Pamela RHEINFRANK, individually, and as Parent and Natural Guardian of M.B.D., Plaintiff-Appellant, v. ABBOTT LABORATORIES, INC.; AbbVie, Inc., Defendants-Appellees, 680 F. App'x 369

ROGERS, Circuit Judge. This state-law product liability case arose from birth defects in the form of serious physical and cognitive disabilities suffered by plaintiff Rheinfrank’s child. The defects were allegedly caused by the antiepileptic drug Depakote, manufactured by defendant Abbott Laboratories, and taken by Rheinfrank while pregnant with the child. Rheinfrank contended that the labeling on the drug was not adequate.

The district court held that Rheinfrank’s Ohio law failure-to-warn claim—to the extent that it was based on Abbott’s failure to warn of the risk of developmental delays—was preempted by federal drug labeling law because the Food and Drug Administration had rejected label changes containing such warnings even well after Rheinfrank’s pregnancy. A jury found for Abbott on several remaining claims, and Rheinfrank appeals the overall judgment for defendant. Each of Rheinfrank’s contentions on appeal is not sufficient to warrant reversal. The district court’s preemption analysis was correct.

With respect to the jury verdict, the district court did not abuse its discretion in putting limits on expert testimony related to the scope of the experts’ respective expertise, or by rejecting requested jury instructions that were repetitive, confusing, or both. I. For nearly her entire life Rheinfrank has suffered from epilepsy. In 1988, after a relapse of her epileptic seizures, she began taking two antiepileptic drugs, one of them Depakote. She continued that course of treatment over the next fifteen years, including the years she was pregnant with each of her first four children.

Late in 2003, while she was still on her daily regimen of the two drugs, Rheinfrank became pregnant with her fifth child, M.B.D. While she was carrying M.B.D. she continued to take both antiepileptic drugs daily. In July 2004 Rheinfrank gave birth to M.B.D., who was later diagnosed with physical deformities and cognitive disabilities, including Fetal Valproate Syndrome. Those disabilities Rheinfrank now blames on her daily use of Depakote.

The Food and Drug Administration (FDA) first approved Depakote in early 1983, and, by the time Rheinfrank was first prescribed the drug in 1988, it had also been designated a Pregnancy Category D drug by the FDA, As of 2003, the drug’s label accordingly included a “Black Box Warning” cautioning about the risks the drug posed to developing fetuses: TERATOGENICITY: VALPROATE CAN PRODUCE TERA-TOGENIC EFFECTS SUCH AS NEURAL TUBE DEFECTS (E.G., SPINA BIFIDA).

ACCORDINGLY, THE USE OF DEPAKOTE TABLETS IN WOMEN OF CHILDBEARING POTENTIAL REQUIRES THAT THE BENEFITS OF ITS USE BE WEIGHED AGAINST THE RISK OF INJURY TO THE FETUS. THIS IS ESPECIALLY IMPORTANT WHEN THE TREATMENT OF A SPONTANEOUSLY REVERSIBLE CONDITION NOT ORDINARILY ASSOCIATED WITH PERMANENT INJURY OR RISK OF DEATH (E.G., MIGRAINE) IS CONTEMPLATED. SEE WARNINGS, INFORMATION FOR PATIENTS. AN INFORMATION SHEET DESCRIBING THE TERA-TOGENIC POTENTIAL OF VALP-ROATE IS AVAILABLE FOR PATIENTS. The “Usage and Pregnancy”

section of the same label further underlined these risks: Usage in Pregnancy ACCORDING TO PUBLISHED AND UNPUBLISHED REPORTS, VAL-PROIC ACID MAY PRODUCE TER-ATOGENIC EFFECTS IN THE OFFSPRING OF HUMAN FEMALES RECEIVING THE DRUG DURING PREGNANCY. THERE ARE MULTIPLE REPORTS IN THE CLINICAL LITERATURE WHICH INDICATE THAT THE USE OF ANTIEPILEPTIC DRUGS DURING PREGNANCY RESULTS IN AN INCREASED INCIDENCE OF BIRTH DEFECTS IN THE OFFSPRING. ALTHOUGH DATA ARE MORE EXTENSIVE WITH RESPECT TO TRIMETHA-DIONE, PARAMETHADIONE, PHE-NYTOIN, AND PHENOBARBITAL, REPORTS INDICATE A POSSIBLE SIMILAR ASSOCIATION WITH THE USE.

OF OTHER ANTIEPILEPTIC DRUGS. THEREFORE, ANTIEPI-LEPSY DRUGS SHOULD BE ADMINISTERED TO WOMEN OF CHILDBEARING POTENTIAL ONLY IF THEY ARE CLEARLY SHOWN TO BE ESSENTIAL IN THE MANAGEMENT OF THEIR SEIZURES. THE INCIDENCE OF NEURAL TUBE DEFECTS IN THE FETUS MAY BE INCREASED IN MOTHERS RECEIVING VALPROATE DURING THE FIRST TRIMESTER OF PREGNANCY. THE CENTERS FOR DISEASE CONTROL (CDC) HAS ESTIMATED THE RISK OF VALPROIC ACID EXPOSED WOMEN HAVING CHILDREN WITH SPINA BIFIDA TO BE APPROXIMATELY 1 TO 2%.

OTHER CONGENITAL ANOMALIES (E.G., CRANIOFACIAL DEFECTS, CARDIOVASCULAR MALFORMATIONS AND ANOMALIES INVOLVING VARIOUS BODY SYSTEMS), COMPATIBLE AND INCOMPATIBLE WITH LIFE, HAVE BEEN REPORTED. SUFFICIENT DATA TO DETERMINE THE INCIDENCE OF THESE CONGENITAL ANOMALIES IS NOT AVAILABLE. THE HIGHER INCIDENCE OF CONGENITAL ANOMALIES IN ANTIEPILEPTIC DRUG-TREATED WOMEN WITH SEIZURE DISORDERS CANNOT BE REGARDED AS A CAUSE AND EFFECT RELATIONSHIP.

THERE ARE INTRINSIC METHODOLOGIC PROBLEMS IN OBTAINING ADEQUATE DATA ON DRUG TERATO-GENICITY IN HUMANS; GENETIC FACTORS OR THE EPILEPTIC CONDITION ITSELF, MAY BE MORE IMPORTANT THAN DRUG THERAPY IN CONTRIBUTING TO CONGENITAL ANOMALIES. Both before and during her pregnancy with M.B.D. in 2003 and 2004, Rheinfrank had received her prescription for Depakote from Dr. Dagmar Lemus, then a resident of internal medicine at Cincinnati’s Good Samaritan Hospital.

Although Lemus claimed that she did not remember prescribing Depakote to Rheinfrank, she nevertheless insisted that she would have relayed the information on the Black Box warning before prescribing the drug. Le-mus further explained that she would never have relied on any promotional or other materials when deciding to prescribe the drug to Rheinfrank. In April 2005, Abbott Laboratories sent a letter to Dr. Russell Katz of the FDA to propose an update to the already-approved label for Depakote. Abbott had learned of some of the early results of a study then being led by Dr.

