Appendix — American Textile Mfrs. Institute, Inc. v. Donovan
Supreme Court brief1981
Ask Donna
What actually matters in this document.
Text
Nos. 79-1429 and 79-1583
IN THE
Supreme Court of the United States
OCTOBER TERM, 1980
AMERICAN TEXTILE MANUFACTURERS INSTITUTE, INC.,
ET AL., Petitioners,
V.
RAY MARSHALL, SECRETARY OF LABOR,
UNITED STATES DEPARTMENT OF LABOR, ET AL.,
Respondents.
NATIONAL COTTON COUNCIL OF AMERICA,
v, Petitioner,
RAY MARSHALL, SECRETARY OF LABOR,
UNITED STATES DEPARTMENT OF LABOR,
Respondent.
On Writs of Certiorari to the United States Court
of Appeals for the District of Columbia Circuit
JOINT APPENDIX
NEIL J. KING WADE H. MCCREE, Jr.
Counsel of Record Solicitor General
A. STEPHEN Hut, Jr. KENNETH S. GELLER
ANDREW N. VOLLMER Deputy Solicitor General
WILMER & PICKERING BARRY SULLIVAN
1666 K Street, N.W. Assistant to the Solicitor
Washington, D.C. 20006 General
(202) 872-6000 Department of Justice
Washington, D.C. 20530
(202) 633-2217
(Additional Counsel listed inside)
PETITIONS FOR WRIT OF CERTIORARI FILED MARCH 14, 1980
(No. 79-1429) AND APRIL 9, 1980 (No. 79-1583)
CERTIORARI GRANTED ON OCTOBER 6, 1980
tg
ROBERT H. BorRK
142 Huntington Street
New Haven, Connecticut 06511
GREGORY B. TOBIN
OGLETREE, DEAKINS, NASH,
SMOAK, STEWART AND
EDWARDS
First National Bank Tower
Two Peachtree Street, N.W.
Atlanta, Georgia 30383
Counsel for Petitioner
American Textile
Manufacturers Institute,
Ine.
ROBERT T. THOMPSON
GARY S. KLEIN
THOMPSON, MANN & HUTSON
The Daniel Building
Suite 2222
Greenville, S.C. 29602
Counsel for Petitioner
Milliken and Company
JOSEPH K. MADDOX, JR.
P. O. Box 5784
Spartanburg, S.C. 29304
Counsel for Petitioner
Spartan Mills
ROBERT T. THOMPSON
GARY S. KLEIN
THOMPSON, MANN & HUTSON
The Daniel Building
Suite 2222
Greenville, S.C. 29602
Counsel for Petitioner
Hermitage, Inc.
SAMUEL K. ABRAMS
BRIAN E. MORAN
BAKER & HOSTETLER
818 Connecticut Ave., N.W.
Washington, D.C. 20006
H. J. ELAM, III
NEIL W. KOONCE
Cone Mills Corporation
Greensboro, N.C. 27405
Counsel for Petitioner
Cone Mills Corporation
CARIN A. CLAUSS
Solicitor of Labor
BENJAMIN W. MINTZ
Associate Solicitor
ALLEN H. FELDMAN
DENNIS K. KADE
DIANE E. BURKLEY
JOHN A. BRYSON
Attorneys
Department of Labor
Washington, D.C. 20210
Counsel for Respondent
Secretary of Labor
GEORGE H. COHEN
Counsel of Record for
Union Respondents
ROBERT M. WEINBERG
JEREMIAH A. COLLINS
BREDHOFF, GOTTESMAN, COHEN,
CHANIN, WEINBERG &
PETRAMALO
1000 Connecticut Ave., N.W.
Washington, D.C. 20036
(202) 833-9340
Counsel for American
Federation of Labor and
Congress of Industrial
Organizations, Industrial
Union Department, AFL-
CIO and Amalgamated
Clothing & Textile Workers
Union, AFL-CIO
LAURENCE GOLD
815 16th Street, N.W.
Washington, D.C. 20006
J. ALBERT WOLL
General Counsel, AFL-CIO
815 15th Street, N.W.
Washington, D.C. 20005
ELLIOT BREDHOFF
General Counsel
Industrial Union Department,
AFL-CIO
1000 Connecticut Ave., N.W.
Washington, D.C. 20036
ARTHUR M. GOLDBERG
General! Counsel
Amalgamated Clothing &
Textile Workers Union
15 Union Square
New York, New York 10003
Of Counsel
DAN M. Byrp, Jr.
J. SPRATT WHITE
P. O. Box 70
Fort Mill, S.C. 29715
Counsel for Petitioner
Springs Mills, Inc.
ROBERT H. BoRK
142 Huntington Street
New Haven, Connecticut 06511
Counsel for Petitioner
Fieldcrest Mills, Inc.
THOMAS A. EVINS
CLYDE H. HAMILTON
BUTLER, MEANS, EVANS &
BROWNE
P. O. Box 451
Spartanburg, S.C. 29304
Counsel for Petitioner
Arkwright Mills
ROBERT T. THOMPSON
GARY S. KLEIN
THOMPSON, MANN & HUTSON
The Daniel Building
Suite 2222
Greenville, S.C. 29602
Counsel for Petitioner
Blair Mills, Inc.
HARLAN H. HUNTLEY
ROGER L. TUTTLE
2291 Memorial Drive
Danville, Virginia 24541
Counsel for Petitioner
Dan River, Ince.
THOMAS A. EVINS
CLYDE H. HAMILTON
BUTLER, MEANS, EVINS &
BROWNE
P. O. Box 451
Spartanburg, S.C. 29304
Counsel for Petitioner
Mayfair Mills
FRED M. RICHARDSON
Lovic A. BROOKS, JR.
CHARLES A. EDWARDS
CONSTANGY, Brooks & SMITH
1900 Peachtree Center Building
230 Peachtree Street, N.W.
Atlanta, Georgia 30303
Counsel for Petitioner
Riegel Textile Corporation
RICHARD H. MONK, JR.
C. POWERS DORSETT
West Point-Pepperell, Inc.
P. O. Box 71
West Point, Georgia 31833
Counsel for Petitioner
West Point-Pepperell, Inc.
CHARLES M. CRUMP
Counsel of Record
APPERSON, CRUMP, DUZANE &
“MAXWELL
2610 100 North Main Building
Memphis, Tennessee 38103
(901) 525-1711
JOSEPH A. Moss
4143 27th Street, N.
Arlington, Virginia 22207
(703) 525-4063
Counsel for Petitioner
National Cotton Council
of America
+
TABLE OF CONTENTS
Page
I. Relevant Docket Entries in the Courts of Appeals.. 1
II. Exhibits Submitted in the Proceeding Before the
FR ee NEST PCL IE TE Shey 7
1. National Institute for Occupational Safety
and Health, “Criteria for a Recommended
Standard: Occupational Exposure to Cotton
Dest” (ix. 1, emcerphed) 9
2. Roach & Schilling, “A Clinical and Environ-
mental Study of Byssinosis in the Lanca-
shire Cotton Industry” (Ex. 6-1, excerpted) ._.. 14
3. Bouhuys, “Byssinosis in the United States”
CS Ren RTL OE ME Pah a eS Teh 15
4. Braun, et al., “Prevalence of Respiratory
Signs and Symptoms Among U.S. Cotton
Textile Workers” (Ex. 6-19, excerpted) 23
5. Bouhuys, et al., “Byssinosis in Cotton Textile
Workers” (Ex. 6-24, excerpted) 24
6. Bouhuys, “Breathing, Physiology, Environ-
ment and Lung Disease” (Ex. 6-27, ex-
GE ee ee ee 25
7. Merchant, et al., “An Industrial Study of
the Biological Effects of Cotton Dust and
Cigarette Smoke Exposure” (Ex. 6-44, ex-
I Seahaatha pascal 8 aa es 27
8. Merchant, “Dose Response Studies in Cotton
Textile Workers” (Ex. 6-51) 29
9. Molyneux & Berry, “The Correlation of Cot-
ton Dust Exposure With the Prevalence of
Respiratory Symptoms” (Ex. 6-55, ex-
MRE USE RCE EL Craton MUM ERBN I” BN Seat > me 56
10. Imbus, “Experience with Medical Surveillance
Programs” (Ex. 6-57, excerpted) 58
i.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
ii
TABLE OF CONTENTS—Continued
Molyneux & Tomblesen, “An Epidemiological
Study of Respiratory Symptoms in Lancashire
Mills” (Ex. 6-66, excerpted) .............0........------
Schilling, et al., “Cardiovascular Disease in
Cotton Workers—Part I” (Ex. 6-73, ex-
IE roiglie ends cate butcstlccsccaneinica cece secdeabeesnspnedsaeiadaaSenck
Research Triangle Institute, Technological
Feasibility Assessment and Final Inflationary
Impact Statement (Ex. 6-76, excerpted) .......
Statement of Dr. Arend Bouhuys (Ex. 11,
I er eit ee hs ea es
ETRE aR, NPE TRO S ONT TK OTE EE, CPOE HERI
Weill, “Report to the American Textile Manu-
facturers’ Institute on a Study Entitled
Assessment of Respiratory Responses in Tex-
tile Workers Exposed to Low Levels of Cotton
Dust” (ix. 154, excerpted) ..................-.......
Statement of Bruce Held (Ex. 15, excerpted)..
Research Triangle Institute, “Technological
Feasibility and Economic Impact of Regula-
tions for Cotton Dust” (Ex. 16, excerpted) ....
Merchant, Memorandum to Grover C. Wrenn:
“Review of Draft Proposal for Cotton Dust
Standard” (Ex. 26, excerpted) ....00000.000..00000....
Statement of Edward J. Baier (Ex. 38a, ex-
ERTS PCR aa LIne Sere aPC
Merchant, “Epidemiological Studies of Respi-
ratory Disease Among Cotton Textile Work-
ers 1970-1973” (Ex. 38D, excerpted) .............
Morgan, et al., “Report on Survey on Preval-
ence of Byssinosis and Respiratory Symptoms
in Three Textile Mills” (Ex. 39, excerpted) ....
Page
61
151
156
157
164
167
174
175
176
179
23.
24,
25.
26.
27.
28.
29.
30.
31.
32.
33.
34,
35.
36.
iii
TABLE OF CONTENTS—Continued
Page
Supplemental Submission to Statement and
Testimony of W. K. C. Morgan, M.D. (Ex.
i IED Soctiesesisccs ceri ar Oe 183
Statement of W. K. C. Morgan, M.D. (Ex 40,
vse, sete he a RC Me Pe Re 185
Statement of Russell A. Harley, M.D. (Ex.
a er an ti EOP NRO 190
Annual Cotton Dust Survey (Ex. 48) 198
Supplemental Submission to Statement and
Testimony of Harold R. Imbus, M.D., Sc.D.
(me. 47d, Qxcerpéed) 200
Statement of M. C. Battigelli, M.D. (Ex. 48,
II ate e 203
“Burlington Industries Took Our Breath
Away” (Ex. 54, excerpted)... 208
Statement of Carolina Brown Lung Assorcia-
tion (Ex. 54a, excerpted)... 217
Statement of Moon W. Suh, Ph.D. (Ex. 55)... 221
Statement of Hovan Hocutt eee 234
Statement of L. K. Fitzgerald (Ex. 69, ex-
IE ee ie i ai aie ti xan. 252
Submission of Amalgamated Clothing and
Textile Workers Union: Vol. 1 (Ex. 78,
IN idekinsiicd etisalat. 266
Statement of Parker C. Reist and Lawrence
D. Kornreich (Ex. 79a, excerpted) ..:.............. 293
Reist, et al., “The Impact of New Production
Equipment on Control Costs to Meet the Pro-
posed Cotton Dust Standard” (Ex. 79c, ex-
sen as ise, Re TO RT CRIED 296
37.
38.
39.
40.
41.
42.
43.
44,
iv
TABLE OF CONTENTS—Continued
Page
Statement of William A. Burgess (Ex. 80a,
CID aces inscicsviesineccwicendstsinsnantacinabcene eas 309
Statement of Eric Frumin (Ex. 82, ex-
IID usscttsenicsanisicctscoi-rveieshecmanmeenaaigineiiioamase esas 316
Statement of U.S. Department of Agriculture
(ie, SGC, Ceeeteees: ..nc ee 325
Statement of John S. Barr, III (Ex. 101, ex-
1} SRNR Ne OPENER ON rt Me UREN Ee Tn NR 327
Comments of the Council on Wage and Price
Stability (Ex. 111, excerpted) .................0..... 331
Statement of Dr. Sidney Wolfe & Peter Greene
(ix. 228, GCE? Lo MPS 343
Bouhuys, et al., “Epidemiology of Chronic
Lung Disease in a Cotton Mill Community”
CHER BG, GOES bck xchat ae 344
Supplemental Submission of Amalgamated
Clothing and Textile Workers Union (Ex. 143,
CIID Siciisccetnccndncdacunsdinataneecsascieaaaea is 348
45. Post-Hearing Comment of the American Tex-
tile Manufacturers Institute, Inc. (Ex. 160,
SU iaeccinisnctisesaicsghcesccinasab biaieaatiin 372
46. Letter from F. S. Love to Eula Bingham (Ex.
EV R\ SIONS | nvisnncccinncicciateces 388
47. Statement of Arthur Thomas (Ex. 62, Attach-
a | a ae NC Be: SOMME Ce AC ae 486
III. Transcript of Hearing Before the Agency .............. 399
1. Excerpted testimony of Dr. Bouhuys .............. 401
2. Excerpted testimony of Dr. Kilburn .............. 410
3. Excerpted testimony of Dr. Weill .........000000000.. 410
4. Excerpted testimony of Mr. Held -.................. 412
5. Excerpted testimony of Dr. LeSourd ............ 414, 418
6. Excerpted testimony of Dr. Lee ~...0000.00000000.... 416
7. Excerpted testimony of Dr. El Batawi .......... 421
IV.
v
TABLE OF CONTENTS—Continued
Page
8. Excerpted testimony of Dr. Merchant....423, 435, 485
9. Excerpted testimony of Dr. Taylor... 434
10. Excerpted testimony of Dr. Baier... 442
11. Excerpted testimony of Dr. Morgan _............ 444
12. Excerpted testimony of Dr. Harley ........... 444
18. Excerpted testimony of Dr. Martin... 445
14. Excerpted testimony of Dr. Imbus ................ 445
15. Excerpted testimony of Dr. Neefus ......__. 449
16. Excerpted testimony of Ms. Norton ......... 451
17. Excerpted testimony of Ms. McCoy _.............. 452
18. Excerpted testimony of Mr. Harrell... 453
19. Excerpted testimony of Mr. Sandlin ......._. 453
20. Excerpted testimony of Mr. Hassell... 454
21. Excerpted testimony of Mr. Baldwin ......___ 455
22. Excerpted testimony of Mrs. Sanders... 456
23. Excerpted testimony of Mr. Graves _............. 457
24. Excerpted testimony of Mr. Hocutt... 458
25. Excerpted testimony of Mr. Figh _............... 473
26. Excerpted testimony of Mr. Chapman ............ 476
27. Excerpted testimony of Dr. Reist 477
28. Excerpted testimony of Prof. Burgess ............ 481
29. Excepted testimony of Mr. Davis _............... 482
30. Excerpted testimony of Mr. Mabe _................. 483
31. Excerpted testimony of Mr. Frumin .....___. 484
Order of Supreme Court granting Petitions for
Writs of Certiorari in Nos. 79-1429 and 79-1583... 487
I. Relevant Docket Entries
in
American Federation of Labor & Congress
of Industrial Organizations, et al.,
Vv.
Ray Marshall, et al.
No. 78-1562 (D.C. Cir.)
(and Consolidated Case Nos. 78-1736,
78-1979, 78-1980, 78-1981, 78-1982,
78-1983, 78-1984, 78-1986, 78-1987,
78-1988, 78-1989, 78-1990, 78-1991,
78-1992, 78-1993, 78-2013, 78-2014,
78-2016 and 78-2018)
Date
Filings, Orders, and Proceedings
1978
June 19
June 19
July 27
August 2
August 3
August 7
August 7
August 10
August 14
August 15
August 16
Filing of Petition for Review of Cotton Dust
Standard by American Federation of Labor
& Council of Industrial Organizations, et al.
(D.C. Cir.)
Filing of Petition for Review of Cotton Dust
Standard by American Textile Manufacturers
Institute, Inc. (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Milliken and Company (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Arkwright Mills (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Spartan Mills (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Blair Mills (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Hermitage, Inc. (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by West Point-Pepperell, Inc. (5th
Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Dan River, Inc. (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Cone Mills Corporation (4th
Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Nationai Cotton Council of
America (6th Cir.)
_ PREVIOUS PAGE WAS BLANK |
4
Date
Filings, Orders, and Proceedings
1978
August 16
August 16
August 17
August 18
October 3
October 6
October 7
October 13
December 15
December 15
December 15
December 15
Filing of Petition for Review of Cotton Dust
Standard by Springs Mills (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Mayfair Mills (4th Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Riegel Textile Corporation (4th
Cir.)
Filing of Petition for Review of Cotton Dust
Standard by Fieldcrest Mills, Inc. (4th Cir.)
Order transferring Cotton Dust Petitions for
Review pending in the Fourth Circuit to the
District of Columbia Circuit
Order transferring National Cotton Council
of America Petition for Review from the
Sixth Circuit to the District of Columbia
Circuit
Order transferring West Point-Pepperell, Inc.
Petition for Review from the Fifth Circuit to
the District of Columbia Circuit
Order consolidating Cotton Dust cases in the
District of Columbia Circuit
Filing of Opening Brief of Petitioners Ameri-
can Federation of Labor & Congress of In-
dustrial Organizations, et al.
Filing of Joint Brief of the American Textile
Manufacturers Institute, Inc., et al.
Filing of Supplemental Brief of Petitioner,
National Cotton Council of America
Filing of Amicus Curiae Brief of American
Farm Bureau Federation
Date
Filings, Orders, and Proceedings
1979
January 26
February 7
February 7
February 7
February 14
October 24
November 7
November 7
1980
January 11
January 11
January 16
Filing of Brief for the Secretary of Labor
Filing of Reply Brief of Petitioner National
Cotton Council of America
Filing of Reply Brief and Brief in Interven-
tion of Petitioners American Federation of
Labor & Congress of Industrial Organiza-
tions, et al.
Filing of Joint Reply Brief of the American
Textile Manufacturers Institute, Inc., et al.
Oral Argument in the United States Court of
Appeals for the District of Columbia Circuit
Entry of Judgment and Opinion of the United
States Court of Appeals for the District of
Columbia Circuit (Reprinted in the Appendix
to the Petition for Writ of Certiorari at 19)
Filing of Petition for Rehearing and Sug-
gestion for Rehearing En Banc of Petitioner,
National Cotton Council of America
Filing of Joint Petition for Rehearing and
Suggestion for Rehearing En Banc of Ameri-
can Textile Manufacturers Institute, Inc., et
al.
Order denying Petitions for Rehearing (Re-
printed in the Appendix to the Petition for
Writ for Certiorari at 103)
Order denying Suggestions for Rehearing En
Banc (Reprinted in the Appendix to the Peti-
tion for Writ of Certiorari at 104)
Motion of National Cotton Council of Ameri-
ca for Stay of Mandate
Date
Filings, Orders, and Proceedings
1980
January 16
January 17
January 17
January 22
March 14
April 9
October 6
Filing of Joint Motion for Stay of Mandate
by the American Textile Manufacturers Insti-
tute, Inc., et al.
