Accessibility Defense, Helping Business Avoid and Defend ADA and FHA Lawsuits

ADA and FHA Defense

  • When it comes to ESAs, don’t ask a doctor for advice about the law

    On July 1, 2026 the American Psychological Association published an article about using animals in therapy. You can find it here.  I’m a lawyer, not a doctor, so I don’t have anything to say about most of the article, which deals with how animals can be used by psychologists when interacting with patients. I can say, however, that the article is wrong when it comes to its discussion of emotional support animals and the law. For example, the author writes:

    Under the Fair Housing Act, ESAs are defined as animals that provide emotional support or alleviate a symptom or effect of a disability.

    There is no Fair Housing Act definition of ESAs, either in the statute or the related regulations. This information comes from a HUD guidance document that never had any binding effect and has now been withdrawn by HUD. The claim that an animal is an ESA is meaningless in legal terms because the words “emotional support animal” do not have a fixed legal definition.  Next:

    Clinicians don’t need specific knowledge about the individual animal to write an ESA letter that is ethically and legally acceptable according to federal guidance, said Ekholm Fry.

    It is true that Ekholm Fry said something like this in a 2021 paper, but it was wrong at the time and became even more wrong after HUD withdrew the guidance documents on which Fry apparently relied. First, federal agencies like HUD do not establish ethical standards for medical providers. Whether an “ESA letter” is ethically acceptable depends on the law of states that license psychologists and on the standards of the profession itself. Second, even the now withdrawn HUD guidance does not say that clinicians can write an ESA letter without knowledge about the individual animal. HUD’s old guidance always insisted that evidence of disability and need for an ESA be “reliable.” The APA’s own publications about ESA’s suggest that clinicians should be careful in prescribing ESA’s to evaluate all the effects it may have on their patient, and it is hard to imagine how that can be done without knowing something about the specific animal.

    But to continue, the author next writes:

    Their (the therapist’s) role is simply to document that the person indeed has a disability, as defined under the Americans with Disabilities Act (ADA).

    The sentence is hyperlinked to the ADA. The problem is that the definition that matters for ESA’s in housing is the definition of “handicap” in the Fair Housing Act. The definition of “handicap” in the Fair Housing Act is not the same as the definition of “disability” in the ADA. A therapist who relies on the ADA definition of disability when writing an ESA letter to help their patient avoid a “no pets” policy at an apartment complex isn’t matching their letter to the law that matters.

    Even more important, “disability” and “handicap” are legal terms with legal definitions, not medical terms with medical definitions. Unless a therapist has enough legal training to relate these legal terms to equivalent medical terms the therapist is not competent to say whether their patient has a “disability” or “handicap” as defined in the ADA and FHA. I will add to be fair that lawyers should not go around making medical diagnoses. I might say in casual conversation that say an opposing lawyer has OCD, but I certainly wouldn’t write a letter asserting that they had OCD based on my legal training and the fact I have a copy of the DSM V on my shelf. Lawyers shouldn’t make medical diagnoses and therapists shouldn’t opine about the law.

    I could go on picking apart this article, but these illustrations make the point. Whether a patient has a “handicap” that triggers application of the Fair Housing Act is a legal question that requires applying the legal definition of handicap to whatever medical diagnosis a therapist is competent to make. Whether keeping an animal in an apartment or home “may be necessary to afford such person equal opportunity to use and enjoy a dwelling” is a legal question that depends on the legal definitions of “necessary,” “equal opportunity,” “use and enjoy” and “dwelling” applied to what the animal does for its owner. This may not require a law degree, but it certainly requires legal training about the meaning of these terms as they have been interpreted by the courts. It also requires some knowledge of HUD’s published memos on this subject, because HUD or the President can, with the stroke of a pen, simply eliminate all of HUD’s previous guidance materials. That is what happened on May 22, 2026, a week before the APA article I’m writing about. Only specialists in disability law are likely to be up to date on these issues. Someone with legal training would know how to become up to date using available legal research tools, but I doubt that most therapists have the training required to use those tools or access to them.

