A.C. Ex Rel. J.C. v. Shelby County Board of Education

711 F.3d 687, 27 Am. Disabilities Cas. (BNA) 1339, 2013 WL 1285578, 2013 U.S. App. LEXIS 6426

• Court: Court of Appeals for the Sixth Circuit
• Decided: April 1, 2013
• Docket: 11-6506
• Status: Published

Opinion

OPINION

ALICE M. BATCHELDER, Chief Judge.

A.C., a minor with Type 1 diabetes, her mother, J.C., and her father, B.C., are the Plaintiffs-Appellants in this case. A.C. attended Bon Lin Elementary School, a school governed by Defendant-Appellee Shelby County Board of Education (“SCBE”). After three years of wrangling between Bon Lin officials and Plaintiffs over Plaintiffs’ requests for certain disability accpmmodations for A.C., things came to a head near the beginning of AC.’s second-grade year when Bon Lin’s princi pal twice made reports to Tennessee’s Department of Children’s Services alleging that J.C. and B.C. were medically abusing A.C. Plaintiffs filed suit in federal court soon after, arguing that the principal’s reports were made in retaliation to their disability accommodation requests and thus violated the Rehabilitation Act (“Section 504”) and the Americans with Disabilities Act (“ADA”). The district court found that Plaintiffs had failed to prove a prima facie element of their case and, even if they had, they could not prove that the reasons given by SCBE for making the child-abuse reports were a pretext for retaliation. It therefore granted summary judgment to SCBE.

On appeal, Plaintiffs argue that the district court misconstrued the prima facie analysis and failed to properly accord Plaintiffs factual inferences that, taken together, could lead a reasonable jury to rule for them. Because Plaintiffs are essentially correct, we REVERSE the grant of summary judgment and REMAND for further proceedings.

I

A

In Type 1 diabetes, the body does not manufacture insulin. Unlike Type 2 diabetes, which is often the result of exercise or eating habits and is characterized by the body’s resistance to insulin, Type 1 “is not a lifestyle disease.” Currently, it can only be managed, not cured. Management consists of balancing between having too much blood sugar (hyperglycemia) and too little blood sugar (hypoglycemia). Because management is an inexact science (particularly for children) and because the dangers posed by low glucose levels are generally greater than those posed by high levels (again, particularly for children), A.C.’s treatment plan for her Type 1 diabetes, like many children’s treatment plans, was designed to err on the side of higher levels.

To help manage A.C.’s diabetes, her parents obtained state-of-the-art monitoring equipment that took automatic readings of A.C.’s glucose levels every five minutes. When her levels were out of balance, the equipment’s sensor wirelessly communicated that to a seashell-shaped insulin pump physically attached to her hip via an IV-like arrangement. The pump could supply a dose of insulin as needed and would beep to warn A.C. and her caretakers to intervene if more attention was required, particularly when A.C.’s blood sugar was getting too low.

Though state of the art, the equipment alone was not sufficient to keep A.C.’s glucose levels in balance. The sizeable glucose fluctuations that are inherent in Type 1 diabetes made several supplementary measures necessary. For instance, around mealtimes, it was common to give A.C. a “bolus” shot of insulin — that is, a larger amount of insulin to compensate for the glucose rush caused by the meal. Also, properly calculating the needed amount of insulin required measuring each gram of carbohydrate intake because that intake determined glucose levels, which in turn determined the needed insulin amounts. Crucial to this case, the precise amount of carbohydrates was all that mattered in this calculation; the source, i.e., whether they came from candy or vegetables, was irrelevant, and high levels of carbohydrates were acceptable so long as they were counteracted by high levels of insulin. In addition to counting A.C.’s carbohydrates, J.C. and B.C. manually checked A.C.’s glucose levels often, “monitoring and adjusting her [treatment] in a very close fashion” just to be sure all was as well as it could be. For even with careful carbohydrate counting and state-of-the-art monitoring, some things were just uncontrollable. As an example, it was common and essentially unavoidable that A.C.’s levels would be high in the morning after breakfast but then dramatically lower by late morning.

The relationship between A.C.’s parents and Bon Lin officials started off on the wrong foot and, unfortunately, never seemed to improve. Just before A.C. began kindergarten at Bon Lin in 2007, J.C. made a series of requests for accommodations for A.C., including that Bon Lin retain a full-time nurse at the school to help with diabetes management, that it make A.C.’s classroom a peanut-free zone due to A.C.’s peanut allergy, and that A.C.’s blood tests — several each day — be done in A.C.’s classroom rather than in the school clinic. Bon Lin’s principal, Sharon “Kay” Williams, had never faced similar requests in her many years as a teacher or principal and found the situation quite frustrating. On August 22, 2007, the day before A.C.’s parents were scheduled to meet with a team of school officials about the accommodation requests, Principal Williams expressed her frustration in a voice mail for Barbara Duddy, the head nurse over the nurses who provided care to SCBE students:

Hey, Barbara. I know we’re having a meeting tomorrow about [A.C.]. This is Kay Williams from Bon Lin. [J.C.] is here causing all kinds of confusion and [A.C.’s teacher] has already broken down and cried. This woman is out to lunch. My teacher had ten minutes for lunch because she’s trying to make sure there are no peanut people by her, and now she claims the kid did sit by her with peanut butter. I mean, yet she doesn’t want the child sitting at another table because she doesn’t want her singled out. I don’t know what to do with this lady anymore. She does not reason or have any common sense. So you know that since I am the one with common sense, I am going to have a little problem with her. But at any rate, love ya, and I’ll see you tomorrow unless you want to call.

The tension between Principal Williams and J.C. was not helped by the fact that this message was accidentally left not on Nurse Duddy’s voice mail but on J.C.’s. The August 23rd meeting started and concluded with apologies from Principal Williams.

Principal Williams was not the only frustrated one. On August 21, the day before the misdirected voice mail, J.C. filed a complaint with the United States Office of Civil Rights (“OCR”), alleging that Bon Lin officials were not responding appropriately to her requests for accommodation. OCR intervened, with the result that SCBE provided almost everything J.C. had requested, including a full-time nurse at Bon Lin and language in the school’s coordinated school health plan regarding the training of teachers on managing Type 1 diabetes. SCBE did not agree to accommodate J.C.’s request that A.C. be manually tested four times a day in her classroom instead of in Bon Lin’s clinic. This remained a point of tension through A.C.’s second-grade year.

Another point of tension came from the nurses who provided A.C.’s care. They were not SCBE employees but were contracted out to SCBE from their employer, the Shelby County Health Department. Normal procedure called for Health Department nurses to write an Individualized Health Plan (“IHP”) each school year for every disabled student under their care. Instead, and with Bon Lin’s acquiescence, J.C. wrote her own IHPs for A.C. The nurses, though, were not pleased, citing both concerns for A.C.’s health and for their own liability exposure. Two nurses who were assigned to Bon Lin eventually quit because of these concerns, and a third- — Constance Brown, A.C.’s nurse dur ing first and second grade — threatened to do the same.