Kimford Meador, investigating the neu-rodevelopmental effects of antiepileptic drugs like Depakote on young children (the “NEAD study”).

That study uncovered preliminary evidence of “possible developmental delay in some children exposed to [Depakote] in útero.” In response to these reports Abbott put together a Prior Approval Supplement (“PAS”) that proposed changing Depa-kote’s label to “provid[e] revised information related to teratogenicity and additional information for developmental delay and include revisions to the WARNINGS—Usage in Pregnancy and the Patient Information Leaflet sections.” Accompanying that supplement were two additional changes: “[a]n outline of safety-related changes for terato-genicity/developmental delay and DDI with topiramate” and “[n]ew information concerning the use of valproate in women of childbearing potential: teratogenicity and developmental delay.” Also among the revisions Abbott proposed for Depakote’s label was the inclusion of new language under its “Usage in Pregnancy” section, reading in relevant part: THERE ARE DATA THAT SUGGEST AN INCREASED INCIDENCE OF CONGENITAL MALFORMATIONS ASSOCIATED WITH THE USE OF VALPROIC ACID DURING PREGNANCY WHEN COMPARED WITH SOME OTHER ANTIEPILEPTIC DRUGS.

THEREFORE, VALPROIC ACID SHOULD BE CONSIDERED FOR WOMEN OF CHILDBEARING POTENTIAL ONLY AFTER THE RISKS HAVE BEEN THOROUGHLY DISCUSSED WITH THE PATIENT AND WEIGHED AGAINST THE POTENTIAL BENEFITS OF TREATMENT. THERE HAVE BEEN REPORTS OF DEVELOPMENTAL DELAY IN THE OFFSPRING OF WOMEN WHO HAVE RECEIVED VALPROIC ACID DURING PREGNANCY. Abbott further proposed new language under the “Information for Women Who Could Become Pregnant”

section of its Patient Information Leaflet: These medications have also been associated with other birth defects such as defects of the heart, the bones, and other parts of the body. Information suggests that birth defects may be more likely to occur with these medications than some other drugs that treat your medical condition. In addition, there have been reports of developmental delay in children born to women taking these medications.

In support of these modifications Abbott attached to its PAS a “White Paper” discussing some of the scientific literature relating to developmental delay in children exposed in útero to antiepileptic drugs, including some of the results of the NEAD study.

In February 2006, however, the FDA responded to Abbott’s PAS by email, notifying Abbott that the proposed sentence discussing developmental delay should not be added to Depakote’s labeling: The sentence “There have been reports of developmental delay in the offspring of women who have received valproic acid during pregnancy” is based on two recent publications (Gaily E et al. Neurology 62(l):28-32, 2004 and Vintén J et al.

Neurology 64(6): 949-64, 2006) that attempted to correlate children’s performance on IQ assessments with maternal prenatal use of valproate but which did not adequately control for maternal IQ and maternal educational attainment. Maternal IQ and maternal educational attainment are known to strongly correlate with children’s performance on IQ assessments and thus would confound any attempt to draw a correlation to maternal prenatal valproate use. Given the studies’ inability to establish this correlation, the proposed sentence should not be incorporated into labeling.

A similar proposed sentence in the Patient Information Leaflet was removed in the Approval Letter for S-032 (January 11, 2006). In May 2007, after Dr. Meador released additional data from the NEAD study, Abbott sent another letter to Dr.

Katz, this time bearing the subject-line, “General Correspondence—Request for Advice regarding Developmental Labeling.” The letter’s purpose was to give the FDA “an updated analysis of the occurrence of developmental delay in the attached white paper, which now includes more compelling data from the [NEAD] study.” Attached to that letter was also a 2007 white paper discussing a recently published abstract from Dr.

Meador’s NEAD study, which the letter claimed provided “the first data with adequate control for maternal IQ using a standard IQ measure and showed a significant developmental delay in 185 two-year-old children exposed to valproic acid during pregnancy.” Pointing to those data, Abbott proposed new language to the “WARNINGS—Usage in Pregnancy”

section of Depakote’s label: “There have been reports of developmental delay in the offspring of women who have received valp-roate during pregnancy.” In March 2008, according to an internal contact report prepared by Abbott, the company held a teleconference with Dr. Katz and another representative of the FDA Division of Neurology Products “to discuss teratogenesis associated with val-proic acid treatment.” The report indicated that “Dr. Katz stated they cannot approve this labeling change at this time,” as the data was “not ‘ripe’ for inclusion in labeling since it [was] based on interim data from Dr. K.

Meador and the Neurodevel-opment Effects of Antiepileptic Drugs (NEAD) Study group.” The report went on to state that: [the] FDA feels that the sample size with VPA compared to other agents is small, some of the data for the 2 year old IQ evaluation was imputed, and there are too many confounding factors 'to believe the data is reliable at this time point in the study. Dr. Katz stated that they want to wait until the study is complete at the six-year time point. Dr.

Embrescia then commented that there have been a number of cases of developmental delay reported through our [p]ost-marketing safety surveillance program, and asked whether that might not warrant the change in labeling. Dr. Katz asked for the number of cases we have, and Jim responded that at the time we submitted the proposed labeling change we had 240 reported cases, although many of those cases are confounded by other congenital abnormalities the patients have or other medications they were on. Dr.

Katz indicated he thought these cases were probably too confounded to assess and that he believes this is the type of event where they want investigation in a formal setting to confirm. In April 2009, Abbott once again requested advice from the FDA about adding developmental delay warnings to the Depakote label. In its letter Abbott noted that since its 2005 and 2007 requests, Dr. Meador had published further results from the NEAD study in the New England Journal of Medicine, and other researchers had begun publishing the results of other studies relating to developmental delay in children exposed to Depakote.

Abbott therefore requested that the FDA “provide advice on the acceptability of these data for use to support an amendment to the current label regarding the risk of developmental delay and/or autism/autism spectrum disorder with intrauterine exposure to valproate.” In September 2009, the FDA Division of Neurology Products held another teleconference with Abbott in response to its latest request for advice. According to Abbott’s internal report of that meeting, the FDA “had expressed concerns with Dr. Meador’s data” and had “plans to conduct an independent review.” Dr.

Katz reportedly stated that the FDA’s “statisticians have raised concerns with the [NEAD] study and the methodology for the collection of data,” so that “before taking regulatory actions with labeling, they needed time to evaluate the data.” Dr. Katz accordingly “stated that they were not yet ready to ‘sign off on labeling language’” concerning developmental delay. Moreover, in reply to an inquiry by Abbott as to whether the FDA “would be open to [Abbott’s] proposal for labeling language in the interim while the [FDA] completes its review,” Dr.

