Wednesday, December 25, 2013

Pets as gifts: should we trust adopters?

A few days ago I tweeted a link to a blog post by the ASPCA’s Dr. Emily Weiss questioning the shelter dogma that animals should not be given as gifts. In this and an earlier post, Weiss describes her research which suggests that animals given as gifts are no more likely to be given up than animals not given as gifts. That article is open access, so you can read it and judge for yourself. It is a retrospective survey, so there is room for more rigorous science about this topic, but the paper definitely opens up interesting space for discussion and further investigation.

My bosom buddy Julie Hecht posted her thoughtful response on her Dog Spies blog and then ruminated more in a letter to her blogging pen pal Mia. Julie and I got into a conversation about it on Twitter, which unfortunately led to me ranting a bit (not in a hostile way, just in a “I thought about this all so much during my shelter medicine internship and I must let you all know everything I learned!” way). I’ve been thinking since then that 140 character spurts is not the best way to get across what I was trying to say.

Here is the story we tell ourselves about animal sheltering: there are irresponsible people out there. Lots of them. And they have animals, which they don’t value as animals deserve to be valued. They bring the animals to shelters, where people who care more and know more do their best to find the animals good homes. It is the job of the shelters to place these animals in the best homes possible, and to that end they should be very careful about every placement, because animals who have been abandoned once deserve never to be abandoned again.

There is a lot that is true in this story, mainly that animals do get the short end of the stick way too often and that, once abandoned, they absolutely deserve for the rest of their lives to be catnip and sunny couches or steaks and tennis balls. What I question is whether the shelter system that we are able to provide today is equipped to manage them well for long enough to find those homes, and whether shelter workers have the information necessary to predict what kind of home a particular adopter is actually able to provide.

Many, in fact most, shelters in this country are overwhelmed and are still euthanizing adoptable animals to provide space for more animals to come in. There are shelters for which this is not true, more and more of them every year. But they are the exception, and they tend to cluster in particular parts of the country. My friends in New England were shocked when I told them that during my internship I saw shelters where euthanizing healthy kittens for space was common. So given this situation, is it better to hold on to animals until you can find them the home that you think is perfect? Or is it better to take a chance and hope that you can get that animal out of a shelter which may have a 50% kill rate?

That leads us to the question of these lovely shelters which are able to place every medically and behaviorally healthy animal, and often even some not so healthy ones. These days there are plenty of those out there too. What should they think about animals as gifts?

I think that Dr. Weiss’s article makes the point that we aren’t really sure that we have all the information necessary to judge a particular adopter. I don’t think that this particular study makes an open and shut case. But I do think it provides evidence that this is a question worth asking. What do we really know about the home any adopter is going to provide? Is it worth denying an animal a potentially loving home because you don’t trust the word of the adopter?

In her post, Julie argued that there are some cases in which animals as gifts are particularly bad ideas, giving the example of bringing a puppy home to your grandmother who does not have the energy to deal with it. I agree. My suggestion is that shelters should consider moving to more case by case evaluations of particular adoptions, rather than operating on policies alone. If an adopter makes a good case, consider the adoption, even if they are planning to do something like give the animal as a gift. Keep an open mind about what constitutes a good home. But in the case where the adoption is patently a bad idea, then yes, talk the adopter through making a better decision, and refuse the adoption if need be.

In academic shelter medicine, where we like to think about changing everything about shelters because we don’t have to actually operate shelters, there has been a lot of discussion about this kind of change. Outdoor cats? No home visit prior to adoption? No adoption fee at all? Maybe those things are all good ideas. Maybe we really don’t know much about what makes a good adoption or a good home. We guessed, for years, and that was all we could do. But there is more and more research in the shelter community these days. We are starting to apply science (SCIENCE!) to these questions. I hope we can all both keep our minds open and evaluate the incoming research rigorously. It is a fascinating time for shelter medicine and shelter research; as one academic shelter veterinary specialist said to me, when I expressed shock at the overturning of some old principle or other, “Everything is on the table.”

Happy holidays to you and your animals from me and mine!