    If a psychologist is asked to write an ESA letter they should, if they have not already done so, research the professional literature for their profession concerning the possible benefits and harms of ESA prescriptions and the likely ethical complications. Then they should do what they are qualified to do. First, determine using appropriate diagnostic criteria diagnose the patient’s medical condition. Second, describe what is known about what the ESA does for their patient. “Know” means know with whatever degree of certainty is needed to treat the patient according to applicable professional standards. And then they should stop. Whether they have diagnosed a handicap or disability is a legal question, not a medical question. Whether the animal is “necessary to afford such person equal opportunity to use and enjoy a dwelling” is a legal question, not a medical question. In short, psychologists and other medical providers should stick to doing what they are trained and competent to do, and leave the legal consequences of their medical conclusions to be decided by those whose training and competence is in the law.

    As a final note, both therapists and lawyers need to understand that AI cannot make up for a lack of training. At least a couple of lawyers are sanctioned every week because they relied on AI to do their legal research. In my disability law practice I have seen lawyers who decided to learn about the Fair Housing Act via Google get their clients in serious trouble.  A psychologist who relies on AI or a Google Search to learn about writing emotional support animal letters is certainly doing their patient a disservice and may be doing the patient a lot of harm. Those who have two legs that work don’t need an artificial leg. Those who have a brain that works don’t need artificial intelligence.


    Share
    , , ,
    , , , ,

  • SBA v DOJ – who’s calling the shots on website accessibility?

    On June 23  the Small Business Administration’s Office of Advocacy called on the Department of Justice to go beyond its recent one year pause on Title II website regulations and simply scrap them entirely. See, Advocacy Recommends. The legal justification is astonishing – SBA claims that under the latest Supreme Court authorities, Loper Bright Enterprises v. Raimondo and West Virginia s. EPA, the Department of Justice cannot make regulations “because Congress has not clearly authorized regulations in this area.” Section 12134 of the ADA requires that the Attorney General “promulgate regulations in an accessible format that implement” Title II of the ADA, which is the part of the ADA that governs state and local government accessibility.  Congress has clearly authorized regulations, so why not website regulations? Perhaps because the ADA doesn’t apply to local government websites.

    More important for most of my readers, this same argument seems to apply to Title III of the ADA, which covers most businesses. In fact, the SBA’s press release prominently mentions the detrimental effect of the regulations on small businesses. Section 12186(b), which is part of Title III, uses the same language as Section 12134 to require regulations from the Department of Justice (although it is a little confusing because there are special provisions related to transportation). If DOJ is not authorized to issue Title II regulations concerning website accessibility then it almost certainly isn’t authorized to issue Title III regulations concerning website accessibility either.

    I would love to see the SBA internal memo that explains its conclusions, but it seems clear that the Trump Administration is not unified on the question of website accessibility regulations. DOJ’s delay in implementing the Title II regulations made sense as a recognition of the technical complexity, but there was no hint that there would be no regulations at all. The postponement of any effort at Title III regulations also made sense, or at least was justified, based on the argument that the Title II regulations would be a testing ground for later Title III regulations. In either case DOJ seemed committed to the idea of regulating website accessibility even if it was slow to do so.

    The SBA recommendation also makes a good point apart from whether the regulations are authorized at all. It writes:

    Advocacy recommends that the DOJ consider exempting small governments with a population of fewer than 10,000 from this rulemaking. Advocacy also recommends that the DOJ provide safe harbors to reduce litigation risks to small governments.

    These points should apply with equal force to small businesses and the need for a safe harbor. ADA website litigation is driven in part by the existence of enormously complicated technical standards that cannot, as a practical matter, be met with a website that has constantly changing content, which is true of almost all business websites. These standards also make no distinction between requirements that are necessary for a disabled person to do what the website is for (such as buy something) and those that are secondary to the main function (such as links that allow a visitor to “Like us on Facebook!”).  DOJ would be doing businesses, especially small ones, a favor if it promulgated regulations that did not simply adopt WCAG 2.2 AA technical standards wholesale as the Title II and earlier proposed Title III regulations had done. Instead, if DOJ is going to regulate, it should adopt regulations that distinguish between elements essential for the use of the website for its purpose and elements that are secondary. This would not completely solve the problem of industrial scale litigation driven solely by the desire of lawyers to make money,  but it would help.

