Appeal from the United States District Court for the Western District of Texas USDC No. 5:23-CV-769

Before Dennis, Graves, and Duncan, Circuit Judges .

James E. Graves, Jr., Circuit Judge :

Treating the petition for rehearing en banc as a petition for panel rehearing (5th Cir. R. 35 I.O.P.), the petition for panel rehearing is DENIED . Because no member of the panel or judge in regular active service requested that the court be polled on rehearing en banc (Fed. R. App. P. 35 and 5th Cir. R. 35), the petition for rehearing en banc is DENIED . The opinion issued November 21, 2025, is withdrawn by the panel and the following is substituted in its place.

Page 2In this appeal, we consider whether an autistic elementary school student’s individualized education program gave him an appropriate education under the Individuals with Disabilities Education Act ( IDEA ). Although the school district’s efforts to aid the student’s academic progress were laudable, his behavioral problems were uniquely severe-sometimes placing his life in danger. So the district shouldered a weighty responsibility to employ effective strategies to reduce the behaviors. It knew that additional extended-school-year services might help, but refused to provide them. The district court concluded that the school district denied the student an IDEA -appropriate education as a result. We AFFIRM .

When the administrative record begins in 2020, the student (“ IM ”) was a second grader in the North East Independent School District (“the School District”) in San Antonio, Texas. With a speech impediment, autism, and an intellectual disability, IM struggles to communicate and socialize. His vocabulary is limited. So he communicates mostly through gestures, facial expressions, and an iPad with a specialized, communication app. His behavior is also disruptive. He often hits walls and furniture, jumps, does handstands, and spins around. By fourth grade, IM 's academic skills had progressed no further than a kindergartener. Most significantly, IM has toileting issues, and frequently escapes from school (elopement).

Naturally, IM ’s conditions impede his public-school education. Staff must supervise him constantly to prevent elopement and his disruptive behaviors. He cannot effectively learn academic concepts or participate in the regular classroom.

To address these educational challenges, the School District provides an independent education program. It monitors and develops this IEP through a committee of its employees and IM ’s mother, Bianca. The dispute here emerged through the IEP Team's meetings, as the School District's and Bianca’s views on appropriate accommodations diverged.

At first, they agreed. When IM ’s second-grade year ended, in June 2021, the IEP Team met to develop next year's program. For third grade, the School District would place IM in a special-education classroom, provide speech and occupational therapy, and transport him to and from school. It would also accommodate him academically and behaviorally. This included a plan to prevent elopement and an electronic communication iPad. Along with regular-year accommodations, the School District offered an extendedschool-year program. The ESY program's goal was to prevent regression between terms.

So after second grade, IM joined the program. For three weeks in June, he attended half-day sessions four days a week. During that time, IM ran away on 18% of school days. But after ESY ended, IM 's educational services ceased until school resumed. IM had 'regressed significantly.' As a third grader, he ran away on more than 40% of school days. And after spring break that year, he regressed further, urinating in the classroom twice, something he did not have problems with before.

Shortly after spring break, the IEP Team met to plan for fourth grade. The team planned for IM to continue in a special-education classroom for most classes . But for some classes he would transition to the regular classroom. He would also continue with speech and occupational therapy. On these accommodations, the School District and Bianca agreed.

Page 4Yet a rift emerged over the ESY program. To address IM ’s regression over breaks, Bianca requested full-day ESY services throughout the summer. The School District refused. While it conceded that additional ESY would help, it deemed full-summer services unnecessary. So it offered only half-day services for six weeks, which it reasoned were sufficient to maintain IM ’s progress.

The IEP Team reconvened a few weeks later, and Bianca asked the School District to reconsider. Again, it refused. So after third grade, IM joined the half-day summer ESY program for six weeks. That program ended in mid-July—leaving a month-long break until fourth grade.

In Fall 2022, IM returned for fourth grade. He had regressed again. In the first two weeks, he ran away at least three times. Although he had no toileting incidents in the ESY program, he had 20 in the first six weeks of the school year.

Bianca requested a meeting to address the regression. She was growing increasingly concerned at IM’s lack of 'meaningful progress.' Chiefly, she worried about regression over even short breaks. The School District responded by offering an in-home training evaluation. Bianca declined. The IEP Team then recessed the meeting so its members could review the information discussed.

The meeting resumed a few months later. Bianca remained concerned. Indeed, IM 's elopement-regression left her in fear for his life. She worried particularly that the School District failed to immediately notify her that IM had run away from the school bus. The School District responded by offering a safety vest for IM to wear on the bus. It also proposed more behavioral interventions.