Katz noted that Abbott “would be within [its] rights to submit a [“changes being effected,” or CBE supplement],” but added that the FDA “would not take action until the review of the data was complete.” Two months later, Abbott submitted a CBE labeling supplement to add developmental delay warnings to De-pakote’s label. In October 2011, years after Abbott’s initial request, the FDA approved the addition of a warning about developmental delay on the Depakote label.

In 2013 Rheinfrank brought this suit in diversity against Abbott, on behalf of herself and M.B.D., asserting several Ohio state statutory claims of strict liability based on design defect, inadequate warning, and nonconformance with representations, as well as a number of state common law claims. Abbott responded with a number of motions, including one for

summary judgment on Rheinfrank’s failure-to-warn claim with respect to the risks of developmental delay and her request for punitive damages under the Ohio Product Liability Act. As to the failure-to-warn claim, Abbott contended that because the FDA had twice rejected its proposed modifications to De-pakote’s label warning of new evidence of development delay, there was “clear evidence” under Wyeth v. Levine, 555 U.S. 555, 129 S.Ct. 1187, 173 L.Ed.2d 51 (2009), that the agency would not have approved a developmental-delay warning before M.B.D.’s conception and birth ten years earlier.

The district court accepted Abbott’s argument, reasoning that: because there is clear evidence the FDA would not have approved a change to the Depakote label adding a developmental delay warning prior to M.B.D.’s injury.

The Court finds the FDA’s February 2006 decision that developmental delay warnings “should not be incorporated into [Depakote] labeling” and the FDA’s 2008 belief that “the data do not provide sufficient evidence to support [Depa-kote] labeling changes at this time” constitute “clear evidence” that when confronted by the issue in 2003, the FDA would have rejected an attempt to add a developmental delay warning. Although the district court thus determined that Rheinfrank’s particular claim that Abbott had failed to warn of the risk of developmental delay was preempted, the court nevertheless denied

summary judgment on Rheinfrank’s broader failure-to-warn claim because other questions of material fact remained. The court did grant

summary judgment for Abbott, however, on Rheinfrank’s request for punitive damages, ruling that she was “barred from recovering for punitive damages because the FDA has not made a finding of either fraud or misrepresentation.” Before trial, the district court also entered an order limiting in various respects the testimony of several of Rhe-infrank’s experts. The first of those limitations involved the testimony of Drs. Michael Privitera (a neurologist), Howard Saal (a geneticist), and C.

Ralph Buncher (an epidemiologist), restricting all three from giving any opinion as to what warnings or information should have been included in Depakote’s label. As to Dr. Privitera, the court concluded that: Dr. Privitera is not qualified to opine on the regulatory aspects of the case, including whether Abbott was required to send a patient package leaflet directly to patients or whether Abbott’s submissions to the FDA should have included certain materials.

Similarly, testimony about what Defendants should have included in the label or what materials should have been submitted to the FDA falls outside the scope of his expertise, as it falls under the regulatory component and is speculative. Thus, Dr. Privit-era also may not testify about whether Depakote should have been contraindicated for all women of childbearing years. On the other hand, testimony in which Dr. Privitera opines on the medical facts and science regarding the risks and benefits of Depakote and compares that knowledge with what was provided in the text of the labeling is admissible. As for Dr.

Saal, the court determined that, although he was “well-qualified ... to testify as to the medical facts and science and compare that information and data with the language of the 2003 Depakote label,” he was “not an expert in FDA regulations,” and thus “lack[ed] the requisite expertise to opine as to the regulatory aspects of the ease, including what ‘should’ have been in the 2003 label and whether the drug ‘should’ have been contraindicated for women of childbearing years, as this assumes regulatory knowledge.” The court concluded much the same about Dr.

Buncher: despite being “qualified to opine on the medical facts and science regarding Depakote and compare that data to a De-pakote label,” the court found that he lacked “specialized knowledge or expertise in the regulatory field, as he has never worked for the FDA, nor is he an expert on FDA labeling regulations.” The court concluded that Dr.

Buncher “laek[ed] the requisite expertise to opine as to the regulatory aspects of the case, including what ‘should’ have been in the 2003 Depakote label and whether Depakote ‘should’ have been contraindicated for women of childbearing years, as this assumes regulatory knowledge.” The court also placed several limitations on the testimony of Rheinfrank’s regulatory expert, Dr. Suzanne Parisian.

As relevant here, the court ruled that, although a medical doctor and former “FDA Medical Officer,” Parisian was nevertheless “unqualified to opine that as of 2003, Depakote was known to be the most teratogenic drug,” as she is “not a teratologist and does not have any specialized knowledge or experience in evaluating [antiepileptic drugs].” Her opinion testimony was thus limited to “what actions Abbott could have taken and/or was required to take with respect to communicating risks to healthcare professionals,” as well to "regulatory requirements relating to the development, testing, marketing, and surveillance of prescription drugs.” Before the case went to the jury on the four remaining causes of action'—strict liability under Ohio law for failure to warn and failure to conform to representations, and negligent failure to warn and negligent design—Rheinfrank requested a number of jury instructions.

Among those were requests for instructions concerning the manufacturer’s standard of care and knowledge (No. 7), federal requirements for drug labeling (No. 8) and for adding safety warnings (No. 11), as well as an instruction concerning strict liability for an inadequate post-market warning (No. 14).

The district court refused to offer the first three of these instructions separately (Nos. 7, 8, and 11), “but incorporated aspects of those proposed instructions where appropriate.” As the court later explained when denying Rheinfrank’s motion for a new trial: As Plaintiffs acknowledged in their Proposal for Supplemental Jury Instructions in Response to [Jury] Question #4, the majority of their Requested Jury Instruction No. 7 was incorporated into page 30 of the Civil Jury Instructions under Claim One: Strict Product Liability—Defect Due to Inadequate Warning.

The Court rejected Requested Jury Instruction No. 8, because Plaintiffs cited no Ohio authority in support of the instruction and because the Court was concerned that an instruction on mis-branding would confuse the Jury. The Court incorporated the first sentence of Plaintiff’s Requested Jury Instruction No. 11 into the instruction on Compliance with Regulations. The Court rejected the remainder of the requested instruction because it found no reason to deviate from the Ohio pattern instructions and that superfluous instructions would be burdensome and confusing to the Jury.

The court also excluded an instruction Rheinfrank had proposed on post-market warnings (No. 14). Rheinfrank had in fact asked for two instructions, Nos. 13 and 14, both dealing with her strict liability failure-to-wam claim. As proposed, Instruction No. 13 would have read: STRICT LIABILITY—INADEQUATE WARNINGS/INSTRUCTIONS M.B.D. claims that Defendants’ drug Depakote was defective due to inadequate warnings or instructions, and as a result caused M.B,D.’s harm, injuries, and losses.