    Finally, the SBA announcement is likely to be more than the opening shot in a battle between agencies about the limits of regulatory authority. I suspect that underlying the argument about the power to regulate is an argument that the ADA simply does not cover websites, an argument accepted by the Courts of Appeal in several circuits with respect to Title III entities. Presented as an argument about regulatory authority the question of whether Congress intended the ADA to cover websites may finally make it to the Supreme Court.


    Share
    , , ,
    , , , , ,

  • HUD’s radical change of position on ESA’s

    On May 23, 2026, HUD’s Assistant Secretary for Fair Housing and Equal Opportunity announced a dramatic change in HUD’s view of accommodation requests for emotional support animals.² Stripped of the usual bureaucratic word swamp the point of the announcement is simple. HUD will not find reasonable cause to believe discrimination occurred for any case involving an emotional support animal. If a tenant or resident who was denied a waiver of a “no pets” policy for an ESA files a complaint with HUD it will fail. The same applies to any other request concerning an ESA. Individuals who claim they have a disability related need for an emotional support animal can still file a private lawsuit under 28 USC §3613, but the government isn’t going to help them out. Here is the exact language of the Memorandum:

    “Effective immediately, for complaints related to animal-related reasonable accommodations, FHEO will find reasonable cause and recommend charges only for those cases involving animals trained to provide disability-related assistance.”

    The key words are “animals trailed to provide disability-related assistance.” Emotional support animals are, by definition, not trained to do anything. It is (supposedly) their mere presence that somehow relieves the symptoms of a disability that interfere with the equal use and enjoyment of an apartment or other home. (more…)


    Share
    , , , ,
    , , , ,

  • Emotional Support Dogs again?

    The March 25, 2026 decision from the Connecticut Supreme Court in Connecticut Commission on Human Rights and Opportunities v Mansions takes a small but significant step toward limiting misuse of the Fair Housing Act as a way to avoid pet deposits or “no pets” rules. For full disclosure purposes, I represented the winning landlord throughout the case.

    To understand the decision, which was based on Connecticut law but should be persuasive in Fair Housing Act cases, requires a little background in how a person can be disabled and how that might affect their right to a reasonable accommodation like allowing an emotional support animal. Under both federal and Connecticut law a person can be “disabled” in three ways. First, they can really be disabled; that is, they have a mental or physical impairment that substantially limits a major life activity. Second, they can have a “record” of a disability. That would be something like a written medical record. It doesn’t matter if the record is correct, only that it exists and the landlord knows about it. Finally, they can be “regarded as” disabled even though they are really not disabled.

    A tenant cannot be really disabled and regarded as disabled at the same time. The whole point of “regarded as” disability is that discrimination is wrong even if it is based on a wrong perception of a disability. It is just as bad to refuse to rent to a tenant because the landlord believes they are disabled as it is to refuse to rent because they are really disabled. And a person who is both really disabled and regarded as disabled doesn’t need a “regarded as” theory of liability because they already have the best theory of liability, a real disability.  At the same time, a tenant can prove they are regarded as disabled without having to prove they are really disabled.

    “Record of” disability floats somewhere between these two. Connecticut law is not quite the same a federal law with respect to “record of” disability, but in principle under both state and federal law “record of” disability is like “regarded as” disability. A landlord should not be allowed to discriminate against a tenant because the tenant has an inaccurate record of being disabled any more than the landlord should discriminate against a person because the landlord has an inaccurate belief about the tenant.¹

    Emotional support animals enter the discussion because under both federal and Connecticut law a person with a disability may be entitled to an accommodation – a waiver of some policy – it it is necessary for their equal use and enjoyment of an apartment, house, or other dwelling. But neither federal nor Connecticut law explicitly limits this theoretical entitlement to real disabilities. The bare words in both laws just says that if you are “disabled” you may be entitled to an accommodation.