Page 5A few weeks later, IM had his most dangerous elopement yet. He escaped and exited campus through an unlocked gate. A busy road lies nearby, which IM ran toward. He was saved only by the intervention of bystanders, who restrained him only after he crossed the road. The principal, who had pursued IM in her car, retrieved him eventually. The episode lasted about thirty minutes.

The IEP Team met again soon after. Bianca shared her alarm about the recent elopement. The School District denied that school breaks caused regression. To the contrary, it opined, “his elopement data [was] trending downward.” So the School District declined Bianca’s request for more ESY services over the summer, and over other breaks. “The [meeting] ended in disagreement.”

Soon after, IM (through Bianca) requested a special education due process hearing. She sought extended-school-year services for any break, a GPS tracking device, and an IEP goal for safe bus riding.

After a two-day evidentiary hearing in April 2023, the hearing officer sided with IM . The officer concluded that the School District failed to provide IM an appropriate education under the Individuals with Disabilities Education Act ( IDEA ). He thus ordered the School District to provide fullsummer ESY services and year-round access to a voice-assistedcommunication device.

The School District appealed to federal court. The district court upheld the hearing officer’s findings. This appeal followed.

We review de novo, as a mixed question of law and fact, a district court’s decision whether an IEP was appropriate under the IDEA . Cypress- Page 6Fairbanks Indep. Sch. Dist. v. Michael F. ex rel. Barry F. , 118 F.3d 245, 252 (5th Cir. 1997). Because IM challenges his IEP , he bears the burden to show that the program was not appropriate. See id.

Underlying factual findings we review for clear error. Boone v. Rankin Cnty. Pub. Sch. Dist. , 140 F.4th 697, 706 (5th Cir. 2025). This includes findings about a student’s educational needs and whether they obtained an educational benefit. Id. Still, when a district court finds facts under a misapprehension of the controlling principles, we review that finding de novo. Klein Indep. Sch. Dist. v. Hovem , 690 F.3d 390, 397 (5th Cir. 2012).

Under clear-error review, we reconsider a factual finding only if the evidence leaves us “with the definite and firm conviction that” the district court is mistaken. Boone , 140 F.4th at 706 (quoting Renee J. ex rel. C.J. v. Hou. Indep. Sch. Dist. , 913 F.3d 523, 528 (5th Cir. 2019)). When a district court chooses between two permissible views of the evidence, there is no clear error. Id. (quoting Anderson v. City of Bessemer City , 470 U.S. 564, 574 (1985)).

Perhaps seeking de novo review of clear-error issues, the School District blurs the distinction between legal conclusions and factual findings. For example, at oral argument, it insisted that its challenges were solely legal. But then argued nearly all factual issues. In any event, where the School District's argument can properly be considered legal, we review de novo. See Hovem , 690 F.3d at 397. Other arguments we review for clear error. Boone , 140 F.4th at 706.

The IDEA offers states federal funds to educate disabled children. Endrew F. ex rel. Joseph F. v. Douglas Cnty. Sch. Dist. RE-1 , 580 U.S. 386, 390Page 7 (2017) (citing 20 U.S.C. § 1400 et seq. ). In return, a state must provide disabled students with “a free appropriate public education.” Id. (citing 20 U.S.C. § 1412(a)(1)). That includes “special education and related services” tailored to a child’s individual needs. Id.

The primary vehicle to ensure an appropriate education is an IEP . Endrew F. , 580 U.S. at 399. That is, a written plan prepared by the school district's representative, 'a teacher, the child's parents . . . and when appropriate, the child.” Est. of Lance v. Lewisville Indep. Sch. Dist. , 743 F.3d 982, 989 (5th Cir. 2014) (citation modified). The team must follow detailed procedures, which “require careful consideration of the child’s individual circumstances.” Endrew F. , 580 U.S. at 391.

An IEP must be 'reasonably calculated to enable a child to make progress appropriate in light of the child’s circumstances.” Endrew F. , 580 U.S. at 399. We gauge this progress using the Michael F. factors. 118 F.3d at 253. Those are: whether “(1) the program is individualized [based on] the student’s assessment and performance; (2) [it] is administered in the least restrictive environment; (3) the services are provided in a coordinated and collaborative manner by the key stakeholders; and (4) [it demonstrates] . . . academic and non-academic benefits.” Id. (citation modified). The fourth factor is “critical” because a child must be likely to progress, not regress or achieve 'trivial educational advancement.' See Boone , 140 F.4th at 707 (quoting Renee J. , 913 F.3d at 529).

