Appeal from the United States District Court for the Eastern District of Wisconsin. No. 84-C-472 -- Robert W. Warren, Judge.
Bauer, Cummings, Wood, Jr., Cudahy, Posner, Coffey, Flaum, Easterbrook, Ripple, Manion, and Kanne, Circuit Judges.
Since 1982 Johnson Controls, Inc. (hereinafter "Johnson Controls" or "Johnson") has maintained a fetal protection policy designed to prevent unborn children and their mothers from suffering the adverse effects of lead exposure. International Union, United Automobile, Aerospace and Agricultural Implement Workers of America, UAW (hereinafter "UAW"), several UAW local unions and a group of individual employees brought suit alleging that this policy violated Title VII, 42 U.S.C. § 2000e, et seq.*fn1 The district court granted summary judgment in favor of Johnson Controls and the plaintiffs appealed. This case was originally argued before a panel of this court and the panel's opinion was circulated among all the members of the court pursuant to Circuit Rule 40(f).*fn2 Prior to publication of the panel opinion, a majority of the members voted to hear the case before an en banc court and, following rehearing en banc, a majority of the court voted to affirm the decision of the district court.
The Battery Division of Johnson Controls, Inc., was created upon Johnson Controls' 1978 purchase of Globe Union, Inc. (hereinafter "Globe" or "Globe Union").*fn3 Globe Union was formed through the consolidation of two battery companies and had been in the battery business for almost fifty years before Johnson's purchase. Globe Union and Johnson Controls have maintained ongoing efforts to improve industrial safety through measures designed to minimize the risk lead poses to those directly involved in the manufacturing of batteries.*fn4
The steps that Globe Union and Johnson Controls have taken to regulate lead exposure have not been focused merely on complying with governmental safety regulations, but originate from their long-standing corporate concern for the danger lead poses to the health and welfare of their employees, their employees' families and the general public. During the period of the 1970's when OSHA's regulation of employee exposure to lead was virtually non-existent, Johnson Controls' predecessor, Globe Union, initiated a large number of innovative programs In an attempt to control and regulate industrial lead exposure. For example, in 1969, Dr. Charles Fishburn, M.D., who later became one of the primary proponents of Johnson Controls' fetal protection policy, instituted programs for monitoring employee blood lead levels.*fn5 In an attempt to manage lead exposure, other safety programs were initiated at Globe and Johnson including a lead hygiene program, respirator program, biological monitoring program, medical surveillance program and a program regulating the type, use and disposal of employee work clothing and footwear to minimize lead exposure. Globe Union also transferred employees out of high lead environments whenever a physician's medical evaluation report established that the individual had a high blood lead level. In the case of such transfers, medical removal benefits were provided to the employee before OSHA required such compensation.*fn6 Globe Union and Johnson Controls have continued to address their serious concern for industrial safety through efforts to design and regulate lead manufacturing areas to reduce employee lead exposure. For example, laminar flow pumps constantly supply a down draft of low velocity clean air to improve the environment of workstations where employees deal with lead. Central vacuum systems and powered floor scrubbers and sweepers are used to keep the manufacturing area as clear of lead dust as possible. Since Johnson Controls' purchase of Globe Union in 1978, it has spent approximately $15 million on environmental engineering controls at its battery division plants.
Globe Union, Johnson Controls' predecessor, established its first policy regarding fetal protection from lead exposure in 1977 as part of its comprehensive efforts to protect its employees from exposure to lead. Globe Union's announcement of the policy in a memorandum to battery plant and personnel managers stated:
"This change [the announced policy] has come about slowly as more and more medical opinion and evidence is persuasive of the risk to the unborn, developing child.
We have stopped short of excluding women capable of bearing children from lead exposure, but do feel strongly that those women who are working in lead exposure . . . and those women who wish to be considered for employment be advised that there is risk, that we recommend not working in lead if they are considering a family, and further that we ask them to sign a statement that they have been advised of this risk."