Filing of Opposition to Motions for Stay of
Mandate by the Secretary of Labor
Filing of Response of Union Petitioners to
Motions for Stay of Mandate
Order directing clerk to issue mandate (Re-
printed in the Appendix to the Petition for
Writ of Certiorari at 106)
Filing of Petition for a Writ of Certiorari by
Petitioners American Textile Manufacturers
Institute, Inc., et al.
Filing of Petition for a Writ of Certiorari by
Petitioner National Cotton Council of America
Entry of Order granting Petitions for Writs
of Certiorari (Reprinted in this Joint Ap-
pendix at 489)
II. EXHIBITS SUBMITTED IN THE PROCEEDINGS
BEFORE THE AGENCY
9
EXHIBIT 1
National Institute for Occupational Safety and Health,
“Criteria for a Recommended Standard: Occupational
Exposure to Cotton Dust”
Feasibility of Control
Control of dust levels in cotton-processing operations
can be achieved by changing or treating the raw mate-
rial which is the source of the dust, by changing the
process which produces the dust, or by removing the
dust from the air once it is generated.
The transition from natural to synthetic fibers in the
past decade has resulted in lower byssinosis-producing
dust levels in those mills using synthetics or blends. Mer-
chant et al [18] reported a median dust level of 0.485
mg/cu m for 493 samples collected in mills working with
pure cotton and a median dust level of 0.163 mg/cu m
for 237 samples collected from mills using blends of nat-
ural and synthetic fibers. The synthetic fibers are vir-
tually trash free and thus contribute little to the total
sample collected. [136] It would be expected that, as the
synthetic content of the raw material increased, the dust
levels in most work areas would decrease.
A second method of changing the raw material is
through the improvement of growing techniques to reduce
the trash content of the cotton. Approaches such as de-
veloping cotton varieties which shed their bracts prior to
maturation and harvest or the development of dwarf
determinant cottons with increased fruiting potential
compared to the production of vegetative parts [137]
offer future potential methods of dust control, but at
present these approaches are not feasible. There is little
doubt that exposure to cotton dust trash has been greatly
augmented by the replacement of handpicking by ma-
chines. [138] Studies are in progress to find new chem-
_ PREVIOUS PAGE WAS BLANK |
‘ 5
10
icals which will more efficiently defoliate cotton and re-
duce the trash content of the harvested seed cotton. [137]
A third method of changing the raw material is by
steaming. Studies of this technique indicated that steam-
ing could reduce the toxic effect of the cotton dust with-
out rendering the cotton unsuitable for processing. [94]
Some investigators conclude that the byssinosis-producing
dusts are not removed or detoxified, but are just made to
adhere more firmly to the cotton fibers; therefore the
byssinosis problem is not solved but only moved from the
opening, picking, and carding areas to the winding and
weaving operations. [139] However, recent studies in a
cotton spinning plant do not show an increase in down-
stream dust levels when processing steamed cotton. [140]
Thus, while steaming may not be at this time a feasible
alternative to dust control, it may become effective as a
supplementary control method after further development.
[140] Work is also in progress on improved ginning
methods to allow for more efficient trash separation
and/or fractionation. [137]
There appears to be little that can be done now to
change the process in which cotton fibers are formed
into yarn and woven into cloth in order to control dust
production. The cotton manufacturing process is essen-
tially one of fiber cleaning and alignment with the de
sired goal of removing all material in the cotton except
the mature fibers. The dust in the cotton is thus an
unwanted byproduct which must be dealt with. It is
possible that the ginning process could be changed, and
work is being actively pursued along these lines [137];
but at present process change does not appear to be a
feasible alternative to dust control.
There are also a number of environmental factors
which can be adjusted to some extent to control dust
levels. As the opening and cotton cleaning machinery
more efficiently removes trash and dust from the cotton
11
stock, the release of this material into the work areas
diminishes. Thus, dust levels in the picking room, card-
room, and subsequent operations can be reduced by better
cleaning of the cotton in the opening and cleaning line.
[141]
Decreasing the production machine density in a work
area should decrease the dust levels in that area also.
Conversely, if the machines are crowded together, higher
dust levels would be expected. Dust emission is also
increased by higher card speeds. Improperly or poorly
maintained production machinery can also lead to higher
dust levels. Hocutt [141] reports that cardroom dust
levels are lower when the cards are well maintained and
properly operated with alert operating personnel using
precise machine settings.
Large central air-conditioning systems will tend to
even out dust concentrations over the entire mill. [141]
Of course, this might create new problems instead of
getting rid of one, since it would cause increased dust
level in some area while decreasing those in others. How-
ever, air-conditioning a textile mill does provide some
dust control, but the control is incidental and the in-
creased dust load in the air-conditioning unit will result
in increased maintenance costs and impede the perform-
ance of the air conditioner. Dust is removed by the
washer, a unit which acts as the primary humidifier and
heat exchanger in the air-conditioning system. The
washer is about 25% efficient for large particles [141]
and much less efficient for smaller ones. Reliance on the
air conditioner for dust control means that the dust must
travel through the work area to reach the air-conditioner
inlet, usually located on one wall of the room. Thus, the
very nature of the system makes it inefficient for dust
reduction or removal.
The capture and removal of dust from the air after it
has been generated from the cotton represents the most
12
widespread and efficient method of dust control at the
present time. Table IV-3 gives summarized results from
a study conducted recently by Barr et al [127] on the
effectiveness of dust controls in mills having different
degrees of dust control and processing varying grades of
cotton. These results serve only to indicate the ranges of
concentrations which may be expected for the different
operations, with and without dust controls.
It appears that in some instances there may be dust
concentrations (<0.5 mg/cu m) near cotton processing
operations without dust control devices, but this would be
the exception rather than the rule.
TABLE IV-3
RANGE OF TYPICAL LINT AND DUST CONCENTRATIONS
Total Dust excluding lint
Operation (mg/cu m) (mg/cu m)*
Picking, no control -— 0.6-1.6
Picking, control 0.4-0.7 0.3-0.4
Opening and picking, no control 1.5-9.1 0.2-1.9
Opening and picking, control —_ 0.3-0.5
Carding, no control §.2-21.2 0.3-5.4
Carding, control 0.5-8.4 0.1-4.2
* As measured by the vertical elutriator from reference 127.
In a study of four cards, Wood and Roach [11] found
dust removal to be less effective than indicated in Table
IV-3; total dust levels ranged from 4.2-5.8 mg/cu m
without dust extraction to 1.3-4.8 mg/cu m with dust
extraction.
Efficient dust removal in cotton processing areas de-
pends on two factors. First, there must be effective dust-
capturing devices located at the points where dust could
be generated. The captured dust is then transported
away from the point of generation to a point where it
13
can either be discharged to the outside atmosphere or re-
moved by some means from the carrier air stream. Direct
discharge of the captured dust to the ouside atmosphere
is not practical for two reasons. The dust is emitted in
sufficient quantity to quickly produce its own air pollu-
tion problem outside, and some will probably find its way
back into the work area negating the effect of the dust
capture mechanisms. A second and often more compelling
reason for not directly discharging into the atmosphere
is the loss of conditioned air which must. be replaced by
new, heated or cooled makeup air. The quantities of air
required for effective dust capture are large enough so
that the usual practice is to recycle this air volume within
the mill. [127]
Air which receives dust at one point and is cleaned at
another point during complete recirculation will with
time reach an equilibrium dust concentration. With ideal
mixing, this concentration will equal the cleaned air dust
concentration plus the ratio of the dust production rate
divided by the recirculated air flow rate. [127] With
good dust capture (low ¢ .st production rate) or a large
recirculating air flow rate the concentration of dust in
the air which is returned to the workroom will eventually
determine the concentration of dust in that space. The
isolation of dusty operations, successful in controlling
dust exposures in other industries, has not been widely
adopted in cotton mills.
& * * * on
14
EXHIBIT 6-1
Roach & Schilling, “A Clinical and Environmental Study of
Byssinosis in the Lancashire Cotton Industry.”
Our suggested grading of workrooms according to dus-
tiness was arrived at after consideration of the preva-
lence in groups exposed to similar concentrations for sim-
ilar lengths of time. We might have deduced permissible
levels of dustiness from dose response curves in a similar
way to that suggested by Roach (1953). However, we
were hesitant to do this because of the possible effects of
selection in those groups with the highest prevalence of
byssinosis. People with byssinosis are generally aware
that the dust affects their breathing and it is likely that
they will leave their jobs more readily than those who are
not so affected. To this extent the people working in a
cotton mill are likely to be a selected population, particu-
larly those groups who work in the dustiest jobs and are
in consequence affected most by dust. To deduce permis-
sible levels of dustiness from dose-response relationships
we would have had to assume that selection operates in
an exactly similar way in all mills or does not operate
at all.
* * + ” *
15
EXHIBIT 6-16
BYSSINOSIS IN THE UNITED STATES
Dr. Arend Bouhuys
J.B. Pierce Foundation Laboratory
Yale University School of Medicine
New Haven, Conn.
I would like to discuss some of the main points con-
cerning the definition of byssinosis. At risk are cotton,
flax, and hemp workers in carding and other types of
dusty work. The symptoms of the disease are chest tight-
ness, cough, and wheezing. The peculiar property of
these symptoms is the timing of their occurrence. In the
first few years of exposure, symptoms occur on Mon-
day, or other days after absence from the work en-
vironment; later, symptoms occur on other days of the
week; and eventually, symptoms are continuous, even in
the absence of dust exposure. This description constitutes
a definition of byssinosis, which many of my colleagues
and I adhere to. That is, byssinosis is a disease which
starts with short lasting acute responses to dust ex-
posure and may progress to chronic lung disease with
more or less severe disability. There has been disagree
ment on the use of the term byssinosis. Some would want
to limit the diagnosis to persons who have evidence of
chronic lung impairment only. This, however, would de-
tract from the urgent need to prevent the development
of chronic lung disease, and we must focus an important
part of our effort on those people who respond acutely
to dust but who do not yet have irreversible lung
damage.
This condition has been observed all over the world.
In Belgium it was studied in some 2,000 cotton workers
and was described as Monday chest tightness in the year
1845. I might add that already in 1837 a young New
York physician, McCready, wrote a book on occupational
16
disease, and in it he made the remark that the New
England cotton mills should do something about the dust
in the workrooms. However, he did not elaborate on the
reasons why he thought that this dust should be removed
from the air.
The disease has been studied on all continents except
Latin America. So far it ranges from Dr. Gandevia in
Australia (who is here today), Dr. Shogo Shima in Japan,
and others in India, Taiwan, and a number of African
countries, several European countries, and to the United
States and Mexico. In nearly all of these studies, a
smaller or larger number of people have been diagnosed
as having byssinosis. Many physicians have been in-
volved in these studies and their results have been re-
markably similar. In most of these countries the disease
was identified in the industry by epidemiologists. It is
a recurring pattern, whenever this disease occurs, that
physicians in hospitals and clinics underestimate the
prevalence of this disease.
The prevalence of byssinosis in the United States,
studied in three mills, has been found to be rather similar
in that some 20% to 30% of the people in the carding
room are affected. The percentages of people in the
spinning rooms vary more. There are technical reasons
for this, but I won’t go into these at this time. Two
main points I want to make about these findings. In
the first place I want to draw your attention to a mill
in the Atlanta Federal Penitentiary, where some 100 men
were exposed to dust for less than one year. Percentage
prevalence of Monday symptoms of byssinosis was similar
in that category and in the total numbers of workers.
The total number of these workers, the duration of ex-
posure to the dust, the number of years that people had
worked in this environment was much less than in the
two commercial mills, where people had on the average
worked for 12 to 18 years. In all these areas the total
17
dust concentration, in milligrams per cubic meter, is not
excessively high compared to what has been found in
earlier studies. This is very likely related, at least in
part, to what Richard Schilling has already shown you
about the effect of lint removal from the air. In all
these mills there were some systems to abate dust and
it is likely that these machines did a rather good job in
cleaning out the visible coarse dust that does not get
into your lungs, but left in the air the invisible small
dust that does get into your lungs. I should remind you
that about the best air filters for small dust particles
are our own lungs; they have so many very fine tubes
where the circumstances for deposition of particles, less
than 5 u in diameter, are really very excellent and these
are much better filters than any of the mechanical de-
vices that are currently used in most textile mills.
How can we detect this disease apart from asking the
workers for their symptoms? One of the best ways of
doing this is by having the worker blow hard into a
spirometer, a measuring device for lung volume changes,
before he starts work on a Monday and again before he
goes home at the end of the shift. At the end of the
Monday the value measured by this device [the forced
expiratory volume in 1 second (FEV 1)] is lower than
in the morning. The results of a group of cotton workers
in Sweden provide data similar to that obtained in many
cotton mills including those in this country. At the end
of the day on Monday these people, on the average, have
significantly lower values of their capacity to breathe out
forcefully, and this decrease is larger in those who com-
plain of Monday symptoms than in those who do not
have such a history. This change means that somewhere
along the airways there is some narrowing of the fine
tubes that conduct the air to our lungs. The people who
do not. complain have a similar response as those who
do. In other occupations, including coal mining and
other dusty jobs, such a response does not occur. The
18
people in coal mining and other dusty jobs do not de-
crease their FEV 1 during exposure and likewise people
in other occupations do not show such a response.
The decrease of FEV 1 in textile workers is caused by
the dust they inhale. Those with the more severe drop
of FEV 1 will complain of chest tightness; those with
less severe responses may not complain.
Another way of looking at the same phenomenon is
provided by the flow-volume curve. These are curves
which I will interpret very simply by saying that the
horizontal axis shows how much you can breathe out.
The length of the horizontal portion of the curve is
proportional to the volume one breathes out. The dis-
tance along the ordinate is proportional to how fast you
can breathe it out. In other words, you have a sort of
performance diagram of the lungs during expiration.
Without going into the physiology of it, I would just
like to state that the rapidity of expiration in a con-
siderable proportion of this curve is quite unrelated to
how hard you try. You have to deliver a minimum of
effort but beyond that you will not improve the flow
rates. These flow rates decrease markedly by the end
of the work shift. An example is given, based on results
in six representative workers. These flow rate decreases
are caused by the narrowing of small airways in the
lungs.
Let me describe two older textile workers. These men
have been exposed to hemp dust for about 25 years or
more and it illustrates what can happen. Two things
can happen to someone who stays in this dust long
enough. On the one hand, we have a man of 58 years
who has a normal pulmonary function. His vital ca-
pacity is normal for his age and height. His flow rates
are also normal for his age and height. This person
has never experienced Monday symptoms of byssinosis
while he worked in the dust. Unfortunately, such people
19
are in the minority. The other man is 59 years of age.
His pulmonary function is severely decreased. His vital
capacity is far too small, but much more important, his
capacity to breathe out rapidly is severely disturbed.
As a result this person is severely disabled. This man
has experienced Monday Symptoms of byssinosis al]
through his working life. Unfortunately, such people
form a much larger proportion of an older population of
these workers than do the non-reactors of which the first
man is an example. There are many disabled cotton
workers in the southern part of the United States. In
1966 Dr. Schilling and I spent altogether one week in
the South and, with the aid of two physicians, we saw
14 cotton workers. Half of these were severely disabled
by respiratory deficiency. I’m convinced that a more
thorough and more systematic study would show this
type of disability is prevalent wherever appreciable num-
bers of people have been exposed to cotton dust for
many years. We have not yet been able to do such a
thorough epidemiological study in the United States, but
we have done this in Spain in a town where a large
proportion of the population used to work in hemp dust.
Hemp workers and controls do not differ in the younger
age category. The difference is highly significant in the
older category where the hemp workers have a severely
decreased FEV 1, which is quite significantly smaller
than that in the control subjects who were living in the
Same area but who were not exposed to the dust. I might
say that these control subjects are quite comparable in
this respect to the data which has been obtained by sev-
eral people in other populations, like Dr. Ferris in British
Columbia. This type of data is pretty much the same
wherever you do it unless there is some special factor at
work, like in this case the hemp dust.
It has been asked, of course, what is the agent in the
dust that causes this acute response which may lead in
20
a certain portion of cases to severe chronic lung disease.
The bracts are the crumbling leaves around the stem
that are dead long before the plant is harvested. When
we make a very simple watery extract of these bracts,
we can demonstrate certain effects on human lung tissue.
In the laboratory we can also have this extract inhaled
by healthy people. Two healthy investigators inhaled
this extract of bracts for ten minutes. In this case we
had boiled the bracts to determine whether or not the
agent was heat stable. Both the subjects experienced
chest tightness after this exposure. Both of them have
a decrease of maximum flow rates—very similar to what
one finds in the cotton and hemp workers exposed to the
industrial dust. We believe that there is a pharmacologi-
cal agent which is in the dust because the dust contains
bract particles, broken up. A substance that is rather
easily water-soluble, so that when dust hits your air-
ways, this substance can transfer to the tissue and exert
its toxic action. Whether this agent is the sole cause of
the long term changes remains to be seen. It is also quite
possible that apart from the one action of this agent,
which we have identified, there are others which con-
tribute to an action. So far, however, we have been unable
to get similar data with any other part of the cotton
plant we’ve tried, so for practical purposes I think that
these leaves are very important. If they were not in the
cotton dust we would very likely have much less of a
problem. Control of this disease is obviously necessary
and I’m sure that this is going to be mentioned by several
speakers in this program and I don’t want to preempt
what they are going to say. I would like to emphasize,
I think there are many good measures that can be taken
on relatively short term at least to prevent poeple who
are now at work in the industry from becoming dis-
abled and those who may have some degree of irrever-
sible decrease of pulmonary function from getting worse.
I think this can be achieved by education, by pre-employ-
21
ment medical examination, and by periodical medical
examinations. There are a number of common sense
measures that can be taken to protect people who are
in the industry. I’m sure that some people who are
affected might be transferred to other jobs. This is not
practical forall of them, obviously, and it is not prac-
tical for all of those who respond to dust. Also, there
are drugs that can prevent the acute action of this dust,
which might be used in suitable cases. On the long
term, there is no substitute for suppression of the fine
dust; in the air. This is a very urgent matter, as Richard
Schilling has also pointed out. Technology will have to
be developed for getting rid of the fine dust which, so
far as I am aware, has not been done in the textile in-
dustry on any large scale.
I think that a large amount of work has to be done,
not only in this particular industrial lung disease, but
in general in industrial lung diseases. My personal
feeling and that of several of my colleagues is that as
a cause of environmentally induced lung disease, indus-
trial emissions are far more important than general air
pollution as a cause of disability and death. They can
be completely prevented and we don’t have to wait for
much further research in order to prevent the diseases.
There is a lot that remains to be learned ehout. these
diseases, but there is also a serious lack of application
of the knowledge that we have and we don’t have to wait
until all the answers are in before we can prevent people
from becoming disabled by industrial lung disease.
The data on which this presentation is based has been
published in the following papers:
Bouhuys, A., L. J. Heaphy, Jr., R. S. F. Schilling,
and J. W. Wellborn. 1967. Byssinosis in the United
States. New Eng. J. Med. 277:170-175.
ne ee
22
Bouhuys, A., R. L. Wolfson, D. W. Horner, J. D.
