressure- time factor f o r t h a t plant. that plant.

The locat ion ai' the X indicates the value for

"Negative" X X X =--X x x . x x x- - - - - -- x x

X X

*Inches Hg/lo&& is incnes Hg absolute pressure Over logarithm of the time i n minutes.

Fig. 15. Absolute pressure-time values.

I

95

110111 llllal m lill m IO llal BAO 0 0 16 6

I

SECTION VI

SUMMARY AM) COmCLUSIONS

A. THE DEFINITION UF ACROOSTEOLYSIS

Any statement as t o the occurrence of acroosteolysis i n the polyvinyl chlorfde Fndustry- must be first preceded by 8 deMnition of t h a t clinical syncIrame. Examinat%& of Table XLI demonstrates that there a re severdl x-ray defects that may be consistent with AOL ye t of thmselves me not diagnostic of this condition. I n addition, the presence or absence of Raynaud's phenm- enon is another. prime diagnostic feature which must be considered.

.

X-RAY DEFECTS ASSOCIATED WITH 1M)IVTDUALS IN DIAGNOSTIC CATEGORIES 3 m 0 U G H 6

Number of IndirLduals

X-Ray Defect D x C a t . 5, 6 . Dx C a t . 3 Dx Cat. 4 (with Raynaud's) (with Raynaud's) (without Raynaud '$1 5 15* - 2-5

1 Multiple Fingers

Concurrence Ronconcurrence

SPngLe Finger

Concurrence Nmc oz1cupreac e

- 8-10 .. .

NO bone defect

TGTAL 25 74 16 *one case did not have Raynaud's phenomenon but did have nniltiple f'inger x-ray defects.

I 96

BAO 0 016 7

c a

'

r Q )

m m rc) In its most patent form (diagnostic categortes 5 , 6), AOL is c lea r ly

characterized by a radiolucent les ion involving one o r more d i s t a l digital phalnn es accompanied by Raynaud's phenomenon.

U w

w OD

B. COURSE OF DISEASE

I. Early

Most. frequently there i s only Raynaud's pheno:ne.non without x-ray changes.

V a r i a t i o n s : (a ) X-ray les ions and Raynaud's phenomenon may be reported simultaneously to the physician. However, close questioning of the pa t ien t usually elicits a history of Raynaud's phenomenon preceding the takFng of the x-ray. The p rec i se ' r e f a t ionh ip of the first x-ray change i n relation t o the t i m e of onset o f Raynaud's phenomenon can never be known for sure. (b) In the present study, one case has been found i n which the questionnaire was answered nega t iveu tn response t o all questions concerning signs and symptoms. x-ray subsequently demonstrated tha t t h i s worker had an active, progressive case of AOL, the poss ib i l i t y ex i s t s that AOL caxi 0c.cur asymptomatically. However, t h i s w a s the sole case out of 25 patient cases of AOL and might also be considered as an instance of an anomalous natural his tory f o r AOL.

. _

Since

2, A c t i v e Disease -

The x-ray changes involve multiple fingers and are associated w i t h signs and symptoms involving the hand and i ts vascular t ree . While signs o r symptoms of whitish pimples (#$), thickening of the skLn (#34), and shiny white patches (#g) a re of re la t ive ly high discriminatory power, i n the absence of Raynaud's they are of such low incidence i n plants with or without cases t h a t they must be considered nondescript. While it first sppears tha t symptoms of tingling, pins and needles (&a), a warm sensation in the hands (#29), a cold sensation of hands and f ingers (#%), a numbness of f ingers (#3l), and stiffness, f'ullness, ar tightness'of f ingers (#%) should be indicators, t h e i r discriminatory power is vftiated by t h e i r high frequency i n the t o t a l population. The symptoms referring t o skin and fingernail l es ions occurred too infrequently in the population w i t h AOL t o be coinpared with the compounding population group-

3. Recovery Phase

During that period when the bone les ions heal, i n most cases the my- m u d ' s phenomenon. tends to remain s t a t i c far many years. tendency indicates that, i n view of several medical and job h i s to r i e s obtained subsequently t o the survey, diagnostic category 6 cases probably represent

(Indeed, t h i s

I

97

i!

cases of AOL that were active in t h e 1950's.) SEaling of the bone lesions r e s u l t i n the development of x-ray defect 5 as a consequence of bony tuft ", dissolution, while reconstruction o f the altered shaft may cause x-ray de- f e c t s 6 through 10. While it is possible that healing of the bone may producE e to ta l ly reformed, nor& appearing, d i s t a l phalanx,this has not been seen by other €nvestlgators. - C. POSSIBLE VARIATIONS IN TfIE NATURE AND COURSE OF AOL