    With that, we can look at the Mansions case. The aggrieved parties were the Pizzoferratos. Ms. Pizzoferrato suffers from “anxiety” and believed her two shitzus helped with that anxiety. It was important to her and the lawsuit that she said she needed both dogs, not just one or the other.

    [Why did I put “anxiety” in quotes? Because “anxiety” is a feeling or symptom caused by some underlying condition. You can be “anxious” about an upcoming test, or trip, or a relative’s health. That kind of anxiety is appropriate and not a disability. You can also be anxious because you suffer from Generalized Anxiety Disorder, which is a mental disorder in which the person feels anxiety when it is not appropriate or to a degree that is not appropriate. Suffering from Generalized Anxiety Disorder is a disability. Ms. Pizzoferrato was an anxious person, but the trial court did not find that she had the kind of anxiety that would be a disability.]

    Mansions agreed to allow Ms. Pizzoferrato to have one dog despite its “no pets” policy, but refused to allow the second. There is no agreement among the various parties about who thought what and whether what they did was reasonable, but the trial court found that Mansions must have “regarded” Ms. Pizzoferrato as being disabled because it allowed one dog.  And for the trial court this was enough. Ms. Pizzoferrato met the first requirement for an accommodation because she had a “disability” under one of the three definitions of disability. She met the “needs an accommodation” requirement because she said, and the trial court believed, both dogs helped her anxiety. With this the trial court believed she had proved she was entitled to an accommodation allowing both dogs.

    Mansions lost the trial and decided to appeal. In a long opinion the Court of Appeals basically followed this train of thought:

    • Real disability is off the table because there was no finding of a real disability by the trial court.
    • A person who is only “regarded as” having a disability is never entitled to an accommodation because they have no real disability that would create a need for an accommodation.
    • Even though the trial court didn’t say it explicitly, the trial court implied found that Ms. Pizzoferrato also had a “record of” a disability.
    • A person with a “record of” disability might need an accommodation, but that possibility didn’t matter in this case because
    • There was no evidence at trial that Ms. Pizzoferrato really needed two dogs.  This conclusion was based on the legal definition of necessity used in various federal appeals court decisions. The Court of Appeals agreed with those decisions and said that under  that legal definition of necessity Ms. Pizzoferrato only “needed” the one dog Mansions allowed.

    Thus, according to the Court of Appeals, even the implied finding that Ms. Pizzoferrato had a “record of” disability would not give Ms. Pizzoferrato a right to have two dogs as an accommodation. Based on this the judgment of the trial court was reversed.

    Of course that was not the end of the story. The Connecticut Supreme Court is always available to decide whether the Court of Appeals made a mistake. In this case both sides thought that the Appellate Court had made mistakes. Both sides appealed to the Supreme Court and after a lot of briefing and a couple of hours of oral argument² we got the opinion this blog is about. The Supreme Court said:

    • Everybody, including the Pizzoferrato’s and the Commission on Human Rights and Opportunities, agrees that Ms. Pizzoferrato did not need an accommodation if she was merely regarded as disabled. Supreme courts hate to commit to broad statements when a narrow statement will do, and the Court leaves open the possibility that under some circumstances a person with a “regarded as” disability might need an accommodation. But not Ms. Pizzoferrato.
    • Because of this the Court of Appeals was right to reverse the trial court’s judgment. Hurray for my client. We win.
    • But, says the Supreme Court, the Court of Appeals had no business implying a finding by the trial court that Ms. Pizzoferrato also had a “record of” a disability. That implied finding was superfluous to the trial court’s result and appeals courts should not imply superfluous findings by trial courts.
    • Similarly, because a “regarded as” disability would not support an accommodation for even one dog, the Court of Appeals’ discussion of what it means to “need” an accommodation and why Ms. Pizzoferrato didn’t “need” two dogs was also superfluous.
    • We don’t want our legal precedents clogged up with unnecessary discussion and findings so the Court of Appeals’ superfluous findings  about “record of” disability and the definition of “need” are vacated.