(Emphasis in original). In its 1977 "Statement of Risks," Globe also observed that at that time scientific and medical evidence had not as yet conclusively established the risk lead exposure posed to the unborn. However, after noting possible risks the 1977 policy statement read:
"We would have to say that it is, medically speaking, just good sense not to run that risk [lead exposure] if you want children and do not wish to expose the unborn child to risk, however small, and so recommend that you counsel with your family doctor and advise us of your wishes to transfer."
Johnson adopted its current fetal protection program in 1982 following its determination, based upon scientific research, that it was medically necessary to bar women from working in high lead exposure positions in the battery manufacturing division. The fetal protection policy applies to work environments in which any current employee has recorded a blood lead level exceeding 30 ug/dl during the preceding year or in which the work site has yielded an air sample during the past year containing a lead level in excess of 30 ug per cubic meter.*fn7 The policy recites that women with childbearing capacity will neither be hired for nor allowed to transfer into those jobs in which lead levels are defined as excessive.*fn8 A grandfather clause in Johnson's fetal protection policy permits fertile women who were assigned to high lead exposure positions at the time of the adoption of the policy to remain in those job assignments if they are able to maintain blood lead levels below 30 ug/dl.*fn9 Those employees who are removed from positions because of excessive lead levels are transferred to another job in Johnson's employ without suffering either a loss of pay or benefits.
The major reason Johnson adopted its current fetal protection policy was the inability of the previous voluntary policy to achieve the desired purpose: protecting pregnant women and their unborn children from dangerous blood lead levels. Between 1979 and 1983, at least six Johnson Controls employees in high lead exposure positions became pregnant while maintaining blood lead levels in excess of 30 micrograms. In addition, at least one of the babies born to this group of employees later recorded an elevated blood lead level. Moreover, Johnson Controls' medical consultant, Dr. Fishburn, testified as follows concerning a specific lead-related incident:
"Q: Now, let's talk in terms of prior to 1983, at any of the other Globe plants, do you have any knowledge or do you have an opinion that in any instances the exposure of the mother -- or the blood-lead level of the mother while she was pregnant had an effect on the fetus and, therefore, the child?
A: [Dr. Fishburn] Well, I'm not aware of any specific instances other than the one, which was in the early 80s, about 1981.
Q: Where was that? What plant?
Q: Where's the Owosso plant?
Q: In just simple terms, what was the nature of the problem?
A: The nature of the problem was hyperactivity and control of the child. And the child had elevated blood-leads and protoporphyrinis.
Q: In your medical judgment was the problem of the child affected in any way by the exposure of the mother during pregnancy?
A: In my opinion the history of the hyperactivity and the difficulty she was having with him could very well and probably was due to the lead that he had."*fn10
In announcing its new, more defined policy, Johnson Controls emphasized its continuing interest in the protection of employees and their families from occupational health hazards and was responding to the increased understanding of the risk of lead exposure that had developed in the five years since it established its former voluntary policy:
"We have over the years developed policies to protect the health of our employees whenever the nature of their duties may expose them to a health hazard and to prevent members of their families from being exposed to a health hazard because of their employment with Globe Battery Division.
Medical research has shown that a woman's exposure to lead can cause ill effects on the health of her unborn child because the lead absorbed in the mother's blood as the result of this exposure can cross the placenta and mix with the child's blood. This has been acknowledged by the Occupational Safety and Health Administration.
Medical research also shows that a risk to the unborn child's health is present at a much lower blood-lead level than an adult. The ill effects to the unborn child can occur during the early stage of pregnancy, before the mother is aware that she is pregnant, and can continue throughout the pregnancy.
Because of this, it is the Division's policy that women who are pregnant or who are capable of bearing children will not be placed into jobs involving lead exposure or which could expose them to lead through the exercise of job bidding, bumping, transfer or promotion rights.
This policy is intended to reduce or eliminate the possible unhealthy effects of lead on the unborn children of pregnant employees and applicants. It does not apply to those women who have medical confirmation that they cannot bear children. However, the policy is in no way intended to support or encourage women of childbearing capability to seek to change this status. Employees are strongly advised against any such action."