1. Raynaud's Phenomenon without Eone'Change (Table XLI, Dx Cat. 3)

Theoretically such a syndrome might be possible. Thus, indfviduals -thin dkgnostic category 3 d g h t be suffer ing e f fec ts of this industrial. exposure. That such might be considered possible i s suggested by several plants, one of which synthesizes vinyl chloride monomer only, having a somewhat higher incidence of Rayaaud's phenomenon than our putat ive llcontrol" population. viously, the use of the data obtained in t h i s study t o determine ''normaUy" occurring Raynaud's phenomenon has severe l imitations. aut t h a t one worker who manifested only Raynaud's phenomenon (Dx Cat. 3) later developed bone lesions. Tlzus, haw many persons with Xaynaud's phenomenon only represent the early phase of AOL one cannot determlne.

Ob-

It must be pointed

The question might be asked whether AOL actually r e s u l t fram exposure to v1: monomer (producing the Raynaud's phenomenon corqponeat) and other agents in the polymerlzer scraptngs which cause the bone changes.

2, %ne Changes without Raynaud's Phenomenan (Table XLX, Dx Cat. 4)

Though one case w i t h active AOL bone lesions denied signs and symptoms,

Revertheless, t h i s possibility must be raised in d e w of the cases of I t s t i l l is difficult to conceiye that this w o u l d obtain over any period OF time. the individuals i n diagnostic category 4 haVFng bone defects suggestive of mL3.d AOL but without; Raynaud's phenomenon. fciUowed through the course a? their employ.

These individuals should be c lose ly

1. As discussed above, any descript;ion of the prevalence of AOL will depend upon haw one defines t h i s condition. Accordingly, we have grouped prevalence data according t o whether the AOL syndrome is definitive (Dx C a t . 5, 6) o r potential cas-es in Dx Cat. 3 plus 5, 6, the resulting prevalence being shown in Table XLIX.

98

BA00016 9

I

M P 0 0 0 t-' 4 0

i

Prevalence Queotlmable or no

Rlotory, Reactor Cleaning

&porting & Reactor cleaning Eqerience, Reporting Reaotor Cleaning

w r i e n a e , 1961-66 by F o l l ~ - u p

Total --yearn, raaotor . l€ia d. I Care per cleurfng sxperienoe = la82 1082 60 mn-years

Definitive md porribla C W e I (h. Cat. 3 md 5,6) - 41

Total workerr at rink, relotor (lSa+lOR) Csro por Clem- - 1007 1047 37 warkern Total aan-years, reaotor Jl€ia+lOa1 1 I Care par ChMing 1082 lOfj2 39 Illan-yearr

. . ~

Total PVC work iorca, M e

Men a t rink = 4751

Tutal pran-yewr:in Pvc pafiation = 21510

Case per man-years Man-years in = = in PVC study lmpula- = - PVC otudy -I 1 Cara per man-years i n

768 PYI: study population 2130 633 tion *'so 3 q 3 powation

a By nnalyrin of rwcy data. 8ae page 100.

nbt anilabh ror Dx cat. 3 pOpUhtiO!h bBased upon port-awey follow-up, rls workerr in Dx cat. 2, 6 were d e t a d n e d to have had relrctor cleaning sxperieqce.

C ~ ~ r ~ ~ r that rix workers in DX. cat. 3 M no reactor cleaning exparianoa.

Similar data

a. Bevalence, def ini t fve cases (Dx Cat. 5 , 6) = 25

(I) Numerators

e W a? e CD Y

W

(a) Those reporting reactor cleaning experience within study period = 18 L

(b) Those found t o have reactor cleaning experience prior to study period 1961-66 = 6

(c) No reactor cleaning experience (possible lab-scale e q e r i - ence) = 1

(a) Total, cases = 25

(2) Denominator

(a) Number of men reporting any experfence in reactor cleaning = 1047

(b) Number of maa-years of reactor experience = 1082

b. Prevalence, potential cases of AOL among vorkers in diagnostic category 3 = 16

(I) Among the workers found in diagnostic category 3 are eight individuals having x-ray defects that a r e possibly compatible with AOL (see Table XLI). Another group within diagnostic category 3 - totalling eight person6 were notea to have 3 or more of synrptoms 28-35 fs addition to Raynaud's phenomenon.

