Dr. Pamela Clay sued her former employer, Holy Cross Hospital, alleging that the Hospital terminated her because of her pregnancy in violation of Title VII. The Hospital moved for summary judgment. The district court granted the motion, concluding that Clay failed to demonstrate that the Hospital's legitimate, non-discriminatory reason for her termination was a pretext for discrimination (the Hospital administration asserted that Clay failed to demonstrate the dedication necessary to grow her practice into a profitable enterprise). Clay appeals.1 We affirm.

Holy Cross Hospital is located in Chicago, Illinois. In 1992, the Hospital embarked on a plan to create a Neighborhood Affiliate Network ("Network") of primary care physicians who would practice in offices in the community surrounding the Hospital. The Hospital's strategy was to subsidize the network of physicians with guaranteed salaries until their practices matured into profitable enterprises, thus eliminating the need for the Hospital's subsidy. A Practice Management Department managed the daily operations of the Network. Bill Seliga, the Hospital's Vice President of Practice Management, headed the Practice Management Department and reported directly to the Hospital's Chief Executive Officer, Mark Clement.

The Hospital hired its first physician for the Network in 1993. From the beginning of the Network's creation, Clement explained to the Network's physicians that they were expected to "hustle" to build a profitable practice because, at some point, they would be on their own.

In June 1996, the Hospital hired Dr. Pamela Clay to work as a pediatrician at its Ford City Neighborhood Affiliate. Clay signed a written employment contract entitling her to an initial salary of $100,000. The contract allowed either Clay or the Hospital to terminate the employment relationship upon 90 days' notice. The Ford City Affiliate was under the administrative control of Julie Rudolph, a clinical operations manager in the Hospital's Practice Management Department.

Clay became pregnant in late October 1997. (She did not disclose her pregnancy to the Hospital's administration until May 1998.) By early 1998, the Network failed to reach a break-even point and continued to need the Hospital's subsidy. Additionally, the Hospital was having financial difficulties of its own at that time. Clement concluded that the Hospital could no longer afford to subsidize the Network at then-present levels, in part due to reductions in Medicare payments caused by the Balanced Budget Act of 1997. In the first few months of 1998, the Hospital was losing approximately $300,000 per month on the Network.

Due to the Network's poor financial performance, on April 14, 1998, Clement met with Seliga to review the Network's subsidy and the fact that some physicians and their practices were not performing as well as expected. Clement asked Seliga to assess the practice of each physician in the Network to determine which physicians "were going to get the hospital where it needed to be," and which physicians should be "prune[d]" from the Network. Clement asked Seliga to assess each physician's practice, including subjective factors like how hard a physician was working, and whether the physician was participating in the Hospital's marketing efforts in order to build a profitable practice.

After his meeting with Clement, Seliga met with several managers in the Practice Management Department, including Julie Rudolph, and Theresa Gaffney, the Hospital's Marketing Director. At that meeting, Seliga told the practice managers that the Hospital needed to reduce the Network's subsidy, and that it would likely have to lay off some physicians. He solicited the managers' opinions on which physicians to terminate. The practice managers commented on each physician's willingness to participate in practice-building efforts, and on their likelihood to grow their practices. The group created a preliminary list of physicians they perceived to be less likely to grow their practices to profitability. That initial list included Clay and eight other physicians.

Between April 14 and April 20, 1998, Seliga reviewed Clay's numbers concerning her revenues, expenses, participation in community events, hours in the office, marketing activities, procedures provided to patients, and patient volume. Then on April 20, Seliga met again with Gaffney and the practice managers to discuss the list of proposed layoffs. According to Seliga, the group reached a consensus that each physician on the list was unlikely to grow their practice to profitability because they lacked the willingness to participate in marketing activities or to make the other efforts required to see, attract, and service more patients. The issue of pregnancy was never brought up in the decision-making process.