You must find for M.B.D. on her claim that Depakote was defective based on inadequate warnings or instructions if the greater weight of the evidence demonstrates that: (

A) At the time Depakote left Defendants’ control, Defendants knew or should have known, in the exercise of reasonable care as an expert in the field, that Depakote posed increased risks of harm to unborn children; and ' (

B) Defendants failed to provide the warnings or instructions that a manufacturer exercising reasonable care would have provided concerning Depakote’s increased risks of harm to unborn children; and (

C) As a result of Defendants’ failure to provide adequate warnings or instructions, Pamela Rheinfrank’s doctor prescribed and Pamela Rheinfrank used Depakote, which resulted in M.B.D.’s harm, injuries, and losses. Rheinfrank’s proposed Instruction No. 14, on the other hand, would have added: STRICT LIABILITY—INADEQUATE POST MARKET WARNINGS/INSTRUCTIONS M.B.D. claims that Defendants’ drug Depakote was defective due to inadequate post-market warnings or instructions, and as a result caused M.BJD.’s harm, injuries, and losses. You must find for M.B.D. on her claim that Depakote was defective based on inadequate warnings or instructions if the greater weight of the evidence demonstrates that: (

A) After Depakote left Defendants’ control, Defendants knew or should have known, in the exercise of reasonable care as an expert in the field, that Depa-kote posed increased risks of harm to unborn children; and (

B) Defendants failed to provide the post-market warnings or instructions that a manufacturer exercising reasonable care would have provided concerning Depakote’s increased risks of harm to unborn children; and (

C) As a result of Defendants’ failure to provide adequate post-market warnings or instructions, Pamela Rheinfrank’s doctor prescribed and Pamela Rhein-frank used Depakote, which resulted in M.B.D.’s harm, injuries, and losses. The court ultimately decided to exclude the requested Instruction No. 14, citing two reasons.

First, the court was “concerned that two, separate strict liability instructions back-to-back would be confusing and inconsistent with Plaintiffs’ other proposed instructions and interrogatories.” As the court explained in rejecting Rhein-frank’s motion for a new trial: For example, [Rheinfrank’s] Requested Jury Instruction No. 15 (Presumption Based on Inadequate Warnings) did not distinguish warning and instruction from post-marketing warning and instruction. [Rheinfrank’s] Proposed Interrogatory No. 1 did not distinguish between warning and instruction and post-marketing warning or instruction.

The court reasoned, moreover, that Rhein-frank “did not justify why [she] proposed two separate instructions on the strict liability failure to warn claim, as opposed to one merged instruction.” On those grounds the court had instead decided to provide a single “Compliance with Regulations in its Civil Jury Instruction,” which explained that “a drug manufacturer bears the ultimate responsibility for the content of its label at all times.

It is charged both with drafting an adequate label and with ensuring that its warning remain adequate as long as the drug is on the market.” The court further noted that, before the charge conference, it had circulated, its Annotated Civil Jury Instructions to both Rheinfrank and Abbott—instructions that did not include Rheinfrank’s requested instruction No. 14. During the later charge conference with both sides’ counsel—a meeting that lasted more than three hours—Rheinfrank raised no objection to the court’s exclusion of Instruction No. 14.

After an 11-day trial, the jury began deliberations on the four remaining claims, during which they submitted to the court a question (Jury Question No. 4) about the merged inadequate-warning instruction: Page 30 of the Civil Jury Instructions, under the Claim One, paragraph two, sentence one states “a prescription drug is defective due to inadequate warning or instruction if at the time the prescription drug left the control of the manufacturer ...” Question: What does “left the control of the manufacturer” mean? When its approval was granted or when it left the factory?

According to the district court, as related in its later opinion rejecting Rheinfrank’s motion for a new trial: When the Court received Jury Question No. 4 and read it to counsel, discussion immediately followed about whether “left the control of the manufacturer” is a defined phrase in case law or statute. [Rheinfrank’s] counsel proposed that she believed the correct answer to be “when it left the factory.” Jury Question No. 4 then took on a life of its own when counsel submitted briefs to the Court on the question. [Rheinfrank] argued that the Jury must be confused about whether Depakote may be defective due to inadequate post-marketing warning or instruction.

They contended that the Court should submit not only their Requested Jury Instruction No. 14, but also Nos. 7, 8, and 11, and direct the Jury to refer to the Court’s Civil Jury Instruction on Compliance with Regulations. [Abbott] disagreed and argued that when something leaves the control of the manufacturer is a fact question for the Jury and that additional instructions would be improper. The Court agreed with [Abbotts’]

interpretation of Jury Question No. 4 and instructed the Jury that “left the control of the manufacturer” was an issue for them, as Jury, to decide based on their assessment of the evidence and instructions that were previously provided. After the jury returned a complete verdict for Abbott, Rheinfrank moved for a new trial, arguing, among other things, that the court’s rulings limiting the expert testimony of Drs.

Privitera, Saal, Buncher, and Parisian were in error and had unfairly prejudiced her case, and that her case had been further prejudiced by the court’s erroneous exclusion of her four requested jury instructions (Nos. 7, 8, 11, and 14). Rheinfrank contended, in addition, that the court’s response to Jury Question No. 4 was improper, and urged the court to reconsider its earlier ruling granting partial

summary judgment to Abbott on its federal preemption and punitive damages claim. The district court, however, stood by its earlier rulings as to the expert testimony, its exclusion of the first three jury instructions (Nos. 7, 8, and 11), and its earlier grant of

summary judgment to Abbott on Rheinfrank’s failure-to-wam claim about the risks of developmental delay and her request for punitive damages. As to the requested Instruction No. 14, the court concluded that, by failing to object to the exclusion of the proposed instruction during the charge conference, Rheinfrank had waived any objection. Rheinfrank countered that she had not waived her right to argue that the requested instruction should be given, because the jury’s Question No. 4 had called into doubt a “key instruction” dining deliberations that required correction, as in Reynolds v.

Green, 184 F.3d 589 (6th Cir. 1999). But the court rejected that argument and its analogy to Reynolds, reasoning that: The language the Jury expressed its confusion over—“left the control of the manufacturer”—-was in both of [Rhein-frank’s] Requested Instruction Nos. 13 and 14. Belatedly charging the Jury in the middle of its deliberations with [Rheinfrank’s] Requested Jury Instruction No. 14 would not have answered [Rheinfrank’s] question.

Further, the proposed scenarios by the Jury— “[w]hen its approval was granted” or “left the factory” provided insight into the mindset of the Jury, demonstrating that they were proposing

definitions for the “left the control of the manufacturer,” none of which included any prodding into “post-market” scenarios. The Court viewed [Rheinfrank’s] argument— a belated attempt to bootstrap several additional jury instructions to an answer to Jury Question No. 4—as inappropriate, unhelpful, and likely to cause greater confusion. The court further noted that, because the instruction the court gave already “covered the differences between [Rhein-frank’s] Requested Jury Instructions Nos. 13 and 14 and cured any deficiencies,” its instructions “read as a whole” were appropriate.

