clock menu more-arrow no yes mobile

Filed under:

The surprisingly weak scientific case for emotional support animals

Increasingly, Americans are bringing pets on planes to destress. But there’s little rigorous evidence to back them up.

svetikd/Getty Images
Brian Resnick is Vox’s science and health editor, and is the co-creator of Unexplainable, Vox's podcast about unanswered questions in science. Previously, Brian was a reporter at Vox and at National Journal.

A college student wanted to bring a hamster on a plane and then flushed it down an airport toilet after Spirit Airlines told her she wasn’t allowed to board with it. A United Airlines passenger attempted to get on a flight with a peacock. Another air traveler took a turkey on a plane. Yet another brought on a duck wearing red booties. Last winter, a dog’s teeth scraped a little girl’s head on a Southwest Airlines flight. Earlier this month, a Delta passenger complained that his seat was covered in dog feces.

These were real events that happened in America: travelers toting “emotional support animals,” claiming they need the ESAs (which are distinct from service animals trained to help those with physical disabilities) to stay calm while flying. And there are more of them out there: back in January, Delta reported that it carries around 700 service or support animals daily and has had to create a special support desk for them.

How is it legal to bring your duck on the plane? Under the federal Air Carrier Access Act, passengers are allowed to bring animals aboard by showing a letter from a mental health clinician or doctor asserting that the pet is part of their therapy. But the law is surprisingly vague about which species can come on board and gives airlines significant discretion. “You are never required to accommodate certain unusual service animals (e.g., snakes, other reptiles, ferrets, rodents, and spiders) as service animals in the cabin,” it reads.

Yet as a quick Google search will show, it’s possible to obtain these letters online for a small fee. Some passengers may very well be exploiting the law to bring pets on planes. (Delta, at least, now also requires a note from a veterinarian.) And stories about peacocks and ducks in booties on planes are increasingly leading ESAs (and their handlers) to be treated as a punchline. In the New York Times, columnist David Leonhardt called the animals a “scam” and “one of the downsides of a modern culture that too often fetishizes individual preference and expression over communal well-being.”

But before we consider these animals a national blight, we should ask: Do they actually work to help people in distress cope? What do we really know about the emotional support value of pets? It’s a question that might be on your mind this holiday traveling season, if you see these companion animals at the airport.

Molly Crossman is a psychology researcher at Yale who published a 2016 review in The Journal of Clinical Psychology of the evidence on using animals to counter psychological distress. Here’s what she found: “The clearest conclusion in the field is that we cannot yet draw clear conclusions.”

Crossman’s research is about finding ways to reduce the mental health treatment gaps in America. “Traditional models of treatment therapy and medication reach a very small proportion of the people who actually need services,” she says.

And given that one in five Americans experience mental illness in a given year, she thinks animals are an intriguing option to help more people. They could also be crucial for combating the loneliness that comes with an aging population.

I called her up to talk about how animals might be beneficial for mental health — a question that, frustratingly, doesn’t yet have a clear answer.

This interview, conducted in February, 2018, has been edited for length and clarity.

I mean, just look at this face.
Martin Poole/Getty Images

Brian Resnick

Do we really need rigorous empirical evidence to know that pets bring comfort to people? Isn’t that kind of obvious? Many, many people have pets. Obviously they bring joy.

Molly Crossman

Yes. I get that question a lot. A lot. There are a few different answers.

One is that there are different standards of evidence.

So if you want to say that “my pet makes me feel good and it’s fun,” that’s great. You don’t really need lots of evidence for that.

But with these emotional support animals, we’re talking about what is essentially a prescription from doctors to people with clinically significant symptoms. When we talk about that, there are very specific standards of evidence for psychiatric and psychological treatment, and these have not met that standard.

Brian Resnick

So is there good evidence that pets reduce emotional distress?

Molly Crossman

A lot of people have this impression that [the evidence] is very well established and we really know that [animals] are beneficial. But what is surprising is that we actually don’t know that at all.