Despite these tensions, both Bon Lin officials and the nurses worked very hard to care for A.C. In addition to their regular teaching duties, A.C.’s teachers closely monitored her glucose levels in the classroom, and both Nurse Brown and Amy Carver, AC.’s second-grade teacher, kept a daily log of the levels. When AC.’s pump would warn that her levels were too high or too low, which happened about five times per day and sometimes up to ten times in a day, her teachers would immediately contact the school nurse. Once, A.C. had an incident at home where her blood sugar dropped so low that she had to receive glucagon (a shot that quickly raises blood sugar) and be hospitalized for several days. When that happened, AC.’s first-grade teacher, Lauren Richardson, was quick to note her absence and to check in on her condition until she returned. Teachers and school officials stayed in close contact with J.C. and B.C., often meeting with them personally to address both health and academic issues (A.C. was markedly behind academically).

A.C.’s parents also remained closely engaged with A.C.’s care at school. For instance, B.C. himself personally came to school to train A.C.’s nurses on using the pump, provided them the pump’s manual, and arranged an in-person instructional session between a representative from the pump’s manufacturer and the nurses. The parents also stayed engaged on a personal level. Bon Lin’s Assistant Principal recalled their regularly attending school functions, field trips, and parties with A.C., and testified that they always kept her clean and dressed very nicely.

Despite this engagement, some of A.C.’s teachers began to worry that more than just diabetes was causing A.C. harm. First-grade teacher Richardson, still untrained about Type 1 diabetes, saw A.C. ■with candy and cookies and drew the conclusion that AC.’s parents were committing medical abuse by failing to feed her appropriately. Richardson shared this belief with Principal Williams, and indicated that she wished to file a report with DCS. To preserve the school’s relationship with AC.’s parents, Principal Williams instructed Richardson to stay quiet. Similarly, Carver nervously shared stories of A.C.’s in-class glucose fluctuations with her paramedic husband. Her fears were not assuaged by her husband’s response that AC.’s parents were lucky A.C. was still alive.

By the start of second grade, A.C.’s third year at Bon Lin, the relationship between A.C.’s parents and SCBE officials continued to be rocky. The school year began with SCBE officials floating, and beginning to implement, the idea of putting A.C. into “homebound services”— which meant removing her from Bon Lin. For reasons that are not apparent from the record, this idea went nowhere, and AC.’s parents did not hear about it until after litigation commenced.

J.C. also strained old tensions with her second-grade IHP for A.C. (which was tardily delivered in late September instead of mid-August), in which she resubmitted the request that A.C.’s blood tests be done in her classroom instead of the clinic. Her renewed request was based on the perception that, because sick children all go to the clinic, A.C. would be more likely to get sick if she were tested there. In response, Bon Lin set up a Section 504 1 meeting for late October to discuss the new IHP.

October 2009 is the crucial month for this case, particularly toward the month’s end. All through October, A.C.’s glucose levels continued to fluctuate frequently and significantly, often outside A.C.’s target glucose range. Early in the month, on October 6, Principal Williams attended a meeting of SCBE principals that dealt with Tennessee’s statutory duty to report child abuse (the “Principals’ Meeting”). The county district attorney appeared via video, informing school officials of their legal duty to report suspected child abuse and advising that there would be no adverse consequences for a mistaken report. He also warned of the consequences of failing to report, highlighting as a cautionary tale the experience of another Tennessee principal who was indicted for not reporting sexual abuse. Principal Williams was already aware of her statutory duty and had made reports to DCS several times in her career.

Then came the last week of October. On the morning of Tuesday the 27th, B.C. met with Nurse Brown and reiterated the classroom-testing request. Later that day, A.C.’s parents and school officials held a Section 504 meeting to discuss A.C.’s academic progress. The officials raised the idea of testing A.C. for a learning disorder, to which her parents objected. That afternoon, Assistant Principal McClellan contacted Nurse Duddy and SCBE Health Administrator Brown-Woods with data on A.C.’s glucose fluctuations, asking for treatment guidance. McClellan again broached the subject of putting A.C. into homebound services.

The next day, Wednesday the 28th, J.C. sent a note to Nurse Brown emphasizing that blood tests should be done in the classroom, not the clinic. In a kind of mute reply, Nurse Brown wrote in her nursing log that “students in [A.C.’s] class may be ill” and “If such a problem, may need to reconsider school itself!” She also noted that she asked J.C. for permission to directly contact A.C.’s endocrinologist, and that her request was denied.

On Thursday the 29th, B.C. met with McClellan for half an hour and repeated the classroom-testing request, this time adding that A.C. had an “open wound” that could contribute to her susceptibility to illness at the clinic. McClellan discussed that request with Nurse Brown, with whom he had also discussed the prior two days’ requests.

Friday the 30th was the fateful day. It started with a common occurrence: A.C.’s blood sugar was low, leading Carver to call in Nurse Brown to check on A.C. Nurse Brown found that A.C.’s level was indeed low, and remarked to Carver that they were fortunate A.C. had not passed out. Here, the parties’ accounts of October 30th diverge.

Principal Williams and Nurse Brown testified that in response to Nurse Brown’s remark, Carver began to hyperventilate, crying in the hallway and attracting a crowd of nearby teachers; she was finally taken to Principal Williams’s office to be calmed. There, Carver, who was then pregnant, expressed not only her fear for AC.’s health, but also her fear that the stress of caring for A.C. might harm her own unborn baby. Even though it was still morning, Principal Williams saw Carver was inconsolable and sent her home.

While Carver testified to a very similar occurrence, she did not recall its taking place on October 30th and instead recalled different facts about that date. Carver remembers having a pleasant exchange with J.C. via e-mail on the 29th that let J.C. know about a classroom Halloween party on the afternoon of the 30th, attending the afternoon party on the 30th, and cleaning up afterwards. She could not recall anything unusual happening that day.

A month after her deposition, Carver filed an errata sheet adding the substantive allegation that she “hyperventilated” on the morning of the 30th. She later filed a declaration explaining that, after her deposition, she had reviewed Nurse Brown’s notes and recalled that, in fact, the hyperventilation episode occurred on the 30th.

What makes this dispute important is that, in SCBE’s chronology, the hyperventilation episode was the event that pushed Principal Williams into making a DCS Report. By contrast, Plaintiffs argue that what really caused the DCS Report was that the repeatedly renewed requests for classroom testing caused simmering tensions to boil over.

Setting aside the hyperventilation dispute, what happened next on October 30th is indisputable: Principal Williams sent out an e-mail to SCBE Superintendent Burton, Angela Hargrave (SCBE’s Director of Student Services), and Shunji Brown-Woods (SCBE’s Director of Coordinated Health) at 10:18 AM. It stated that Carver was “having an anxiety attack from the constant harassment of [AC.’s parents],” that Carver was worried about A.C.’s “roller coaster [glucose] levels,” and that AC.’s parents were “not monitoring [A.C.] at home.” The e-mail also noted that A.C. “comes in [to school with] very high [glucose] and then crashes.” Williams concluded the e-mail by asking for guidance on how to proceed and advising that she was “ready to report the family to DCS” because “[w]e care about this child and the parents do not.”

