Intervening Misconduct, Honest Belief, and Strict Comparator Fit Defeat ADA Pretext at Summary Judgment

Case: Mark Bergens v. Diverse Concepts, LLC, et al. (6th Cir. Mar. 18, 2026) (unpublished)
Court: United States Court of Appeals for the Sixth Circuit
Disposition: Summary judgment for employer affirmed

1. Introduction

This appeal arose from an Americans with Disabilities Act (ADA) and parallel Tennessee-law discrimination suit brought by Mark F. Bergens, a hospitality manager employed by Diverse Concepts, LLC (and affiliated entities operating the Timberwood Grill and Blue Moose Alcoa). Bergens was hired as an assistant manager with an expected promotion to general manager of a new restaurant, Blue Moose Alcoa. Days after receiving the formal offer, Bergens suffered a stroke, missed two weeks of work, and returned with accommodations.

The employment relationship ended after what the opinion labels the “moonshine incident,” in which video footage showed Bergens searching co-workers’ bags without permission while looking for moonshine. Management terminated him, citing loss of trust and poor judgment. Bergens argued the bag-search episode was a misunderstood prank and that the real reason was disability-based concern about his post-stroke capacity to open the new restaurant.

The key appellate issue was narrow: under the McDonnell Douglas burden-shifting framework, could Bergens create a genuine dispute that the employer’s stated reason was pretext for disability discrimination?

2. Summary of the Opinion

The Sixth Circuit affirmed. The parties agreed (for summary-judgment purposes) that Bergens could establish a prima facie case and that Diverse Concepts articulated a legitimate, nondiscriminatory reason for discharge: loss of trust resulting from the bag search and Bergens’s response when confronted. The case therefore turned entirely on pretext.

The court held that Bergens failed to raise a genuine issue of material fact on pretext. The video and testimony established the incident occurred, and the decisionmaker’s conclusion was protected by the honest-belief rule. Bergens’s comparator evidence failed because proposed comparators were not similarly situated and did not engage in substantially identical conduct (and, as to one comparator, management lacked contemporaneous knowledge). Other circumstantial evidence—timing, alleged planning changes, accommodation/transfer theories, criticisms of the investigation, progressive-discipline arguments, and remarks—did not cumulatively permit a reasonable jury to find the stated reason was a cover for discrimination.