Tbose among these 16 potential cases those reporting experience in =actor cleaning during 1961-66 = 10

(2)

(3) No follow-up is available t o define this group's work experience prior to the study group. reactor cleaner assignments for 1961-t;T.

However, the remaining 6 do not show

2, Discussion: Because there was no concerted follow-up of the 74 ins- vidn&ls.-in diagnost;ic category 3, the significance of the cel ls at the r i g h t of %bPeXLIIis open to question. -

E, RELATIONSHIP BETWEEN JOB A S S I G " T ! S AloD PIE OCCWiREmCE OF AOL

For the pwoses of M e r discussion, we shaU be referrfng to a ~ o o s t e o l y s i s (AOL) i n t e r n Of usual Syndrome, L e . , Raynsud's phenomenon, multip2.e hand symptoms, f a w e d by radiolucent changes i n the distal phalanges

I . 100

-. .

or elsewhere, follared by heal ing after removal from the job but w i t h per- s i s t e n t R w u d ' s phenomenon.

1, Reactor Scraping

This job apeears t o be t he greatest single fac tor in the prec ip i ta t ion o

In sop~b few cases outside thi6 Stuer, AOL has been seen to oc- AOL. c8p be found, cur in maintenance mechanics. The cleening of lines from the reactor tanks may afford these workers occasional exposures qua l i t a t ive ly similar t o that of reactor cleaners.

In the overwhelming majority of cases, a history of work i n this job

2. Compounding Operations

No definitive cases of AOL Were found i n operutions involving compounding or fabrication of the f inished polymer. phenomenon occurrence i n such plants. (t.t., 557 individuals) does not prove, of course, that the disease never occurs Fn this operation. population is not unexpected i n v i e w o f the common occurrence of such border- line x-rays i n the control pop la t ion .

Neither was there undue Raynaud's The relatively small population involved

'

The one guestionabXe x-ray (Rx Cat. 4) seen i n th€s

. . - ..

F. MATERIAL EXPOSURE AND AOL

1. Despite campater assisted comparison, no one compound or class of compounds clearly or exclusively w a s associated wL?;h the plants having AOL i n comparison to plants withuut AOL.

2, As regards Raymud's phenomenon, there is a possibility that exposure t o vinyl chloride 2er se may precJpi ta te a response. While it. is understood that the -0-e to -Vir;yl chloride in such unique operations is minimal, nevertheless, the rather unexpected higher incidence of Raynaud's phenomenon in Plant G is noteworthy.

If this turns out t o be the case, the implications for t h e e t iology of Accordingly a close surveilLance of VC monomer production AQL are obvlous.

p u p l a t i o n may prove valuable.

G. WORK PRACTICES AM) AOL

1. Recovery Phase '

The use of a short period for recovering unreacted monomer from t h e pay- mer might lead t o more residual monomer i n the reac tor (e,g., i n voids of. W C

I ,

301

BAO 0 0 1 7 2

1

1

i I

4 I

I

I 1 1

I

I

9

em residue. on reactor W a l l s ) . A study of the temperature and time charac te r i s t ics of the recwery phase in several plants failed t o show differences i n practices betireen plants with AOL and those without.

2 c.l

W Q -P CCI

2. A b Flush of' Reactor Eefore Entry of Cleaner

While operating procedures have been established by plant s u p e h s i o n , it appears that these are being shortcut by hourly eaployees resu l t ing Ln in- creased potential for exposure. - -

3. Reactor Cleaning Procedures

Hand scraping of reactors has been practiced in a l l plants except one that started operation in 1966 and which has always used solvent cleaning,

Among the seven plants wfth AOL, all hand scrsped the reactors, generally after every batch, though one of these started using water j e t s i n 1965 and another cleaned a portion of the reactors w i t h water je ts from 1962;

Amang the 17 plsnt s without AOL, only 10 hand scraped t h e reac tors after every batch and one of these changed t o solvent cleaning i n 1965. mammlly cleaned reactors every 2 t o 10 batches. The plant t ha t s t a r t e d opera- +Aon in-1966 with solvent cleaning has had no cases.

O t h e r s

In the two plants with questionable cases, one hand scraped the reactors after every batch and one after "several" cycles. cleaning intermit tent ly for a portion of the reactors i n 1966.