All three recipients answered promptly via e-mail. Brown-Woods responded first, recommending a DCS report. She based her recommendation on A.C.’s “erratic” fluctuations, SCBE’s discomfort with the IHP, and the perception that A.C. was “in danger” because she was “being poorly managed.” Brown-Woods also referenced concerns about being able to retain Nurse Brown, and noted that SCBE had “lost two previous nurses as a result of working with this family.” Superintendent Burton then concurred, though he emphasized that Principal Williams must “make sure there is documentation” to back up her allegations. Hargrave followed up last, sharing that she had contacted a DCS supervisor about the situation (albeit without naming A.C. or her parents), and the supervisor said that DCS “would pull this report and make certain that it is thoroughly investigated.”

Minutes after receiving the last of the responses, Principal Williams called DCS. She identified A.C.’s parents as abusers and provided two basic rationales for the allegation of medical child abuse. First, she said that A.C. was being “sent to [Bon Lin] with cookies, Kool-Aid, and candy, which makes her sugar shoot up to a very high level.” Second, she reported that A.C.’s “diabetes is not being monitored at home, and [that she had] documentation on this neglect.” Principal Williams emphasized that she worried the abuse could be “fatal” and that two previous nurses had left in fear that A.C. would die, with the third nurse now also thinking of leaving for the same reason.

Ten days later, on November 9, Principal Williams escalated her allegations considerably. By this point, DCS had already initiated its investigation by going to A.C.’s house repeatedly, questioning her parents, and interviewing her privately. A.C.’s parents were terrified, immediately hiring a lawyer and sending a request to their endocrinologist that he confirm that they were not medically abusing their daughter. A DCS investigator then contacted Principal Williams, who alleged, inter alia, that “the family wants something horrible to happen to [A.C.] at school.” She explained that A.C. “could die at school,” and that “the parents are just looking for a law suit.” Both Principal Williams and Assistant Principal McClellan testified that they believed, and had discussed, exactly that view of A.C.’s parents: that the parents genuinely wanted something “horrible” to happen to A.C. at school so they could file a lawsuit.

DCS continued its investigation, doing more at-home meetings with A.C. and her family, running background checks on both J.C. and B.C., obtaining A.C.’s medical records from her doctors, interviewing A.C. and each of her siblings privately, and incorporating an assistant attorney general into the investigation. Early on in the process, DCS officials recognized that J.C. and B.C. “appear[] to have great knowledge of [Type 1 diabetes] and how to care for it.” They concluded the inquiry on December 17, closing the case based on their finding that the “medical maltreatment” allegation was “unfounded.”

Just after the end of A.C.’s second-grade school year, Plaintiffs filed this lawsuit, alleging that the DCS Reports were made in retaliation for the attempts of A.C.’s parents to secure accommodations for A.C.’s disabilities.

B

The parties agreed to proceed before a magistrate judge. After discovery, SCBE filed for summary judgment, which the magistrate (hereafter “district court”) granted.

In the portion of its analysis relevant to this appeal, the district court focused entirely on the DCS Reports. 2 Using the McDonnell Douglas burden-shifting analysis, see McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), the district court held that Plaintiffs met their burden on the first two elements of a prima facie retaliation claim by showing that (1) their requests for accommodation were “protected actions” under the Rehabilitation Act and the ADA; and (2) SCBE clearly knew of those requests. But, the district court held, Plaintiffs failed to meet the third prima facie element because they did not prove that the DCS Reports were an “adverse act.” Relying on Cox v. Warwick Valley Central School District, 654 F.3d 267 (2d Cir.2011), the district court distinguished between child abuse reports that are made for protective purposes and those that are made for punitive purposes, treating only the latter as adverse. It derived from Cox the rule that, to prove that the making of a DCS report was an adverse action, a plaintiff must make a “clear showing of retaliatory or punitive intent.”

Further, following Cox, the district court found that state law substantially influenced its analysis. Tennessee requires its school officials immediately to report child abuse or neglect, Tenn.Code Ann. § 37-1-403, and clothes them with immunity for all good-faith reports, id. § 37-1-410(a)(5)(A). To strengthen that immunity, the statute presumes that all reports are made in good faith. Id. § 37-1-410(a)(5)(B). Quoting Bryant-Bruce v. Vanderbilt University, Inc., 974 F.Supp. 1127, 1139-41 (M.D.Tenn.1997), the district court held that a plaintiff can overcome the statutory presumption only by proving the defendant “acted in bad faith by clear and convincing evidence.”

Cobbling Cox and Tennessee law together, the district court found that — at the prima facie stage — Plaintiffs must show “by clear and convincing evidence” that “[Principal] Williams harbored a retaliatory or punitive intent when making the report.” The district court then summarily rejected the evidence that Plaintiffs provided of retaliatory intent, finding, for instance, that the record did not show that Principal Williams lied or that she ever stated that the parents “wanted to kill their child.” The district court agreed there was evidence that Nurse Brown was resistant to testing A.C.’s blood sugar in the classroom and that SCBE officials had been planning to put A.C. into the home-bound services program, but found that evidence “insufficient to overcome” the good-faith presumption.

The district court then considered SCBE’s evidence that Principal Williams had acted with a protective purpose: (1) the existence of Tennessee’s statutory reporting duty; (2) the Principals’ Meeting, where Williams was encouraged to make child abuse reports; (3) Richardson’s 2008 concerns with “medical abuse” and stated desire to make a report; (4) Nurse Brown’s repeated comments to Principal Williams about concerns with A.C.’s glucose fluctuations; (5) Carver’s similar concerns about fluctuations, including her husband’s “lucky to be alive” comment; (6) the events of October 30, where Carver began hyperventilating and had to be comforted by Principal Williams; and (7) alleged counsel from DCS that Hargrave passed on to Principal Williams that reporting was mandatory. The court concluded that Plaintiffs “fail[ed] to establish a prima facie case of retaliation.” In a paragraph at the conclusion of its opinion, the district court also held that, even if it had reached the pretext phase of the McDonnell Douglas analysis, Plaintiffs’ claims would have failed for the reasons stated in its prima facie analysis.

On appeal, Plaintiffs argue that the district court improperly forced them to rebut SCBE’s non-discriminatory rationale at the prima facie phase, that they met their prima facie burden, and that at the pretext stage the district court failed to draw all factual inferences in Plaintiffs’ favor.

II

We review de novo the district court’s grant of summary judgment. Spees v. James Marine, Inc., 617 F.3d 380, 388 (6th Cir.2010). Summary judgment is proper where the moving party shows both that there are no genuine issues of material fact and that it is entitled to judgment as a matter of law. Id. (citing Fed.R.Civ.P. 56(c)(2)). In reviewing this case, we must view the facts and all reasonable factual inferences in Plaintiffs’ favor. Id. (citing Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986)). Summary judgment should be reversed “if ‘there is a genuine need for trial,’ which turns on ‘whether the evidence is such that a reasonable jury could return a verdict for’ ” Plaintiffs. See Bazzi v. City of Dearborn, 658 F.3d 598, 602 (6th Cir.2011) (quoting Weigel v. Baptist Hosp. of E. Tenn., 302 F.3d 367, 375 (6th Cir.2002)). Ultimately, the question before us is “whether the evidence presents a sufficient disagreement to require submission to a jury or whether it is so one-sided that one party must prevail as a matter of law.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 251-52, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986).