Prior to adopting its updated fetal protection policy, Johnson seriously considered alternatives to the exclusion of women with childbearing capacity from high lead exposure positions, but after research and consultation with medical and scientific experts found itself unable to structure and implement any alternatives which would adequately protect the unborn child from the risks associated with excessive lead exposure. Johnson's experience demonstrated that the voluntary exclusion program was ineffective. To date neither Johnson nor any other battery manufacturer has been able to produce a lead free battery, or to utilize engineering research and technology to implement a system or procedure capable of reducing the lead exposure of its employees to acceptable levels for fertile women. Limitation of the fetal protection policy to women actually pregnant was found ineffective because there is the very definite possibility that lead exposure will occur between conception and the time the woman discovers her pregnancy.*fn11 Such a limitation is further inadequate because reduction of blood lead levels following removal from a lead exposure area requires a significant length of time that frequently extends well into the pregnancy term. Limitation of the policy to women planning pregnancy also was not found to be a suitable alternative because of one of the exigencies of life, the frequency of unplanned or undetected pregnancies. Permitting fertile female employees to attempt to maintain a blood lead level below 30 ug/dl or utilizing the mean or median blood lead levels of current workers as a measure of whether a woman should be permitted in a position would also not effectively protect the unborn child. The reason these actions would be inadequate is that an employee's risk of high lead levels is usually greatest immediately after commencement of work in a high lead environment.*fn12
Dr. Fishburn, Johnson Controls' medical consultant, noted that Johnson and other corporations manufacturing batteries accept and routinely follow these medical policies:
"Q: Now, you testified as to what other industries had done in the past with respect to women working -- women capable of bearing children working in high lead areas. Are you familiar with other companies as to their practices?
A: Well, I'm familiar with a good many companies that I've worked with here in the area in foundry industry. I'm familiar with General Motors, Dow Chemical, Ford Motor Company, Owens-Corning. There's a large number of large and small companies that I'm familiar with, and where there are occupational standard programs within those companies I've exchanged the information with them.
Q: And what problems do they have in regard to this problem?
A: In regard to lead even the doctors -- as I said Dr. Kehoe or Dr. Bellmap here -- any doctor that has worked with lead, either in the mines as I did with Bellmap, or smelters, primary smelter we did not, and -- and I was never taught to place a reproductive female in the average work exposure of lead. And furthermore, before the 60s, to my knowledge, no women were working in lead exposures. It was Kehoe's opinion at the time that any doctor that would allow this to happen was committing malpractice
Q: Now this was--now my question was: Are you familiar with what other companies are doing now?
A: They are restricting women from lead exposure who can have children."
In altering its fetal protection policy to more effectively protect the unborn child and its mother, Johnson responded to the most recent medical evidence which established that lead exposure in utero presents a substantial health risk to the unborn child, as well as its female employees, and believed that Title VII would allow it to address this risk.
Proper analysis of the Title VII issues this case presents requires a thorough understanding of the following fundamental question: Does lead pose a health risk to the offspring of Johnson's female employees? In considering the evidence in the record on this subject it is important to note that both the UAW and Johnson Controls agree on appeal that a substantial health hazard to the unborn child in the womb has been established. The UAW admits in its brief that "it is clear that . . . substantial risk of harm to the fetus . . . has been established." UAW Brief at 33. Similarly, Johnson Controls states that "[the] evidence in the record on [substantial risk of harm to the fetus] is overwhelming." Johnson Controls Brief at 22.