After some additional meetings between Seliga, Clement, and other Hospital administrators, the list of physician layoffs was finalized on May 1, 1998. The reduction-in-force ("RIF") was termed the "Neighborhood Affiliate Reorganization" and was expected to save the Hospital approximately $2 million. As the head of the Practice Management Department, Bill Seliga had the ultimate authority, subject to Mark Clement's review, to decide which physicians would be terminated under the RIF.

Seliga testified that he decided to terminate Clay because he concluded that she would not achieve a financially-self- sufficient practice regardless of the amount of assistance she received from the Hospital. According to Seliga, he based his decision on his own knowledge of Clay's performance and on the advice he received from other Hospital employees, including Julie Rudolph and Theresa Gaffney. Seliga testified that he was advised by Gaffney that Clay was uncooperative with the Hospital's marketing efforts, and that she failed to participate in any of the Hospital's 25 marketing events. Seliga also states that he was told that Clay had low patient accessibility. Patient accessibility means the amount of time that a physician is in her office and her receptivity to walk-in patients. It is undisputed that patient accessibility is "the key" to building a successful practice. According to Seliga, Gaffney and Rudolph told him that Clay was unwilling to see patients outside of her scheduled office time, and that she would turn patients away who were waiting in her office at the end of her schedule. Seliga also testified that Rudolph informed him that Clay was reluctant to take advantage of opportunities to grow her practice by covering for other physicians who were unavailable. And Seliga determined that Clay's practice had low growth and low revenue, and that she had a generally uncooperative attitude. Therefore, Seliga claims that he concluded that Clay failed to demonstrate the entrepreneurial spirit and commitment necessary to grow her practice into a profitable enterprise.

Clement, Rudolph and Gaffney corroborated Seliga's testimony. According to Clement, physicians were selected for termination because "what they might have been looking for in a position may not have been consistent with what we were looking for in a physician." Clement testified further as to the kind of physician that the Hospital was not seeking: "Some physicians, some employees are looking for a j-o-b, an entitlement; I'll do my 40 hours; I'm not going to be responsible for growing the practice; I'm not going to be an owner." Rudolph affirmed that the physicians selected for termination were assessed to be less likely to grow their practices than those who were retained. Rudolph also testified that she commented to Seliga about Clay's reluctance to see patients after hours. And Rudolph noted Clay's inadequate revenue, as her net collections amounted to $22,453 for the six-month period from July 1997 to December 1997, while she received a $50,000 salary for that same period. Furthermore, Gaffney testified that she told Seliga about Clay's unwillingness to participate in the Hospital's marketing efforts.

On May 14, 1998, Seliga met with Clay about her termination. Seliga testified that he first learned of Clay's pregnancy at that meeting. The following day, on May 15, the Hospital formally announced the Reorganization.

After the announcement of her termination, Clay sued the Hospital, alleging that she was terminated because of her pregnancy in violation of Title VII. The Hospital moved for summary judgment, arguing that Clay could not establish a prima facie case of pregnancy discrimination because Seliga, the Hospital's decision-maker, was unaware of Clay's pregnancy when he selected her for termination. The Hospital also argued that Clay could not show that the Hospital's legitimate, non-discriminatory reason for her termination (Seliga's belief that Clay failed to demonstrate the potential to grow her practice to profitability) was a pretext for discrimination.

In her response to the Hospital's motion, Clay argued that she established a prima facie case because Seliga knew about her pregnancy before he selected her for termination. Clay became pregnant in late October 1997, and was pregnant for a full nine months until she delivered her baby on July 16, 1998. According to Clay, she started to become visibly pregnant in February 1998. She testified that it is her "belief" that Seliga knew of her pregnancy because she and Seliga had attended a few Hospital conferences and meetings in February, March, and April of 1998. But Clay also testified that she did not disclose her pregnancy to the Hospital's administration until May 6, 1998, when she faxed her request for maternity leave to the Practice Management Office.