And even if they were not, the court concluded, “based on the evidence and the Jury’s total defense verdict, there would have been no difference in the outcome of the case.” The district court accordingly denied Rheinfrank’s motion for a new trial, and she now appeals. II. The first of Rheinfrank’s several claims on appeal is that the district court prejudiced her case by erroneously limiting the testimony of several of her expert witnesses, allegedly entitling her to a new trial.

But because each of those rulings was based on the district court’s reasonable assessment as to the limits of those experts’ respective expertise, and all were consistent with the rulings of other courts, the district court could reasonably have limited the testimony as it did without exceeding the “great latitude” this court must afford such evidentiary rulings, see Taulbee v. Wal-Mart Stores, Inc., 5 Fed.Appx. 361, 363 (6th Cir. 2001) (quoting Rye v. Black & Decker Mfg. Co., 889 F.2d 100, 101-02 (6th Cir. 1989)).

Those rulings therefore did not amount to an abuse of the district court’s discretion, the standard for this court’s review here, Decker v. GE Healthcare Inc., 770 F.3d 378, 391 (6th Cir. 2014). Rheinfrank first argues that the district court erroneously limited the testimony of several of her medical experts— Drs. Privitera, Saal, and Buneher—to the medical facts about Depakote and how they compared to the details contained on its label, because, Rheinfrank claims, those experts were also qualified to opine as to the “adequacy” of that label.

But as the district court explained, the expertise of each of those witnesses—neurology in Pri-vitera’s case, genetics and dysmorphology in Saal’s, and epidemiology in Buncher’s— did not extend to regulatory questions like the ones Rheinfrank proposed for them to answer: “what ‘should’ have been in the 2003 Depakote label and whether Depa-kote ‘should’ have been contraindicated for women of childbearing years.” None of them, after all, had ever worked for the FDA—unlike Dr. Parisian, who did testify for Rheinfrank on the regulatory aspects of Depakote’s label.

None had any specialized training or expertise in FDA labeling regulations. Under the standard set forth in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and its progeny, the district court need only have decided, by a preponderance, that the expert witness was qualified to proffer her testimony, see Decker, 770 F.3d at 391. In this case, given that several other courts have similarly limited the testimony of medical experts to questions within their specialized medical ken, see In re Gadolinium-Based Contrast Agents Prods. Liab.

Litig., No. 1:08 GD 50000, MDL No. 1909, 2010 WL 1796334, at *19 (N.D. Ohio May 4, 2010); In re Diet Drugs (Phentermine, Fenfluramine, Dexfenfluramine) Prods. Liab. Litig., No. MDL 1203, 2000 WL 876900, at *11 (E.D. Pa. June 20, 2000), the court could reasonably have found that Rheinfrank’s experts were likelier than not unqualified under Daubert to opine on regulatory areas in which they had neither specialized training nor professional experience—including opinions as to what ‘should’ have appeared on Depakote’s label, subject as that must be to an essentially technical, regulatory judgment.

The district court consequently did not abuse its discretion by so limiting their testimony. Rheinfrank’s arguments to the contrary, moreover, are without merit. Rheinfrank appears to contend that the district court’s ruling excessively limited these experts’ testimony by' making the question of “adequacy” into a “regulatory matter,” thus prejudicially excluding their other opinions as to the “adequacy” of Depakote’s label in a wider, non-regulatory sense. But that is a misstatement of what the district court ruled. As the district court explained in limiting Privitera’s testimony: Dr.

Privitera is not qualified to opine on the regulatory aspects of the case, including whether Abbott was required to send a patient package leaflet directly to patients or whether Abbott’s submissions to the FDA should have included certain materials. Similarly, testimony about what Defendants should have included in the label or what materials should have been submitted to the FDA falls outside the scope of his expertise, as it falls under the regulatory component and is speculative. Thus, Dr.

Privit-era also may not testify about whether Depakote should have been contraindicated for all women of childbearing years.

The court placed much the same limitation, moreover, on Saal’s and Buncher’s testimony, ruling that they could not testify as to “what ‘should’ have been in the 2003 Depakote label and whether Depakote ‘should’ have been contraindicated for women of childbearing years, as this assumes regulatory knowledge.” In no case, that is, did the court exclude these experts’ opinions as to the “adequacy” of Depa-kote’s label from the point of view of their respective medical expertise, but only from the point of view of the label’s regulatory adequacy.

In fact, Privitera was allowed to “opin[e] on the medical facts and science regarding the risks and benefits of Depakote” and was likewise able to “com-parte] that knowledge with what was provided in the text of the labeling,” and he did so. Similarly, both Saal and Buncher were allowed to “testify as to the medical facts and science” and likewise were able to “compare that information and data with the language of the 2003 Depakote label,” and they did so.

Thus the limitation the district court placed on each of these experts’ testimony as to the “adequacy” of Depakote’s label was itself limited, and only to the regulatory aspects of the labeling. As explained, that limitation was reasonable enough to fall within the “considerable leeway” of discretion that has been confided to the district court when making such rulings, see Meridia Prods. Liab. Litig. v. Abbott Labs., 447 F.3d 861, 868 (6th Cir. 2006) (quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999)).

Rheinfrank also challenges the limitation placed on her regulatory expert, Dr. Parisian, who was not permitted to testify “that as of 2003, Depakote was known to be the most teratogenic drug.” But this limitation also was reasonable and therefore not an abuse of discretion. The district court concluded that Parisian was not qualified to give that testimony because even though she had an extensive background as an FDA Medical Officer and is a medical doctor, she is “not a teratologist and does not have any specialized knowledge or experience in evaluating [antiepileptic drugs],” id.

The court concluded that such a specific opinion as to the teratogenic effects of Depakote relative to other antiepileptic drugs therefore “falls outside the scope of her expertise,” limited as that is by her regulatory background with the FDA and as a physician without any specialized understanding or expertise in antiepileptic drugs. As Rhein-frank appears to concede, Parisian is no specialist in teratology and therefore has no more specialized or expert understanding of antiepileptic drugs than any other licensed physician.

The court could therefore reasonably conclude that her opinion about the relative teratogenicity of Depa-kote exceeded her expertise. As that is all we have required to uphold a ruling like the one the district court made here, see Bye, 889 F.2d at 101, that decision was not an abuse of discretion. But even if the district court did err by so limiting Parisian’s testimony, that error was harmless.

First, Rheinfrank was in fact able to elicit testimony at trial from Parisian that covered the relative teratogenicity of Depakote, the very testimony the exclusion of which Rheinfrank now claims prejudiced her case: Q: And as of this time on August 30, 2002, did the Depakote label contain information that the risk of malformations was higher with Depakote than with other AEDs? A: No. Q: And is that what information was included in the report from Dr. Holmes to Abbott? A: Yes.