The School District challenges the district court’s conclusion that it denied IM an IDEA -appropriate education. Because the district court found for the School District on the second and third factors, it challenges only the Page 8first and fourth.¹ Those are the program's individualization and its benefits. See Michael F. , 118 F.3d at 253. After considering each, we conclude the district applied the right legal principles, and did not clearly err in its factual findings.

On the first factor, the district court found that the School District failed to individualize the IEP sufficiently because the program neglected IM 's toileting issues and 'alarming' elopements. The School District disagrees. Because it does not meaningfully argue that the district court misapprehended the controlling principles, we review for clear error. See Hovem , 690 F.3d at 39; Boone , 140 F.4th at 708.

From our review, we cannot conclude that the district court clearly erred in finding that IM ’s IEP was insufficiently individualized.

When a child’s behavior impedes their learning, the IEP team must “consider [using] positive behavioral interventions and supports, and other strategies, to address that behavior.” 20 U.S.C. § 1414(d)(3)(B)(i). Yet an IEP need not be perfect. E. R. ex rel. E. R. v. Spring Branch Indep. Sch. Dist. , 909 F.3d 754, 769 (5th Cir. 2018). “[T]he question is [instead] whether the IEP is reasonable . . . .” Endrew F. , 580 U.S. at 399. It must only 'aim to enable the child to make progress.' E. R. , 909 F.3d at 769 (citation modified).

As the School District points out, IM ’s IEP reflected myriad behavioral interventions and supports to address elopement and toileting—including a behavior intervention plan. For elopement, the IEP offered strategies like an incentive system and a 'stop sign' cue. For Page 9toileting, it employed visual reinforcement to encourage IM to eliminate in designated areas.

The School District claims this was enough. Granted, with these interventions, IM ’s elopements reduced, and his toileting improved. Over second grade, his elopement diminished from 83% of school days to only 23%.

Yet after any break, IM regressed. When second grade began, he ran away on 46% of school days. After spring break, he had two toileting incidents. The IEP Team thus resolved that breaks caused IM to regress 'in behavior and skills related to independence (i.e., toileting).' Indeed, the team observed that he generally “has a documented history of regression caused by the interruption of services.” So the team concluded that he would benefit from extended ESY .

In short, the School District knew that IM regressed over breaks. And it identified the solution: to extend the ESY program. The School District thus extended the program by two weeks, and IM clearly benefited. In the ESY program after third grade, he ran away on only 25% of days and had no toileting incidents.

But even with the extra weeks, IM kept regressing. He had 20 toileting incidents in the first two months of fourth grade. Puzzlingly, the School District stopped systematically tracking elopement-impeding precise comparisons. Yet the string of incidents suggests severe regression. In the first two weeks, IM ran away on more than 30% of days. He also ran away from the bus several times. Most seriously, he escaped from school and crossed a busy road. Save for the intervention of bystanders, he might have been seriously injured or killed. Naturally, this incident and the bus escapes, placed Bianca “in fear for her son’s life.”

This fourth-grade regression shows that the summer, six-week ESY program did not fully address IM ’s toileting-and-elopement regression. And Page 10the School District did not offer ESY services during the other breaks, even though those breaks caused regression too. Nevertheless, the district maintained the ESY status quo, and limited IM ’s access to six weeks.

These facts track our recent Boone decision neatly. There, we held that when an IEP fails to adequately address elopement in an autistic child, it is insufficiently individualized. See Boone , 140 F.4th at 708. The school district “had a behavior intervention plan in place . . . which addressed elopement.” Id. But this did not adequately address this 'major issue.' Id. Even though the school district knew 'the elopement accommodations [the student] required, [it did not have a] plan to implement them.' Id. So the district court found the IEP insufficiently individualized. Id. at 708-09. We affirmed. Id.

True, the School District here tried harder to address elopement than the Boone district. But IM ’s elopement posed a more serious threat. Boone notes only one elopement, which was controlled at the student's current school. Id. at 704-05. By contrast, IM ’s elopements posed a present, grave danger. “When [ IM ] elopes, he is unconcerned with his safety and is likely to cross streets without awareness of traffic . . . [and] [h]e does not take his [communication device] when he elopes.” This danger required more from the School District.

Overall, even though the School District implemented behavioral strategies, IM 's breaks from school frustrated their benefit. The district knew the summer ESY program improved toileting and elopement, and admitted it was too short. It refused to offer full-summer services even so. Weighing this evidence, the district court found the IEP insufficiently individualized. We see no clear error.