Brain, and E. Zuskin. 1969. Byssinosis in cotton
textile workers. Respiratory survey of a mill with
rapid labor turnover. Annals Int. Med. 71 :257-269.
Zuskin, E., R. L. Wolfson, G. Harpel, J. W. Well-
born, and A. Bouhuys. 1969. Byssinosis in carding
and spinning workers: Prevalence in the cotton
textile industry. Arch. Environ. Health, 19 :666-
673
The prevention and control of byssinosis and of indus-
trial lung diseases have been discussed in:
Bouhuys, A. and J. M. Peters. 1970. Control of en-
vironmental lung disease. New Eng. J. Med. 283:
573-582
Bouhuys, A., J. C. Gilson, and R. S. F. Schilling.
1970. Byssinosis in the textile industry; research,
prevention and control. Arch. Environ. Health 21:
475-478.
23
EXHIBIT 6-19
Braun, et al., “Prevalence of Respiratory Signs and
Symptoms Among U.S. Cotton Textile Workers”
* * « * &
The workers who complain of tightness do not neces-
sarily have a loss in FEV,.., or vice versa, as stated in
the literature.
* * = e *
‘ad
24
EXHIBIT 6-24
Bouhuys, et al., “Byssinosis in Cotton Textile Workers”
* - * * .
No general agreement has as yet been reached on
these questions. Our tentative answer to the first ques-
tion is affirmative: If lung function tests show that air-
way obstruction develops in a worker during textile dust
exposure, we view this as objective evidence that the dust
affects his airways. For several reasons such persons
ray not always mention subjective symptoms of res-
miratory distress during a questionnaire interview. When
interviewed on Monday after work, symptoms may be
mentioned by men who denied them during a previous
interview conducted at a time that they did not experi-
ence the dust effect (22). Secondly, some textile workers
markedly dissimulate symptoms (28), possibly because
of fear of losing their job. Thirdly, as in other diseases,
the threshold of awareness of distress may vary indi-
vidually. We believe that byssinosis should be diagnosed
in all workers in whom evidence of an effect of textile
dust on the lungs is obtained.
* * a * *
25
EXHIBIT 6-27
Bouhuys, “Breathing, Physiology, Environment and
Lung Disease”
* * * * *
As a working hypothesis, one may assume that the re-
peated microinsults to the lungs that occur each time a
worker is exposed to toxic dust, especially on Monday,
have a cumulative damaging effect. The damage mechan-
ism is not known. If the histamine release from mast cells
during acute exposure involves damage to cell walls, this
process might represent such a microinsult. Chronic lung
damage might result from these and possibly other micro-
insults on the cellular elements, if regularly repeated dur-
ing many years.
* * * * *
The practical significance of the functional grades (see
Table 17-3) is as follows:
F 0: no demonstrable acute effect of the dust on
ventilatory capacity; no evidence of chronic ven-
tilatory impairment.
F %: slight acute effect of dust on ventilatory ca-
pacity; no evidence of chronic ventilatory impair-
ment.
F 1: definite acute effect of dust on ventilatory ca-
pacity; no evidence of chronic ventilatory impair-
ment.
F 2: evidence of slight to moderate irreversible im-
pairment of ventilatory capacity.
F 3: evidence of moderate to severe irreversible im-
pairment of ventilatory capacity.
Increasing grades indicate increasing severity of func-
tional impairment. F % and F 1 indicate acute dust
26
effects. Workers in grade F 1 should, if at all possible,
be removed to jobs in nonrisk areas.
* = £ x +
Table 17-3
Recommended Functional Grades in Byssinosis
A FEV, + FEV,,,
Grade * (liters) (% of predicted )
FO —0.05-0; or + >80
F% — 0.06- —0.20 >80
Fl >- —0.20 >80
F2 — 60-79
F3 — <60
* If the grades based on (A FEV,,, and FEV,,, differ, assign the
highest of the two grades.
} Difference between FEV,,, before and after work shift on a
first working day of the week.
{ FEV,,, in the absence of dust exposure (2 days or longer) ; use
value postisoproterenol whenever this drug can be used.
27
EXHIBIT 6-44
Merchant, et al., “An Industrial Study of the Biological
Effects of Cotton Dust and Cigarette Smoke Exposure”
Until recently, little awareness had been expressed in
the U.S. medical literature about respiratory disease ac-
quired from working in the cotton textile industry.
* * * * *
It is the purpose of this paper to describe and quanti-
tate individual and combined effects of cotton dust and
cigarette exposure as associated with byssinosis preva-
lence and four parameters of ventilatory capacity. Be-
cause of the constraint of time, only white males em-
ployed in preparation and yarn processing areas will be
considered for statistical analyses.
Methods
Selection of the Population: Initially, a total of 22
North Carolina textile manufacturing plants were visited
to determine the grade and count of yarn manufactured,
plant layout, ventilation systems, machine exhaust sys-
tems and distribution of workers by sex, race, age and
work area. Twelve plants processed primarily cotton, five
cotton-synthetic blends and five either synthetic or wool
material. Based on information gained through these
visits and from a previous survey* which provided esti-
mates on the distribution of smoking habits, three strata
of exposure (cotton — high risk, cotton-synthetic blend
= moderate risk, synthetic-wool — low risk) were se-
lected for a three factor factorial design to fill adequately
four categories of age (15-29, 30-39, 40-49, 50-70), sex
and three categories of smoking habit (never smoked,
current smoker, former smoker). Initially two plants at
extremes of exposure were picked for the first two sur-
veys. From these surveys better estimates of demographic
characteristics and disease prevalence were sought to help
select other plants to fill categories of the factorial de
NE EF ee
28
sign. Other selection criteria were age of the plant,
whether it was associated with a mill village and prox-
imity to other mills under consideration. At least thirty
miles were allowed between plants.
Three cotton mills, all processing strict low middling
cottons, were selected. Mill A processed cotton with a
mean count of 6. Mill B with a mean count of 31, and
Mill C with a mean count of 33 plus 25% synthetic ma-
terial. Two cotton-synthetic blend mills processing pri-
marily strict middling cottons were selected. Mill D proc-
essed 50% synthetic material and 50% middling cotton
with a mean count of 29, and Mill E processed 49% syn-
thetic and both strict low middling and middling cottons
with a mean count of 32. The grade, count and mix of
materials had been relatively constant in the few years
preceding the study in all five cotton mills. Two synthetic
processing mills, Mill G and Mill H, processed only syn-
thetic material and one mill, Mill F, processed 100%
wool. All had previously been cotton mills but none had
processed cotton for more than 20 years. All mills selected
had once been associated with a mill village.
Within each mill up to four primary work areas were
defined. The preparation area included opening, blending,
picking, carding, drawing, combing and roving. The yarn
processing area included spinning, winding, twisting,
spooling and warping. Slashing and weaving operations
were combined to form a third primary work area. Other
employees studied were those employed in the warehouse,
shops, labs, supply areas, inspecting area, the yard and
administrative personnel. None of these employees could
be assigned a dust level since their dust exposure was
intermittent and variable.
29
EXHIBIT 6-51
Merchant, et al., “Dose Response Studies in Cotton
Textile Workers”
There is evidence of awareness of the fundamental as-
sociation between increased dust level and increased prev-
alence of respiratory disease among textile workers since
early in this century when efforts were made to minimize
disease through suppression of dust from the carding
engine.’ Although this concept of dose-response was ap-
preciated, it was not quantitated until 1960 when Roach
and Schilling * reported their study of the association be-
tween three fractions of cotton dust and its character-
istic biological response. They found the strongest linear
association between the protein content of the middle
- fraction of dust (7, to 2mm-Hexlet) but also a strong
linear correlation with total lint and dust. Because of this
strong association with gross dust concentration and the
simplicity and rapidity of sampling for total dust, they
recommended 1 mg/m* gross dust as a reasonably safe
level of occupational exposure. This recommendation was
adopted by the American Congress of Government In-
dustrial Hygienists and is the current threshold limit
value for cotton dust.®
Roach and Schilling recognized that some fine dust
(< 7) and medium dust (7u-2mm) was trapped on the
lint mat collected on the 2 mm screen, which. may have
reduced the correlation coefficients for these fractions.? 4
Subsequent observations showed that the fine fraction of
dust (< 7u) could account for nearly all of the prevalance
of byssinosis,® ** and that gross dust levels, particularly in
yarn processing areas (spinning, winding, twisting) could
be misleading indicators of biological effect. °° In recog-
nition of these findings, new standards for safe levels of
cotton dust exposure are being considered in Great Britain
and the United States.
The study described in this paper offered the opportun-
ity to observe biological effects in a large working popula-
30
tion over a wide range of dust exposure. The primary
objectives of this portion of the study were to evaluate
the vertical elutriator cotton dust sampler and to de
velop dose-response relationships to help establish the safe
level of exposure to lint-free cotton dust.
Methods
Definitions, the study design, and methods used in col-
lecting data are found in the Methods section of the pre-
ceding paper.“ As previously described, each worker was
assigned a mean, median and a geometric mean dust level
of his work area which was used to develop dose-response
curves. Although little difference was apparent in these
three measures of central tendency (Table 1), the median
Table 1.—Vertical Elutriator Dust Samples by Mill Type,
Work Area, Measures of Central Tendency and Range
North Carolina, 1970-71
Work Area
Mill Type Preparation Yarn Slash/Weave Total
Cotton
Dust Samples, n 260 131 102 493
Workers, n 215 494 367 1076
Mean .899 .303 .985 .654
Median .669 .257 .798 485
Geometric Mean .639 .201 .828 411
Range .090-3.56 .018-1.30 .227-2.74 .013-3.56
Blend
Dust Samples, n 130 61 46 237
Workers, n 164 398 146 708
Mean .264 .122 .682 .270
Median 163 .088 .683 .163
Geometric Mean 194 .097 .666 .169
Range .006-1.37 .006-.715 .292-.959 .006-1.37
Synthetic-Wool
Dust Samples, n 134 121 — 255
Workers, n 212 586 — 798
Mean .269 .157 — 191
Median .238 .186 — .186
Geometric Mean .232 114 — 138
Range 017-788 .013-.847 — 013-.847
31
dust level was chosen because it reduced the effect of out-
liers and provided the best distribution of the population.
Mathematical dose-response curves were fitted to data
on byssinosis prevalence by median dust level. Six groups
of workers (Table 2), including black and white men and
Table 2.—Six Worker Groups by Smoking Status
Byssinosis
Group Smoking Status Grades Exposure
z Current Smokers Grade 4, 1,2 Cotton/blend mill prep-
aration and yarn areas
= Never Smoked Grade 42, 1,2 Cotton/blend mill prep-
aration and yarn areas
3. All Workers Grade 42,1,2 Cotton/blend mill prep-
aration and yarn areas
4. All Workers Grade 144,1,2 Cotton/blend mill slash-
ing and weaving areas
5. All Workers Grade 2 Cotton/blend mill prep-
aration and yarn areas
6. All Workers Grade 1 and 2 Cotton/blend mill prep-
ar:.tion and yarn areas
women, in two primary exposure areas ( primarily cot-
ton dust and cotton dust plus sizing) by smoking status
are considered:
Raw data was categorized by arbitrary dust level
groups to provide a wide and balanced (as well as pos-
sible) distribution of workers by dust level (Tables 3, 4).
Because the number of workers above 1 mg/m? repre-
sented less than five percent of the total population and
was therefore likely to result in unstable rates, and be-
cause those exposed to these high dust levels were likely
to be a select, relatively more resistant group of workers,
they were eliminated from consideration for calculation of
linear regressions (except for Group 4). The truncated
regression plots and correlation coefficients are displayed
for this data in Figs 2, 3 and 4.
32
Table 3.—Demographic and Smoking Characteristics of Preparation and
Yarn Processing Workers in Cotton «>? Blend Mills by Median Dust Level
Dust 0 -057 0839 .1255 .1878 .2809 .4202 .6287 .9405 1.408
Level to to to to to to to to to to
-056 -0888 .1254 .1877 .2808 .4201 .6286 .9404 1.407 2.105
Men
n 70 34 104 144 61 122 69 65 80 42
Mean Age 39.5 36.9 38.7 42.1 33.6 87.8 $8.8 $7.1 41.4 38.7
% Current
Smokers 67.1 64.7 56.7 54.9 718.8 53.3 73.9 55.4 73.3 66.7
% Former
Smokers 14.3 23.5 19.2 22.2 9.8 18.0 10.1 16.9 18.3 14.3
% Black 5.8 8.8 7.7 19.0 26.2 29.2 27.5 36.9 13.3 64.3
Women
n 15 87 89 135 17 85 17 0 1 0
Mean Age 41.5 43.9 44.4 42.9 39.6 88.8 37.7 21.0
% Current
Smokers 44.0 $2.4 25.8 86.3 47.1 86.5 $2.5 0 0 0
% Former
Smokers 12.0 0 4.5 6.7 5.9 12.9 10.4 0 0 0
% Black 19.2 10.8 5.6 17.9 52.9 16.9 27.8 0 100.0 0
Percent change in FEV, with six hours of dust ex-
posure for cotton/blend mill preparation and yarn area
workers and synthetic/wool mill workers by dust level is
shown in Table 5. Linear regressions truncated at 1
mg/m were calculated and the regression lines and cor-
relation coefficients shown in Figure 6.
Kenneth A. Busch, National Institute for Occupational
Safety and Health, analyzed prevalence data using a pro-
bit model,’?** and developed Table 6 through 9 and the
probit curves in Figs 2 through 4. His protocol for an-
alysis was as follows: For each set of data, the method
of maximum likelihood was used to fit a straight line to
Y= probit (prevalence ratio) vs. X= log, (median dust
level in mg/m*). A chi-square goodness-of-fit test was
made to test jointly for both nonlinearity and excessive
variance about the fitted curve as compared to theoretical
results expected under the probit model. Ninety-five per-
cent statistical confidence limits were put upon the true
33
dose-response curve as well as upon predicted dust levels
corresponding to eight arbitrary levels of hypothetical
byssinosis prevalence: p= 1%, 2%, 3%, 4%, 5%, 10%,
25,% and 50%. Pairs of curves for Groups 1 and 2, and
for Groups 3 and 4, respectively, were tested for parallel-
ism and, if found to be parallel, a “relative toxicity”
metameter R was estimated along with its 95% confidence
limits. By “relative toxicity” is meant the ratio between
dust levels which produce the same prevalence of byssin-
osis in the two groups. Thus an R-value of 1.0 would
imply the two curves were coincident. When curves for
two groups are found to be non parallel, the difference be-
tween them is difficult to interpret since relative toxicity
would not be constant at all levels of response. In such
a case, a conditional relative dose metameter Rp was
calculated for several p-levels of prevalence of byssinosis
in the two groups.
Results
Table 2 summarizes 985 vertical elutriator (Fig 1)
dust samples by mill type, work area, measures of central
tendency and range. The greatest number of samples was
collected in the preparation areas of the cotton and blend
mills. The highest dust levels are recorded in the slash-
ing and weaving areas of cotton and blend mills where
sizing is used and contributes to the dust concentration,
Where sizing is not used, the dust levels in the prepara-
tion area were consistently much higher than in the
yarn processing areas. The mean dust levels were gen-
erally higher than the median or geometric mean levels,
reflecting the effect of high out-liers. In synthetic and
wool mills there was little difference in dust level between
preparation and yarn processing areas. The highest dust
levels in these mills were found in the preparation area
of the wool mill and the preparation area of a synthetic
mill where gasoline driven machinery was used.
34
Demographic and smoking characteristics for men and
women in cotton preparation and yarn areas by dust
level are shown in Table 3. No clear age trend is ap-
parent among men or women, nor is there a consistent
difference in the proportion of current or ex-smokers be-
tween dust subgroups. Among men, a greater precentage
of blacks were found with increasing concentration of
dust exposure. This was less apparent among women who
rareiy work in the dustiest areas of the mills.
Busch concluded from his analysis that the log-probit
model fitted the data well in every case. The points were
scattered randomly about the fitted dose-response curve
and the variance of deviations from the curve was not
significantly greater than would be expected based upon
an assumption of binomially-distributed prevalence ratios
at each dust level.
Slopes (b) and intercepts (a) of the fitted log-probit
dose-response curves, Y= a + bx, are shown in Table
6 together with three calculated points (ordinates Y) for
each fitted curve and their 95% confidence limits. The
curve equation was then inverted to predict dust levels
corresponding to arbitrary prevalence levels. Table 7
shows the predicted dust levels and their 95% confidence
limits for the six groups of workers for prevalence levels
of 1%, 2%, 3%, 4%, 5%, 10%, 25% and 50%.
Chi-square tests for goodness-of-fit of the individual
curves and for parallelism of pairs of curves (1 vs. 2
and 3 vs. 4) are shown in Table 8. Curves for current
smokers (Group 1) and those who never smoked (Group
2) were found to be significantly different although
nearly parallel but not coincident. A relative toxicity of
R= .56 with 95% confidence limits of R= .35 to R— .83
was found (Table 8) ; this suggests that only 56% (35%
to 83%) as much cotton dust is associated with any
given prevalence of byssinosis among smokers as among
¥
35
those who never smoked. Curves for all prepartion and
yarn workers (Group 3) and for all slashing and weav-
AIR FLOW
GONTROL SAGE
(LIMITING
ORIFICE)
Diigo: ©
VACUUM
—— PUMP
| SETTLING
— CHAMBER
VERTICAL ELUTRIATOR
COTTON DUST SAMPLER
~Fig 1.
FY
36
ing workers (Group 4) were not found to be parallel.
Therefore a single R-value was meaningless and it was
necessary to calculate R,-value as a function of eight
arbitrary prevalence levels which are shown in Table 8.
Curves for Groups 3 and 4 were found to be significantly
different but yielding R,-values which ranged from .087
at a byssinosis prevalence of 1% to .56 at a prevalence
of 50%. Therefore, only approximately 9% (4% to
15%) as much cotton dust is associated with a 1% bys-
sinosis prevalence for workers in the preparation and
yarn areas of cotton mills as compared to the amount re-
quired for the same prevalence for cotton slashing and
weaving workers. At a prevalence of 5% an average of
15% as much dust is required, as a prevalence of 25%,
32% as much, and at the 50% prevalence level 56% as
much dust is required.
The probit-dose response curves are shown in Figs 2,
3, 4, together with the linear regression plots and cor-
relation coefficients. The truncated linear plots are ob-
served to fall within the 95% confidence limits of the
probit curves below a dust level of 0.5 mg/m*. Both dose
response curves for cotton slashing and weaving workers
fell nearly on the same line through 2.0 mg/m? of dust.
The strength of the linear association between dust level
and byssinosis prevalence was uniformly high except for
those who had never smoked (Group 2) where the cor-
relation coefficient was .52; however, one subgroup within
this population contained only seven workers, none of
whom had symptoms of byssinosis. Expected byssinosis
prevalence levels are shown in Table 6 and the curves
plotted for Groups 3, 5 and 6 in Figure 4. Prevalence of
Grade 2 byssinosis (Group 5) ranges from 1.3% (0.7-
2.3) at 0.1 mg/m* to 3.0% (2.1-4.3) at 0.2 mg/m to 8.0
\6.1-10.2) at 0.5 mg/m*, as computed from the probit
curve. When Grade 1 and 2 byssinotics are combined
(Group 6) the probit prevalence levels are 2.1% (1.3-
3.3) at 0.1 mg/m’, 5.0% (3.9-6.6) at 0.2 mg/m? and
13% (10.7-15.8) at 0.5 mg/m‘,
72
37
Table 5 and Fig 5 show percent change in FEV,, with
increasing levels of exposure among cotton preparation
and yarn area workers and synthetic and wool workers.