Clay also responded to the Hospital's motion for summary judgment by presenting evidence that the Hospital ranked her number 32 out of 70 physicians in the Network according to the financial performance of her practice. Thus, Clay presented evidence that she treated more patients and generated more revenue during her first year of practice than other physicians who were not terminated by the Hospital.

The district court granted summary judgment for the Hospital, concluding that Clay failed to create a genuine issue of material fact that the Hospital's proffered reason for her termination was a pretext for discrimination. The court determined that Clay failed to "squarely rebut" the reason offered by the Hospital for her termination: the Hospital administration's belief that Clay lacked the willingness to do what was necessary to make her practice profitable, like participate in the Hospital's marketing efforts, increase her patient accessibility, and see patients at other locations. Clay appeals.

Clay argues on appeal that the district court erroneously granted summary judgment for the Hospital. We review de novo the district court's decision to grant summary judgment, construing all facts, and drawing all reasonable inferences from those facts, in favor of Clay, the non-moving party. Oest v. Illinois Dep't. of Corrections, 240 F.3d 605, 610 (7th Cir. 2001). Summary judgment is proper when the "pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." Fed.R.Civ.P. 56(c).

According to Clay, the Hospital violated Title VII when it terminated her because of her pregnancy. "The Pregnancy Discrimination Act amended Title VII of the Civil Rights Act to clarify that pregnancy discrimination is included in Title VII's prohibition on sex discrimination."2 Ilhardt v. Sara Lee Corp., 118 F.3d 1151, 1154 (7th Cir. 1997). Clay presents no direct evidence of discrimination, and thus she proceeds under the burden-shifting approach of McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973). According to this method, Clay must first present a prima facie case of discrimination by establishing that: (1) she was pregnant (a member of a protected class) and her employer knew that she was pregnant; (2) she was performing her duties satisfactorily; (3) she was discharged; and (4) similarly situated employees not in the protected class were treated more favorably. Ilhardt, 118 F.3d at 1154-55. If Clay presents a prima facie case, then the burden of production shifts to the Hospital to articulate a legitimate, non- discriminatory reason for her termination. See Bekker v. Humana Health Plan, Inc., 229 F.3d 662, 672 (7th Cir. 2000). Once the Hospital has proffered a legitimate reason, the inference of discrimination disappears, and Clay must prove by a preponderance of the evidence that the Hospital's proffered reason was a pretext for intentional discrimination. Id. The ultimate burden to prove intentional discrimination remains with Clay. Id.

The district court bypassed consideration of whether Clay established a prima facie case, stating that "[b]ecause [Clay] has made no showing of pretext, this court need not decide whether she has established a prima facie case of discrimination." Clay argues on appeal that she presented sufficient evidence to create a genuine issue of material fact that the Hospital's proffered reason for her termination was a pretext for discrimination. Pretext "means a dishonest explanation, a lie rather than an oddity or an error." Kulu mani v. Blue Cross Blue Shield Ass'n., 224 F.3d 681, 685 (7th Cir. 2000). "A 'pretext for discrimination' means more than an unusual act; it means something worse than a business error; 'pretext' means deceit used to cover one's tracks." Id. at 684. "On the issue of pretext, our only concern is the honesty of the employer's explanation." O'Connor v. DePaul University, 123 F.3d 665, 671 (7th Cir. 1997). Thus, even if Seliga's reasons for Clay's termination were "mistaken, ill considered or foolish, so long as [Seliga] honestly believed those reasons, pretext has not been shown." Jordan v. Summers, 205 F.3d 337, 343 (7th Cir. 2000).