Even apart from Parisian’s testimony, moreover, Rheinfrank was able to elicit the same testimony from her other two experts, Privitera and Buncher, who each testified to the higher teratogenicity associated with Depakote relative to other AEDs. This court has indicated that where “the substance of the excluded testimony [is] furnished by other witnesses,” that exclusion is not prejudicial. Hines v. Joy Mfg. Co., 850 F.2d 1146, 1153 (6th Cir. 1988). Thus, even if the district court erroneously limited Parisian’s testimony, that error would not have been prejudicial. Nor, as a result, would it warrant reversal.

See McCombs v. Meijer, Inc., 395 F.3d 346, 358 (6th Cir. 2005). III. Rheinfrank next contends that she is entitled to a new trial because the district court erred in rejecting four jury instructions that she requested, resulting in confusion for the jury and prejudice to her case. But because the instructions that the court did give either substantially covered Rheinfrank’s requested instructions or did not prejudice Rheinfrank’s case in any event, the court’s refusal to provide those four instructions does not amount to reversible error, see Decker, 770 F.3d at 396.

Those rulings, reviewed for an abuse of discretion, Cummins v. BIC USA, Inc., 727 F.3d 506, 510 (6th Cir. 2013), therefore do not warrant a new trial. Rheinfrank argues that the district court erred by refusing to give her three instructions relating to the manufacturer’s standard of care and knowledge (No. 7) and requirements for drug labeling (No. 8) and adding safety warnings (No. 11) under federal law, because all three were correct, and their omission confused and misled the jury and thus were prejudicial.

But as the district court explained as to Rheinfrank’s requested Instructions No. 7 and 11, both were partially incorporated into the instruction the court actually gave.

Moreover, at least two of the three differences that Rheinfrank cites between the given instruction and the two she requested (Nos. 7 and 11)—that the given instruction failed to explain what would make a label inadequate under federal law and did not identify the procedures by which the manufacturer would update its labels—do not amount to substantial departures from the given instructions, which made clear the gist of those procedures, and explained what would make a prescription drug warning adequate.

Although Rheinfrank believes those explanations were deficient, for not having indicated “what ability Abbott had to change its label should it become inadequate,” in fact the given instruction made that ability clear, at least implicitly: a drug manufacturer bears the ultimate responsibility for the content of its label at all times. It is charged both with drafting an adequate label and with ensuring that its warning remain adequate as long as the drug is on the market. Abbott could hardly bear the duty of maintaining the adequacy of its label if it lacked the power to effect those changes.

Moreover, the same compliance instruction explained, in relation to Depakote’s Black Box warning, that “[t]he manufacturer may make suggestions, but the FDA makes the ultimate determination regarding the content of the Black Box warning”—implying, once again, that Abbott had at least some ability under federal law to amend its labeling. As the given instructions substantially covered Rheinfrank’s requested Instructions No. 7 and 11, the district court did not err by excluding them as separate instructions, see Decker, 770 F.3d at 396.

Furthermore, Rheinfrank’s requested instruction on the federal requirements for drug labeling (No. 8), which would have further defined the “adequacy” of a “label” under federal law, also substantially overlapped with the given instruction on the adequacy of prescription drug warnings. Indeed, if anything, the adequacy instruction given by the court was more detailed than the one Rheinfrank herself proposed.

Ultimately, the only substantive differences between the requested Instruction No. 8 and those the court gave come down to the omitted definition of a “label” and Rheinfrank’s proposed definition of “mis-branding.” But as the district court explained, including those details—which were not supported by any Ohio authorities—would only have confused the jury, especially as misbranding did not appear to be a focus of the testimony at trial.

In any event, Rheinfrank has given no reason to believe that the exclusion of the definition of a “label” from the court’s instructions prejudiced her case; On the contrary, even though the proposed definition would have made clear that promotional and other marketing materials were also “label-ling,” in addition to the label that was clearly addressed at trial, Rheinfrank’s own physician explained that those materials did not figure in her decision to prescribe Depakote to Rheinfrank.

Given, then, that Rheinfrank’s requested Instruction No. 8 either overlapped in large measure with instructions given by the court or was otherwise not prejudicially excluded, the district court did not err by rejecting it as a separate instruction, see Decker, 770 F.3d at 396.

Rheinfrank also contends that the district court erred by refusing to give her requested Instruction No. 14, which would have provided a separate strict liability charge for “post-market warning/instruction,” because, Rheinfrank claims, the instruction given by the court failed to explain adequately that “the manufacturer’s duty is the same both before and after the product leaves its control.” But there are two independently fatal flaws in Rhe-infrank’s claim.

First, as with the other instructions Rheinfrank requested, the instructions given by the district court already covered, not just substantially but entirely, her requested Instruction No. 14.

As noted, the court gave an instruction that made clear that “a drug manufacturer bears the ultimate responsibility for the content of its label at all times," and that it must “ensur[e] that its warning remain[s] adequate as long as the drug is on the market.” Indeed, the given instruction on the strict product liability failure-to-warn claim further explained that “[t]he manufacturer has the duty to remain reasonably current with scientific knowledge, development, research, and discoveries concerning the product,” and that, consequently, it “must communicate its superior knowledge to those who, because of their own limited knowledge and information, would otherwise be unable to protect themselves.” Read as a whole, as jury instructions must be, see Pivnick v.

White, Getgey & Meyer Co., LPA, 552 F.3d 479, 488 (6th Cir. 2009), these instructions made clear that what Rhein-frank had proposed in her requested Instruction No, 14 was already covered by the instructions given by the court: Abbott had a duty to maintain the accuracy of its label at all times, even after Depa-kote had gone on the market and after the doses of the drug that Rheinfrank took had left the factory. That overlap is enough to clear the district court of any accusation of error, see Decker, 770 F.3d at 396.

But even if the district court had erred in rejecting Rheinfrank’s Instruction No. 14, she confronts still another difficulty with her claim: by failing to object during the charge conference and before jury deliberations, that objection is now forfeited, Howe v. City of Akron, 723 F.3d 651, 660-61 (6th Cir. 2013) (citing Fed. R. Civ. P. 51(c)(1)). In either case, then, Rheinfrank cannot succeed on her claim that the district court abused its discretion by refusing to give her requested instruction. The question submitted by the jury during deliberations does not alter this conclusion.

Rheinfrank contends that once the jury asked about the meaning of “left the control of the manufacturer,” it became clear that the jury had mistakenly believed that “the timing of when Depakote’s label left Abbott’s control” bore on the company’s liability. Thus, Rheinfrank argues, once the jury expressed its confusion on this “key instruction,” under Reynolds v. Green, 184 F.3d 589, 594 (6th Cir. 1999), her right to raise an objection was revived and her original requested Instruction No. 14 became proper.