Both the ADA and Section 504 prohibit retaliation against any individual because of his or her opposing practices made unlawful by the Acts or otherwise seeking to enforce rights under the Acts. See, e.g., 42 U.S.C. § 12203 and 28 C.F.R. 35.134 (ADA); 29 U.S.C. § 794(a) and 29 C.F.R. § 33.13 (Section 504). The Acts have a similar scope and aim; for purposes of retaliation analysis, cases construing either Act are generally applicable to both. Andrews v. Ohio, 104 F.3d 803, 807 (6th Cir.1997); accord Burns v. City of Columbus, 91 F.3d 836, 842 (6th Cir.1996).

The only actions that Plaintiffs claim were retaliatory are the making of the DCS Reports. Their evidence is directed toward demonstrating retaliatory intent or motive in the making of those reports. Plaintiffs claim that the retaliation was directed against their requests for accommodations,- and primarily against the repeated requests for classroom testing made in the last week of October.

Plaintiffs do not present any direct evidence of retaliation, so their retaliation claim falls under the McDonnell Douglas burden-shifting approach. 3 DiCarlo v. Potter, 358 F.3d 408, 415 (6th Cir.2004). The initial burden falls on Plaintiffs to present a prima facie case of retaliation. Id. at 414. That requires them to establish that: (1) they engaged in activity protected under Section 504 and the ADA; (2) SCBE knew of this protected activity; (3) SCBE then took adverse action against Plaintiffs; and (4) there was a causal connection between the protected activity and the adverse action. See Gribcheck v. Runyon, 245 F.3d 547, 550 (6th Cir.2001). “The burden of establishing a prima facie case in a retaliation action is not onerous, but one easily met.” Nguyen v. City of Cleveland, 229 F.3d 559, 563 (6th Cir.2000).

Plaintiffs’ successful carrying of their easy burden would create a presumption that SCBE used the DCS Reports to retaliate against Plaintiffs for their protected conduct. DiCarlo, 358 F.3d at 414. To rebut this presumption, SCBE must show that it had a legitimate, non-discriminatory basis for the DCS Reports. Id. And if it can do so, the burden shifts back to Plaintiffs to “prove by a preponderance of the evidence that the legitimate reasons offered by [SCBE] were not its true reasons, but were a pretext for” retaliation. Id. at 414-15.

Viewing the facts and drawing all inferences in Plaintiffs’ favor, we conclude that the district court erred in granting summary judgment to SCBE. Its first error was the premature placing on Plaintiffs at the prima facie stage the burden of overcoming SCBE’s stated reasons for its actions. Its second error was the dismissal of Plaintiffs’ evidence of retaliatory motive as “insufficient” to outweigh SCBE’s countervailing proofs of motive. Plaintiffs provided evidence that could have permitted a reasonable jury to find that the DCS Reports were not made in good faith but were motivated by a desire for retaliation. “ ‘[C]laims involving proof of a defendant’s intent seldom lend themselves to summary disposition,’ ” and this case is no exception. See Center for Bio-Ethical Reform, Inc. v. City of Springboro, 477 F.3d 807, 823 (6th Cir.2007) (quoting Bloch v. Ribar, 156 F.3d 673, 682 (6th Cir.1998)). Since the question of retaliatory motive is one that a reasonable jury in this case could resolve against SCBE, summary judgment is not appropriate.

1. The Prima Facie Analysis

The parties do not dispute that Plaintiffs’ requests for accommodation, including the classroom-testing requests, were protected under Section 504 and the ADA. Both this circuit and most others agree that requests for accommodation are protected acts. See Bryson v. Regis Corp., 498 F.3d 561, 577 (6th Cir.2007); accord Baker v. Windsor Republic Doors, 414 Fed.Appx. 764, 777 n. 8 (6th Cir.2011) (noting that a “good-faith request for reasonable accommodations” is a “protected act” under the ADA). 4 The parties similarly do not dispute that SCBE was fully aware of Plaintiffs’ requests.

The sticking point is the third prima facie element — whether the making of the DCS Reports was “adverse action.” At first glance, this does not appear to be a difficult question. To be adverse, a retaliatory action must be enough to dissuade a reasonable person from engaging in the protected activity; “petty slights or minor annoyances” cannot qualify. Burlington N. & Santa Fe Ry. Co. v. White, 548 U.S. 53, 68, 126 S.Ct. 2405, 165 L.Ed.2d 345 (2006). Having a government official appear at their door armed not only with the power to take their disabled child away but also with allegations that they are actively and nearly fatally abusing that child would surely be enough to dissuade many reasonable parents from seeking accommodations at school. The type of intrusive investigation that Tennessee officials must conduct of such parents’ homes and children to complete a DCS investigation, see, e.g., Tenn.Code Ann. § 37-1-406, could be powerfully dissuasive in its own right and only cements that conclusion. Not surprisingly, then, this Court and other courts have treated the making of such reports as adverse. Jenkins v. Rock Hill Local Sch. Dist., 513 F.3d 580, 588-89 (6th Cir.2008) (listing “making a false report to Children Services” among the actions that would “chill a person of ordinary firmness” from engaging in protected activity); Weixel v. Bd. of Educ. of N.Y., 287 F.3d 138, 149 (2d Cir.2002) (finding that “threatening (and instituting) child welfare investigations in response to plaintiffs’ medically excused absences from school” constituted retaliatory adverse action under Section 504 and the ADA). Considering these factors, and the record evidence that A.C.’s parents were shaken by the whole situation, the actions of the Principal in making DCS Reports rose well above a petty slight or minor annoyance and constituted adverse action.

The district court, however, found a difference between “protective” DCS reports and “punitive” ones. Relying on the Second Circuit’s decision in Cox, the court required Plaintiffs — at the prima facie stage of the case — to disprove SCBE’s argument that Principal Williams made the DCS Reports protectively and in good faith.

There are a number of problems with this approach, but the most salient is that the district court apparently misunderstood the law. Cox was a First Amendment retaliation case, not an ADA/Section 504 retaliation case, and so did not involve the McDonnell Douglas burden-shifting approach. See 654 F.3d at 270. Thus, Cox never suggested that, for purposes of clearing the low bar set for a prima facie showing of retaliation, making a DCS report could not constitute adverse action. And by incorrectly pulling Cox out of its legal context and “eonsider[ing] [SCBE’s] alleged nondiscriminatory reason ... when analyzing the prima facie case,” the district court ignored this Court’s clear instructions against “bypass[ing] the burden-shifting analysis.” Wexler v. White’s Fine Furniture, Inc., 317 F.3d 564, 574 (6th Cir.2003) (en banc). The district court should rather have examined Plaintiffs’ prima facie proofs “independent of the nondiscriminatory reason ‘produced’ by the defense as its reason” for the DCS Reports. See Cline v. Catholic Diocese of Toledo, 206 F.3d 651, 660 (6th Cir.2000). SCBE may raise Cox in its defense, but at the pretext stage, not the prima facie stage. Id. at 660-61. For the same reason, the district court erred in relying — -at the prima facie stage of the proceedings— on Tennessee law governing the duty to report child abuse. In short, we conclude that Plaintiffs adequately established the adverse-action element of their prima facie case.