The l a t t e r u t i l i z e d solvent

Size differences i n p lan t populations between t h e "positive" and "negative" plants reduce the val ld i ty of rneaningf'ul comparisons between the t w o p u p s as related t o cleaning methods.

Use of high pressure water S e t s and solvents for reac tor cleaning g r e a t l y reduced t h e t i m e that workers w e r e instde the reactors following discharge of the batch and a l s o t h e i r contact with materials remaining i n the reactor.

H. CONTROL MEASURES IN PEE"CrENa0N OF ACROOSTEOLYSIS

In suggest- control methods, it would seem i n order t o reduce exposure

Decrease i n the t i m e spent scraping reactors could t o suspect m a t e r i a l s even though the ident i f icat ion of them as the causative agents i s not conclusive. be accomplished by more extensive u t i l i za t ion of high pressure w a t e r or solvents in cleaning procedures. . -1 chloride and also of personnel on the charging f loo r should be kept a t a minimum, desirably below 50 pa r t s per million as suggested by Torkelson e t a1.9 .

f I The severi ty of exposure of reac tor cleaners to

I 102

9 .

m

rather than merely below the current threshold l i m i t v a k e o f 500 parts per million. c-

u - -a, Q To assure that atmospheric concentrations are below this level, air

analysis instruments of adequate sens i t iv i ty should be used. The flammeble vapor detector a t most of the plants gives a positive reading only down t o 400 t o 800 par t s per million of vinyl chloride. A more s e n s i t i v e instrument of t h i s type I s commercially available tha t reads to a tenth of these levels. Also 8 portable gas chromatograph has recently become commercially available t h a t has w e n greater sensit ivity. Both of these instruments give immediate readings o f the. concentration present.

The current practice of venting the reactors by means of a portable blower a a c e d on the charging f loor with the vinyl chloride laden air from the reactor being discharged i n t o the general char- room atmosphere may result 3n excessive p y l chloride concentrations i n t h i s area occupied by t h e reactor operators and others. :Since the odor threshold -Ls 4000 parts per million, freedom from odor does not assme saf'e concentration. Certainly the concentration varies greatly depending on the enclosure, general ventilatian, and the number of' reactors being vented during the day. Air analysis using ade- quately sensitive methods should be conducted fn these u e a s . IY vapor can- centrations a t locations where personnel is exposed are abwe 50 parts per million f o r any appreciable duration, consideration should be given t o provi- sion of an exhaust system t o remove residual Vinyl chloride from the reactor to the outside of the building or t o a vapor recovery system.

I

BAO 0 0 174

I

1. The associat ion between reactor cleaning and the occurrence of AOL is 6ufficientI.y clear-cut that steps should be taken to minimize the exposure of workers responsible for t h i s operation. substituting water-jet or solvent cleaning for t h e present hand scraping opera- eions, ventilation should be provided to reduce t h e Vinyl chloride cancentratton below

These steps may take the form of

* Where it is necessary for workers t o enter the reactor tanks, sufficient

50 PPm*

2. Because of evidence that the disease AOL affects bony structures other than the phalanges, and because of information made available by other Snvesti- gators that there may be other cqnnecttve tissue involvement, reco-ized cases should be reinoced from -her exposure and should be examined at intervals t o determine whether there is any progresston of the disease.

3. Attempts t o produce the disease i n experimental animals should be continued to isolate the agents responsible for the disease,

k. Plants t ha t had no cases of' the disease i n the present survey should not assume that they are irmmule. Ix many instances failure to demonstrate cases may have been the result o f the small number c t f empLoyees exposed or limLted experience rather than due to a difference in practices. these factors, plants that do not: now appear to have cases o f AOL should perform regular periodfc medical surve i l l ances since this study cannot predict unequivocally that cases of AOL w i l l not Occur in the future, For the 6me reasons, it is recommended that a l l plants, whether or not they have had cases, adopt the practices outlined in item 1 above.

Because of

-

f . 104

211169 . 3 VCT5 VRD 201 2930 - 201 3046

Epidemiological lnvesitgation of the Polyvinyl Chloride Industry in Reference to Occupational Acroosteolysis

keyword: NTL expert: binder: Acro 3

attachments:

Page 101 of 198

I lllllll IllU 1111 11111 111 IllU 11111 Ilm 1111 111 - PBS I lllllll IllU 1111 11111 111 IllU 11111...

Documents

Transcript of I lllllll IllU 1111 11111 111 IllU 11111 Ilm 1111 111 - PBS I lllllll IllU 1111 11111 111 IllU 11111...