As the district court explained, however, the same fact that distinguished Reynolds also made Rhein-frank’s proposed instruction inappropriate as a substitute: the very language apparently at the root of the confusion over the court’s instruction (“left the control of the manufacturer”) also appeared, nearly verbatim, in Rheinfrank’s requested Instruction No. 14 (“[ajfter Depakote left Defendants’ control”).

Thus the district' court concluded not only that Reynolds was inapplicable, but, more importantly, that swapping in Rheinfrank’s instruction would add nothing but more confusion for the jury, failing as it did to answer the question they posed in the first place. Even assuming that Rheinfrank’s objection was not waived, then, her proposed instruction would still have failed to cure the jury’s confusion and would arguably have compounded it.

It was therefore not error—and thus not an abuse of discretion— for the district court not to give Rhein-frank’s requested Instruction No. 14 even in the face of the jury’s question. IV. Rheinfrank further contends that the district court erred in granting

summary judgment on Abbott’s preemption defense against Rheinfrank’s failure-to-warn claim, because genuine issues of material fact remained as to whether the FDA would have approved a developmental delay warning prior to MB.D.’s suffering the injuries she did in útero. This error, Rheinfrank argues, also entitles her to a new trial. But because the evidence in the record reveals that the FDA twice rejected Abbott’s attempts to strengthen Depakote’s label to add a developmental delay warning, there was clear enough evidence under Wyeth that the FDA would not have approved any such change in Depakote’s label. The district court’s grant of

summary judgment on Abbott’s preemption defense, reviewed de novo, see Int’l Union v. Cummins, Inc., 434 F.3d 478, 483 (6th Cir. 2006), was therefore proper. Because the FDA twice refused Abbott’s attempts to strengthen Depakote’s label, based on its own review of the evidence that the drug adversely affected the development of children exposed to it in útero, Rheinfrank’s failure-to-warn claim is preempted by federal drug labeling law.

In Wyeth, the Supreme Court reaffirmed the “central premise of federal drug regulation that the manufacturer bears responsibility for the content of its label at all times,” requiring the manufacturer to “craft[ ] an adequate label” and to “ensur[e] that its warnings remain adequate as long as the drug is on the market.” 555 U.S. at 570-71, 129 S.Ct. 1187.

However, because the FDA also has the authority to reject any labeling changes—under both its unilateral “Changes Being Effected” (CBE) and its “Prior Approval Supplement” (PAS) regimes —the Court also indicated that the manufacturer could “show by ‘clear evidence’ that the FDA would have rescinded any change in the label and thereby demonstrate that it would in fact have been impossible to do under federal law what state law required.” PLIVA, Inc. v. Mensing, 564 U.S. 604, 624 n.8, 131 S.Ct. 2567, 180 L.Ed.2d 580 (2011) (citing Wyeth, 555 U.S. at 571, 129 S.Ct. 1187).

But the only way the FDA can “rescind” a proposed label change is either by rejecting it after it has been added through a CBE supplement, see 21 C.F.R. §§ 314.70(c)(6)(iii)(A)-(C), (c)(7), or by declining it in the first place, under the PAS regime, see id. at 314, 119 S.Ct. 1167.70(b). Together Wyeth and Mensing therefore indicate that “a court cannot order a drug company to place on a label a warning if there is ‘clear evidence’ that the FDA would not approve it.” Robinson v. McNeil Consumer Healthcare, 615 F.3d 861, 873 (7th Cir. 2010).

That evidence would be clear enough, moreover, in cases where the FDA has rejected a proposed label modification in a “submission to which the agency was responding”—such as in response to a. PAS. Id. That makes sense, as the Seventh Circuit concluded in Robinson, for “it would be odd to think” that a manufacturer “had a legal duty to guarantee against a risk that the FDA thought not worth warning against.” Id. In this case, because Abbott has produced precisely that kind of evidence Abbott has met its burden under Wyeth’s clear-evidence standard.

In its email response to Abbott’s 2005 PAS submission, the FDA made clear that a developmental delay warning “should not be incorporated into [Depakote’s] labeling”—a response that even Rheinfrank has characterized as a rejection by the FDA of the proposed labeling. Two years later, moreover, in response to another request for advice from Abbott, the FDA’s Dr.

Katz once again explained that “the data do not provide sufficient evidence to support [Depakote] labeling changes at this time.” Indeed, the FDA would only approve Abbott’s proposed developmental delay warnings in 2011—some six years after Abbott first suggested them. Given, then, that as of 2008 the FDA did not believe the state of the data supported a developmental delay warning, it stands to reason that as of 2003, with even less data to go on, the FDA would similarly have rejected a developmental delay warning—even a CBE, judged as that is by the same standard as a PAS, see 21 C.F.R § 314.70(c)(6)(iii)(A).

It would accordingly have been impossible for Abbott, prior to M.B.D.’s injury in 2003-2004, to comply with Ohio’s product liability law to provide warnings about developmental delay while respecting the FDA’s apparent unwillingness—and their authority not—to accept them. Under Wyeth, that is enough for federal law to preempt a failure-to-warn claim raised under state law. Rheinfrank raises a series of objections to this conclusion, but none is persuasive.

Rheinfrank first contends that the evidence Abbott provided showing that the FDA would have rejected any change to Depakote’s label was too informal to be binding, and thus, under Wyeth, was too little and too “fleeting” to constitute clear evidence that the FDA would not have accepted a proposed label change, under either the CBE or PAS regime for label modification. But the Court in Wyeth did not say that for evidence to be clear it must result from a formal procedure of approval or disapproval.

Indeed, to require as much would appear to require rewriting the Court’s chosen test—from whether “the FDA would not have approved a change” to a drug’s label under a CBE or PAS to whether the FDA had not approved it. Wyeth, 555 U.S. at 572, 129 S.Ct. 1187 (emphasis added).

As the Court has since clarified, given the FDA’s authority to rescind any unilateral CBE, all that Abbott need have done—and did do here— is show that “the FDA would have rescinded any' change in the label,” Mensing, 564 U.S. at 624 n.8, 131 S.Ct. 2567, a showing that does not appear to exclude the kind of informal communications from FDA higher-ups that Abbott provided.

Furthermore, Rheinfrank misreads Wyeth for her claim that unless communications between a drug manufacturer and the FDA occur under formal channels, they would be too “fleeting” and insufficiently “final” to qualify as “clear evidence.” As the Court instead explained, the real problem with Wyeth’s claim—that “the FDA intended to prohibit it from strengthening the warning about” its pharmaceutical—was that there was “no evidence in [the] record that the FDA or the manufacturer gave more than passing attention to the issue” surrounding the proposed warning.

Wyeth, 555 U.S. at 572, 129 S.Ct. 1187 (emphasis added and internal quotation marks omitted). Here, by contrast, there is considerable evidence in the record showing that Abbott followed up on new data surrounding developmental delay and Depakote, that Abbott contacted the FDA multiple times to propose modifications to reflect that data, and that each time officials from the FDA unit responsible for reviewing those modifications unequivocally stated that they were inappropriate at the time.