What the best evidence has to say about animals and distress

Kerrick/Getty Images

Brian Resnick

Overall, what are the strongest claims we can make about animals and mental distress?

Molly Crossman

Yeah. Well, I’ll qualify it first by saying that most of the research in this area is on dogs. There is some on horses as well, and a few studies on other species.

But in terms of the dog studies, we can say that, probably, interactions with animals don’t make stress-related symptoms worse. So that’s good.

It also seems they convey sort of small to medium reductions in stress and stress-related symptoms. That’s the strongest thing I’m willing to say.

Brian Resnick

How is the research flawed?

Molly Crossman

We actually don’t know that it’s the animals specifically that are producing these small to medium reductions in stress. It might be other components of the interventions in which they’re evaluated.

It might just be people get better over time. And if you have no control group at all, you can’t say the animals were more beneficial than just people coping on their own.

In cases where [researchers] do include control groups, they’ve often been what we call “no treatment” control groups. They’re just sort of a waitlist. [With that design, it’s hard to know the benefits are directly attributable to the animals.] Is it the handlers of the animals, who often interact extensively with participants? And we know that social interaction with people is essential for mental health. So there’s all these questions about how important are the animals [in these therapies].

In our research, we’re trying to get to the point where we can say there seems to be something here specific about dogs. That it’s not the other people who are involved; not just the passage of time.

Brian Resnick

I imagine the research is also made difficult by the fact there are so many different programs that use animals to reduce distress. There are some where animals with handlers are brought to hospitals with therapies. Which is different than owning a pet and having it at home.

Molly Crossman


Brian Resnick

So you have reviewed a great amount of the research on using animals to reduce distress. Is there research specifically on “emotional support animals”?

Molly Crossman

They really haven’t been studied. Research on pet ownership should speak to them. But there’s very limited, if any, research on emotional support animals specifically.

The research on dogs is inconclusive. The research on emotional support peacocks and hamsters doesn’t exist.

Dragan Todorovic/Getty Images

Brian Resnick

So why are clinicians allowed to prescribe “emotional support animals” despite the lack of evidence?

Molly Crossman

In psychology, we don’t license treatments — we license providers. So there’s no FDA for psychotherapy. Once you’re licensed, within a certain range, you kind of have the freedom to use whichever treatment you want. And even though some treatments are considered evidence-based and some are not, there’s nobody regulating which you use.

Brian Resnick

I can see why the concept of “emotional support animal” is appealing for mental health clinicians. When I talk to clinicians in reporting, they sometimes lament how their time with patients is so limited to just a few hours of appointment. An emotional support animal is a full-time treatment.

Molly Crossman

It is extremely compelling, and I think that’s one of the draws, is this idea of full-time support.

Brian Resnick

Backing up: Do we even have good data on whether dog ownership increases well-being overall?

Molly Crossman

No. And actually, I think that’s one of the areas where the evidence is relatively weak, and that’s just because it’s a very hard question to answer. You need very, very big data sets, where you can sort of control lots and lots of variables in people’s lives. Isolating the effect of a pet in the context of all the other factors that influence a person’s mental health is so hard, so the evidence there is really, really mixed.

Brian Resnick

So most of this research is on dogs, and it’s not very conclusive. I’m guessing it’s not right to generalize to other species, like hamsters.

Molly Crossman

It’s not.

Probably different animals are good at different things. So, for example, we’ve seen just within dogs, different dogs have different personalities, and they convey slightly different effects. We start with dogs, and I think a lot of other research groups do as well, because it seems like dogs are sort of specially designed, for lack of a better word, to interact with people, to understand our social and emotional cues.

Emotional support animals could be harmful in some cases. And those effects need to be studied.

Fox Photos/Getty Images

Brian Resnick

Is there potential for harm with emotional support animals?

Molly Crossman

I do have concerns about the potential for harm, which isn’t to say that I think they’re definitely harmful. I just think it’s an important possibility, and it’s a little bit of a complicated explanation, but I’ll try.