    Now from my viewpoint the fact that I won the case for my clients (who really won it for themselves by making the right decisions) is reason enough to shout from the rooftops, but is there any other reason to think this case was important? As it turns out there is. While the Supreme Court was unwilling to say more than the bare minimum needed to get to the right outcome, this opinion appears to be the first by any state or federal court that recognizes what should be obvious: you can’t need an accommodation if you are not really disabled.†  This matters to landlords because many will grant an accommodation request they don’t believe in just because it is easier and cheaper than fighting a HUD or state investigation. If it turns out the accommodation was far more trouble than they thought – as for example when the tenant decides that they are also entitled to take their 90 pound Rottweiler swimming in the pool off leash – the landlord can go back to basics and say the accommodation was never right in the first place because there was no disability. A court may say that by granting the accommodation the landlord “regarded” the tenant as disabled, but that won’t help the tenant because “regarded as” is not a ticket to keeping the dog at all, let alone in the swimming pool. A landlord can accept the bogus accommodation letter purchased on the internet without admitting the diagnosis is correct.

    Also, even though the discussion of necessity by the Court of Appeals was vacated, I think later courts will look at the Court of Appeals’ discussion of “necessity” and find it persuasive. This ruling wasn’t vacated because it was wrong; it was only vacated because it was unnecessary. I think that other judges will be impressed by what the Court of Appeals said in a long and thoughtful opinion.

    Where do we go from here? More than one kind of sloppy thinking about the FHA has contributed to the ESA problem faced by landlords. The cases on “necessity” relied on by the Court of Appeals did not concern animals, and a solid case explaining why emotional support animals should almost never be “necessary,” even for those who really are disabled, would be helpful. What is needed most, however, is for both the health care profession and the courts to give some hard thought to what an ESA must do to be a “necessary” accommodation. Does the ESA help a person with a disability overcome the limitations created by their disability so they can more equally use and enjoy their apartment in the same way a guide dog helps a blind person overcome the limits of blindness? Or is it just a furry narcotic that helps the owner not care about how their disability limits them?  Courts should reject the notion that an accommodation can be “needed” when it does nothing to improve equality of use and enjoyment of a dwelling.³ The purpose of both the Connecticut and federal fair housing statutes is to improve equality, not just make those with disabilities feel better about inequality.

    ++++++++++++++++++++++++++++++++++

    ¹ If it seems like this is needlessly complicated well, it is. The problem is that when defining “disability” (or “handicap” under federal law) the legislature was writing laws to prohibit employment discrimination as well as housing discrimination. In an employment context employers often get medical records that might show a prospective employee is disabled when they have never met the prospective employee. “Record of” disability was created to cover that situation. Of course if the employer or landlord discriminates based on the record they clearly also regard the employee/tenant as disabled, so it is somewhat redundant, but there is a long pre-FHA history to these definitions and so the state and federal legislatures were not really starting from scratch.

    ² If you are interested you can hear the argument at the Connecticut Supreme Court’s website, https://www.jud.ct.gov/supremecourt/Audio/OralArgumentsAudio.aspx. I have not listened to it myself because  I never sound as brilliant in the recordings as I remember I sounded at the time.

    † To be perfectly accurate, a tenant almost certainly can’t need an accommodation if they are merely “regarded as” disabled. However, we are talking about emotional support animals, and the opinion, I think, excludes the possibility that a “regarded as” disability can support a need for an ESA.

    ³ A brief mention of “use and enjoyment.” The word “enjoy” here does not mean have fun or feel good. It means take advantage of what an apartment offers; shelter, heat, water and so forth. Lots of people who are not disabled don’t have fun or feel good in their home, but they still “enjoy” it as long as the roof doesn’t leak and the heat and water work.


    Share
    ,
    , , , ,

  • ADA 30 Days to Comply Act – an additional thought.

    I say “an additional thought” because while I was thinking about what to say concerning the newly introduced ADA 30 Days to Comply Act William Goren wrote his own analysis of the law, which you can read here – Goren Blog. Briefly, the law imposes a 30 day notice requirement as a prerequisite to filing suit and requires that the notice be meaningful; that is, specific enough to help a business owner correct whatever ADA violations are said to exist.¹ One of the most common complaints from businesses who are the targets of this kind of litigation is that they didn’t know there was a problem and they got no chance to fix things. This law aims to remove that complaint.