The record very clearly establishes that once lead is deposited in a mother's blood, it crosses the placenta and affects her unborn child. Because the fetus' blood system is nourished by the mother, the unborn child possesses approximately the same blood lead level as the mother.*fn13 It is similarly undisputed that the unborn child "is medically judged to be at least as sensitive, and, indeed, is probably even more sensitive to lead than the young child." Affidavit of J. Julian Chisholm, M.D. (hereinafter Dr. Chisholm Aff.) at para. 6.*fn14 See Affidavit of Dr. Anthony R. Scialli, M.D. (hereinafter Dr. Scialli Aff.) at para. 5 ("[Because] of the extremely rapid development of the central nervous system during gestation, the fetus may be even more sensitive to toxic effects of lead than the young child");*fn15 Affidavit of M. Donald Whorton, M.D. (hereinafter Dr. Whorton Aff.), at para. 8 ("[Recent medical studies] suggest that even relatively low levels of lead exposure to the fetus can cause damage to the higher brain function resulting in decreased neuro-behavioral development in the child");*fn16 Deposition of Marvin S. Legator, Ph.D. (hereinafter Dr. Legator Dep.) at 49 (UAW witness) ("There is no question about the sensitivity of the fetus. . . . I don't think one could argue about the fact that lead affects the fetus");*fn17 Deposition of Ellen Silbergeld, Ph.D. (hereinafter Dr. Silbergeld Dep.) at 21-23 (UAW witness) (Noting that in the area of central nervous system impairment a young child is probably more sensitive than an adult and that the consequences for the young child and the fetus are similar).*fn18 The Centers for Disease Control summarized these basic facts in a document questioning the efficacy of current OSHA standards in protecting the unborn child and implying that an unborn child is adversely affected by lead levels lower than the 30 ug/dl reflected in Johnson Controls' fetal protection policy:
"In a pregnant woman, lead crosses the placenta and lead concentrations in umbilical cord blood are nearly equal to those in maternal blood. Since the growing brain of the fetus is likely to be at least as sensitive to the neurologic balance of lead as the brain of a young child, umbilical cord blood levels should be at least below 25 ug/dl. Therefore, the OSHA standard is probably not sufficiently strict to protect the fetus. Further study is needed to define acceptable lead levels among women of childbearing age.
"Ideally, engineering features should prevent workers from being exposed to lead dust and vapors. When workers are exposed, compliance with Occupational Safety and Health Administration (OSHA) regulations appears to be effective in preventing them from transporting lead home to children. . . . The prevention of lead exposure to the fetus needs special emphasis. Women of childbearing age should be excluded from working at jobs where significant lead exposure occurs."*fn19
The chief reason why an unborn child's lead exposure is of such great concern is that it has been medically established that lead attacks the fetus' central nervous system and retards cognitive development. See Dr. Chisholm Aff. at para. 6 ("[Harm] [to the fetus] includes . . . retarded cognitive development which may result in learning deficiencies and behavioral disorders"); Dr. Scialli Aff. at para. 7 ("The potential damage to the central nervous system of the fetus from lead exposure includes intellectual and motor retardation, behavioral abnormalities and deficiencies in learning abilities. It is my medical opinion that such damage may be permanent"), Dr. Whorton Aff. at para. 8 ("[Recent medical studies] suggest that even relatively low levels of lead exposure to the fetus can cause damage to the higher brain function resulting in decreased neurobehavioral development in the child"); Affidavit of Dr. Paul B. Hammond, Ph.D., Professor, Environmental Health, University of Cincinnati (hereinafter Dr. Hammond Aff.) at para. 3 ("[As] the director of a study that is currently being conducted in Cincinnati on the health effects of lead exposure on children . . . [we] analyze[ ] the subsequent mental development of children who have been exposed to lead both in utero and postnatally. The initial results of this ongoing study . . . establish that exposure of the fetus to maternal blood lead levels in excess of 12 micrograms per deciliter of whole blood creates a significant risk of low birth weight and a clear decrement in the subsequent mental development of the infant"); Dr. Silbergeld Dep. at 64 (Describing studies of child exposure to lead, in which she participated, which are "highly consistent with reports of hyperactivity, decreased attention span, learning problems which have been described in [lead exposed] children"), and at 49-50 (Reporting that studies of prenatal lead exposure in which she participated found similar effects to post-natal lead exposure. These effects are "substantial irreversible cellular and functional damage to the brain").*fn20
Unlike physical birth defects, such as those associated with thalidomide,*fn21 lead's sometimes subtle damaging effects may not fully manifest themselves until the child is diagnosed as having learning problems in a school setting some five to six years after birth:
"What we are worried about are very subtle things, the ability to really affect learning ability. And so far as impairing the child's progress, they really aren't evident until he gets into school. He discovers that he can't remember, that his brain cannot pay attention, what our psychologists here called deficits in auditory processing, which is a fancy way of saying they can't understand what they hear, can't process it, and use it effectively. And those things will impair a child perhaps toward the end of the first grade, particularly in the second grade."