The truncated linear regression (Fig 5) for the cotton
population again showed a strong linear association (r —
.82) between biological effect and dust level. No decre-
ment in percent changes in FEV,, is seen with increasing
levels of exposure to synthetic and wool dust.
Discussion
Dust Sampling.—In the year preceding this study, sev-
eral cotton dust sampling techniques were evaluated."
The sampling device desired was to be relatively easy to
operate, portable, durable, operable unattended for a full
shift and provide a lint free dust sample. Of the sam-
pling devices evaluated, the vertical elutriator developed
by Lynch and Lumsden * was found to best fulfill these
criteria. The instrument was designed to sample dust
with a mass median aerodynamic diameter of 15. and
less which effectively eliminated biologically inert lint
while retaining particles of the size expected to be in-
haled. Since high correlations between byssinosis symp-
toms and dust level have been observed with the middle
fraction (7. to 2mm) of dust, *" it was thought that
larger particles (7y-15,) as well as “respirable” parti-
cles (<7) should be collected. An added practical ad-
vantage of including larger particles is the accumulation
of a weighable sample in a shorter interval. A potential
disadvantage was the possibility of including fine but
inert “linters” in the sample. This has been observed
rarely in our experience and appears to occur only under
the dustiest of conditions when exact dust levels are not
as important.
In the absence of a specific assay for the etiological
agent(s) responsible for the biological effects of cotton
dust, the most important consideration in evaluating a
38
Table 4.—New Data on Byssinosis Prevalence vs.
Median Dust Levels for Six Groups of Workers
Mid-Point * of Range Sample Cases of
of Median Dust Levels Size Byssinosis Prevalence (%)
(mg/m) n r p = r/nx 100
Group 1. Cotton Preparation and Yarn Areas
All Grades—Current Smokers
.05 167 10 6.0
15 159 21 13.2
25 82 14 17.1
35 64 14 21.9
45 55 15 27.3
55 26 11 42.3
75 31 16 51.6
1.1 12 5 41.7
1.5 12 7 58.3
1.9 26 12 46.2
Group 2. Cotton Preparation and Yarn Areas
All Grades—NonSmokers
15 127 3 2.4
15 145 9 6.2
.25 44 2 4.5
35 ? an 16 29.6
45 49 12 24.5
55 7 0 0
75 18 5 27.8
1.1 2 0 0
1.5 8 2 66.7
1.9 8 3 37.5
Group 8. Cotton Preparation and Yarn Areas
All Grades—All Workers
.0458 145 5 3.4
.0686 71 2 2.8
.1026 193 14 7.3
.1535 279 27 9.7
.2296 78 10 12.8
3435 208 89 18.8
.5139 147 87 25.2
-7689 65 80 46.2
1.150 81 14 45.2
1.722 42 17 40.5
* Arithmetic mean of end-points used for groups 1, 2, 4.
Geometric mean of end points used for groups 3, 5, 6.
39
Table 4.—Continued
Mid-Point * of Range Sample Cases of
of Median Dust Levels Size Byssinosis Prevalence (%)
(mg/m?) n r p = r/nx 100
Group 4. Cotton Slashing and Weaving Workers
All Grades—All Workers
35 15 1 6.7
45 136 7 5.1
55 2 0 0
75 195 14 7.2
1.1 34 6 17.6
1.5 87 23 26.4
1.9 44 16 36.4
Group 5. Cotton Preparation and Yarn Areas
Grade 2
.0458 145 0 0
.0686 71 1 1.4
.1026 193 2 1.0
.1535 279 9 3.2
.2296 78 1 1.3
3435 208 8 3.8
.5189 147 14 9.5
-7689 65 11 16.9
1.150 81 6 19.4
1.722 42 6 14.3
Group 6. Cotton Preparation and Yarn Areas
Grades 1 & 2
.0458 145 0 0
.0586 71 2 2.8
.1026 193 2 1.0
.1535 279 11 3.9
.2296 78 2 2.6
3435 208 23 11.1
.5189 147 22 15.0
.7689 65 15 23.1
1.150 31 8 25.8
1.722 42 9 21.4
* Arithmetic mean of end-points used for groups 1, 2, 4.
Geometric mean of end points used for groups 8, 5, 6.
40
OVSSImS 1S PwL VaR CECE OF MOLAR GUST LK WiR fora COTTOm FFF ita
AAD TAGK An(A mUPEL PS B00 COTTCM WASHi at dau sf beim, My :
LIRGAR MLGALSSICHS O00 FITETO PLSiT (org -PELPMELE CRY —<
70r AAO [MEiR GSE COMPILE LIMITS.
moate Cary ima, 19/0-7!
-
eer
ws | wa GROUP 3
$ ae (r*.99)
7° po =
a Ya ora
- or a
= 50 - me m
< ° Pr as we
< a ” nn“ oa?
= a a oo”
« SOF ZA
= ,
s a
4 7
= 30 a
= y/ A Pa
a 4 a ogi
sO
e 4 7
ne 4 of
¢
° 20- 4 4
c i
a 14 ‘7
*
a 1p Group 3 (0) CPY |
Group 4 (a) CSW
*
f
£ “
a. waa - ; oe s
On' 2.345 l 5 —
Medion Dust Level (mg/m*)
Fig-2
gravimetric sampling technique is the degree of correla-
tion with biological indicators over the range of exposure.
As shown in Figs 2 through 5, a strong linear associa-
tion was found in this population, particularly below 0.5
mg/m‘*, the level below which the majority of the popula-
tion is exposed and where attention is focused concerning
establishment of a reasonably safe exposure level.
As a field instrument the vertical elutriator cotton
dust samplers have proven to be most practical and dur-
able. The 25 samplers built for this survey have now been
in use for two years and have collected over 3000 sam-
ples. No modification of the instrument or the sampling
procedure has been necessary. Its design lends itself to
use within the manufacturing plant; at no time has it
ie
OTSSIMOSIS POEVAL ENCE BY PCOL AM OUST CIVIL AMONG CyeetaT coOctes
FAD THISE oO SETCR SHOKTD, COTIOM PREPARATION AMO YAPH AREA WORKERS:
RiALAR AIGAESSIONS AMO FITTCD POOSTT TOSE-aLSPONSE CURVES.
WORTM CAROLIAA, 1970-71
7Or
60
°o _
=
-
a
ae om ow
50- GROUP | po na pail
(ce.98) an Same,
-
= an °
o* -”
a rd
- ° -*
40 a
ail “4
GROUP 2
30 dr .52)
20
Group | (©) Smokers
Group 2(4) Non-smokers
Cyssinosis Prevalence (%)-All Grades
=
ee, Pee ak eS i =" 1 i fe
staat & 1 ie) 2
Medion Dust Level! (mg/m)
Fle 2X
interfered with the manufacturing process or personnel.
Sampling with this instrument can easily be learned and
could be used by manufacturing personnel to monitor
their working environment.
Dose-Response.—In developing dose-response curves for
byssinosis, tables were developed by sex, smoking groups,
work areas (preparation, yarn production and slashing/
weaving) and mill type (cotton and blend). A strong
linear association was repeatedly observed in these sub-
populations. Since there appeared to be no difference in
age between individual dust level subgroups (Table 3),
there was no reason to age adjust the data. Similarly,
no difference in byssinosis prevalence has been found in
men and women, allowing both to be considered together.
42
Preparation and yarn processing areas are justifiably
combined since the dust is of the same composition, the
dose-response curves for each area are similar, and the
areas are frequently continguous. When the yarn arrives
in the slashing department, sizing is added to the yarn
and has been found consistently to increase the concen-
tration of lint free dust in the slashing and weaving
areas. Therefore, the biologically active airborne material
in these workrooms is diluted with biologically inert. siz-
ing, making it necessary to consider this dust separately
when considering these significantly different dose-re-
sponse relationships. Cigarette smoking has been found
to significantly increase byssinosis prevalence * 1! 1°17 and
must therefore be considered a potential source for a sec-
ondary association. Table 8, however, shows no apparent
difference in smoking habits between dust level sub-
populations. Probit analysis revealed that byssinosis
prevalence for smokers was significantly higher than for
non-smokers. Although it may be argued that the most
Susceptible group (smokers) should be most closely con-
sidered when recommending safe levels of exposure, the
majority of this working population smoke cigarettes and
the dose-response curves for current smokers and all
workers regardless of smoking habit are similar. For this
reason, and because a standard generally applies to the
entire population at risk, consideration of the entire pop-
ulation regardless of smoking habit appears to be most
appropriate.
As initially observed by Roach and Schilling,” and later
confirmed by El-Batawi,'® Molyneaux,”® and our group in
an independent study,”® dose-response curves for the asso-
ciation between dust level and byssinosis prevalence and/
or change in FEV,, are strikingly linear. This was found
repeatedly in this data no matter how the groups were
divided, unless the number of workers in the dust level
sub-populations was so low as to produce unstable rates
43
(Group 2, r = .52, Fig 3). With this exception, correla-
tion coefficients were consistently above 0.9. Similarly,
the association between percent change in FEV,, and
median dust levels was also strong (r = .82, Fig 5). A
source of increase variance in this measurement is di-
urnal variation in expiratory flow rate which was ob-
served but not adjusted out. The distribution of workers
by shift within dust level sub-populations was similar and
therefore unlikely to produce a secondary association.
Above a median dust level of 1 mg/m® the dose-response
curves tend to flatten. We suspect that selection, leaving
relatively resistant workers in these dustier areas, is re-
sponsible for this decrease in prevalence and percent
change in FEV,.,. As reviewed in the preceding article,
there is substantial evidence of selection away from dust
exposure in this industry.®**! In Table 3, a consistent
trend toward a higher percentage of black men was found
with increasing dust level. Blacks have been found to
have lower rates of chronic bronchitis than whites of
Of Crore 61 iL Iee Ort
wOteded Gnd Semied i itsmde woeeert.
‘
7
aon, tony ~ bes — ond °
‘orn Workers (0)
4 Fad £2.82
*%O,FEVIO
: ;
4
n
+2
L
1 i _, i i — i
Oo 04 02 03 04 05 06 0.9 | 1.3
MEDIAN OUST LEVEL (mq/m?)
Fig 4
44
Table 5.—Raw Data on Percent Change in FEV,,, vs. Median Dust
Level among Preparation and Yarn Area Workers in Cotton/ Blend
Mills and Workers in Synthetic/Wool Mills
Cotton/Blend Synthetic/Wool
Mid-Point of Range of Mills Mills
Median Dust Levels n mean §.D. n mean S.D.
(mg/m?) % AFEV,,, % AFEV,.,
0.05 299 0.4 9.4 241 0.7 5.5
0.15 309 0.5 9.8 212 0.2 7.0
0.25 118 —1.8 10.0 141 6.7
0.35 130 —4.2 6.1 112 0.7 7.0
0.45 95 —2.6 16.2 7 0.2 3.3
0.55 85 —2.0 8.8
0.75 61 —4.9 10.1
1.10 146 —0.9 10.3
1.50 © 12 —2.7 6.0
1.90 385 —5.6 10.9
Table 6.—Parameters of Log 10-Probit Dose-Response Curves and Three Points
on Each Fitted Curve with Their 95% Confidence Limits
Intercept Slope Expected Prevalence (%)
Group (a) (b) -1 mg/m*® .2 mg/m? -56 mg/m?
4.826 1.108 10.0% (7.3-13.38) 17.1% (14.2-20.5) 30.6 % (25.9-85.6)
1
2 4.641 1.301 4.9% (2.8-7.9) 10.2% (7.5-13.6) 22.7 % (17.0-29.8)
3 4.718 1.226 6.5% (5.0-8.5) 12.7% (10.8-14.9) 25.8 % (22.5-29.3)
4 3.966 2.134 0.08% (0.01-0.7) 0.6% (0.1-2.1) 4.7% (2.7-7.7)
5 3.944 1.170 1.8% (0.7-2.3) 3.0% (2.1-4.3) 8.0% (6.1-10.2)
6 4.269 1.304 2.1% (1.3-8.3) 5.0% (3.8-6.6) 13.0 % (10.7-15.8)
the same age and environmental exposure.”? The rela-
tively greater number of blacks with relatively lower
response at high dust levels in this textile population
may represent another example of ethnic difference in
susceptibility to the biological manifestations of inhal-
ants. Further analyses of this data are being done to
determine whether this is a sociological or a biological
problem.
Another important observation concerning these dose-
response curves is that there appears to be no threshold
beneath which no one with Monday chest tightness was
found (Table 4). This indication that low dust concentra-
tions result in measurable effects suggests that cotton
45
(¥°L-8°Z) 9°8 (9°82-6°8) 0° (¥°9°3°S) T'S (L°O-3'T) Lt (S°S-T'T) 6'T (9°S-L6") OT %03
(9°T-P8") T°T (9°F-P'T) 1S (6T-S'T) 9°T (69°-0") 8h" (L6°-Th*) LS" (L¥°-82") 98° %3
(9F°-18") 88° (86°-6)") b9° (06°-29") LL" (8T°-Z1") ST" (93°-b1") 02° (¥T°-990") OT" ot
($Z°-S1") 02° (1h"-b2") 18° (¥9°-98") 29° (OT*-¥90") LL0° (ST"-L90") OT° (1L0°-8Z0") LbO° %s
(12"-Z1") 91° (88°-81") 92° (89°-18") 9P° (¥80°-8F0") 890° (Z1°-#F0") 980° (690°-610") 880° %y
(LT°-160") 81° (92°81) 02° (Z9°-92") OF" (890°-Z80") 090° (0T"-180") 890° (L¥0°-810") 620° yd}
(81°-890") L60° (61°-980") PT" (Sh°-61") 88° (19°0-1Z0") 980° (180°-020") 090° ($80°-800") 020° 2%
(980°-F80") 090° (Z1°-2b0") 280° (98°-Z1") 93° ($80°-Z10") 120° (990°-010") 180° (2Z0°-600") L10° Mt
9 dno g dnoin p dnoin ¢ dnoiy Z dnoary I dnowy aouspeaaig
(gui/Bul) jaa] ysnq pezorperg
SSOUJSEA JO BoUBTBArIG JO SOAs] AIBIZIGQIY 103 syUII'] EOUePYUOD %96 IIEYL, pus spear] yong p2}o1perg—"), a1qBI,
46
dust is a highly biologically active inhalant. British stud-
ies reported by Molyneaux and Berry™ also found no
threshold for the biological effects (byssinosis and simple
bronchitis) of respirable dust (<7,-Hexlet) and middle
fraction dust (7. to 2mm-Hexlet). Although these dust
fractions and the vertical elutriator dust fraction are not
strictly comparable, they do contain much of the same
dust distribution. When their dose-response regressions
are plotted with those presented in this paper, there is
good general agreement both in origin and slope. At 0.2
mg/m*, both sets of curves reveal roughly 15% with some
grade of byssinosis. As has been observed before, surpris-
ing uniformity in byssinosis prevalence has been found
in cross-sectional surveys despite differing populations,
working conditions, and frequently somewhat different.
methods of study.
OTISIMOSIS POL VALE RCE BY COSTE Am) OF ole -- A... Ate
70r AMO8S COTTIOS PRL PSRAT ICN On9 TAPe Byta
RPMEAR ALCALSS 10nd Ory FITTEG beretl COM. me pOrse Connts.
sOuin Cama ina, 1970-7)
om
o
T
-_s
--"
on
--
w
Oo
oo
a
af GROUP 3+ 2 Gredey (0)
(e*39)
a
2]
°
PP a
ow”
-" GROUP 6- Gredes 182 (a)
(r*.98)
ul
o
---"
--—~ nour s- - Ge
-_ ede 2 (5)
(r*.$6)
Byssinosls Prevalence (%)
tv
oO
)
1
—
Median Dust Level (mg/m3)
Fig 5.
47
Table 8.—Chi-Square Tests for Goodness-of-Fit of the Log-Probit
Model and for Parallelism of Pairs of Fitted Lines.
Goodness-of-Fit Parallelism
Group x? df P x? df P
1 6.0 8 .65
2 14.1 8 .08
3 7.5 8 52
4 3.8 5 .58
5 8.3 8 .60
6 10.0 8 .26
1&2 20.1 16 21 0.53 1 AT
3&4 11.3 13 .59 6.0* 1 .015
Symbols: df=degrees of freedom
P=probability of x? as large or larger than observed
value
* = significant at the .05 probability level
Table 9.—Relative Toxicities and Their 95% Confidence Limits
Byssinosis Prevalence Relative Toxicity
Groups P R 95% Confidence Limits
1&2 All levels 56 .35 to .83
3 & 4+ 01 .09 .04 to .17
02 ll .06 to .19
.03 12 08 to .20
.04 14 .09 to .21
.05 15 -10 to .22
10 .20 -15 to .26
25 32 .24 to .43
.50 56 32 to .96
Symbols: P=No. of subjects with byssinosis/total No. of subjects
R=ratio of dust levels producing same prevalence (eg.,
dust level for group 1/dust level for group 2)
+=Note: Dose response curves for groups 3 and 4 are
non-parallel, so that relative toxicity varies
with level of prevalence
48
The log-probit model was found to fit this data well
and offers the opportunity to consider prevalence at fixed
dust levels and conversely, calculate dust levels from any
given level of prevalence. This is clearly very useful when
considering data for setting standards. Probit analysis
(Table 6) shows the expected byssinosis prevalence for
all grades of byssinotics at 0.2 mg/m* to be 12.7%, while
only 3.0% for Grade 2 byssinotics. In order to determine
what a reasonably safe level of lint free dust might be,
the question is raised whether all grades of byssinosis or
only Grade 2 byssinosis prevalence is the appropriate
indicator. We concluded that overall byssinosis prevalence
best estimated byssinosis risk since it was likely to be
less affected by selection than Grade 2 byssinosis. In our
experience, workers usually do not consider selecting
themselves out of exposure if they have only occasional
chest tightness or tightness which is not severe and con-
fined to Monday. Those with tightness on Monday and
other days (Grade 2) frequently also complain of dyspnea
and fatigue; as a result it is not uncommon to find that
these workers request a change in job location. There-
fore, those with Grade 2 byssinosis symptoms are likely to
represent a highly selected group which would very likely
result in an underestimation of risk.
Also to be considered is the sensitivity of the biological
indicators used in this study; the standard indicators in
studying the effects of cotton dust have been byssinosis
symptoms and change in FEV,.,. There is now evidence
that use of flow-volume loops ***** and measurement of
the leukocyte response * increase sensitivity in detecting
biological effects. Therefore, because of selection and be-
cause the indicators of response in developing this dose-
response data although well standardized are probably
still dull tools, we conclude that, for cotton and blend mill
preparation and yarn areas, a reasonably safe level of
cotton dust exposure is 0.1 mg/m*. Even at this low level
an expected byssinosis prevalence of 6.5% was found by
49
probit analysis. A separate level of 0.75 mg/m is sug-
gested as a reasonably safe level in slashing and weaving
areas. It will be necessary to include in any cotton dust
standard provisions for periodic testing of workers to
detect those most susceptible and to avoid placement of
those with impaired lung function in dusty areas.