The last element of the prima facie case turns on whether there is a causal connection between Plaintiffs’ protected activities and the allegedly retaliatory DCS Reports. Again, at the prima facie stage, Plaintiffs’ burden is minimal, requiring merely that they “put forth some evidence to deduce a causal connection between the retaliatory action and the protected activity.” Nguyen, 229 F.3d at 566. This evidence must be sufficient to allow “an inference ... that the adverse action would not have been taken had the plaintiff’ not engaged in protected activity. Id. at 563. Temporal proximity can often help meet this causal burden, id., and where the adverse action comes “very close in time” after the exercise of protected activity, “such temporal proximity ... is significant enough” to meet the burden alone, Mickey v. Zeidler Tool & Die Co., 516 F.3d 516, 525 (6th Cir.2008). The evidence in this case, including both temporal proximity and evidence calling into question the truthfulness of the reports themselves, is sufficient to meet Plaintiffs’ minimal burden to put forth evidence permitting a reasonable inference of causation.

Plaintiffs argue that the short time span between their requests for classroom glucose testing and the initial DCS Report helps show causation. Even though the classroom-test requests were hardly new at the end of October 2009, we agree that the temporal proximity here helps permit an inference of causation. Plaintiffs first made the request at the beginning of the 2007 school year, and Bon Lin both denied the request and enacted a school policy that made the clinic the presumptive place for testing, albeit permitting some degree of variance. But, if reasonable factual inferences are made in favor of Plaintiffs, the classroom-test-request issue ceased its long-term simmering and boiled over in late October, when the school held a meeting to discuss the IHP’s renewal of the request, J.C. and B.C. both made repeated classroom-test demands several days in a row, the school apparently resisted those demands, and Nurse Brown noted in response to the demands that A.C. should perhaps stop attending Bon Lin. Then, the day after B.C. made the last request in a face-to-face meeting with Assistant Principal McClellan, and just after Principal Williams sent an e-mail to SCBE officials referencing the parents’ “constant harassment,” Williams reported the parents to DCS. We consider the temporal proximity between that report and the request-based tensions evidence that permits an inference of causation.

Plaintiffs do not rely on temporal proximity alone, though. They also argue that the fundamental elements of Principal Williams’s DCS Reports are false, and that this falsity shows retaliatory intent. In her initial DCS Report, Principal Williams made two basic factual allegations: first, that A.C.’s parents were abusing A.C. by sending her to school with “cookies, Kool-Aid, and candy”; and second, that the parents were committing potentially “fatal” neglect in monitoring A.C.’s diabetes (which Principal Williams claimed she had “documentation” to prove). Principal Williams’s follow-up DCS Report — during the course of an active child-abuse investigation that she had initiated — went on to allege that the parents “want[ ] something horrible to happen to [A.C.] at school,” that they “are just looking for a lawsuit,” and that they thus did not care that “the child could die at school.”

Plaintiffs argue not only that all of these allegations are false or based on a false view of Type 1 diabetes, but also that Principal Williams ought to have known (or even did know) that they were false. First, Plaintiffs point out that someone with Type 1 diabetes can enjoy cookies and sweets just as any other person can so long as the sugar is measured and compensated for with insulin, which Plaintiffs say the record shows they did. Principal Williams had received mandatory diabetes training, and Plaintiffs argue that this training informed her that, unlike the way Type 2 diabetes can be managed with “improved nutrition choices,” Type 1 diabetes “can’t be prevented and has nothing to do with lifestyle choices.” Thus, Plaintiffs’ conclude, Williams should have known that A.C.’s being permitted to consume sweets was not evidence of abuse. Further, Williams had a duty to train A.C.’s teachers with the same diabetes materials that she had received. Given that Richardson was the source of the allegation that A.C.’s parents “abusively” gave A.C. sweets, Plaintiffs argue that if Williams had fulfilled her duty, then Richardson would have known better than to consider sweet snacks the equivalent of medical abuse. Richardson admits that she never received training on diabetes.

Next, Plaintiffs point to several reasons why Principal Williams’s allegation that the parents were not monitoring A.C.’s diabetes was untrue. Most obviously, A.C. had a state-of-the-art, around-the-clock monitoring system literally attached to her body. As Plaintiffs note, this system was in constant use: pump logs from October 2009 alone show that A.C.’s glucose levels were monitored thousands of times automatically, and hundreds more times manually. Plaintiffs also point out that Principal Williams even admitted that her statement to DCS that she could document the alleged lack of monitoring was untrue.

Finally, Plaintiffs argue that Principal Williams’s last allegation — that A.C.’s parents wanted something “horrible” to happen to A.C. at school, even including death, so they could sue the school — is indefensible. They argue that Principal Williams knew from personal experience that A.C.’s parents regularly attended Section 504 meetings, requested accommodations for A.C., sought protection from the Office of Civil Rights for A.C., and refused to settle for means of earing for A.C. that might leave her vulnerable. Plaintiffs further note that Principal Williams also knew A.C.’s parents obtained a state-of-the-art monitoring system for A.C. and went out of their way to personally train Bon Lin nurses on how to use it and to provide professional follow-up training. And they point out that the record also shows that A.C.’s parents came to her school functions and kept her both clean and very neatly dressed. Simply put, they contend, there is no room in the record to conclude that A.C.’s parents were anything but caring toward A.C. For its part, SCBE does not try to defend Principal Williams’s last allegation.

Relying on Nguyen, 229 F.3d at 563, SCBE responds that so long as Principal Williams would have made a DCS Report regardless of the accommodation requests, Plaintiffs cannot establish a causal connection between those requests and the DCS reports. But this response simply does not negate the fact that Plaintiffs have carried their minimal burden to show evidence of causation at this stage of the proceedings. What drove Principal Williams to make the DCS Reports is not only a complicated question of fact that, on this record, must be resolved by a jury, it is also the question of fact on which this case turns. See, e.g., Center for Bio-Ethical Reform, 477 F.3d at 823-24 (concluding that where there were genuine issues of material fact regarding whether the defendants would have taken the same action absent the plaintiffs’ protected activity, summary judgment was inappropriate). Indeed, the jury will consider not “a” hypothetical DCS Report, but rather the DCS Reports that Principal Williams actually made, reports which did not focus on alerting DCS officials to the glucose fluctuations that SCBE so emphasizes on appeal but rather pinned the blame for those fluctuations on A.C.’s parents’ alleged neglect and active misconduct, allegedly born in part out of a desire to see A.C. harmed. It would be permissible for a reasonable jury to conclude that those reports would not have been made inevitably, without the accommodation requests.