    My additional, and perhaps cynical, thought is that even if passed² the bill will provide only modest relief from the industrial scale litigation model that dominates enforcement of Title III of the ADA. This is how that model works:

    • A plaintiff who essentially works for a law firm drives around looking for obvious ADA violations. These are not hard to find because older businesses never had appropriate parking and ramps while newer businesses often suffer from deferred maintenance problems.
    • The law firm files a dozen or more suits each month in the name of that plaintiff.
    • The lawsuits can be readily defended because as Bill notes in his blog, there are often real questions about standing to sue.
    • The defendant also has the option of mooting the lawsuit by fixing every ADA problem.
    • But instead the defendant almost always settles. Why? Because the cheapest way out of the lawsuit is a motion to dismiss and the legal fees to prepare and file that motion are greater than what it costs to settle the case.  For the defendant the analysis is simple. It costs $x to settle the case and $y dollars to defend the case. The defense may fail, settlement is a sure thing. As long as settlement is cheaper to defend then settlement is always the rational choice.

    The 30 Days to Comply Act can’t change this analysis. Suppose a plaintiff files suit without sending the required notice. The defendant now has a stronger motion to dismiss than they would otherwise, but it will still cost money to prepare and file the motion and that cost will still likely be more than cost to settle.  If the plaintiff sends a perfunctory notice that doesn’t meet the statutory standard, which is easy enough, the motion to dismiss becomes weaker because what the statute requires isn’t black and white. Title III ADA litigation is mostly driven by the simple economics of cost to settle versus cost to defend and the 30 Days to Comply Act does not change those economics.

    Industrial scale ADA litigation is driven by the economics of settlement and the ease of finding ADA violations. There are some positive things Congress could do to change these drivers, although it does not seem inclined to do so:

    • Spend taxpayer money to make sure all businesses understand their Title III accessibility obligations. I am frequently called by businesses and even other lawyers who believe businesses with fewer that 15 employees are exempt from Title III or that older businesses have been grandfathered. Both notions are wrong. There is no occasion in the process of opening a new business that qualifies as a public accommodation for a business owner to learn that they have ADA accessibility obligations. The IRS communicates with every new business (and every established business as well) so there is a natural channel for sending this information at minimal cost.
    • Eliminate the judicially created doctrine that merely seeing an ADA violation creates an injury sufficient to meet the standing requirement in Article III of the Constitution. The Supreme Court may do this some day, but Congress can do it now.
    • Provide for an award of defense attorneys’ fees if a lawsuit is dismissed under Rule 12(b)(1). This would make the attorneys who drive these cases and their “clients” think twice about filing suits based on a quick drive-by with no possible injury and standing.

    The first of these is the most important because, as plaintiff attorneys love to point out, the ADA is decades old and by now  we should not have common problems like accessible parking, ramps and sidewalks in sufficient numbers for making large scale litigation possible.  Congress decided, when it passed the ADA, to cheap out on helping businesses understand and comply with the law, leaving private lawsuits (and the publicity they cause) as the most common ways for businesses to learn what the ADA requires.  I’ve written about this before³ but it is worth saying again; industrial litigation only exists because of widespread non-compliance with the requirements of the ADA. Many of the “problems” are trivial in terms of their real impact on accessibility, but that is the subject for another blog. In most cases basic compliance is relatively inexpensive and the real problem is ignorance. Congress should think about attacking that problem before it tries one more time to impose a 30 day notice requirement on ADA lawsuits.

     

    +++++++++++++++++++++++++++++++++++++++++

    ¹ The law concerns standards for physical access to “places of public accommodation;” meaning for the most part businesses open to customers.

    ² Similar bi-partisan bills have consistently failed to pass for more than a decade, even when one party controlled both houses of congress and the presidency.

    ³ See, for example my earlier blog Disabled Access Credit Act.


    Share
    , , ,
    , , , ,