Dust Control.—There is now substantial evidence that
biological effects result from low levels of lint-free cotton
dust exposure, even with relatively crude indicators of
response. To make the cotton textile mill working envi-
ronment reasonably safe, very great strides must be
taken in controlling fine dust. Initially, this requires a
reorientation in thinking regarding dust handling from
suppressing primarily lint to suppression of fine dust as
well. Currently, machine exhaust and ventilation and
filtration systems are designed primarily to control lint
and large dust particles and not fine dust. As a result,
the workroom frequently appears lint free and relatively
clean, yet lint free dust levels may be relatively high. To
solve this problem, two basic approaches may be taken.
One is the traditional approach of dust control after the
dust has been introduced into the mill; the second is re-
moval of dust from lint prior to manufacturing. Clearly,
the most desirable approach would be a method to harvest
cotton without contamination with trash. New methods
to improve picking are now under consideration. Dust
reduction at the ginning stage would be the next most
desirable location to control dust. There is now experi-
mental evidence that the application of steam to cotton is
compatible with manufacturing and reduces lint free
dust levels and biological activity by roughly one half.’
A plant wide intervention trial is now underway to test
the effectiveness of steaming under manufacturing condi-
tions. The feasibility of steaming cotton at a gin is also
being tested. Although steaming may provide improve-
ment in environmental conditions, it is apparent that
50
much improved dust control, particularly in preparation
areas will also be required.
Table 2 shows that lint free dust levels are particularly
high in the preparation areas of cotton mills. Review of
dust levels from six cotton mills (three from this study;
670 samples) revealed vertical elutriator median dust
levels of 1.50-1.59 mg/m* in opening and blending areas,
1.60-1.69 mg/m* in picking areas, 1.70-1.79mg/m* in
carding areas, then dropping to 0.70-0.79 mg/m* in
drawing, 0.40-0.49 mg/m* in roving, 0.20-0.29 mg/m? in
spinning, winding and twisting, and then up again in
weaving to 1.00-1.09 mg/m*. In eleven blend mills (50%
cotton or less), two of which are part of the study re-
ported in this paper, 1232 vertical elutriator dust samples
found a median dust level of 0.30-0.39 mg/m* in opening
and blending, 0.50-0.59 mg/m* in picking, 0.60-0.69
mg/m in carding, 0.30-0.39 mg/m* in drawing, 0.10-0.19
mg/m? in roving, 0.00-0.09 mg/m? in spinning. 0.10-0.19
mg/m’ in winding and twisting and 0.50-0.59 mg/m* in
weaving areas.”° In these eleven blend mills, the spinning
and weaving areas show median dust levels that may be
considered reasonably safe and levels in roving, winding
and twisting approach this level. Since less than 10% of
the work force is employed in areas preceding roving,
roughly 90% of those working in these eleven blend mills
could be considered as working in a reasonably safe or
marginally safe working environment. Those remaining,
although exposed to hazardous dust levels, are working in
a third or less the dust concentration of their counter-
parts in cotton mills.
By contrast, it appears from these figures that there is
no work area in the cotton mills that could be considered
reasonably or marginally safe, although yarn processing
and weaving areas are not far from these levels. The
areas clearly in acute need of attention are those of
opening, blending, picking and carding. Because drawing
51
and roving are almost invariably in the same work area
as the carding engine, much of the dust exposure in these
latter two areas probably arises from the carding engine.
To protect these workers, the carding area should be
partitioned from drawing and roving and each area pro-
vided with an independent ventilation system. Under
such conditions the dust levels in drawing and roving are
expected to more closely approximate those in spinning.
Close attention to recirculation of lint free dust (< .05
mg/m*)* should further reduce dust levels to more ac-
ceptable levels. Preprocessing removal of fine dust by
steam may also contribute to achieving safe levels in
these areas. But even if steaming were to reduce dust
levels by a half, the processes of opening, blending, pick-
‘ng and carding will require efficient exhaust systems.
“‘ortunately, in each of these machines, the area in which
the fiber is most vigorously processed is in a relatively
enclosed part of the machine. Further enclosure of each
of these machines with well designed exhaust systems
does not appear to be insurmountable. Efficient removal
of fine dust, without recirculation, should markedly im-
prove these preparation areas and perhaps could also
remove some dust which previously was released from
the yarn in subsequent processes, thereby contributing to
control in these work areas.
In summary, the following statements can be made:
1. This study confirms the finding of others, that a
strong linear association exists between prevalence of
byssinosis and decrement in expiratory flow rate with
concentration of lint free dust.
2. The log-probit model, curves of which followed the
linear regression plot below 0.5 mg/m*, fit this data well
and provided both expected byssinosis prevalence and
conversely expected dust levels, both of which proved to
be useful in interpretation of the dose-response relation-
ship.
52
8. Based on these curves, it is concluded that a reason-
ably safe level of lint free cotton dust is 0.1 mg/m’, a
level at which nearly 94% of the population exposed were
found to have no symptoms of byssinosis. A separate level
of 0.75 mg/m is suggested for slashing and weaving
areas.
4. Probit dose-response curves for smokers and those
who never smoked, showed that smokers had a signfi-
cantly higher prevalence of byssinosis.
5. The vertical elutriator cotton dust sampler, over a
period of two years, has proven to be a durable and prac-
tical instrument which collects a biologically active lint
free fraction of dust linearly associated with indicators
of biological response.
6. Lint free dust levels by work area suggest that no
work areas in the cotton mills sampled had reasonably
safe dust levels and that the areas of opening through
carding had very high levels. By contrast, in blend mills
all areas beyond drawing had reasonable safe or margin-
ally safe dust levels while levels preceding roving are
considered only moderately elevated.
7. It is recommended that carding machines be isolated
from drawing and roving processes by partitioning and
the use of independent ventilation systems.
8. Serious attention should be given to more complete
enclosure of opening, blending, picking and carding ma-
chines and design of a highly efficient exhaust system to
remove fine dust.
9. At the present time, a successful occupational health
program for the cotton textile industry should include
efforts to remove or reduce dust prior to processing, effi-
cient machine exhaust and ventilation systems, and med-
ical surveillance to detect susceptible workers before they
acquire permanent pulmonary impairment.
53
The authors express their appreciation for the guid-
ance of Drs. Ben Drake and Martin Hines, the advice of
Drs. Carl Shy and Al Tyroler, data processing provided
by Joe Rouchard and Jim Holmes, the advice and contri-
bution of Howard Ayre, J erry Lynch and Kenneth Busch
in analysis of data, and for the cooperation and partici-
pation of the employees and management of Burlington
Industries.
References
1. Collis EL: Industrial pneumoconioses with special
reference to dust phthisis (Milroy Lectures, 1915), Pub-
lic Health 28:252-253, 1915; 29:11-20, 37, 44, 1916.
2. Roach SA and Schilling RSF: A clinical and envi-
ronmental study of byssinosis in the Lancashire cotton
industry, Brit J Industr Med 17 71-19, 1960.
3. Committee on Threshold Limit Values for Airborne
Contaminants, Threshold Limit Values for 1971, Ameri-
can Conference of Governmental Industrial Hygienists,
Cincinnati, Ohio, 1971.
4. Schilling RSF: The history of byssinosis and the
British experience. Transactions of the National Con-
ference on Cotton Dust and Health, pp. 7-12, 19-20, Uni-
versity of North Carolina, Chapel Hill, North Carolina,
1970.
5. McKerrow CB, et al: The size of cotton dust par-
ticles causing byssinosis: An environmental and physio-
logical study, Brit J Industr Med 19:1-8, 1961.
6. Wood CH, Roach SA: Dust in cardrooms: A con-
tinuing problem in the cotton-spinning industry, Brit J
Industr Med 21:180-186, 1964.
7. Molyneaux MKB, Tombleson JBL: An epidemio-
logical study of respiratory symptoms in Lancashire mills,
1963-1966. Brit J Industr Med 27 :225-234, 1970.
54
8. Hammad YY, Corn M: Hygienic assessment of air-
borne cotton dust in a textile manufacturing facility,
Amer Industr Hyg Assoc J 32:662-667, 1971.
9. Merchant JA, et al: Byssinosis and chronic bron-
chitis among cotton textile workers, Ann Int Med 76:423-
433, 1972.
10. Lumsden JC, et al: Cotton dust sampling. In
preparation.
11. Merchant JA, et al: An industrial study of the
biological effects of cotton dust and cigarette smoke ex-
posure. Proceedings of the 1972 Skytop Conference on
Respiratory Disease in Industry, J Occ Med 15:1973.
12. Busch KA: Probit analyses of data on prevalence
of byssinosis in working populations exposed to cotton
dust. National Institute for Occupational Safety and
Health Memo. February, 1972.
13. Finney DJ: Probit Analysis, A Statistical Treat-
ment of the Sigmoid Response Curve, Cambridge Uni-
versity Press, 1962.
14. Lynch JR: Air sampling for cotton dust. Transac-
tions of the National Conferences on Cotton Dust and
Health, pp. 33-48. University of North Carolina, Chapel
Hill, North Carolina. 1970.
15. Molyneaux MBK, Berry G: The correlation of
cotton dust exposure with the prevalence of respiratory
symptoms. Proceedings of the International conference on
Respiratory Diseases in Textile Workers. pp. 177-183.
Alicante, Spain, 1968.
16. Schilling RSF: Epidemiological studies of chronic
respiratory disease among cotton operatives, Yale J Biol
Med 37 :55-74, 1964.
17. Carey GCR et al: Byssinosis in flax workers in
Northern Ireland. HMSO, Belfast, 1965.
55
18. Batawi MA El, et al: Byssinosis in the Egyptian
cotton industry: changes in ventilatory capacity during
the day, Brit J Industr Med 21:18-19, 1964.
19. Merchant JA, et al: Preprocessing cotton to pre-
vent byssinosis. In press, Brit J Industr Med.
20. Gandevia B, Milne J: Ventilatory capacity changes
on exposure to cotton dust and their relevance to byssi-
nosis in Australia, Brit J Ind Med 22 :295-304, 1965.
21. Elwood PC: Respiratory symptoms in men who had
previously worked in a flax mill in Northern Ireland, Brit
J Industr Med 22 :38-42, 1965.
22. Densen PM et al: A survey of respiratory disease
among New York City postal and transit workers,
Environ Res 1:265-286, 1967.
23. Bouhuys A, et al: Maximum expiratory flow rates
in induced bronchoconstriction in man, J Clin Inv 48:
1159-1168, 1969.
24. Merchant JA, et al: Assessment of clinical indi-
cators of response to cotton dust. International Confer-
ence on Biological Responses to Organic Agents. Trans-
actions of the New York Academy of Sciences, in press,
1972.
25. The Advisory Committee to the National Confer-
ence on Cotton Dust and Health. The Status of Byssi-
nosis in the United States, Arch Environ HIth 23 :230-
234, 1971.
Supported in part by a grant from National Institute
of Environmental Health Sciences. Grant No. 2TO1ES-
00124 and by a grant from the National Institute of
Occupational Safety and Health, Grant No. SRO1
0H00302.
BYSSINOSIS PREVALENCE (per cent)
56
EXHIBIT 6-55
Molyneux & Berry, “The Correlation of Cotton Dust
CN
Oo
t
40.
nO
o
o
Exposure With the Prevalence
of Respiratory Symptoms”
* * * « *
FIGURE 4
CORRELATION. BYSSINOSIS IN FIVE OCCUPATIONS
945 Subjects
z= 0.766 (excluding ringspinners)
i + Po wingspinners
ay
, 4
0.2 0.4 0.6 0.8 1.0
MEDIUM DUST (mg/M*)
DYSSINOSIS PREVALENCE (oer cent)
SIMPLE BRONCHITIS PREVALENCE (per cent)
57
FIGURE 5
CORRELATION. CYSS:INOLIS IN SPELNFRAIME TENTERS
60.
40.
20 -
0,2 0.4 0.6 0.8 1.0
MEDIUM DUST (m9/ta*)
FIGURE 6
CORRELATICN SUPLE DRONCHITIS in non sras! cers (35-54 years)
60 + 17i Suijects
x= 0.619 he
40 -
20 ||
CH ringsdinners
0
: q AY ‘ Ce ? oe
0, 2 0.4 0. 6 0.5 4 .
RESPIRACLE OUST (ms/At’)
58
EXHIBIT 6-57
Experience with Medical Surveillance Programs
By: Harold R. Imbus, M. D., Sc.D.
Medical Director
Burlington Industries, Inc.
* * * * *
Nevertheless, one can see the marked difference in the in-
cidence of symptoms of tightness in the chest and in the
decline of FEV, in these employees. Only one employee
has Grade I byssinosis. This is now becoming a man-
ageable situation in which medical surveillance can iden-
tify susceptible individuals, monitor their respiratory
function periodically, advise regarding respirators, and
smoking habits, and even consider transfer if necessary.
Table II illustrates the card room of another plant. In
early 1971 an initial survey was done. Dust levels were
high. and Column 1 shows clearly that there is a sig-
nificant problem and something needed to be done.
Clearly there was little more that a medical surveillance
program could do than to identify the problem, which was
too much dust. Control measures were instituted shortly
thereafter with installation of new card cleaning equip-
ment. You can see from Column 2 that dust levels were
lowered by about two-thirds and so was the percentage
with symptoms of byssinosis. Decrement in FEV, was
also decreased significantly. However, further improve-
ment here is highly desirable. Unfortunately, many tex-
59
tile plants find themselves in this position, having in-
stalled ventilation equipment, in the last few years, at
large expense, which though what was generally available,
is not adequate to control the byssinosis problem. At this
time, an entirely new installation of chute feed carding
is in process in this plant’ and dust levels will be much
lower in the near future. Likewise, it is anticipated that
the number of reactor employees will be much lower.
Table III, left side, illustrates a survey of a card room
and drawing area in a rather large cotton blend plant
which has spun cotton yarn for many years. Dust levels
in bot areas are approximately .4 milligram per cubic
meter. In the card room, no employee has symptoms of
byssinosis and only one (3%) had a decrement of 10%
or more in FEV,. It is seen that average FEV, actually
increases during the working day. In the drawing area,
3.2% have byssinotic symptoms and 6.5% have a decre-
ment of FEV, of 10%, an average decrement of 48ce,
about what we have at our Corporate Headquarters.
This plant uses a higher grade of cotton blend and it ap-
pears that we have an entirely manageable situation.
With medical surveillance, any employee having problems
can be detected early and protected either with a respira-
tor or by transfer.
On the other hand, Table III, right side, shows a small
area in another plant with dust levels somewhat com-
parable, but incidence of symptoms and decrement of
FEV, is much greater. Other areas in this plant were
found to have even higher dust levels, but I show these
areas because dust levels are in the same “ball park.”
This is a 100% cotton plant. Even though dust levels
here do not look inordinarily high, it is obvious that
further action needs to be taken to lower them, and
medical surveillance is strictly an interim measure.
* * * * od
60
The point I am trying to make by using these examples
is that in order for a medical surveillance program to
provide any real protection you must first of all have a
manageable situation.
* * * + e
We believe that medical surveillance programs can pro-
vide protection for employees when combined with en-
vironmental control. They offer a valuable tool, in addi-
tion to dust measurement, for evaluation of the effect
of that environment upon employees. They can identify
the employee who has a problem and provide him with
medical assistance and in receiving benefits that are due
him. They can identify the employee who is an increased
risk, inform him, and help him to take necessary meas-
ures to protect himself. Medical surveillance programs
are not a substitute for diligent efforts to control dust.
61
EXHIBIT 6-66
An Epidemiological Study of Respiratory Symptoms
in Lancashire Mills, 1963-66
M.K.B. MOLYNEUX and J.B.L. TOMBLESON
Department of Occupational Mealth, University of
Manchester and H.M. Medical Inspectorate of Factories
Molyneux, M.K.B., and Tombleson, J.B.L. (1970). Brit.
J. industr. Med., 27, 225-234. An epidemiological study
of respiratory symptoms in Lancashire Mills, 1963-66.
An epidemiological study of card and blowroom workers
in 14 cotton spinning and two man-made fibre spinning
mills in Lancashire had been carried out on a prospective
basis of six-monthly examinations over three years. The
number of operatives to be included was decided so as to
give a sufficient sample for the statistical assessment of
fall in FEV, at the same time allowing for population
movement. The examination of each worker included a
history, a questionnaire of respiratory symptoms, and a
measurement of forced expiratory volume in one second.
The results in this paper, which will be followed by
others on other aspects of the surve » give the prevalence
of both byssinosis and bronchitis, according to the defini-
tion given, in the 1,359 cotton workers and 227 man-
made fibre workers, seen at least once, and also the dust
levels in the mills. Eight of the mills processed coarse
and six medium cotton.
The total prevalence of byssinosis, as defined in 26.9%,
being higher in coarse than in medium cotton mills, and
the occupational groups most affected are strippers and
grinders, carders and undercarders, and draw frame
tenters. In coarse mills symptoms develop in some men
and women within the first four years of exposure, and in
medium mills between five and ten years’ exposure. Re-
peat questionnaires in about half the population, two
62
years after the first questionnaire, showed the develop-
ment of symptoms of chest tightness in an appreciable
number not previously affected. The incidence of bron-
chitis is increased in operatives with symptoms of bys-
sinosis, but is influenced by age and smoking.
Total dust levels averaged 3.1 mg/m® in coarse mills
and 1.2 mg/m® in medium mills. The findings indicate
that dust control measures, though they have produced
considerable improvement, are not now fully effective
with present methods of production.
TABLE 3
BYSSINOSIS: PREVALENCE (%) IN POPULATION SEEN
All cotton Man-made fibre
Grats Tide Female. fetal Male Female Total
Yo 6.9 6.7 68 1.0 2.3 18
I 138 12.7 132 31 1.5 2.2
Il 7.6 6.2 68 0.0 0.8 0.4
Total 283 25.6 26.7 341 4.6 4.4
No. 566 793 1359 97 130 227
$ ol
63
(4c)
Finvuodyyg
ov iV
ee
a |
(2)
qr
$Z
801
(0)
(0)
0s
Lk 4
Lol
88
ts
0L
8¢
16
9L
ss
se
$6
os
-_—
. "ON
pris
0-91
8-re
0-52
6-0r
Z-0€
L$7
$0!