Sunday, December 22, 2013

New research on dog bite fatalities

We’ve known for a while what kinds of dogs are at risk of biting humans: not any particular breed, but dogs who are not well socialized and not well cared for. Dogs living in houses with people are much less likely to bite than dogs living outside in yards or on chains.[1, 2] So why is this new paper about dog bite fatalities important?

Patronek G.J., Sacks J.J., Delise K.M., Cleary D.V. & Marder A.R. (2013). Co-occurrence of potentially preventable factors in 256 dog bite–related fatalities in the United States (2000–2009), Journal of the American Veterinary Medical Association, 243 (12) 1726-1736. DOI:
Unlike previous researchers, who mostly approached the question of who gets bitten and what kinds of dogs bite by combing through old records, these authors monitored current events and followed up on every dog bite related fatality that was reported, for ten years (2000-2011). They interviewed law enforcement officers who were involved with these cases. They interviewed medical examiners and coroners. They followed current news articles about cases. This is all information that becomes very difficult to find when you’re trying to learn about a dog bite fatality years after the fact. As the authors write:

In our opinion, the present study represents the most comprehensive analysis of factors...associated with dog bites to date. Personal interviews with credible investigators were successfully conducted in 221 of 256 (86.3%) cases... Law enforcement personnel provide first-hand information not reported in the media and often identified errors of fact in the media reports.

Some information was still very difficult to obtain, and the most interesting part of the paper for me may have been the description of the lengths the investigators went to in their attempts to ascertain the reliability of reports of what breed some of these dogs were. They note that “the source of breed descriptors in media reports is usually unknown” and therefore not trustworthy. Interestingly, this paper never put that comment into context, but it is hard to read it without thinking about how challenging it can be to visually identify the heritage of a mixed-breed dogs, and all the implications that this has for news stories which seem to reflexively identify aggressive mixed-breed dogs as “pit bulls.”

In the context of the debate about whether pits get disproportionately named in media reports about dog aggression, this paper provides some interesting fodder. The authors calculated how often media reports contradicted each other: 21.6% of the time in reports about incidents involving single dogs, 36.4% in incidents involving multiple dogs. How often media reports differed from the animal control officer’s report: 34.9% in incidents involving single dogs, 43.3% in incidents involving multiple dogs. In the rare cases when a pedigree or DNA testing was available, that data disagreed with media reports in 7/19 cases for single dog incidents and 7/28 cases for multiple dog incidents.

What this paper found overall was mostly a vindication of what we already believed: there is no single factor that leads a dog to bite a human. But one very important factor is whether the dog is a “family” dog or a “resident” dog. The paper provides some lovely verbiage on the difference:

A resident dog was a dog, whether confined within the dwelling or otherwise, whose owners isolated them from regular, positive human interactions. A family dog was a dog whose owners kept them in or near the home and also integrated them into the family unit, so that the dogs learned appropriate behavior through interaction with humans on a regular basis in positive and humane ways.

Later in the paper, they add:

Dogs that are deprived of human interaction or direction are denied access to accurate information about appropriate behaviors with humans. Consequently, dogs in stressful, potentially dangerous situations or when maltreated may behave in ways primarily to protect themselves.
In other words, dogs who are not given a chance to learn how to interact appropriately with humans may not act appropriately with humans.

The rest of the paper is packed with nice statistics which I am not going to try to reproduce here. Suffice to say I expect to see excerpts from it on slides in presentations about canine aggression for years to come. I do want to explicitly point out that this paper only covered dog bite fatalities, not dog bites alone; fatalities due to dog bites are extremely rare (this paper found 256 in the United States over a 10 year period), whereas dog bites alone are quite common. I think it’s easy when reading this paper to want to extrapolate all this lovely data about the causes of fatal dog bites out to the causes of non-fatal dog bites. That’s understandable but a little dangerous: it usually requires repeated bites to kill a human, so I imagine such an attack to be different from the more common single bite. But I still believe all this data is very relevant to how we keep our dogs and how to prevent bites. The message the authors give is: be responsible with dogs and they will treat you well. Don’t, and you might be on dangerous ground.