Plaintiffs’ evidence of falsity in the reports of abuse coupled with the temporal proximity of those reports to Plaintiffs’ requests for accommodations is sufficient to permit an inference of causation. Because Plaintiffs established all four elements of their prima facie ease, the district court erred in ruling otherwise.

2. SCBE’s Non-Retaliatory Reason for the DCS Reports

As noted above, the district court held that even if Plaintiffs had made out a prima facie case of retaliation, they could not sustain their burden of rebutting SCBE’s non-retaliatory reasons for making the DCS Reports, of which the court identified seven. On appeal, SCBE repeats those seven, with an emphasis on Nurse Brown’s glucose-fluctuation concerns. It notes that the fluctuations had been recorded by the school, were very often outside A.C.’s prescribed target ranges, and that even A.C.’s endocrinologist agreed they were causes for concern.

SCBE also adds three supplementary proofs to support the DCS Reports of child abuse. First, J.C. and B.C. refused to allow the nurses to contact A.C.’s endocrinologist directly. Second, Richardson was aware that A.C.’s blood sugar once fell so low while she was at home that she had to receive a glucagon shot and be hospitalized. Third, A.C.’s parents failed to send flow-chart information about A.C.’s glucose levels into school each morning, as the original IHP called for.

Altogether, SCBE has presented ten reasons for making the reports to DCS. Broadly, these fall into three categories: first, reasons related to the duty to report itself (i.e., the Tennessee statute, the Principals’ Meeting); second, reasons relating to A.C.’s health generally (i.e., the presence of glucose fluctuations); and third, reasons relating to J.C. and B.C.’s treatment of A.C. (i.e., refusing to allow direct contact with the doctor, etc.). SCBE has thus carried its burden of articulating a non-retaliatory basis for the DCS Reports.

3. Plaintiffs’ Pretext Response

The burden therefore shifts back to Plaintiffs to prove pretext. They can do so by demonstrating, by a preponderance of the evidence, DiCarlo, 358 F.3d at 414-15, that SCBE’s reasons (1) lack a basis in fact, (2) did not actually motivate the DCS Reports, or (3) were insufficient to motivate the DCS Reports. See Vincent v. Brewer Co., 514 F.3d 489, 497 (6th Cir.2007). The district court concluded that, based on the same analysis it used for the prima facie phase, Plaintiffs failed to show that SCBE’s non-retaliatory reasons were pretextual.

Plaintiffs begin by taking aim at SCBE’s general reasons, the most central of which was that A.C.’s glucose levels fluctuated often and dramatically. But this alone, Plaintiffs say, could not have motivated the DCS Reports. First, concerns about a child’s struggle with a genetic illness do not justify the logical leap to the conclusion that those struggles are the result of child abuse. Notably, Principal Williams’s DCS Reports did not focus on the fluctuations but rather alleged that A.C.’s parents were committing specific acts of misconduct to cause the fluctuations, which could certainly imply that Williams recognized that fluctuations alone could not justify her Reports. Indeed, Nurse Brown, the individual who primarily raised the fluctuation concerns, never recommended making a report to DCS. Second, dramatic fluctuations are the calling card of Type 1 diabetes. Any management system will have difficulty controlling them and they are naturally more pronounced and less controllable for young children. Thus, Plaintiffs conclude, it would be irresponsible to report parents as child abusers just because their child experiences what is sadly typical for the disease. 5 For the same reasons, Plaintiffs argue, SCBE’s claimed reliance on Carver’s and Nurse Brown’s concerns about glucose fluctuations is not sufficient to motivate the Reports.

For the October 30 incident, though, where the fluctuations allegedly caused the Carver-hyperventilation incident that — according to SCBE — triggered the DCS Reports, Plaintiffs have a more specific counter-argument: the incident never happened. They point to Carver’s inability at her deposition to recall anything unusual that happened on the 30th, and her testimony that she was at school for a Halloween party that afternoon in her classroom. This testimony casts doubt on Principal Williams’s account that Carver had hyperventilated on the morning of the 30th, created a scene in the hallway that distracted other teachers from their classrooms, had to be comforted in Williams’s office, and then was sent home. Carver tried to supplement her testimony with a substantive addition on the errata sheet and an affidavit explaining that she had not recalled at her deposition that her hyperventilation incident had in fact occurred on the morning of October 30. But this does not eliminate, at the summary judgment stage, the factual issue that has been raised by Carver’s original deposition testimony. Just as “ ‘a party who has been examined at length on deposition [cannot] raise an issue of fact simply by submitting an affidavit contradicting his own prior testimony,’ ” Biechele v. Cedar Point, Inc., 747 F.2d 209, 215 (6th Cir.1984) (quoting Perma Research & Dev. Co. v. Singer Co., 410 F.2d 572, 578 (2d Cir.1969)), so also we think it clear that a party moving for summary judgment cannot extinguish a genuine issue of material fact simply by filing an errata sheet and affidavit to counteract the effect of previous deposition testimony. Of course, SCBE will have an opportunity at trial to present Carver’s live testimony, subject to cross-examination, and the jury will be able to evaluate any explanations that may be elicited or offered for the asserted incompleteness of Carver’s deposition testimony.

Next, Plaintiffs target the various proofs of misconduct or suspected misconduct by A.C.’s parents. They refused to allow Nurse Brown to contact A.C.’s doctor directly, Plaintiffs explain, because their friction with the school had made them hesitant to allow direct contact. Nor was this simply a trust issue: as A.C.’s primary caregivers, J.C. and B.C. wished to ensure that they were the primary point of contact for every matter concerning A.C.’s health. But, they say, they would have been willing to allow written questions to go through them from the nurses to AC.’s doctor, as that would have left them in the loop for every communication.

Regarding their failure to send in flow charts every morning, Plaintiffs point out that the 2009 IHP deleted the requirement for those charts. The reason for this deletion was, as B.C. explained to Assistant Principal McClellan on October 29, that A.C.’s blood sugar was invariably high in the mornings due to breakfast, rendering the charts useless. Thus, in October 2009 SCBE had no reason to be expecting the charts, much less suspecting child abuse from their absence.

Regarding Richardson’s allegation that A.C.’s parents were medically abusing A.C. by allowing her to have sweets, Plaintiffs refer to the argument they made before: this is a basic misconception of Type 1 diabetes that Principal Williams should have known was unfounded and about which Williams should have trained her teachers.