7-85
L-8y
0-0r
8-9f
8-€1
0:97.
seposd yy = fj aposn [pos aporn
Lt 00
0-0 0-0
oF 0-0
Le Le
0-0 0-0F
L9 $02
‘09 0-6
8-r1 8-1
| +6
0-0 Vet
92 6L
Cot l-ve
Vet v2
601 $v
6-L LZ
ze t+
09 ozt
a | 2s1P05D
-— — | wnipsy
a> asuro>
00 =| uinipayy
Le 255805
ais WNIpryy
asur0>
wnipay
>
asuro>
wnipay
21805
wnipay
2005
wripay
281805
wnIpasy
2007)
wnipay
‘981205
WIP r,
w~ ©
t+ of
=
oo mM
oe
—
AT AA
mn ae
bie
or
owe
= Oh ee ee - ee se
(%) aausporasg
WwW
SONNE oo OC
ose
oe
Se ee eee
coels (ajeus) ucwoisey
(ajrw) soancqr’y
** (ayeusay) souuids Sury
(ajeuss;) s3)u91 s2qwo>
(9jeusd§) 42;u9} Swesjpradg
(ojewsay) 32)U9) wey neg
ee (2;eway) J33u3} pied
(aut) J9puisd pue sodding
(ajew) Japsessopun ‘sapsey
uonvdnr»O
(2/twW) J9QWwey> UoO}}02 pur Wool no;g
STUW 3SHVOD GNV Wha NI SNOILVdNDIO O} NI (%
¢ JIGVL
) BINIIVAIAY SISONISSAG
64
EXHIBIT 6-73
Schilling, et al., “Cardiovascular Disease in Cotton Workers”
* * —_ o *
Sources of Unreliability in Mortality Rates
Many doctors certify multiple causes of death from
which the Registrar General must select one for com-
puting death rates for specific diseases. The practition-
ers often cannot verify their diseases either by necropsy
or special investigation. Thus both the Registrar Gen-
eral and the general practitioner introduce possible
sources of error into the mortality rates ascribed to
specific diseases. And as the habit of multiple certifica-
tion has become more common in recent years, the secular
trends in the mortality rates of certain diseases may
therefore be seriously misleading.
Multiple Certification—When the medical practitioner
gives more than one cause of death on the certificate the
Registrar General selects one according to certain rules.
In 1921-23 and 1930-32 any definite diseases of the heart
or kidneys was preferred to any disease of the respiratory
system when both were mentioned together. In 1939 the
Registrar General’s selection rules were revised. Prefer-
ence is now given to the disease which, as far as can be
ascertained was, in the opinion of the certifying medical
practitioner, the starting point of the sequence leading
up to the immediate cause of death. The effect of adjust-
ing bronchitis and cardiovascular death rates for 1921-
23 and 1930-32 in the light of this revision cannot be
shown accurately, but it is possible to get some idea of
the likely change from two different sources—the Regis-
trar General’s statistics for 1939 (the year of change),
when he classified all the death certificates according
to both new and old rules, and from a review of the death
certificates of card room workers, applied by the trade
unions. The latter is a source of information often avail-
able but seldom used.
69
The relevant conclusion of the Registrar General (1947)
as regards our particular problem is:
“Bronchitis and asthma death rates have been
affected considerably . . . by the increasing mention
of myocardial degeneration as a contributory cause
of death and consequent classification by the (old)
rules of selection, of increasing numbers to heart
disease.”
He gives conversion factors so that an estimate may
be made of the number of deaths before 1939 from
diseases, of the myocardium among the general popula-
tion which would be put back to respiratory deaths if
the new selection rules applied. For bronchitis in all
males there is added 0.237 times the number of deaths
which were previously ascribed to diseases of the myo-
cardium. But for groups with a high respiratory mor-
tality a higher proportionate transfer should be made.
There is, however, another method available for the very
necessary interpretation of the Registrar General’s rates
of cardiovascular and respiratory mortality before 1939.
But it must be emphasized that it is a highly tentative
procedure. In 1939, when deaths were given classified
under both rules, the bronchitis deaths under the new
classification were increased by transference from dis-
ease of the myocardium, endocarditis, and arteriosclerosis
from amounts varying from 30% at ages 25-34 to 114%
at ages 65-69. If these age-specific conversion factors
are applied at the appropriate ages, the death rates from
cardiovascular causes in 1930-32 are reduced by the fol-
lowing amounts for ages 25-69 (Tables C and E).
ESN Lae 8%
SE Ce 10%
Strippers and grinders ................ 24%
Ee 10%
66
We have assumed that practitioners combined bron-
chitis and heart diseases on their death certificates in
much the same proportions in 1930-32 as in 1939 and
that the practice was similar among practitioners in cot-
ton areas as elsewhere.* We have no means of verify-
ing these two proportions, but if our assumptions are
reasonable, cardiovascular death rates for strippers and
grinders would be reduced under the new system of
classification by more than twice as much as the similar
rates for all males and the other cotton groups.
Death certificates of 256 card and blow room workers,
or known ex-card and blow room workers, of all ages in
the Rochdale, Oldham, and Bury areas, obtained from
the trade unions, were examined. These covered a period
from 1941 to 1948. In Fig. 4 the effect of the change in
methods of classification is shown for the 25-69 age
group in which there were 164 deaths.
Of the 24 additional deaths from respiratory causes
under the new system of classification, 19 were trans-
ferred from diseases of the myocardium, three from
endocarditis, one from “other heart diseases’, the re-
maining one coming from chronic rheumatism. This
transference lowers the cardiovascular death rate for
ages 25-69 by
23
67
This compares with the 24% estimated by the other
method. +
x 100 = 34% **
* As multiple certification continued to increase between 1930
and 1940 we may have overestimated the reduction.
** In addition to transferring 23 deaths from cardiovascular to
respiratory causes, two other deaths were transferred to the cardio-
vascular group from other causes, giving a net loss of 21 deaths as
shown in Fig. 4.
+ For all ages, the transference lowered the cardiovascular death
rate by 30%.
67
Figure 4
se
«
DF
64 CARD AND BLow ROOM WORKERS
SSIFICATION OF CauUSe OF DEATH
!
—_——
I941 - 1948
Aces 25- 69
”? 9
- |
&
p | oc ae
2 e
Vv ] J
s| | &R:
° ° 36 zx
= -
° °
3 y
whe
7 3
£ 4
42 a $
‘io
ot be e{ 2] +
$ z ey
= - Fd
O.o SYSTEM New System
oF OF
CLASSIFICATION CLASS! FICATION
[ree 1935] [Posy 1935)
The Registrar General (1949) records that in the
1921-30 decennium 33% of all death certificates for
non-violent deaths gave “multiple” causes. By 1935 this
proportion in all death certificates had risen to 43%. It
is reasonable to infer that there was a substantial in-
crease in multiple certification between 1921-23 and
1930-32, which would mean that a higher proportion of
respiratory deaths would be allocated to cardiovascular
deaths in the latter than in the former triennium. This
may be a possible explanation of the opposite secular
68
trends in cardiovascular and respiratory mortality, and
in fact both these death rates may have been falling.
The North IV (Lancashire and Cheshire) region in
1930-32 showed a mortality excess from respiratory dis-
eases over all males at all ages. Some of the cardio
vascular mortality excess for this region can be similarly
explained. But, even after reclassification by the meth-
ods given above, death rates for heart diseases and
nephritis for males in North IV are still significantly
greater than rates for all males at ages 45 and over, but
spinners and strippers and grinders have higher rates
than North IV at ages over 55 (Table D). These high
cardiovascular death rates for North IV cannot be
explained.
Errors in Diagnosis.—It is not possible to be certain
about the effects on mortality rates of using vague terms
to describe causes of death, or of frank errors in diag-
nosis. The cardiovascular group of diseases includes
myocardial diseases and degeneration which are vague
labels used for diseases in which the exact pathology is
not known. The use of “heart failure” as a cause of
death is not favoured by the Registrar General, and for
these diseases of uncertain pathology in which cardiac
failure is the terminal result, the general practitioner
undoubtedly uses “myocarditis” and the like in order to
avoid enquiries about his certificates.
We have considered the reallocation to respiratory
deaths of certificates on which both cardiovascular and
respiratory causes are mentioned; but it is probable that
there are some death certificates with no mention of
respiratory disease, even though such a disease was the
starting point of the pathological changes leading to death.
And it is also probable that such an event occurs even
t It is of interest that in 1947-8 a similar excess for heart disease
is still present, in spite of the new methods of classification.
69
more often among cotton workers exposed to dust than
among all males or men of the same social class, for rea-
sons which will now be discussed.
Byssinosis has been repeatedly described as a disease
which closely simulates chronic bronchitis and emphy-
sema, but O’Sullivan found that among 33 men suffering
from disabling byssinosis 12 had no clinical evidence of
bronchitis. Shaw Dunn, and Sheehan (1932) did nec-
ropsies on 10 cotton mill workers, nine of whom had been
exposed to dust; while all of these nine had some evi-
dence of chronic bronchitis and emphysema, six showed
hypertrophy of the right ventricle, four of whom died
of congestive cardiac failure. Thus cotton workers with
disabling byssinoisis, with its characteristic asthmatic
symptoms, may present a clinical picture of cardiac
asthma or congestive cardiac failure, which so masks
the respiratory disease that it is not mentioned on the
death certificate.
For nephritis it is possible to make more definite
assumptions about mistakes in diagnosis.
Platt (1947a) suggests that most of the deaths as-
signed by the Registrar General to nephritis in the older
age groups are not renal deaths at all. In his own 161
cases for which full records were available, the main
incidence of deaths from nephritis was under the age
of 45. The Registrar General’s figures for age incidence
are totally different, with the main mortality over the
age of 45. Platt’s own experience is that many cases
of hypertensive cardiac failure with albuminuria are
diagnosed by their general practitioners as chronic in-
terstitial nephritis.
Nephritis death rates are particularly high in certain
occupations. Greenwood and Russell (1988) showed that
in the 1921-28 triennium the textile trades, pottery dip-
pers and file cutters, all of whom had high nephritis death
70
rates also had a high risk of dying from respiratory
disease.
For all occupations there is a significant correlation
between the standardized death rates for bronchitis and
nephritis in 1921-23 (Table 2). In 1980-32 the same
type of occupations appear to have high nephritis death
rates—blow room workers and foremen, strippers and
grinders, and cutlers—but for this period, as shown in
Table 2, there is a significant correlation between the
standard mortality ratios of bronchitis and nephritis only
for Social Classes III and IV. A possible explanation for
this relationship is that many of the men in these trades
died of congestive cardiac failure caused by respiratory
disease and were certified as nephritis deaths.*
Thus there is some evidence that medical practitioners
may fail to distinguish renal disease from other dis-
eases which in their terminal stages simulate renal
disease. It seems possible, therefore, that the mortality
excess from nephritis in cotton workers may be due to
deaths from respiratory disease, which terminates as
congestive cardiac failure. If, however, occupational in-
fluences cause hypertension, as has been suggested, there
will be another reason for high “nephritis” death rates
in cotton workers.
Conclusions
In order to get a clearer picture of any occupational
mortality excess due either to cardiovascular and respira-
*In the same paper Greenwood and Russell drew attention to
the high nephritis death rates in barmen and inn and hotel keepers
for whom an excess of alcohol is probable. In the triennia 1921-23
and 1930-32 there is a significant correlation between standardized
death rates for nephritis and cirrhosis of the liver. It is possible
that this correlation is due to some cases of cirrhosis of the liver
which terminally have oedema being misdiagnosed as nephritis.
Fishberg (1939) quoting his own experience and that of others is
very sceptical of the role of alcohol in producing renal disease.
_.
71
tory diseases for the triennium 1930-32, cardiovascular
and respiratory deaths are given in Fig. 5, added to-
gether, for the cotton trades, for all males, and for males
of similar social classes to the cotton workers. The occu-
pational hazard of the cotton trades, particularly of the
stripper and grinder, is obvious. We know that most of
this is respiratory and that in the past respiratory mor-
tality of cotton workers has almost certainly been under-
estimated; but when a correction is attempted for the
respiratory deaths allocated to the cardiovascular group
of diseases, there still seems to be an excess of deaths
for the cotton workers and particularly so for the strip-
pers and grinders. Thus there may be a real excess of
heart disease due to occupational influences. The latter
possibility is so important in view of the very high na-
tional death rate from cardiovascular diseases and the
absence of information on its relation to environmental
factors that this clue must be followed up. As Morris
(1951) said, cardiovascular disease of middle and old
age is one of the darkest areas of vital statistics and at
present the main source of information must be the
mortality rates. In another paper the results of a clinical
investigation of more than 300 men employed in cotton
textile mills will be described.
TABLE 2
CORRELATION BETWEEN NEPHRITIS AND BRONCHITIS
(AGEs 20-64 YEaRs)
Nephritis and Bronchitis rT No. of Groups P
1921-23* 38+ .04 164 P<.001
1930-32+ .22+.12 71 Not significant
1930-32 .44+.13 59 P<.01
(Social Class III-V)
*From Registrar General’s Decennial Supplement, based on
C.M.F.’s of all trades and referring to chronic nephritis only.
+ Based on S.M.R.’s of trades with 10 or more deaths, and refer-
ring to acute and chronic nephritis.
72
Figure 5
DeatH RATES IN 1930-32
2
A
DEATHS FROM NEMIRITIS, CARDIOVASCULAR
B
+60 CARDIOVASCULAR DEATHS LESS AN CSTIMATED
®ND RESPIRATORY OISEASES NUMBCK TRANSFERRED TO RESPIRATORY CAUSES
AS GIVEN IN THE R.G.3 DECENNIAL SUPT oT THROUGH CHANGED METHODS CF CLASSIFICATION
+T 5°
Act MAMLED x=
Sem. cusss Or FL
Soca, Cosss OF SS
Weavaes =. + 40
SPimm ees =.
Semrrcas £ Cunoeas FQ
sa 2
=
5 +20
=
=
= +10
a =
no” => bh.
7 = = oo,," v
FN Fe : . 2
NSS = LL EYN TS &
45- 65-69 pe 45
Figures above columns for coulion workers arc number of deaths upon which rates ure bused. (See Appendix, Tables C and E.)
Summary
Since 1891 the Registrar General has recorded high
death rates from cardiovascular diseases among cotton
workers.
It has been suggested that this mortality excess may
be due to hyperpiesis caused either by the inhalation of
textile dusts or by inbreeding in Lancashire cottun towns.
In this paper the Registrar General’s figures are ana-
lysed principally for the triennium 1930-32 (the last to be
published). It is shown that in the three occupational
groups studied, strippers and grinders, weavers and
spinners, there is an association between mortality and
the extent of dust exposure.
The significantly high death rate of the strippers and
grinders (the occupation with the highest dust exposure)
from cerebral vascular lesions supports the hypothesis
73
that hyperpiesis may be the cause of this mortality
eXcess.
Two sources of unreliability in mortality rates are
discussed. (1) Since 1920 medical practitioners have
entered multiple causes of death with increasing fre-
quency on death certificates. The Registrar General makes
rules for selecting one cause of death from certificates
with multiple causes. Before 1939 these rules exaggerated
cardiovascular mortality and reduced respiratory mor-
tality. Since 1939 more reliable methods of selection have
been used, and some of the cardiovascular mortality of
cotton workers who it is known have high respiratory
death rates from byssinosis can be explained simply by
methods of book-keeping. (2) The use of convenient
though vague terms such as “myocarditis” to describe
causes of death, and errors in diagnosis, particularly of
nephritis in the older age groups, may also exaggerate
cardiovascular death rates.
When all possible corrections are made for the exag-
geration of cardiovascular mortality through the errone-
ous allocation of respiratory deaths to it, there still re
mains an excess of cardiovascular mortality particularly
for strippers and grinders. This may well be real and
may be due to occupational influences.
We should like to thank Dr. J. N. Morris of the Social Medicine
Research Unit of the Medical Research Council, Dr. W. P. D. Logan
of the General Register Office, Mr. G. Hart of Rugby Mill, Hollin-
wood, and our colleagues inside and outside our Department for
advice and criticism; Mr. A. Robertson, Mr. H. Chorlton and Mr.
J. Meadowcroft of the Card and Blowing Room Operatives and
Ring Spinners Association.
74
EXHIBIT 6-76
Research Triangle Institute, Technological Feasibility
Assessment and Final Inflationary Impact Statement
‘Table I-6. Total Installed and Annualized
Compliance Costs: -all Sectors
(Costs in Millions of Dollars)
Industry Exposure .|Install-| Annualized [Direct Energy —
Sector Limit ed Cost Capital Operating |Cost {| Annualized
(mg/m) Charge Cost | cost
Yarn Production | 0.5 211.8 33.8 6.9 20.6 61.5
0.2 984.4 158.3 15.6 67.9 241.6
0.1 2,802.7 450.6 ra 147.0 620.6
Cotton Ginning 0.5 16.9 2.7 4.9 1.8 9.4
0.2 292.2 47.0 13.2 30.3 90.4
| 0.1 343.6 55.2 14.7 35.4 105.3
Cotton Weaving 0.5 9.1 1.5 1.7 0.6 3.7
0.2 1,387.9 223.1 40.7 86.3 350.1
0.1 3,939.1 633.2 112.6 245.1 990.9
Waste Processing | 0.5 17.3 2.8 2.6 2.5 7.9
0.2 32.0 5.2 2.9 4.7 12.8
0.1 56.1 9.0 3.4 8.2 20.6
Total 0.5 255.1 40.8 16.1 25.5 82.5
0.2 2,696.5 433.6 72.4 189.2 694.9
0.1 7,141.5~- {1,148.0 154.0 435.7 1,737.4
* * +. * a
III. BENEFITS OF THE PROPOSED STANDARD
1. Introduction
This chapter identifies occupational groups exposed to
cotton dust, discusses the inherent problems of benefit
estimation, and estimates the benefits expected to result
from implementation of the proposed standards. Yarn
preparation workers and other cotton industry workers
are discussed separately herein.
2. Occupational Groups Exposed to Cotton Dust
Two large groups of workers known to be exposed to
cotton dust are considered in this evaluation. They are:
75
1) yarn preparation workers, and 2) workers engaged
in weaving, ginning, cottonseed oil milling, and industries
using cotton waste.
a. Yarn Preparation Workers
In August 1971, 338,310 production workers were
employed in cotton and man-made fiber mills* in the
United States according to a Bureau of Labor Statistics
(BLS) survey [1]. Cotton mills employed 193,014 pro-
duction workers, while man-made fiber mills employed
145,296 such workers. Of the total production workers,
an estimated 22,609 were employed in bleaching, cloth
dyeing and finishing, and fabricating departments; hence,
they were not considered part of the regular textile
operations work force. The regular textile operations
work force then becomes, by subtraction, 315,701.
At the time of the BLS survey, slightly less than one
half of the industry’s production workers were employed
in mills manufacturing all or most of their products
entirely of cotton fibers and approximately one-fifth were
in mills manufacturing products of cotton and man-made
fibers (blends). The balance of the production workers
(thirty percent of the production work force) were em-
ployed in mills manufacturing all or most of their prod-
ucts entirely of man-made fiber.
The BLS survey estimated the number of workers in
cotton mills and all textile mills by selected production
occupations. The estimates for cotton mill yarn prepara-
tion workers were excerpted without change from the
BLS report and are shown in Table III-1. The numbers
of yarn preparation workers in blend plants, also shown
in Table III-1, were obtained by multiplying the BLS
estimates for all textile mills by 0.2 (the fraction of
* Includes SIC 2211, Weaving Mills, Cotton; SIC 2221, Weaving
Mills, Manmade Fiber and Silk; SIC 2281, Yarn Mills, Except Wool;
SIC 2282, Throwing and Winding Mills; and SIC 2284, Thread Mills.
76
workers in mills manufacturing products of cotton and
man-made fibers). Maintenance and miscellaneous worker
estimates for both cotton and blend mills were prorated
from figures for all textile production employees. The me-
dian dust concentrations assigned to the severa] produc-
tion occupations listed in Table III-1 are from Merchant,
et al. [2], and are measured dust levels from six cotton
mills and eleven blend m*'ls. The dust samples were of
approximately six hours duration and were collected with
the Lumsden-Lynch vertical elutriator.