Deposition of Dr. J. Julian Chisholm, Jr., M.D., Director, Lead Program, John F. Kennedy Institute and Associate Professor of Pediatrics, Johns Hopkins School of Medicine (hereinafter Chisholm Dep.), at 27.
Probably the worst aspect of lead's influence upon an unborn child's future intellectual development is that its effects have frequently been found to be irreversible.*fn22 Further, the most recent research suggests that the unborn child may be affected at lead levels previously believed safe. See J.M. Davis & D. Svendsgaard, Lead and Child Development, 329 Nature 297 (1987) (Collecting results of recent studies in this area).
Lead exposure can also pose other physical threats to the unborn child such as reduction of the infant's birth weight, premature delivery, and stillbirth. See Dr. Chisholm Aff. para. 6. Lead may also affect the other vital fetal organs including, but not limited to, the liver and kidneys.*fn23
The danger resulting from lead exposure cannot simply be avoided through removing a pregnant woman from lead exposure promptly after the discovery of pregnancy. Dr. Chisholm, a recognized expert in the research field of treatment and prevention of lead poisoning in young children, observed that "excluding only woman who are actually pregnant from work areas where there are elevated blood lead levels would not sufficiently protect the health and safety of the unborn child." Dr. Chisholm Aff. at para. 10. This is true because lead continues to exert an effect upon the mother and her unborn child for a significant period of time after she has been removed from lead exposure. Dr. Chisholm's uncontroverted affidavit explained:
"[Substantial] medical evidence . . . establishes that lead remains in the body for a significant period of time after removal from a high lead environment. Lead builds up not only in the blood and soft tissues, but is also stored in the bones. Following removal from the high lead environment, as the lead built up in the blood and soft tissues leaves the body, the lead in the bone begins to turnover, thus maintaining high blood lead levels even long after removal. As a general rule of thumb, it takes approximately two or three times as long for the blood leads to decrease as it did for such blood levels to increase. Therefore, if a woman is exposed to blood lead levels in excess of 25 or 30 micrograms for any length of time, such levels will not decrease sufficiently to avoid damage to the fetus, even if she is removed when the pregnancy is discovered."
Dr. Chisholm Aff. at para. 10. See also Whorton, supra note 16, at 8 ("Since lead is an accumulative toxicant which is stored in the bone, with a half-life in the body of 5 to 7 years, a woman with a significant blood lead burden would pose a potential hazard to any conceptus for many years after exposure"); Dr. Silbergeld Dep. at 27-28 ("For all purposes there is continuing exposure to lead even after removal from sources of lead . . . . If it were possible to take a person into a lead-free environment after an episode of exposure, the turnover of lead is probably on the order of 100 days. So within a year or so there would be a reduction in lead. There is also, of course, a rapid removal of lead from the blood into the bones [and] soft tissue that is no longer available for circulation. But considering that most people are not removed to lead-free environments, even if they go from areas of relatively high exposure to lower exposures, there is continuing cyclic . . . exposure to lead") (emphasis added). Furthermore, because lead is stored in the mother's bones, and "because the bones may decalcify during pregnancy in order to provide the fetus with calcium, there may be an additional danger of transfer of stored lead to the fetus." Dr. Chisholm Aff. at para. 10. See also Whorton, supra note 16, at 7-8 ("[Dr. W.] Manton [in a 1985 edition of the British Journal of Industrial Medicine] reported on a longitudinal study of blood lead levels in a woman before, during and after pregnancy. He reported a doubling of prepregnancy blood levels during pregnancy and suggest[ed] mobilization of lead from bone during pregnancy as the mechanism") (footnote omitted). These conclusions are consistent with research that OSHA relied upon and quoted in establishing its 1978 lead standard:
"The placenta also has considerable storage capacity, and during the first few months of pregnancy, it grows tremendously in size while the fetus remains relatively small. Calcium along with other substances is stored in the placenta to be used in the later months of pregnancy for growth by the fetus. It could be expected that lead would be similarly stored."