Plaintiffs also address the conversation between Angela Hargrave, SCBE’s Director of Student Services, and the DCS Supervisor. Hargrave presented A.C.’s situation to the Supervisor without naming A.C. or her parents and then reported the Supervisor’s response to Williams. Plaintiffs offer two reasons that this report should not be credited. First, Plaintiffs argue, Hargrave gave to the DCS Supervisor information that was inevitably tainted by the erroneous information (“parents don’t care for this child”; “parents aren’t managing A.C.’s diabetes at home”) that Principal Williams had given her. SCBE counters, and heavily emphasized at oral argument, that Hargrave merely informed DCS of a hypothetical situation where a child was experiencing glucose fluctuations like A.C.’s, and that on the strength of that hypothetical alone DCS told her that a child abuse report was mandatory. This leads to Plaintiffs’ second argument: the record does not clearly show either what Hargrave discussed with DCS or what information she delivered from DCS to Williams. Hargrave’s only testimony in the record is a declaration that does not state that Hargrave’s description of the situation to DCS was wholly limited to fluctuations. And while Hargrave’s declaration says that she “advised” Principal Williams that DCS had said Principal Williams was “obligated” to report, the only recorded communication on October 30 between Hargrave and Principal Williams was via e-mail. All Hargrave said in that e-mail was that DCS promised to investigate the report, not that Principal Williams was obligated to report. There is no evidence in the record that Hargrave contacted Principal Williams in any other way before Williams made her report. Given that it was only a matter of minutes between Principal Williams’s receipt of Hargrave’s e-mail and her call to DCS, a reasonable jury could doubt that any other contact occurred.

That leaves only the Tennessee statutory duty to report and the October 6, 2009, Principals’ Meeting discussing that duty. 6 The meeting’s discussion of the state law featured several report-motivating comments by the district attorney, including that failure to report abuse “is a crime under Tennessee law” and that teachers with suspicions of abuse should not “attempt to investigate” personally and instead should just report their concerns. Id. Plaintiffs point out, though, that this presentation specifically addressed only sexual or physical abuse (using as examples rape, sexual molestation, video recordings of a minor student engaged in sexual conduct, incest, visible extension cord marks on a child’s legs, and visible physical injuries of sufficient severity that they required medical attention), and never once addressed the more nuanced question of when to report parental “medical” abuse, and particularly not the yet more difficult issue of what constitutes medical abuse of a child with Type 1 diabetes.

There is another reason that a reasonable jury could discount reliance on either the statutory duty or the Principals’ Meeting as the impetus for the October 30, 2009, call to DCS: Principal Williams’s delay in making her report. Williams testified that she was aware of her reporting duty and that she had made a number of DCS reports over her twenty-two-year career. And at the October 6 meeting, the district attorney emphasized that, by law, reports must be made “immediately” and exhorted his audience that they should not wait to report. See TenmCode Ann. § 37-1-403. But Principal Williams both had discouraged Richardson from filing a DCS Report in 2008 and had waited for over three weeks after the Principals’ Meeting to file her own report, even though she had long possessed virtually all of the data that she included in her DCS Reports and most of the data that she brings to our attention on appeal.

Plaintiffs offer several more independent proofs to show pretext, two of which have already been covered: first, that the DCS Reports were both arguably false and suspiciously timed, see, e.g., DeBoer v. Musashi Auto Parts, Inc., 124 Fed.Appx. 387, 393 (6th Cir.2005) (“[Sjuspicious timing is a strong indicator of pretext when accompanied by some other, independent evidence.”), and, second, that Carver’s testimony is inconsistent with SCBE’s account of October 30th. They add three more proofs, all of them in an attempt to show that Principal Williams and her advisors, like Nurse Brown, had a retaliatory motive. The first two proofs both come from the week of October 30, and both suggested that the school was looking to remove A.C. from Bon Lin School: (1) Nurse Brown’s emphatic note, in response to the repeated classroom-testing requests, that A.C. may need to stop attending school; and (2) Assistant Principal McClellan’s renewed move to put A.C. into homebound services. The last proof they offer is Principal Williams’s errant voice mail which, they argue, shows “bias and hostility toward accommodations by the same person who ultimately placed the DCS call.” 7

We note another factor that could show pretext. The DCS Reports and the e-mail precursor to them consistently reflected a concern with the requests for accommodations themselves, apart from A.C.’s health. Thus, for instance, the very first sentence in Principal Williams’s e-mail on the 30th to SCBE officials is that Carver was having an anxiety attack “from the constant harassment of [A.C.’s parents].” In her October 30 DCS Report, Williams referred again to institutional concerns, noting that two nurses had already quit and another was threatening to do the same. And in her November 9 DCS Report, she again made this allegation, clarifying that the nurses quit to avoid liability concerns— i.e., concerns about A.C.’s demanding parents, not about A.C.’s health.

We conclude that a reasonable jury could find by a preponderance of all of this evidence that SCBE’s stated concerns about A.C.’s health were pretextual, and that the DCS Reports were actually motivated by the school’s well-established displeasure with A.C.’s parents and their accommodation requests. Of course, a reasonable jury could decide that SCBE’s concerns were not pretextual, but on review of a grant of summary judgment, “the issue is whether the plaintiff has produced evidence from which a jury could reasonably doubt the [defendant’s] explanation. If so, her prima facie case is sufficient to support an inference of discrimination at trial.” Chen v. Dow Chem. Co., 580 F.3d 394, 400 n. 4 (6th Cir.2009). To survive summary judgment, Plaintiffs need only rebut, not disprove, SCBE’s proffered reasons for the DCS Reports. See Blair v. Henry Filters, Inc., 505 F.3d 517, 532 (6th Cir.2007), overruled on other grounds by Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 129 S.Ct. 2343, 174 L.Ed.2d 119 (2009).

This conclusion does not yet conclude the pretext enquiry. As SCBE correctly argues, if Principal Williams honestly believed the reasons she cited for reporting A.C.’s parents to DCS, Plaintiffs cannot succeed. The “honest belief’ rule, as it is called, comes in at the pretext stage and places an additional evidentiary burden on Plaintiffs. See, e.g., Braithwaite v. Timken Co., 258 F.3d 488, 494 (6th Cir.2001). This circuit follows a modified form of the rule, asking whether the defendant “made a reasonably informed and considered decision before taking an adverse ... action.” Smith v. Chrysler, 155 F.3d 799, 807 (6th Cir.1998). Where the defendant “ ‘made an error too obvious to be unintentional,’ ” a fact-finder may infer that “‘it had an unlawful motive for doing so.’” Id. (quoting Fischbach v. D.C. Dep’t of Corr., 86 F.3d 1180, 1183 (D.C.Cir.1996)).

Here, the material facts are in dispute. A reasonable jury could conclude that neither of the core allegations made in the initial DCS Report was “reasonably informed” and that both comprised errors “too obvious to be unintentional.” In particular, a jury could find that, based on her training, Principal Williams knew or should have known that sending A.C. to school with sweets was not per se child abuse. And a jury could find that because she had easy access to Nurse Brown, who knew that A.C. was constantly monitored via the state-of-the-art sensor-and-pump arrangement and also knew how to access stored records in the pump revealing whether the monitoring continued when A.C. was at home, Principal Williams could have “reasonably informed” herself about whether there was any basis for alleging that A.C. was not monitored at home. 8 Finally, a jury could find that the record does not substantiate Williams’s claim that A.C.’s parents wished A.C. to be hurt so that they could file a lawsuit against the school.