Inspection of Table III-1 reveals that 105,193 and
21,230 yarn preparation workers are exposed to cotton
dust in cotton and blend mills, respectively, and are at
risk of developing byssinosis.*
* * * * s
A commonly used measure of the benefit of health
programs is the value of future earnings lost due to
premature death. Since byssinosis rarely appears as a
cause of death on a death certificate, and further, since
the relationships are not well defined between cotton dust
exposure, byssinosis, and chronic bronchitis-emphysema
deaths which are more commonly recorded, lost earnings
due to premature death could not be calculated. While
excess mortality ** due to bronchitis and pneumonia has
been observed in English workers exposed to cotton dust
[8], it has not been seen in the limited number of studies
of mortality in the U.S. textile industry. Thus, Enterline
and Kendrick [9] studied 6,281 white male cotton textile
workers aged 15 to 64 as a control population in an
examination of mortality in asbestos workers. Using the
U.S. white male population as the standard, a Standard-
ized Mortality Ratio (SMR) ([Observed deaths/Expected
* Office and yarn dyeing workers are not included.
** More deaths than would be expected in a given population dur-
ing a given period.
° e * “OE 2- O22 “dd (C26 44?}4)
£ ON “SL “LOA **pay *dnary t *.S434404 PLEIXOL VOIIOD Uy Saipnys Jsuodsay asog, * ue “32 “Queyrsay *y'¢
eG {921119 Su Bhd Wa49709_ “SN =" 2°0_“vorfuyysen * tog, us291 109
aes 126i 35 SOL} axel *AdAIng Obey Alisnpay *uogey $0 QUawIsedAQ “$"M zaDuN05 eIeZ
414 *pecuey pue voII0D
*“USsd4d Ysnp 4ayV0 pue ysnp 1103309 gO aungxiy
3
“SI_dwes 4oRessgNLa LertqIaA anoy FIS,
“$$9{ JO W03302 quarzad as Guy $sa2014)
a ————
Of2" tz - CE SOt Widl
t9l"0 BEL*2 seb"o rl6°9 SNOIUEL[TISjm Pur arueudzUy ey;
610 = o1"0 t10"s 62°0 = 02°0 602" ze 2905
Sit'9 O82 ce usek *surrein
0S> --- (42413 Aga | S29zSIMId]
F 80t"t 126‘! aura zGuya *sacpusy s0asing
Guy sing ep *furpuy: "Gut poogs
60°0 - 00°09 hv ae 4 62°0 - 02°0 Cel‘os Lear gKs
zi ses Bupa wey, 49yI0 ‘awesy *suduMeds
SiC*p 210°62 outsy Guys *suauuids
¢28 itg*2 ($49x}4) UAW yory>I=>>
00s‘2 960°8t mats Bujyvutds *sagy jig
Gurusyds
lti‘2 col*2t L&203qNs
6t°0 - OL’O veo" 69°0 - 0b°0 659° Wwesp Buoy ‘ssapuay saqqnis
6I°C - Ot’o 6SI 60 - Ob'0 ee Psepueys *sagpud, sagqnis
6€°0 - ot'0 £06 62°0 ~ 02°0 v08'2 S4APUI Dwiesy-Guimesg
6€°O0 - Of’O e9t 62°C - 02°0 8S9 S4apuay 49qu99
69°0 - 09°0 80l 62° - O2°t 2is‘t (S4aysjuty) ssapuoy paey
69°0 - 09°0 zt GL*t - OL t2t S4addisys psep
69°0 - 09°0 ttt 62" > 22° C2" S4apuisG pied
Buimesp puc Guipsez
. 2202QN5
ls y90°2 t
oun % 5 rrr S4dPUI} Jd4IIg
6$"0-05°0 ze §9°1-09"t a t $aapua} sauadg
6£°0-0£°0 lez 6S 1-0S'I 6us4240 pur buyvodg
2q! 2/64) U¥0}3e43U22U0) $49440% 30 sdquny ( t/fan) yop .04 420009 S4PYAO}|] JO saquny
33g uP | PeW SS aseq_weypoi NOLIVILSISSY1? OC ONY ANZHLHVE3C
eo IH CN319 STII kOLLOD ‘
. {vezz pue ‘2922 *iezz ‘1222 “1122 S.I1S) vosiersjissei gor xq
95NQ ¥O}}0) 03 aunsodx3 30 293699 pue pasodx suaxs0y UOl{esedaag UIA JO Aaquny PIPES, “Le}TT seqry
A
Tale T11-2.
Other Workers Exposed
to Cotton Oust ahd Assuncd [zposures
Assumed Esposure
Nuaber e }
sic Title a Workers Cpoeration Oust Concentration (wg/a°) Method Country Reforerce
0724 — Cotton ginning 40,436" gin stand 1.05 (medtan) vertical U.S.A. r)
elutrlator
baling press 0.48 do éo ¢o
other 0.59 do 6o eo
overall 0.67 do do 6o
2074 = Cottonseed of mills soo® seed house 73-$90 total dust filter thimble Egypt 4
linter 5-18 de Greenburg- Australts s
Saith impinger
baler 20-37. de do
221 Veaving mills, cotton slashing and warping 1.0-1.03 feedtan) vertical U.S.A, 2
- es elutriator
4 weave rous 1.0-1.09 do do do
83,216 cloth roca 1.0-1.09 do do do
pa incenance 1.0-1.09 do éo éo
2221 Weaving mills, sansude slashing and werpthg 1.0-1.09 do éo an 2
fiver and silk weave roca 0.59 do do Je
cloth room 0.59 do co eo
maintenance 0.59 do do 60
224) Marrow fabric mills 23,500 weave room, blend afl! 0.89 do vertical U.S.A, 2
: elutristor
weave rous, cotton af}! 1.0-1.09 do ¢o vo
2293 Paddings and upholstery 3,500 garnett roos 6.06 (mean) éo ¢o +
fillings
2294 Procesied teatlle waste 3,100 sorting 6.5) (respirable horizontal Toyrt 7
‘ éust) elutrisetor
fiber loosening 6.84 do do 6o
carding . $.02 ¢o do do
splaning 1.12 eo vo
felt processing 3.20 de 6e 60
2296 = Tire cord and fabric 2,214¢ weave roca 0.59-1.09 (medtan) vertical U.S.A, 2
: elutristor
2658 Mattresses and bed- 23,200 gernett roca 6.06 (mean) o U.S.A. ]
springs tepe edge 0.36 do éo 40
quilt 0.28 do .& do
boa spring 0.43 @& 6o 6o
hand tie and 0.24 do do oO
coll tack
everall $.02 do do bo
&@ Odtelned from 1972 Census of Manufactures unless nutet otherwise.
bd If a ain crew ts 12.5 workers (Mattonal Cotton Courcil), then tn 1972 there would have been 12.5 workers/gin X 3.517 active gins
* 43,953 workers. Wowever, 8 percent (3,517) of these are assueed te be office workers, Wence, the workers exposed to.cotton .
Gust fs estimated as 40,446, ;
¢ Twenty percent of production exployees (4,500) Is assume! based on conversations with {ndustrial representative. ¢
e@ Estrured.
ot 1977, Part § Cottua
79
deaths] X 100) of 59.4 was calculated for respiratory
disease, indicating that respiratory disease mortality in
white male cotton textile workers is less than that of the
U.S. white male population. This finding was confirmed
and extended in a recent analysis [10]. The results of a
study of mortality in female cotton textile workers re-
ported by Daum, et. al. [11] are suggestive of increased
emphysema deaths. Given the large number of female
workers in the cotton textile industry (about 55 percent
of the work force), it is especially unfortunate that hard
mortality data are not available. SMR’s for chronic
bronchitis and emphysema in both male and female cot-
ton textile workers would be especially appropriate for
use in this analysis; unfortunately they are not avail-
able at this time.
* o * * *
4, Engineering Controls
Engineering controls are defined herein as substitution
of a less toxic substance for a more toxic one, modifica-
tion or isolation of a process, or local or general exhaust
ventilation for the purpose of reducing or eliminating a
worker’s exposure to a potentially harmful substance.
Thus, dust concentrations in the textile industry may be
controlled by modification of the dust-producing processes
or operations, or removal of the dust from the environ-
ment by ventilation.
a. Yarn Production
1) Process Modification
The yarn production process is essentially one of fiber
cleaning and alignment with the desired goal of remov-
ing all material in the cotton except mature fibers, which
are then spun into yarn. The dust associated with cot-
ton processing is thus an unwanted residue which is
inherent in the operation. Textile processes have changed
80
little over the past several decades although there have
been modifications of some operations, primarily to in-
crease productivity; hence, process modification, other
than improved maintenance, has had little effect on dust
concentrations. The design of high speed cards did re-
quire substantially improved dust capture to avoid pro-
duction problems arising from the larger volume and
greater agitation of materials being processed. On the
other hand, open end spinning, which eliminates roving,
may produce more dust than ring spinning.
Decreasing machine density in a work area, which is
mentioned in the literature as a means of reducing ex-
posure, does not, in fact, reduce the concentration of
dust in the room air. Area dust levels would build up
more slowly in the larger volume of room air, but would
eventually reach the same equilibrium levels if. other
controls were unchanged. This technique is, therefore,
felt to offer little to the solution of the problem of dust
control in the textile industry. Replacement of older,
low-speed production equipment with more modern, high-
speed machines can: reduce the generation of dust. How-
ever, very fine yarns cannot be produced by the higher
speed equipment, so replacement is not a universal solu-
tion. Only in instances where large amounts of lower
quality yarn are desirable would replacement be ap-
plicable. Presently, process modification appears to be
a feasible but limited alternative for dust control [2].
2) Ventilation
The capture and removal of dust from the air at the
point of generation (i.e., local exhaust ventilation) ap-
pears to be the most efficient method of dust control at
the present time. Effective dust removal in cotton process-
ing areas depends on two factors: capture devices located
as close as possible to the point of dust generation and
efficient filtration methods [2]. Considering only the tech-
nological feasibility of efficient capture devices and filters,
81
the scope of ventilation control at the dust source could
range from complete enclosure of each machine (including
total enclosure, ducting, and filtration) to application of
local exhaust ventilation hoods and ducts and adequate
collection devices.
Most older textile equipment has very few enclosed
dust generating points. When local exhaust ventilation
systems are included, they are usually retrofitted devices.
The efficiency of these devices can be improved signifi-
cantly by providing as much enclosure of the machine as
practicable. By confining the contaminant to a smaller
volume, air flow requirements, and hence, energy costs
are reduced. Since the air volume required for the major-
ity of the dust capture devices is directly proportional
to the capture distance, a reduction of one-half of the
capture distance results in a corresponding decrease of
one-half of the air volume required, and hence, corre-
sponding reductions in energy requirements. Many of the
more modern machines are well enclosed and are equipped
with dust capture devices as a standard feature.
There is little variation in the physical configuration
of the capture devices required to attain the levels of
control ranging from 1.0 to 0.2 mg/m*. Vendors of
exhaust ventilation systems currently prefer to add dust
capture pickups, filtration devices, and/or increase air-
flows to reach desired room concentrations.
The level of dust control effected is generally related
to the number and design of local exhaust hoods, air
volumes, and the number of filter Stages through which
the contaminated air passes. By employing a sufficient
number of well-designed dust capture devices (e.g., hoods,
slots, and plenums), sufficient air volumes, and efficient
single-stage filtration, dust levels in the range of 1.0
mg/m* (total dust) can generally be expected in the
working environment; with two-stage filtration (and air
washing) , levels not exceeding 0.5 mg/m® (V.E.) may be
82
expected; if efficient third-stage filtration is added (and
air washing), workroom dust concentrations should not
exceed 0.2 mg/m* (V.E.) fcr an eight hour time-weighted
average (TWA) for yarn-producing mills. Controls to
comply with a 0.1 mg/m?’ (V.E.) level of control are cur-
rently technologically feasible, but may require complete
equipment enclosure and stringent three-stage filtering of
both recirculated and make-up air in the ventilation sys-
tem since dust in the ambient make-up air may on occa-
sion exceed 0.1 mg/m* (V.E.).
Control of cotton dust concentrations in the yarn manu-
facturing industry to levels as low as 0.1 mg/m* (V.E.)
thus appears to be technologically feasible by application
of existing technology. This assessment is based on
RTI’s understanding of the processes and operations in-
volved, the amount and velocity of particle release from
various dust generation points, and the general capabili-
ties of current engineering controls. It appears that con-
trol of dust levels to meet the 0.1 mg/m exposure limit
will require: (a) complete enclosure of openers and
pickers, cards, draw frames, combs, roving frames and
spinning frames with air exhaust; (b) filtration of the
exhaust air from the enclosures to levels approximating
0.1 mg/m*; (c) recirculation of the filtered exhaust air
through the central air conditioning system, where addi-
tional cleaning takes place; and (d) filtration of air con-
ditioning make-up air to a level not exceeding 0.1 mg/m’.
Such a system is conceptually feasible although no em-
pirical verification can be provided since it apparently has
not been tried in any installation. The configuration of
the enclosures may be expected to provide the greatest
design problems, since it will be necessary to provide
numerous access ports and a means to check for operating
malfunctions either by direct inspection or by automatic
indicators.
83
2) Yarn Processing
Yarn processing is a term used to cover those opera-
tions through which yarn may pass after leaving the
spinning frame. These include winding, spooling, twist-
ing, warping, and weaving, all of which are described
in Appendix A.
Engineering controls are available for four of the five
processes cited above; the exception is weaving. Based
on information obtained from vendors, local exhaust ven-
tilation systems could be used to maintain median dust
concentrations of 0.5 and 0.2 mg/m® in areas housing
these processes. RTI estimates that the 0.1 mg/m* limit
could be achieved by adding filtration of the make-up air
for the air conditioning system to the local exhaust ven-
tilation systems.
Commercially available engineering controls applicable
to weaving processes were not identified and it is, there-
fore, assumed that none exist. Some experimental work
has been done with the concept of using general room
ventilation as a dust control measure in weave rooms.
This basically consists of modifying existing ventilation
systems to produce air currents of such speed and direc-
tion as to capture a portion of the dust emitted from
the weaving process and then liltering the dust from the
air. No information on cost or effectiveness of this con-
trol technique is available.
Although the residue of particles of bract, stem, and
leaf (the material suspected of being the causal agent in
byssinosis) in weaving yarns is relatively very small,
there is substantial evidence that enough exists in the dust
found in weave rooms to pose a significant hazard for
workers exposed over long periods. The bulk of weave
room dust, however, apparently is sizing used to treat
the yarn and the total airborne dust normally will exceed
the 0.1, 0.2, and 0.5 mg/m* exposure limits considered
84
in this analysis, although cotton dust may be only a small
portion of the total.
No documentation of engineering controls for looms
was found in the course of this study. The discussion that
follows is, therefore, entirely speculative. Three kinds
of engineering controls may be suggested. Perhaps the
simplest and most obvious is to isolate the weave room
from other dusty areas. It was observed that in inte-
grated mills the doors connecting the weave room with
yarn production areas may stand open during production,
with substantial dust drifting into the weave room.
Elimination of such a dust source is easy and poses no
cost penalty. Similarly, room ventilation systems should
be separated so that dust is not introduced through recir-
culation from dusty areas.
The room ventilation system may be the primary
means of lowering dust concentrations in the weave room.
A very high air flow through the room, combined with
high efficiency filtration, would appear to be appropriate.
The use of floor vents drawing air at high rates from
under looms may also be used. Another engineering con-
trol technologically feasible for control of most looms,
apparently is local exhaust ventilation. It is probabie
that hoods can be installed above most looms and that
quite high rates of air flow can be employed to capture
dust released from the loom or blown off by travelling
cleaners where they are employed. Without knowing the
range of possible specifications for air flows or distances
from hood to dust source (which would require specific
preliminary design analysis), it is not possible to estimate
the collection efficiencies that can be attained with local
exhaust ventilation. Air filtration is less difficult to con-
ceptualize, since high efficiency can be achieved with
multi-stage filter systems.
Good housekeeping procedures provide one other obvi-
ous control measure that can contribute significantly to
85
maintenance of low dust levels in the weave rooms. Care-
ful and thorough cleaning of the room weekly to remove
accumulated dust from equipment and from the room in
general tends to prevent the build-up of excessive dust
levels. This should be done at a time when production
is shut down.
The aggregate effect of a combined control program
such as discussed here is hard to estimate. RTI engineers
are of the opinion, however, that such a program may
reasonably be expected to achieve an exposure limit of
0.2 mg/m*® TWA. It is not clear whether a limit of 0.1
mg/m* can be attained in this way.
* * * * *
b. Reduction of Workers Exposure: Yarn Production
Mills
A wide range of engineering controls are technologically
feasible to control dust levels in yarn mills. The industry
naturally has chosen methods that are most cost effective.
The controls specified herein as most probable for com-
pliance with the proposed standards are those now in use,
modified to achieve the required efficiencies. Their feasi-
bility has been demonstrated and they are readily avail-
able from vendors. These controls reduce the dust con-
centration in the work area by capturing and removing
the dust as it is generated through the use of local
exhaust ventilation on production equipment. A typical
system consists of dust capture devices (such as hoods
or enclosures over dust emission sources) , ducts for trans-
porting the contaminant, and a filtration system for ¢iimi-
nating the dust from the airstream. These systems are
discussed in Chapter IV and Appendix A.
Basically, the physical configuration of local exhaust
ventilation systems is independent of the dust concentra-
tion considered. However, differences do arise in the
additional stages of filtration required to control at the
lower dust concentrations and in the dust capture devices
and air volumes. The plant areas in which the systems
86
are employed also affect the control strategy required
to comply with different levels of control. To achieve
dust concentrations not exceeding 1.0 mg/m’, it is as-
sumed that local exhaust ventilation systems are now
employed on opening, picking, carding and drawing opera-
tions and that a single stage of filtration is used. To
achieve dust concentra.ions not exceeding 0.5 mg/m!
(V.E.), RTI’s analysis indicates that, in addition to the
above operations, local exhaust ventilation will also be ap-
plied to combing, roving, and spinning operations and
that two stages of filtration will be employed. To achieve
dust concentrations not exceeding 0.2 mg/m® (V.E.),
these data indicate that ventilation systems will be ap-
plied to the same operations with a third stage of filtra-
tion added, consisting of either an electrostatic precipita-
tor or a high efficiency bag filter. Control to the 0.1
mg/m* limit requires enclosure of equipment, exhaust to
high efficiency filters, and filtering of air conditioner
make-up air.
In addition to engineering controls, a program to re-
duce worker exposure through improved specific work
practices designed for each work position will be insti-
tuted. Although no specific modifications in work prac-
tices are suggested, for costing purposes it was assumed
that a study of the work practices for a total of 21 work
positions will be instituted by firms in the industry to
identify possible improvements and workers will be given
appropriate training based on the findings of the study.
The type of work practice changes that may be instituted
are discussed in Chapter IV.