3. Analysis

3.1 Precedents Cited (and How They Shaped the Decision)

  • Pemberton v. Bell's Brewery, Inc., 150 F.4th 751 (6th Cir. 2025): The court framed ADA causation as “but-for” causation, underscoring that Bergens needed evidence that disability was determinative, not merely a background concern.
  • McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973): Provided the operative framework. Because steps one and two were undisputed, the opinion’s reasoning is a pretext-focused application of McDonnell Douglas.
  • Cardenas-Meade v. Pfizer, Inc., 510 F. App'x 367 (6th Cir. 2013) (per curiam): Used to align the Tennessee-law claims with the same analytical standard as the ADA claim.
  • Miles v. S. Cent. Hum. Res. Agency, Inc., 946 F.3d 883 (6th Cir. 2020): The court repeatedly relied on Miles for (i) the definition of pretext as a “commonsense inquiry,” (ii) the familiar three ways to show pretext, and (iii) the strict “all relevant respects” comparator standard.
  • Babb v. Maryville Anesthesiologists P.C., 942 F.3d 308 (6th Cir. 2019): Cited to confirm that pretext can be shown through other evidence beyond the three common categories, but only with a “cognizable explanation” connecting evidence to pretext.
  • Pelcha v. MW Bancorp, Inc., 988 F.3d 318 (6th Cir. 2021): Central to rejecting “no basis in fact” arguments and discounting ambiguous remarks. The opinion adopted Pelcha’s focus on whether the employer “made up” its reason and its treatment of isolated/ambiguous comments.
  • Hieber v. Oakland Cnty., 136 F.4th 308 (6th Cir. 2025); Blizzard v. Marion Tech. Coll., 698 F.3d 275 (6th Cir. 2012); Escher v. BWXT Y-12, LLC, 627 F.3d 1020 (6th Cir. 2010): These cases anchored the honest-belief rule: an employer can defeat certain pretext showings by demonstrating a reasonably grounded, honest belief in its stated reason, even if the investigation was not “perfect.”
  • Chattman v. Toho Tenax Am., Inc., 686 F.3d 339 (6th Cir. 2012): The comparator-evidence template: the plaintiff must identify employees outside the protected class who engaged in substantially identical conduct and were treated better; also used to stress the relevance of management’s knowledge of comparator incidents.
  • Madden v. Chattanooga City Wide Serv. Dept., 549 F.3d 666 (6th Cir. 2008); Jackson v. VHS Detroit Receiving Hosp., Inc., 814 F.3d 769 (6th Cir. 2016): Reinforced that “substantially identical conduct” is the operative comparator touchstone; exact identity is not required, but similarity must be meaningful in the relevant respects.
  • Bashaw v. Majestic Care of Whitehall, LLC, 130 F.4th 542 (6th Cir. 2025): Used twice: to validate termination based on undermined trust, and to reiterate strict comparator requirements (similar in all relevant respects and substantially identical conduct).
  • Seeger v. Cincinnati Bell Tel. Co., 681 F.3d 274 (6th Cir. 2012); Kuhn v. Washtenaw Cnty., 709 F.3d 612 (6th Cir. 2013): These cases limited reliance on temporal proximity. Kuhn was pivotal: an intervening legitimate reason breaks the inference that timing alone suggests discrimination.
  • Grizzell v. City of Columbus Div. of Police, 461 F.3d 711 (6th Cir. 2006): Deployed to reject speculative narratives (e.g., hidden plans to replace Bergens) unsupported by record evidence.
  • Londo v. UP Health Systems-Marquette, No. 21-1290, 2021 WL 6112971 (6th Cir. Dec. 27, 2021): Cited for the proposition that even a failure-to-accommodate theory does not automatically prove that a stated discharge reason was pretextual.
  • Kean v. Brinker Int'l, Inc., 140 F.4th 759 (6th Cir. 2025): Informed the treatment of progressive-discipline deviations—insufficient alone, sometimes part of a “constellation”—but here the policy’s discretionary language neutralized the argument.
  • Phelps v. Yale Sec., Inc., 986 F.2d 1020 (6th Cir. 1993); Hayes v. Clariant Plastics & Coatings USA, Inc., 144 F.4th 850 (6th Cir. 2025); Willard v. Huntington Ford, Inc., 952 F.3d 795 (6th Cir. 2020); Vincent v. Brewer Co., 514 F.3d 489 (6th Cir. 2007): These cases supplied the standards for evaluating remarks: whether made by a decisionmaker or influencer, whether isolated/ambiguous, and whether sufficiently derogatory to imply discriminatory motive.

3.2 Legal Reasoning

The court’s reasoning was structured around the third McDonnell Douglas step: pretext. It addressed (i) “no basis in fact,” (ii) comparator evidence, and (iii) a bundle of “other” circumstantial evidence.

A. “No Basis in Fact” and the Honest-Belief Rule

Diverse Concepts stated it terminated Bergens because he searched co-workers’ bags without permission, upsetting them and undermining trust, and then initially denied the conduct when confronted. The court found that rationale was grounded in “particularized facts”: video footage, reports from the general manager, and the call with Bergens. That foreclosed the claim that the employer fabricated the event.

Even accepting Bergens’s attempt to reframe the incident as a prank, the honest-belief rule independently protected the employer: it was enough that the decisionmaker honestly and reasonably believed the misconduct occurred and justified termination; the law does not demand a perfect investigation.

B. Comparator Evidence Must Be Tight-Fitting and Decisionmaker-Relevant

Bergens argued three employees committed similar or worse misconduct but were not terminated. The court rejected each comparator:

  • Robert Cartree: different supervisory chain and materially different conduct; additionally, no evidence upper management knew of the incident before terminating Bergens.
  • Daniel King: misconduct not “substantially identical” to breaching co-workers’ privacy and trust by searching bags; also not similarly situated in role/status.
  • Jack Coppinger: different type of misconduct; and, in any event, his behavior later resulted in demotion, undermining the premise of preferential treatment.