Clay first argues that she established pretext because she presented facts demonstrating that Seliga lied when he testified that he did not know about Clay's pregnancy before selecting her for termination under the RIF, and thus Seliga lacks credibility. In support of her contention, Clay testified that she "started becoming visibly pregnant in February 1998," and that it was her "belief" that Seliga knew about her pregnancy because she and Seliga attended approximately two or three Hospital conferences or meetings in early 1998. Clay also asserts in her response brief that her "husband announced his wife's pregnancy to Bill Seliga" at a conference in March 1998. According to Clay, these facts demonstrate that Seliga lacks credibility, and thus "it should follow that everything he [Seliga] says regarding Dr. Clay is either suspect or wholly without merit." Hence, Clay contends that her case turns on the issue of Seliga's credibility, which is best resolved by a jury.

The district court considered Seliga's credibility as an issue pertinent only to whether Clay had established her prima facie case. The court chose not to sort out the question of whether Seliga did or did not know Clay was pregnant; instead, the court bypassed the prima facie analysis and proceeded to the question of whether Clay raised a triable issue of fact regarding pretext.3

Although the court left that step of the prima facie case unanswered, Clay insists that the timing of Seliga's knowledge of her pregnancy is pivotal to her pretext claim. Yet a review of the record does not establish the inference that Seliga knew about her pregnancy before he selected her for the RIF. No doubt there were earlier occasions when he had an opportunity to make that discovery. But Clay's assertion that her husband announced her pregnancy to Seliga lacks support in the record. Mr. Clay's affidavit states that he "discussed [his] impending fatherhood in the company of Bill Seliga" at a conference in March 1998. Mr. Clay does not allege that Seliga heard his conversation or acknowl edged it in any way. Dr. Clay testified that the conference was held in a "large room" and was attended by "approximately 30 or 40" people.4 Just because Seliga was in the same large room with a number of people does not justify the inference that he heard Mr. Clay's discussions with others about his wife's pregnancy.5 Moreover, Dr. Clay testified that she "was trying to keep things quiet"6 about her pregnancy from the Hospital's administration (including Seliga) until May 6, 1998, when she faxed in her request for maternity leave. Precisely because Dr. Clay was "trying to keep things quiet" about her pregnancy from Hospital officials until May 6 supports the inference that she was not certain that Seliga knew about her pregnancy at that time. We conclude that the record does not support Clay's allegation that Seliga knew about her pregnancy before he selected her for the RIF.7

But even if Clay presented sufficient evidence to create a triable issue that Seliga lied about when he became aware of Clay's pregnancy, Clay's pretext argument is still unavailing. To prove pretext, Clay must present facts that cast doubt on the Hospital's specific reasons for her termination. Paluck v. Gooding Rubber Co., 221 F.3d 1003, 1013 (7th Cir. 2000); see also Sweeney v. West, 149 F.3d 550, 557 (7th Cir. 1998) ("The plaintiff must call the employer's honesty into question by rebutting the reason given."). Therefore, Clay's pretext claim turns on Seliga's credibility on his stated reasons for her termination only after Clay has offered "specific evidence from which the finder of fact may reasonably infer that the proffered reasons do not represent the truth." Collier v. Budd Co., 66 F.3d 886, 893 (7th Cir. 1995). Clay's "burden is to squarely rebut the articulated reason for [her] discharge." Plair v. E.J. Brach & Sons, Inc., 105 F.3d 343, 349 (7th Cir. 1997). Moreover, she must present facts to rebut each and every legitimate, non-discriminatory reason advanced by the Hospital in order to survive summary judgment. See Adreani v. First Colonial Bankshares Corp., 154 F.3d 389, 399 (7th Cir. 1998) ("The existence of a genuine issue of triable fact with respect to some of the reasons for discharge proffered by the employer is of no consequence as long as at least one reason is uncontested."). Clay has not met her burden in this case.

The Hospital's stated reason for Clay's discharge under the RIF is that the administration (especially the decision- maker, Bill Seliga) honestly believed that Clay "lacked the drive, work ethic, and dedication to grow her practice to profitability in the future." More specifically, the Hospital asserts that Seliga believed that Clay had a generally uncooperative attitude, failed to participate in the Hospital's marketing events, was unwilling to maintain a high level of patient accessibility, was reluctant to cover for other physicians, and maintained a practice that had low growth and low revenue.