During cleaning operations and certain other opera-
tions, such as waste collection and bale opening, which
require manual handling of cotton fiber, the described
ventilation controls are not adequate to maintain the
exposure of the individuals involved in these operations
at acceptable levels (see Chapter IV). Therefore, the
probable methods of compliance include the use of per-
87
sonal protective devices for these individuals. Specifically,
it is assumed that respirators are used by all individuals
involved in blow down, stripping, grinding, dry sweeping,
and manual trash collection. It is also assumed that
disposable respirators are the most probable choice of
the industry.
* x € * *
4. Compliance Costs
Costs associated with the implementation of controls
to attain proposed cotton dust standards fall into two
general categories:
® costs associated with the installation/operation of
engineering controls
© costs associated with ancillary programs, e.g.
medical surveillance, environmental monitoring,
respirators, recordkeeping, and training.
All of the cost estimates shown are calculated as the
incremental costs required for compliance with the pro-
posed cotton dust standards, at alternative dust exposure
levels. It is assumed that the existing regulation limiting
worker exposure to dust concentrations not in excess of
1.0 mg/m* (TWA) applies to yarn production and
processing industries and that for these industries there
is no incremental capital cost for engineering controls
to comply with an exposure level of 1.0 mg/m* (TWA).
At the time this study was performed it was evident
that not all firms were, in fact, in compliance with the
existing regulation, although the actual level of controls
operating in the cotton yarn industry could not be deter-
mined. In order to provide a reference point for the
incremental costs presented in this report, the costs of
going from zero dust control to compliance with the ex-
isting 1.0 mg/m* exposure standard was estimated at
$143.3 million, in 1974 dollars.’ Other costs associated
with additional requirements of the 1.0 mg/m* standard
scenario, such as medical, posting, personal protective
7%
88
equipment, employee information, work practices, and
monitoring and recordkeeping are addressed and are
shown in the appropriate sections.
There is some ambiguity regarding the application of
the existing regulation to the other industries for which
costs are shown. It is Department of Labor policy that
no regulation will apply to an agricultural operation un-
less it is so stated explicitly in the regulations. Since
cotton gins are not explicitly covered by the existing reg-
ulations, it is assumed that industry is not covered. It
appears that the cotton waste processing industries and
the cottonseed oil industry have not been required to
comply with the existing regulations although it might
be construed technically to apply to them. For this analy-
sis, they were not assumed to be in compliance with the
current standard.
Costs associated with compliance at the 0.5, 0.2 and
0.1 mg/m*® (TWA) (TWA) are shown as the increments
required for compliance at those exposure levels assum-
ing compliance with the current regulations, where
applicable.
a. Yarn Production
1) Engineering Controls
For engineering controls, capital costs include the pur-
chase price of the control systems, labor and materials
required for the installation of the control systems, and
costs associated with building and utilities modification
required for the installation of the control systems. Oper-
ation and maintenance costs include the labor and mate-
riais required for the day-to-day operation of the control
system, including repair parts and filter replacement.
Energy costs, expressed as kilowatt hours, are reported
separately and reflect the incremental increase in energy
consumption resulting from the operation of the control
system and the increased load on existing air conditioning
systems. Additional labor requirements (man-hours) to
operate and maintain the control system are also pre-
—
89
sented. These cost elements are integrated in estimating
annualized costs in the economic analysis of Chapter VI.
Engineering costs were initially computed on the basis
of cost per cubic foot per minute (CMF) of air required
to control dust generation from yarn production and
processing equipment. These were then converted to a per
machine basis, and subsequently to a cost per pound of
raw cotton processed through the equipment. Data used
in determining costs were obtained from control system
vendors, textile plants in which control systems are in
operation, and from data published by the R.. S. Means
Company pertaining to the costs of various types of con-
struction. Table V-1 contains the basic cost data by
equipment type for yarn mills. Details as to how these
data were developed are presented in Appendix B.
Equipment inventory data were obtained from US.
Bureau of Census publications. Equipment capacities and
throughputs were obtained from various literature
sources. A summary of these data is shown in Table
V-2. Details as to how the data in Table V-2 were
generated also appear in Appendix B.
In order to provide estimates of the cost. of engineer-
ing controls for yarn manufacturing processes to achieve
0.1 mg/m' limits, it was necessary to extrapolate from
the costs for less stringent controls. This procedure is
valid as a means of providing a first estimate of coat
because (a) the control techniques postulated to achieve
the 0.1 mg/m limit are essentially extensions of the
techniques used for lesser control, i.e., ventilation and air
filtration, and (b) the cost data for controls at 1.0, 0.5,
and 0.2 mg/m’, as well as RTI’s experience with similar
controls in other applications, indicates that there is a
reasonably stable functional relationship between control
cost and level of control. Mathematically, this calculation,
for any machine type, is as follows:
where: X = control cost at the 1.0 mg/m® limit,
Y = control cost at the 0.5 mg/m limit,
Z = control cost at the 0.2 mg/m* limit, and
U = control cost at the 0.1 mg/m? limit.
Solving for U gives the expression:
_ 56 (ZY)?
“@ YX
The incremental control cost at the 0.1 mg/m? limit,
therefore, is U - X.
3) Overall Industry Costs
Industry-wide costs, initial and recurring, are shown
in Table V-5. These costs are based on applying feasible
controls on all cotton/yarn equipment regardless of type
of fiber being processed. Exclusion of equipment process-
ing man-made fibers only could reduce these costs by as
much as 30 percent. Presumably, a mill processing syn-
thetic fibers exclusively will not be subject to regulation
under the proposed standard. It is not clear, on the other
hand, just how the proposed standard will be applied in
mills that process both cotton and synthetics. The equip-
ment used in both processes is essentially the same and is
convertible from cotton to synthetics and vice versa. The
conservative estimate was made, therefore, that controls
would be applied to all the production equipment in mills
processing cotton and cotton-synthetic blends, even if part
of their product is pure synthetic.
+Z.
91
Costs per pound of raw cotton are shown in Talie V-6;
manpower requirements in Table V-7. Where a range of
costs is presented, the lower figure reflects the cost of
controlling equipment on which cotton or cotton blends
is processed; the higher figure reflects the cost of con-
trolling all yarn processing equipment regardless of fiber
being processed.
’ . * * «e *
2) Weaving
a) Engineering Controls
Cost estimates for engineering controls for yarn proc-
essing equipment in weaving mills were obtained by
soliciting estimates from vendors of dust control equip-.
ment. Costs were estimated using the same analytical,
steps as described for the yarn production industry (See
p. V-10).
Table V-9 gives the ventilation requirements to control
the median dust concentrations to 0.5 and 0.2 mg/m!
(V.E.), and gives the cost per machine, the annual main-
tenance and operating costs, and the annual energy con-
sumption for dust control systems for the processes indi-
cated. As indicated in Table V-9, no controls are esti-
mated to be required in twisting, weaving and warping
to achieve a median dust concentration of 0.5 mg/m®
(V.E.).
* r * * *
Additionally, based on vendor data, it is estimated that
dust control systems for winding and spooling would be
required in only 20 percent of the installations to achieve
0.5 mg/m* (V.E.). These installations are those in which
the yarn being processed was manufactured from very
low quality cotton or which for some other reason have a
greater than average dust problem. To estimate the cost
per machine for dust control systems, the costs obtained
from vendor data were multiplied by the estimated frac-
tion of installations which would require dust control
*“spurpueys pasodetd [pe 4ol OOO'Gios Aue SzSS> ius
Jazjy “S[ouquod BuiruaautGua jo uopze;,e}3Ssul Guppuad ‘,os3u0d o4az Cutunsse ‘pojued eduetjduoz-aud Buranp £450) iy
“ILTPLLeA® PZED YsazelL ay ‘LOL QdUINSNOD 1107169 MeL F/6, UO P3seq BuP UMS $3569 491170
Liv “Lape ZeYR JO SZULBAZSUOD BYR YZIM YUazs;suod 8q OF [A AaZdeY UL LOpoM yndyng-ynduy Jy OJUL pauszus sua
$}]Ss03 assay] *(@Leq/q, OBP) SaLeq GzZ‘Z 40 YOLyZdwWNSUOD UOR}OD Me Z/GT UO paseg aie [ ] SJANXDeUG UL UMOYS S}509 3
‘passadoud uaqij Jo Ssajpuebau quauidinba swazshs u0z3209 {Le 4oj pauinbaa sjosjU07 R
*$3SOD UOLJeULUZIOpU; pue BuLULe4y au4Z-su0 pue *SzS0> quawdinbzZ [eqLde) Sapniou] fi
4snrvad bur peusdo Dae S_GAQWT? butasvutous
Ci°vdt] (8°28) [2°22] (2° tvez] [8666 ] [2°stz]
O'Esl L*bZ L'b2 8°SLSz 9°88 6061 /gierol
O'T o'l o'r - ~ fgtesboud 407 eALdsay)
“4 ’ - 10 10 10 Butzsog
v‘0 “v0 "0 S*? S‘t S*t HulsozpUuOW LeyUaWUOILAUG
ra | 2°L rae | et + 2% ra | BURL LLaAINS [PIL pay
(2°tSt] e°vet | [s°69] s"t9 [6°02] Sst - - - /g53800 A6s0N3
& 5°6 1 (2°9 ) (v°2 ) (s*8Zet) (e°96b ) (€°66 ) saures4 Buyuurds
BT (O° ) (0 ) (e*sst ) (o°00T ) (e*se ) Buy aoy
10 (T°0 ) J (et) i we} e:4 squio)
£°0 (p50 ) (to ) (6‘b2t ) (6"6€ ) (tz) sawesy Meg
‘7s (S*I ) (s*0 ) (2°602 ) (2°22 ) (12 ) spue)
vil (6°0 ) (v0 ) (9° tet ) (e'e9 ) (pat ) 6ujuea,) pue Buyuado
2°91 ‘ol Le l'orsz 8°68 2°S8l [zurudynby Loryues
git/Ou 1-0 git/Ou 2°0 gt/ou so gives to clu/bu 2°0 g@/m S°0 quawdinhz yo adty
(S42, 100 JO SUOL{LIW) $3809 Buysunday (sue_{og JO Suot Lith) Ai $3SO) Let ztuy 4
UOLZINPOAg UALA 2SZLW] Junsodxz a[qisspuuiay sAaLYDy 03
$380) Gulusnsay pue LeLzpUy api M-Aagsnpuy LeqUsUM10.UZ PazeU3sz “S-A aiqzyL
93
systems. Because of this, the cost figures appear quite
low. This method of presenting costs was chosen to permit
total industry costs to be obtained by multiplying the fig-
ures in the table by the total equipment in place. A simi-
lar discussion applies to the figures given for the 0.2
mg/m* exposure limit. In this case, based on vendor
supplied data, controls are estimated to be required in
80 percent of all installations. Installations not requir-
ing controls are those which process yarn made from very
high quality cotton. Consequently, the fraction which re-
quires dust control systems is much greater than for the
0.5 mg/m*® limit. Data on the number of units of each
piece of equipment now installed in the industry, shown
in Table V-10, were obtained from [3] and [4].
Costs for weaving looms are based on engineering con-
trols estimated ‘to be applicable to the operation. For cost-
ing purposes, these consist of local exhaust ventilation
systems over the area of the loom where the shuttle passes
back and forth. It is estimated that a hood for this
application would need to be aproximately two feet by
five feet, based on an |
* * * * a
estimated average loom width of 54 inches. It is further
estimated that an air velocity of 200 feet per minute at
the face of the hood would be needed to adequately cap-
ture the dust, which results in a requirement for 2,000
cfm per loom. The costs per cfm for application of hoods
in such systems are estimated to be about the same as in
the yarn production industry. The figure used for deter-
mining costs was $3.00 per cfm. Using this figure and an
air requirement of 2,000 cfm along with the estimated
209,792 looms [4] processing cotton or blends of polyester
and cotton, a total industry capital cost of $1,258.75 mil-
lion is derived. Annual operating and maintenance costs
are estimated at 3 percent of capital costs, or $37.76 mil-
lion. Energy consumption is estimated at 12.5 Kwh/
94
ig0‘'2S*t | €2°S6e*bl Z9°St2°* Ol] 98° BLS‘ S9K*6ES O8'byz"S} 95°292°2G §86S2°8S2°T nnad woee ooc- ---- swoo]
940° 061 81°09 €SO&e*L €16°601 826°99 162 692 0S° 890 202° 8€ none aes win ovee Suadsem
613°SZL 66°€19 09° LEZ‘ 862° bEL 806° 19 O2E * LEZ Sf Efe v9b' lb —-- ——- es sic SAITSIM]
Sy9*2bl 2b° L6¥ £5666 0/2001 o1s*os Lel'sdt 09° 1SE 90€°SE Lis‘9 eet “ie L4LS°S¥ 880 °9 $43, 00ds
S20‘ tt 2£°88 20°SSL v6L°L1 006‘E OLOLE 665° #S $92°9 609‘ 1 Ose *s £92°tt 029°t S4apuim
RNA oT (247,018) | (01S) (24 7sq-wew (ax 7yny gt} (447,018) (018) [[4%7say-wew]( aX 7 gt} (447,018) (018)
4oqe} Afisduj suo{jeuedg 1e3}de9 409e} A630} suo} 3euadg 1e34de9 40qe} ABs9uz | Suojjesadg| jez;de9 -
pue pue pue quad ynb3
ja2ueuaz ul PW J 2UeUS2 UY} PW a2ueuazU I ey
—-—---—— :
w/6u {° ' °
i cf/ t°o ct/Su 2°0 ct/bu S“0
(Gujaeam) s043u0) bursaaus6uz 404 $350) Aagsmpuy (e370, “[ I-A e191
95
cfm/yr., which is the approximate average of systems
in the yarn production industry at the 0.2 mg/m* (V.E.)
level. Cost estimates for the 0.1 mg/m® (V.E.) level
were obtained by assuming the increment over the costs
of 0.2 is equal to that determined for the yarn production
industry. The costs obtained for each level are given in
Table V-11.
+ * * * *
The Long-term Arrangement Regarding International
Trade in Cotton Textiles (LTA) provided the vehicle for
import quotas on cotton textiles and apparel and was of
major importance for a number of years. The sharp rise
of man-made fiber products in the late 1960’s led to ex-
panded trade agreements in 1971, and finally to the Multi-
Fiber Arrangement Regarding International Trade in
Textiles (MFA), which dealt with cotton, wool and man-
made fiber products. These arrangements have tended to
smother the effects of a small change in the relative prices
of domestic versus foreign textile products [See Section
10 of Appendix C.I].
* + * * *
4) Impact of OSHA Cost-Related Cotton Yarn Price
Increases
An important intermediate step in determining the
final impact is to determine the impact upon plant
profitability of the different exposure limits. This exami-
nation leads to other considerations such as industry-
wide price increases resulting from control expenditures
and the availability of labor force, energy and capital
for investment in control equipment. Estimations of
potential plant closures also resulted from this analysis.
This analysis relies upon the assumption that the com-
panies affected by the regulations will pass on fully their
increased costs via price increases and thus escape the
cost burden of the proposed OSHA regulation. However,
*) xppuaddy 40 2-9 A1qe, pue ey-7A aiqey UO paseq pazeino,e> sabsey> pue s3s07 :9UN0sS
*40399S JIS yoea uy sqududiys yo anjea Aq eb-IA a1qP]
Up UMOYS S3SOD BDUL} [dwod | e303 a4 BuLpLALp Aq paujezqo aue ayes yo 4ellop vad s3s0> aoue;,dwod abeuaay
*
2600" LT10° 8100° LSE0° E222" TO
2610° ¥S00° 2100° 9210° 18Z0° 2°0
600° 9100° . S000° £200° 8910" S‘0
8°9092T WILOL
8800° 2200° <600° £900° 26£0° 10
S£00° 0100° 2000° £200° 2bI0° 2°0
6000° £000° T000° $000° 2£00° s‘0
1°S89 9622
020° £00" 2000 ° 8SI0° 1860° T'0
6200° 2200° $000° £S0u° 82£0° 2°0
6100° 9000° 2000° 0100° £900° s‘0
v'Sbe 822
0v00° 8000° €1000° 0£00° 8810" To
S100° 000° 60000° 0100° £900° 2°0
000° 1000° 0000 ° 2000° 2100° S‘0
8°8082 L522
= S010° 8910° 9200° T1S0° I8te* TO
$120° 2200° 8100". 0810" 6III* 2°0
0200° £200° 8000° 6£00° 1¥20° S‘0
9°0S22 1822
6820° 6900° TI00° 0120° SOET* io
€1to° Z£00° 2000° ¢200° 6S%0° 2°0
6200° 0100° £000° 9100° 6600° S‘0
£°9S8E 1222
S22I° 162Z0° 900° 8880" S2SS° T°0
LLbv0° vETO° 0£00° z2t£0° £v6t* 2°0
220° Tv00° vI00° £900° 61v0° s‘0
9°0992 Tt2z
380) 380) 3809 abiey) $30) (01$) w / Bus 31S
pez}, enuuy Abuauz Buy zesadg © Lezidey pe, leqsuy Squoud ys re
1230) 220419 paz, | enuuy Aaysnpuy aunsodx3
#5805 JO 42((0g/s4e| 10g $0 ONLRA
sapes 40 Ae, (0g
4a $380) Bue} doy paz}lenuuy pue pai_ezsuy jeuoy3ippy “S-1A aiqey
97
they would face the contractions of demand for their
products due to price increases.
In order to estimate the potential cotton yarn price
increase, it is necessary to estimate the rates of return
on investment, both before and after new exposure limits,
for each of the six industry sectors. The level of return
on investment after implementation and the relative
change in this level from the pre-control level are the
bases upon which determinations of adverse economic
impact have been made. It is assumed that firms in these
industries will attempt to maintain the pre-compliance
rate of return on investment, and the average price level
in the six industries would increase sufficiently to main-
tain the pre-compliance rate of return on investment.
Two approaches other than the rate of return approach
used here may appear reasonable: (1) raising output
prices in the same proportion as costs increase (a simple
cost pass-through approach) or, (2) increasing output
prices after compliance to the level required to maintain
earnings per share of common stock ( earnings per share
approach). However, the rate of return approach is pre-
ferred over the simple cost pass-through approach be-
cause, as will be shown later, the latter under estimates
the compliance cost-related output price increases, As for
the earnings per share approach, data problems are often
prohibitive. For example, it could not be assumed that
firms in these industries would attempt to maintain
profits per share of common stock, since adequate data
were not available on privately owned small firms.
* ” * * +
Table VI-8 summarizes the estimated change in profit-
ability, measured in dollars per dollar of industry ship-
ments, as a result of incurring three different levels of
exposure limit compliance costs. The first row of Table
VI-8 shows the pre-implementation profitability levels,
which are assumed to be applicable to all six textile in-
98
"O/-1A pue G-IA S2UqeL Uy eJep UO paseq Jue SUOJZRINDI[e) :Sa2uN0S
6110°0 1£0°0 2020 °0 £96'°0 1s9 0 10
8b00°0 £€0°0 1€20°0 8560 929'0 2°0
£100°0 0v0'O Sbz0 0 SS6°0 S190 S‘0 9622
€820°0 610°0 Z€10°O 5/6°0 O12 ‘0 to
9010°O zt0'0 2020°0 296 °0 Sv9'0 Z2°0
$200 °0 6£0°0 0vz0°0 9560 8190 S‘0 b822
This text is long and has been trimmed here. Open the source document for the complete record.
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.