C

One last issue remains for our resolution. Ordinarily, as we have already discussed, Plaintiffs must prove pretext by a preponderance of the evidence. DiCarlo, 358 F.3d at 414-15. The district court, however, held them to a much higher standard, requiring Plaintiffs to prove by “clear and convincing evidence” that the DCS Reports were made in “bad faith.” The district court’s two related reasons for this were the Cox court’s conclusion and the state statutory presumption of good-faith immunity for reporters.

Cox is distinguishable for several reasons. It dealt with a situation in which a school official both temporarily removed a student from class and made a child-abuse report based on the student’s casual comments and written reports about killing himself and others. 654 F.3d at 271. The student sued, alleging that both the temporary removal and the report were in retaliation against his speech. Id. The Cox court rejected the claim regarding the temporary removal by finding that courts must give school officials “unusual deference” in § 1983 cases concerning “a real threat to school safety and student learning,” deference that could be overcome only through “a clear showing of retaliatory or punitive intent.” ■ Id. at 274. In its disposition of the child-abuse report claim, the court noted that it could not find such a report retaliatory because the student failed to provide “any evidence of retaliatory or punitive intent.” Id. (emphasis added). Thus, Cox is not only distinguishable because it addresses a different area of law (First Amendment claims versus ADA/Section 504 claims) and differing legitimate concerns (broad issues of school safety versus a specific child’s health), it is also inapplicable because Plaintiffs have provided evidence of retaliatory or punitive intent.

The applicability of Tennessee law is a much different question. The state statute was clearly created precisely for situations such as this and clearly would apply were this a challenge based in state law. But it is not, and the district court incorporated the state evidentiary standard without any explanation of why enforcing that standard over the much-lower federal preponderance standard is permissible. The district court did cite another federal district court case out of Tennessee that appears to have incorporated the state-law standard into the federal-law analysis, Bryant-Bruce v. Vanderbilt University, Inc., 974 F.Supp. 1127 (M.D.Tenn.1997). But whether Bryant — Bruce actually did so is unclear and, in any event, the case provides no analysis on the preemption question because the plaintiffs conceded the issue. See id. at 1139.

The parties also provide no real argument on this point. Plaintiffs make only a fairly cursory statement that the federal standard should prevail over Tennessee law because of the Supremacy Clause, providing no supporting case law or argument to guide us through the applicable, and labyrinthine, preemption doctrine. And SCBE essentially makes no response at all.

When recently faced with a very similar ADA/Section 504 preemption issue, this Court noted that deciding “whether the Supremacy Clause trumps possible concerns by the [defendant school board] about avoiding tort liability under [state] law” raised “complicated” issues that required better briefing than the parties had given it. R.K. v. Bd. of Educ. of Scott Cnty., Ky., 494 Fed.Appx. 589, 598 (6th Cir.2012). Accordingly, we directed the issue to be fully briefed on remand. Id. We take the same approach here. Because of the very sensitive concerns at issue — concerns of federalism, of constitutional interpretation, of children’s safety, of teachers’ duty, and of parental rights— deciding what appears to be an emerging issue on an essentially blank slate is imprudent. For purposes of this appeal, we assume without deciding that Tennessee’s good-faith presumption applies. Given the generous view of facts and inferences required at the summary judgment stage, the nature of Principal Williams’s statements to DCS, and the fact that the core issue here turns on Principal Williams’s intent in making those statements, we conclude that a reasonable jury could still find for Plaintiffs. See Center for Bio-Ethical Reform, Inc., 477 F.3d at 823; accord Barrett v. Harrington, 130 F.3d 246, 263 (6th Cir.1997) (holding that where the record reveals “the possibility that ... [the defendant] did have a retaliatory motive,” this creates a “question of fact and avoidfs] summary judgment”).

Ill

For the foregoing reasons, we REVERSE the district court’s grant of summary judgment and REMAND for proceedings consistent with this analysis.

1 . Throughout this opinion, "Section 504” refers to Section 504 of the Rehabilitation Act of 1973. Section 504 is currently codified, as amended, at 29 U.S.C. § 794, which prohibits discrimination by reason of disability under federal grants and programs.

2 . The district court also analyzed each allegedly offensive act — Principal Williams’s voice mail, Nurse Brown's note — to determine if each one, alone, constituted retaliation; it found they did not. Because Plaintiffs concede on appeal that, setting aside the DCS Reports, such individual acts were not in themselves retaliatory, we do not address that portion of the district court’s opinion.

3 . Plaintiffs state that Nurse Brown's note — “If such a problem, may need to reconsider school itself!” — is direct evidence of retaliation. We disagree. See Smith v. Chrysler Corp., 155 F.3d 799, 805 (6th Cir.1998) (noting that direct evidence of discrimination is, for instance, when an employer says “I fired you because you are disabled”); see also Cesa-ro v. Lakeville Cmty. Sch. Dist., 953 F.2d 252, 254 (6th Cir.1992) (requiring the adverse actor's discriminatory comments to be related to the ultimate adverse action). Nurse Brown’s note indicates displeasure with the classroom-testing request but neither says nor hints anything about the DCS Reports, the alleged retaliatory act.

4 . See, e.g., Mayers v. Laborers' Health & Safety Fund of N. Am., 478 F.3d 364, 369 (D.C.Cir.2007); Freadman v. Metro. Prop. & Cas. Ins. Co., 484 F.3d 91, 106 (1st Cir.2007); Cassimy v. Bd. of Educ. of Rockford Pub. Sch., 461 F.3d 932, 938 (7th Cir.2006); Coons v. Secy of U.S. Dep’t of the Treasury, 383 F.3d 879, 887 (9th Cir.2004); Heisler v. Metro. Council, 339 F.3d 622, 632 (8th Cir.2003); Weixel v. Bd. of Educ. of N.Y., 287 F.3d 138, 149 (2d Cir.2002); Haulbrook v. Michelin N. Am., Inc., 252 F.3d 696, 706 n. 3 (4th Cir.2001); Selenke v. Med. Imaging of Colo., 248 F.3d 1249, 1266 (10th Cir.2001).

5 . Further, most of the fluctuations outside the target range were hyperglycemic. Thus, while they were undesirable, the higher levels were less immediately dangerous and reflected A.C.'s treatment plan of erring on the side of too much blood sugar.

6 . Plaintiffs do not respond to SCBE’s argument that Richardson's knowledge of a single severe hypoglycemic incident supported the school’s child-abuse concern. While a response might have been helpful, we do not find its absence dispositive.

7 . Plaintiffs also argue that an irregularity at Nurse Brown’s deposition supports their pretext argument. Resolving that argument is not necessary to provide Plaintiffs the relief they seek, so we do not consider it here.

8 . SCBE notes that Plaintiffs responded to an interrogatory request for all of these records from the previous three years by complaining that responding to the request required expert skill and technical information, and argues that this means it is unreasonable to expect Nurse Brown to know how to access the records. But not only is that argument a non sequitur (comprehensively accessing several years of records is not the same as checking ■back on a couple weeks of records), it is also without basis in the record: Nurse Brown testified that she knew the pump kept records of all its tests and that she knew how to access those records.