Clay attempts to prove pretext by presenting evidence that the financial performance of her practice ranked number 32 out of the 70 physicians in the Network, and that she treated more patients and generated more revenue during her first year of practice than other physicians who were not terminated by the Hospital.

While Clay's evidence indicates that her practice may have performed better financially than other physicians who were not terminated by the Hospital, Clay does not show that her practice was profitable, as it was losing money for the Hospital. Therefore, compared to a profitable practice, Clay's practice did indeed have low revenue. Moreover, Clay's evidence does not rebut Seliga's other reasons for his decision to select Clay for termination under the RIF. See Adreani, 154 F.3d at 399.

Seliga selected physicians for the RIF according to his assessment of each physician's potential to build his or her practice into a profitable enterprise, as well as on the current financial performance of the practice. It is undisputed that Clement asked Seliga to assess each physician's practice, including subjective factors like how hard a physician was working and whether she was participating in the hospital's marketing efforts to build a profitable practice. It is also undisputed that Seliga solicited the opinions of the Hospital's marketing and practice managers on the "physicians' participation and receptivity to volume building, and the physicians' likelihood to grow their practices." Clay's evidence does not rebut these reasons regarding Seliga's assessment of her efforts at building her practice into a profitable enterprise.

Clay presents no evidence to rebut Seliga's testimony that Gaffney advised him that Clay was uncooperative with the Hospital's marketing efforts, and that Clay failed to participate in any of the Hospital's 25 marketing events. Clay attempts to refute this reason with her own testimony that she solicited new patients by participating in a radio broadcast, appearing at a mall, visiting a local factory and joining a community board. But this evidence does not indicate that Clay participated in the Hospital's marketing events, and thus it does not rebut Gaffney's assertion, or the fact that Gaffney notified Seliga about her frustration with Clay's lack of participation. Thus, Clay's testimony does not demonstrate pretext because it does not show that Seliga did not honestly believe that Clay was unwilling to participate in the Hospital's marketing events. See O'Connor, 123 F.3d at 671.

Clay also fails to show that Seliga did not honestly believe that she was unwilling to increase her patient accessibility. According to Seliga, Gaffney and Rudolph informed him that Clay would refuse to see patients outside of her scheduled office time. Rudolph's deposition testimony confirmed Seliga's assertion. And Clay presents no evidence to rebut the basis of Seliga's belief, or to show that he did not honestly believe that Clay was unwilling to increase her patient accessibility.

Another reason for Seliga's decision to terminate Clay was his belief, based on Rudolph's information, that Clay was reluctant to build her practice by covering for other physicians. Clay attempts to refute this reason by identifying excerpts of her deposition testimony where she stated that she had in fact covered for other doctors. But Clay made no such argument at summary judgment, and thus it is waived. See Arendt v. Vetta Sports, Inc., 99 F.3d 231, 237 (7th Cir. 1996) ("'We have long refused to consider arguments that were not presented to the district court inresponse to summary judgment motions.'") (quoting Cooper v. Lane, 969 F.2d 368, 371 (7th Cir. 1992)). Moreover, her testimony does not show pretext because it does not rebut the basis of Seliga's belief (Rudolph's comments that Clay was reluctant to cover for other physicians), or that Seliga honestly believed that Clay was lacking in this area.8 Therefore, Clay has failed to prove pretext because she has not presented facts to refute Seliga's reasons for her termination.

We also note that the Hospital did not rely solely on Seliga's testimony, as Clement, Rudolph and Gaffney corroborated Seliga's description of his methodology in conducting the RIF, and the basis for his belief that Clay had failed to demonstrate the dedication necessary to build her practice into a profitable enterprise. It is also undisputed that the issue of pregnancy was never brought up in the decision-making process, and that Seliga collaborated with other practice managers to compose the list of physicians to be terminated. We conclude, therefore, that Clay's pretext arguments are unavailing.