The opinion thus reinforces that comparator analysis is not a generalized “bad acts” balancing test; it is a role-, context-, and conduct-specific inquiry tied to what the decisionmakers knew and did.

C. “Other Evidence” Did Not Add Up to Pretext

  • Temporal proximity: under Kuhn v. Washtenaw Cnty., the moonshine incident was an intervening legitimate reason that dispelled any inference from timing.
  • “Coppinger was the plan all along” theory: the court deemed it speculative under Grizzell v. City of Columbus Div. of Police, given testimony explaining bonus ranges and placement decisions.
  • Accommodations discussions: the court treated the argument as misdirected and, citing Londo v. UP Health Systems-Marquette, emphasized that even accommodation issues do not automatically prove discharge pretext.
  • Failure to transfer another employee; alleged planning exclusion: the court found no record basis to infer discriminatory motive, noting ongoing communications and a planning document still listing Bergens as general manager shortly before termination.
  • Investigation adequacy: under Seeger v. Cincinnati Bell Tel. Co., employers need not conduct an “optimal” investigation; viewing video and obtaining key reports was enough.
  • Progressive-discipline policy: under Kean v. Brinker Int'l, Inc., deviation alone is not enough—and here there was no deviation because the policy reserved discretion to terminate without prior steps.
  • Remarks: “one arm bandit” comments were infrequent and not tied to the termination decision; senior leaders’ expressed concerns about rehabilitation progress and ability to train were treated as ambiguous and potentially job-related, not unambiguously discriminatory.

Finally, the court considered cumulative effect (as in Kean v. Brinker Int'l, Inc.) and still found the record insufficient for a reasonable jury to infer pretext.

3.3 Impact

Although unpublished, the opinion is a clear synthesis of Sixth Circuit pretext doctrine in ADA discharge cases:

  • Intervening misconduct matters: where a discrete, documented workplace incident occurs after protected activity/health events, it can sever the inference drawn from timing (Kuhn v. Washtenaw Cnty.).
  • Video evidence plus a consistent decisionmaker narrative is powerful: documented misconduct and a credibility issue during the termination call made “made-up reason” theories difficult to sustain, especially under the honest-belief rule (Hieber v. Oakland Cnty.).
  • Comparator evidence remains demanding: the court reiterated that comparators must be similar in all relevant respects and have engaged in substantially identical conduct; status/supervision differences and employer knowledge can be dispositive (Miles; Chattman; Bashaw).
  • Progressive discipline arguments often fail when policies reserve discretion: a handbook that explicitly permits immediate termination can neutralize “policy deviation” as pretext evidence (Miles; Kean).

4. Complex Concepts Simplified

  • “But-for” causation (ADA): the plaintiff must show the adverse action would not have happened but for disability—i.e., disability was the determining cause, not merely a factor in the background.
  • McDonnell Douglas framework: a three-step indirect-proof method—(1) plaintiff’s initial showing, (2) employer’s legitimate reason, (3) plaintiff’s proof that the reason is a pretext.
  • Pretext: evidence that the employer’s stated reason is not the real reason. Common routes include showing the reason is factually false, did not actually motivate the decision, or was insufficient to justify the decision.
  • Honest-belief rule: even if an employer is mistaken, it can still win if the decisionmaker honestly and reasonably believed the stated reason based on particularized facts available at the time.
  • Comparator evidence: using other employees as benchmarks. It works only when the comparators are truly similar in relevant job circumstances and committed substantially the same misconduct but were treated more favorably.
  • Temporal proximity: closeness in time between disability-related events and termination. It can suggest discrimination, but an intervening legitimate event (like misconduct) can break the inference.

5. Conclusion

The Sixth Circuit affirmed summary judgment because Bergens could not create a triable issue that Diverse Concepts’s stated reason—loss of trust stemming from a documented bag-search incident and an initial denial—was a cover for disability discrimination. The decision emphasizes three durable points in Sixth Circuit employment discrimination law: (1) documented intervening misconduct can defeat timing-based inferences; (2) the honest-belief rule protects reasonably grounded decisionmaking even without a perfect investigation; and (3) comparator evidence must be tightly matched in role, context, conduct, and (often) decisionmaker knowledge to be probative of pretext.