As these communications reveal more than the “passing attention” that the Court agreed left Wyeth’s claim wanting, and as all make clear that the FDA would have rejected the proposed change to Depakote’s label, they are clear enough to satisfy Wyeth's standard.

Rheinfrank nevertheless contends that, even if the FDA signaled that it would not have approved a stronger developmental delay warning on Depakote’s label, a finding of “clear evidence” would still be precluded, either because Abbott misrepresented the state of the evidence to the FDA in its communications about its proposed labelling changes, or because there was evidence about Depakote’s effects on developmental delay that Abbott should have known but failed to obtain. Each of these' arguments, however, is unavailing.

First, even if the evidence that Abbott submitted was as misleading or incomplete as Rheinfrank alleges, the evidence in the record establishes that the FDA undertook its own review of the relevant empirical literature apart from Abbott’s various analyses. It was that review, moreover, that ultimately led Dr.

Katz to conclude “that [the FDA was] not yet ready to ‘sign off on labeling language’ ” concerning developmental delay,” and that the FDA “would not take action [on Abbott’s submitted CBE] until the review of the data was complete.” Thus, whatever may have been Abbott’s omissions or misrepresentations in its communications with the agency, the FDA clearly explained that, in its own view of the data, the developmental delay warning that Abbott had proposed was not yet warranted. As explained, that evidence is also clear enough to satisfy Wyeth’s standard.

Rheinfrank further argues that Wyeth makes relevant to a “clear evidence” inquiry not just what Abbott knew but also what it should have known, and that under that standard Abbott fell well short of its responsibility under federal labeling law by refusing to fund or conduct studies probing Depakote’s effects on developmental delay. But this argument is too conjectural to defeat preemption.

As the Court has explained, speculation as to what “a third party or the Federal Government might do” that would “make[ ] it lawful for a private party to accomplish under federal law what state law requires of it” cannot thwart a claim of preemption, as evidence of that kind would make “most conflicts between state and federal law illusory,” and thus “render[ ] conflict pre-emption all but meaningless.” Mensing, 564 U.S. at 620-21, 131 S.Ct. 2567.

That, however, is exactly what Rheinfrank asks this court to accept: a series of speculations as to what the FDA could have done with different evidence that Abbott might have collected if it had run its own studies. Because such speculations are not enough to undermine the clear evidence that the FDA would have rejected a strengthened warning on Depakote’s label prior to M.B.D.’s injury, Rheinfrank’s failure-to-warn claim is preempted by federal law. V. We therefore affirm the district court’s grant of

summary judgment and its denial of a new trial. . Abbott Laboratories separately incorporated its pharmaceutical business as "AbbVie” in 2013. Both are named defendants, and for convenience we refer to them as ‘‘Abbott Laboratories” or "Abbott” in this opinion. .

Under this court’s standard for reviewing a district court’s refusal of a jury instruction, such a decision is "reversible error if (1) the omitted instruction is a correct statement of the law, (2) the instruction is not substantially covered by other delivered charges, and (3) the failure to give the instruction impairs the requesting party's theory of the case.” Decker, 770 F.3d at 396 (internal quotation marks and citation omitted). .

The “Adequacy of Prescription Drug Warning” instruction reads: You may find a warning to be unreasonable, hence inadequate, in its factual content, its expression of the facts, or the method or form in which it is conveyed. The adequacy of such warnings is measured not only by what is stated, but also by the manner in which it is stated. A reasonable warning not only conveys a fair indication of the nature,of the dangers involved, but also warns with the degree of intensity demanded by the nature of the risk.

A warning may be found to be unreasonable in that it was unduly delayed, reluctant in tone or lacking in a sense of urgency. Rheinfrank’s requested Instruction No. 8 reads, in relevant part: Directions for use and warnings are inadequate if they are misleading based on what the manufacturer said, how the manufacturer said it, or what the manufacturer did not say. If the labeling for a prescription drug does not contain such adequate directions for use and warnings, the drug is misbranded.

Similarly, if a drug is dangerous to health when used in the dosage, manner, frequency, or duration prescribed, recommended, or suggested in the labeling, the drug is misbranded. The sale of misbranded prescription drugs is unlawful. .

As this court has explained these two labeling-modification regimes: After initial approval of a drug, branded-drug companies may seek modification of their labeling in two ways: first, through a "Prior Approval Supplement,” which requires submission to and approval by the FDA prior to distribution of the product, and applies to most labeling and other changes with "potential to have an adverse effect on the identity, strength, quality, purity, or potency of the drug product,” 21 C.F.R. § 314.70(b), and second, through a "Changes Being Effected” (CBE) supplement, which must be submitted 30 days before distribution, but does not require prior FDA approval. 21 C.F.R. § 314.70(c).

Label changes "[tjo add or strengthen a contraindication, warning, precaution, or adverse reaction” may be made through the CBE process. 21 C.F.R. § 314.70(c)(6)(iii)(A). Branded-drug companies, therefore, are free to update their labeling, subject only to subsequent FDA disapproval, Wyeth, 555 U.S. at 569, 129 S.Ct. 1187. Fulgenzi v. PLIVA, Inc., 711 F.3d 578, 581 (6th Cir. 2013) (footnote omitted). . Rheinfrank contends, based on one of Abbott's FDA contact reports dating from November 2005, that the FDA had in fact approved a CBE adding a developmental delay warning.

But as the district court explained, the FDA contact report in question—stating that the “FDA was in agreement with the latest wording provided to them regarding ... developmental delay”—appears to have been referring to a revision that Abbott made to its 2005 PAS earlier that summer at the FDA’s request. Indeed, the next sentence makes that clear: "[The FDA officials] noted that we [Abbott] had removed a sentence about developmental delay from the Patient Information Leaflet, and said that was OK.” .

Rheinfrank questions the admissibility of Abbott’s internal contact reports with the FDA, which supply the only evidence that Dr. Katz had twice made clear that the FDA did not think labelling about developmental delay was appropriate in 2008 or 2009. But Rhein-frank does not appear to have raised the evidentiary challenge below or in her opening brief, nor was it considered by the district court. As a result, those arguments must be deemed forfeited, at least absent some showing of a "gross miscarriage of justice.” Wiley v. United States, 20 F.3d 222, 226 (6th Cir. 1994).

Rheinfrank has not made that showing here. . Rheinfrank also challenges the admissibility of this evidence, but she does not appear to have raised this objection below or in her opening brief, nor was it considered by the district court. As a result, this challenge, too, must be deemed waived. See Wiley, 20 F.3d at 226; Kuhn v. Washtenaw County, 709 F.3d 612, 624-25 (6th Cir. 2013). . Because the district court properly denied Rheinfrank’s motion for a new trial, we need not reach her claim for punitive damages.