Clay has failed to demonstrate that the Hospital's proffered reason for her termination was a pretext for intentional discrimination. Clay has not shown that the Hospital's decision-maker, Bill Seliga, did not honestly believe that Clay lacked the drive, work ethic, and dedication to grow her practice into a profitable enterprise. We thus AFFIRM the district court.

DIANE P. WOOD, Circuit Judge, concurring in the judgment.

It may be true that there are some people so obtuse that they cannot recognize the condition of a woman six or seven months along in her pregnancy, and that there are some pregnan- cies that are not detectable until the day of delivery. These, however, are the rare cases. I make this point in order to disagree with the majority's conclusion, ante at 1007, that "the record does not support Clay's allegation that Seliga knew about her pregnancy before he select- ed her for the RIF." To the contrary, as the district court recognized and as even the major- ity's opinion tacitly acknowledges, there are genuinely disputed issues of fact about the state of his knowledge.

The record shows that Pamela Clay became preg- nant in late October 1997 and that she delivered a full-term baby in mid-July 1998. It also shows, according to Clay's affidavit, that she had become visibly pregnant by February 1998. This is not only a fact of which Clay obviously had personal knowledge; it is one she was willing to back up with contemporaneous photographs of herself, which she proffered at the summary judgment stage. Beyond that, common experience suggests that most women are "showing" well before that time, as a brief glance at the patrons in a maternity clothing store would confirm. The record, taken in the light most favorable to Clay's position, also shows that she attended either two or three hospital conferences or meetings with Bill Seliga between February 1998 and before April 1998, when he became en- gaged in the RIF process. The majority's opinion does not take issue with any of these points. See ante at 1004. Instead, it appears to place decisive weight on the fact that Clay did not "disclose" (verbally) her pregnancy to Seliga until May 6, 1998, approximately two months before she deliv- ered the baby. A trier of fact might believe Seliga's story that he (a hospital manager) was oblivious to Clay's visible pregnancy at the time of those meetings, but a trier of fact with the photographs in front of it equally might find such a story incredible. I see no warrant for the majority's decision to resolve this issue of fact in the Hospital's favor, at the summary judgment stage. Indeed, in light of its subsequent discus- sion of pretext, this part of the opinion is little more than dicta in any event. I would find that Clay successfully established her prima facie case of discrimination.

This case turns, as so many do, on the issue of pretext. If we now assume that Seliga knew that Clay was pregnant, lied about his knowledge, and chose her for termination in the RIF procedure, is this enough to cast doubt on the other reasons the Hospital proffered for asking Clay to leave? That, I believe, is a much closer question. The Hospital points to substantial evidence all of which shows that Clay was not the kind of "team player" or financial performer it wanted. Clay has made two mistaken assumptions here. First, she apparently believes that her case is won if she can prove that Seliga was lying about his knowledge of her condition; that, however, is not true. The question remains whether that lie is enough to taint all the rest of the Hospital's evidence about its reasons for including her in the RIF. It is not, as the majority explains, if an untainted reason remains that independently supports the employment decision. Second, Clay appears to think that the way to raise a genuine issue of fact on the pretext question is to present evidence that would refute the Hospital's ultimate conclusion that she was not likely to be a good doctor, a good business generator, and a good financial risk. That, too, is not correct; the question instead is whether there is any evidence that suggests that these were not bona fide reasons for the Hospital's actions. If the Hospital wanted people to attend its own marketing events rather than addressing the topic of marketing with a free-lance approach, it was entitled to take this position. Clay's earnings hardly made her a star among the doctors; she was instead squarely in the middle, ranked financial- ly as number 32 out of 70. For these reasons, as well as the others the majority discusses in the opinion, ante at 1007-09, I would affirm the district court's decision on the basis that Dr. Clay has not raised a genuine issue of material fact on the question of pretext. I therefore concur in the judgment.