Occupational Health and Safety Administration, U.S. Department of Labor, Final Standard for Occupational Exposure to Lead: Attachments to Preamble, 43 Fed. Reg. 54,395 (1978) (quoting study of Dr. Vilma Hunt, Associate Professor of Environmental Health, Pennsylvania University, contained in OSHA's record).*fn24
The overwhelming evidence in this record establishes that an unborn child's exposure to lead creates a substantial health risk involving a danger of permanent harm. This evidence clearly approaches a "general consensus within the scientific community," and certainly "suffices to show that within that community there is [a] considerable body of opinion that significant risk exists to the unborn child from exposure to lead." Wright v. Olin Corp., 697 F.2d 1172, 1191 (4th Cir. 1982). Next we consider the proper legal standards to be applied when employees bring a Title VII sex discrimination action challenging an employer's response to this serious health risk.
Having considered both the nature of the risk of harm that lead exposure presents to the unborn child and the mother and the policies Johnson implemented in response to this problem, we now turn to the question of the proper legal analysis to be applied to Johnson's fetal protection program under Title VII. The question presented is should we follow the lead of the Fourth Circuit, the Eleventh Circuit and the EEOC in determining that these policies can be justified with a "business necessity" defense or must we conclude that these policies may only be justified with a bona fide occupational qualification defense.
In approaching this issue we are cognizant of the mandates the United States Supreme Court has recited on two occasions concerning the necessity of avoiding rigid application of proof patterns to particular factual situations. The Court's concern was first set forth in Furnco Construction Corp. v. Waters, 438 U.S. 567, 576, 57 L. Ed. 2d 957, 98 S. Ct. 2943 (1978), where the Court noted that the formula it had devised for demonstrating a prima facie case of disparate treatment in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S. Ct. 1817, 36 L. Ed. 2d 668, (1973) "was not intended to be an inflexible rule." The Court expanded on this same subject in International Brotherhood of Teamsters v. United States, 431 U.S. 324, 358, 52 L. Ed. 2d 396, 97 S. Ct. 1843 (1977), when it observed:
"Our decision in [McDonnell Douglas] did not purport to create an inflexible formulation. We expressly noted that '[the] facts necessarily will vary in Title VII cases, and the specification . . . of the prima facie proof required from [a plaintiff] is not necessarily applicable in every respect to differing factual situations.' [McDonnell Douglas, 411 U.S. at 802 n. 13]."
(Footnote omitted). The thrust of these repeated Supreme Court pronouncements is that courts are required to avoid inflexible application of judicially devised proof patterns in cases that present factual circumstances different from those encountered previously. See also Wright v. Olin Corp., 697 F.2d 1172, 1184 (4th Cir. 1982). Any proof scheme a federal court applies is useful only if it assists the court in properly identifying the employment practices Congress intended to prohibit under Title VII. These concerns are particularly important in a case of this nature where the interest in financial reward is balanced against a medically established risk of the birth of a medically or physically deprived baby and where the challenged distinction is based upon the reality that only the female of the human species is capable of childbearing.
Two other federal courts of appeals and the Equal Employment Opportunity Commission have addressed the question of the defenses available to an employer under Title VII in a case challenging a fetal protection program. The first court of appeals to address this question was the Fourth Circuit in Wright v. Olin Corp., 697 F.2d 1172 (4th Cir. 1982). That case involved a fetal protection program very similar to the one Johnson instituted, in that it forbade any fertile woman from working in a job which "'may require contact with and exposure to known or suspected abortifacient or teratogenic agents.'" Olin, 697 F.2d at 1182. In considering which of several possible theories of claim and defense should apply in a Title VII analysis of a fetal protection policy, the Fourth Circuit observed:
We must start by conceding that the fact situation [the fetal protection policy] presents does not fit with absolute precision into any of the developed theories. It differs in some respects -- either in its claim or defense elements -- from each of the paradigmatic fact situations with which the different theories have been centrally ...