On the fourth Michael F. factor, the district court found that even though IM progressed academically, he did not adequately progress toward Page 11nonacademic goals. The School District counters that the district court could not have found against it on the fourth factor, while finding that IM received both academic and nonacademic benefits.

We read the School District as asserting two errors. That the district court engaged in “disability remediation in disguise.” And that the district court erred when it did not accord IM ’s academic benefits enough weight.

The School District claims that the district court focused solely on disability remediation. Because this challenges the legal principles the district court relied on, we review de novo. See Hovem , 690 F.3d at 397.

At its core, an IEP is 'a plan for pursuing academic and functional advancement.” Endrew F. , 580 U.S. at 399. So the fourth, and critical, Michael F. factor asks whether 'the child received positive academic and non-academic benefits.” A.A. ex. rel. K.K. v. Northside Indep. Sch. Dist. , 951 F.3d 678, 691 (5th Cir. 2020) (emphasis added). The analysis must be “holistic.” See Hovem , 690 F.3d at 399. Hence defining the educational benefit 'exclusively or even primarily in terms of correcting the child's disability” is error. Id. at 397. That is pure “disability remediation.” Id. at 399. Even so, “remediation may often be part of an IEP .” Id. at 397. Take behavioral modification, for example, which is a 'strategy that may remediate . . . while also being necessary to confer educational benefits.” Id.

Applying these principles, we reversed a district court that focused on remediation alone. Id . at 399. In Hovem , the disabled child (Per) was a 'high[ly] intelligen[t]' high-school student in regular classes, but he had attention deficit disorder and difficulty writing. Id. at 392. In most subjects, Per performed acceptably, but in English, he struggled. Id. So his school granted accommodations to improve his writing but denied greater Page 12accommodations. Id. at 393. Still, he functioned well enough to graduate and go to college. Id. at 399-400.

The district court found that Per’s school had denied him an IDEA -appropriate education because he received “no academic benefit tailored to his disability.' Id. at 396-97 (citation modified). We reversed. Id. at 400. When a disabled child attends a regular classroom, an IEP must 'be reasonably calculated to enable [them] to achieve passing marks and advance from grade to grade.' Id. at 397 (quoting Bd. of Educ. v. Rowley , 458 U.S. 176, 204 (1982)). In other words, “the system itself confirms the . . . educational benefit.' Id. at 399 (citation modified). Hence the Hovem district court erred when it focused on lack of progress in one domain (writing), while discounting Per’s successful graduation. Id. at 397-400.

The School District reasons from Hovem that the district court focused only on remediating IM ’s disability. For two reasons, we disagree.

First, Hovem is distinguishable on its own terms. That panel carefully differentiated Per’s IEP from one 'where [the child] regressed educationally or could not measure up to ordinary grade-level standards.” See Hovem , 690 F.3d at 399. Those children require a different approach. Id. When a child cannot fully be integrated into the regular classroom and cannot achieve at grade level, a court must consider all their circumstances and ensure that their IEP is still 'appropriately ambitious.' Endrew F. , 580 U.S. at 402. After all, 'every child should have the chance to meet challenging objectives.” Id.

When the hearing officer issued his decision, IM was a fourth grader academically functioning at a kindergarten level. This required the district court to look beyond traditional achievement. See Endrew F. , 580 U.S. at 402. It did so, addressing a different issue than Hovem .

Page 13Second, the district court did not focus solely on disability remediation. Instead, it carefully weighed IM ’s progress in different domains. It recognized IM ’s “clear academic progress.” But it also found “the record . . . replete with instances where [ IM ] experienced documented regression after school breaks, especially [with] elopement and toileting.' The district court thus found that IM ’s behavioral regression outweighed his academic progress.

So the district court did not, in fact, focus on remediation, or any other consideration, to the exclusion of others. Instead, it conducted the holistic analysis that Hovem prescribes—no error. See Hovem , 690 F.3d at 399. The School District’s mischaracterization of the district court order adds little.

The School District next argues that the district court erred when it found “academic and non-academic benefits . . . in the summer [after] the disputed ESY ” yet ruled against it on the fourth factor. We review a finding whether a student obtained an educational benefit for clear error. Boone , 140 F.4th at 706.

The School District relies on Hovem ’s holding that academic benefit clearly “militates in favor of a finding that an IEP is appropriate.' Hovem , 690 F.3d at 399 (quoting Adam J. ex rel. Robert J. v. Keller Indep. Sch. Dist. , 328 F.3d 804, 810 (5th Cir. 2003)). But it miscites Hovem to say that ''academic benefit' alone ' clearly' militates' in favor of an appropriate IEP . (emphasis added). This assuredly departs from Hovem and Adam J. Indeed, the statement suggests that academic benefit can overwhelm nonacademic factors.