Basically, in treatments for anxiety that work, we ask people to face their fears. We work with them to gradually approach the things they’ve avoided. These treatments work really well. They’re some of sort of the best mental health treatments that we have. A concern we have in those kinds of treatments is that people will feel like, “I can only approach this terrifying situation if they have my mom with me, or my blankie,” or whatever.

If someone who’s very afraid of heights is only able to approach heights with their emotional support dog, they might start to say, “Wow, I can do this because of Buddy.” We want you to learn you can do it, and we don’t want it to be about Buddy.

While Buddy makes the fear better in the short term, he’s making it worse in the long term. So that’s the concern, is that you’re reinforcing fear in the long term.

Brian Resnick

The animal becomes a crutch.

Molly Crossman


Forcing some animals to provide emotional support could be bad for their welfare

Think this turtle like flying?
Getty Images/EyeEm

Brian Resnick

Could an animal be harmed as the result of supporting a person’s emotions? I could imagine emotional support tortoises hate being on airplanes.

Molly Crossman

We’re talking about treatments that involve live animals, and we have an obligation to those animals to make sure that if we’re sort of using them for our benefit, that the benefits are really being conveyed. Because, for example, being something like a service animal is hard work. Being on a plane can be stressful for an animal; these kinds of things really raise the stakes in my eyes for making sure that they work.

I think importantly, there are some species in which it’s probably not wise to be asking the questions in the first place, because [the] animals themselves are not well suited and will not enjoy doing this work, and there are risks of harm to the people and to the animals. In general, it’s a problem when people advertise benefits of interaction with species that are not domesticated, because there are risks to those interactions, and then people get upset when bad things happen.

My favorite example is [Washington University in St. Louis] brought a bear cub to their campus as part of a pet therapy program, which should be such an obviously bad idea, right? A bear?! Yeah.

But it’s not obvious, and there was this media frenzy when this happened because the bear cub attacked a bunch of students, and then there was this rabies scare. And, like, we just can’t blame the bear in that scenario. It should not have been brought to a campus to play with students. That was not a great idea.

How to move the research forward

Wouldn’t you want to know if these pups are good for decreasing distress? Let’s do the studies to find out.
Catherine Ledne/Getty Images

Brian Resnick

How do we find out if dogs are beneficial?

Molly Crossman

I think to the question of does that study exist, no, because no single study is gonna tell us that. It’s an accumulation of evidence, just like in any field. So I think our group is trying to do that. There are a number of other research groups around the country trying to do this as well, and I think we’re getting there. So we have a few recent studies where we’ve shown that dogs, interacting with a dog, outweighs the effects of just kind of being exposed to it, viewing it, so the interaction is important. The effects of interaction with a dog exceed those of other common coping strategies that people use, and also exceed the effects of sort of any intervention that people are told will work.

Brian Resnick

What does the coverage of emotional support animals in the media miss?

Molly Crossman

I think what most of the media coverage misses is that the central issue here, from my perspective is, do these animals work. I don’t know how to put it in a more refined way, but do they convey the effects that we hope? I think until we answer that question, all of these other arguments [such as whether they’re a sign of a fragile American psyche] are a little bit peripheral.

From the perspective of psychologists, we should answer that question first and then worry about what it says about millennials.

Brian Resnick

In your review paper, you write, “Animals are not equipped to overhaul the limitations of the mental health care system.” What do you mean by that?

Molly Crossman

I think animals can reach lots of people outside of traditional models of treatment. Animals might bring a few extra people in because they’re appealing and can sort of facilitate entry into treatment, [but] they’re not gonna close that gap between. ... Roughly 30 percent of people in need of treatment get any kind of treatment, and I would say that’s a pretty conservative estimate. Animals are not gonna take that from 30 to 100 percent. That’s just not gonna happen.

But, like I said, 70 percent, almost, of homes in the US have pets, so if pets are beneficial and we can somehow leverage those benefits, they might reach lots and lots of people that way.

Sign up for the newsletter Today, Explained

Understand the world with a daily explainer plus the most compelling stories of the day.