Quite the contrary. An IEP review must be “fact-intensive, individualized, [and] holistic.” H.W. ex rel. Jennie W. v. Comal Indep. Sch. Dist. , 32 F.4th 454, 468 (5th Cir. 2022) (collecting cases). To be sure, Page 14progress toward academic goals aids this review. Id. at 468–69. “[B]ut no one factor can overwhelm it.' Id. at 469. So academic progress militates 2 in favor of the fourth factor, but it does not predominate. See id.

Nor has the School District persuaded us to rely on Lathrop R-II School District v. Gray , 611 F.3d 419 (8th Cir. 2010). We are cautious to rely on Gray for two reasons:

First, the Eighth Circuit applies a different standard. We use the Michael F. factors. 118 F.3d at 253. But Gray relied on an Eighth Circuit standard that “even if ‘more positive behavior interventions could have been employed, that fact is largely irrelevant’ [if] the school district made ‘a good faith effort' to help the student achieve . . . educational goals.' Gray , 611 F.3d at 426 (quoting CJN v. Minneapolis Pub. Schs. , 323 F.3d 630, 639 (8th Cir. 2003)). Our court views educational benefit differently.³

Second, Gray is distinguishable. There, the student’s behaviors were disruptive, whereas IM endangered himself. In Gray , the child would bite his finger, have loud outbursts, and sometimes exhibit sexual behaviors like touching his penis. 611 F.3d at 422. This impeded his learning of course. Id. at 426. But not nearly so much as IM , who had more than 20 toileting incidents at the start of fourth grade. And IM ’s elopements placed him in danger—as best evidenced by his 30-minute escape across busy road. The Page 15seriousness of IM ’s behaviors required a greater response from the School District than the behaviors in Gray required.

While the School District cautions us against disability remediation, it asks us to narrow IEP review to another single criterion (academic benefit) in the next breath. That approach contradicts our precedent and the requirement that an IEP be “reasonably calculated to enable a child to make progress appropriate in light of the child’s circumstances.” See Endrew F. , 580 U.S. at 399. The district court inquired holistically into the IEP ’s academic and nonacademic benefits—as it ought to have done.

The School District objects even so, maintaining that the district court could not find against it on the fourth factor while observing that IM made educational progress. Just as with the first factor, we see no clear error.

The record assuredly reflects that the School District made considerable effort with IM -both academically and nonacademically. Academically, the district court found that the IEP reflected “clear academic progress.' Indeed, IM 'appropriately progress[ed], and even master[ed]' some of his third-grade goals.

But the district court found that IM ’s failure to make “adequate[]” nonacademic progress outweighed his academic progress. Most importantly, it noted the many times when IM regressed after school breaks with toileting and elopement. As we discuss above, the record bears this out. IM regressed with toileting, and his elopement grew more frequent (and more dangerous) after every break. The School District knew that the interruption of services caused this regression.

The School District counters that IM 's behavior intervention plan guaranteed him an appropriate education. The district court and the hearing officer took a different view. We do not disturb their finding.

Page 16For one thing, the record reflects that the behavioral interventions were not working. Even though the School District implemented strategies for toileting and elopement over several years, IM regressed nevertheless. His worst period was in fourth grade when he had his most dangerous elopement yet. Although the School District provided minor behavioral interventions, his regression on toileting and elopement shows that more was needed.

For another, we review for clear error. That means that the School District must go further than 'posit[] a dueling view of the evidence.' Boone , 582 F.3d at 712. At best, the School District asks us to adopt its view of the evidence over that of the hearing officer and the district court. On clear error review, we cannot. See id.

Ultimately, this record is unique—like many IEP cases. The School District made a laudable effort to ensure that IM progressed academically. The district court never doubted that. But IM 's behavioral issues were uniquely severe. The elopement particularly so, because it threatened IM ’s safety—a serious impediment to his education by any measure. These severe deficits placed great responsibility on the School District to employ effective strategies. Because the School District failed to employ a strategy—known to be effective-to ameliorate IM 's regression, the district court and the hearing officer found that the regression outweighed the School District’s efforts. We are not left with a firm conviction that they were wrong.

After considering the two contested Michael F. factors, we conclude that the district court applied the right principles, and its findings were not clear error. We thus AFFIRM the district court’s judgment.