[1] Patronek, Gary J., et al. "A community approach to dog bite prevention." Journal of the American Veterinary Medical Association 218.11 (2001): 1732-1749.

[2] CDC. Home and recreational safety. Dog bites.

Monday, December 2, 2013

The epigenetics of fear

I learned something new today about fearfulness, which it turns out has an even more complicated set of causes than I had previously known. And I had previously thought that fearfulness (made up of a whole lot of little genetic causes as well as almost impossible to fully comprehend environmental causes) was pretty damn complicated. The findings I’m going to describe are in mice, but this stuff is totally relevant to fearful dogs, at least in the opinion of this dog zombie.

My story begins earlier today when I received email from an ex asking if a recently published study is too crazy to be for real. (I do actually enjoy being the translator of Nature Neuroscience articles for the ex-boyfriends of the world.) My ex had encountered a National Geographic Phenomena article which covers the Nature Neuroscience article “Parental olfactory experience influences behavior and neural structure in subsequent generations." That is quite a title — let’s try it again. “When mice are trained to fear a particular smell, the brains and behavior of their offspring are affected.” (The Phenomena article, by the way, is detailed and provides some nice snippets of interviews with the researchers who did this study, but misses some of the nuances of the experimental setup. So while I do recommend you read it if you’re interested in this study, you should probably take it with a grain of salt.)

Dias B.G. & Ressler K.J. (2013). Parental olfactory experience influences behavior and neural structure in subsequent generations, Nature Neuroscience, DOI:

These researchers took a group of mice and fear conditioned them to the smell of a chemical called acetophenone. Then they bred them and tested their offspring. The offspring could detect acetophenone at lower concentrations than other mice; they had more receptors in their nose for detecting acetophenone than other mice did; and they were more reactive to loud noises after having been exposed to acetophenone. The smell was inherently scary to these offspring mice, even though they had not previously encountered it.

For the record, I am totally down with the first few changes. Offspring are adapted to the parent's environment by being better at smelling a relevant smell? Freaky as hell, but that is what epigenetics is and why we all find it so fascinating. But a change in behavior? That is beyond the usual freakiness of epigenetics. That's not just passing along more scent receptors for a particular smell. That's passing along the emotional content of the parent's experience with the smell. How is it possible!

Well, first, some details about the experiment:

How severe was the fear conditioning? Not all that severe, it turns out (which makes the results even more surprising to me). Mice were only trained over three days, with only five trials each day. A trial consisted of exposure to the odor, followed by a “mild” foot shock. I don't have a feel for how traumatic this experience was for the mice, and I'd be curious to know more. Was the shock really “mild”? You know, according to the mice, not according to the researchers, because we have all seen instances in which the animal's perception differs from the human's. Was being in the training chamber itself somewhat traumatic? Maybe the animals hadn’t been out of their home cages before. And so forth. But it was certainly a short period of training.

To test the startle response, the researchers put offspring mice into a startle chamber. The mice were habituated to the chamber for a few days before testing began. Then a few startle trials were run, in which the mice were exposed to sudden loud noises, and their responses were recorded. Next the mice were exposed to acetophenone, and then some more startling noises. The difference in their response was what was important: how much more did they startle after having been exposed to the smell, as compared to before exposure to it? Note that the mice were not actually startling just after exposure to the smell; there was a loud noise which triggered the startle. But they seem to have been primed by their reaction to acetophenone to startle more at subsequent noises.

Now, there are a zillion different possible explanations for why these mice could have appeared to be afraid of a smell that they had never encountered before. "Because my dad was afraid of it” is not the first thing that comes to mind, and the researchers tested a whole lot of other possibilities.

Were these mice particularly reactive to all smells? The researchers actually tested two groups of mice on two different smells. The group which was descended from mice trained on smell A reacted to smell A and not smell B. And vice versa for the other group. It really was just that particular smell.

Were these mice more anxious over all, possibly due to their father’s experience, having nothing to do with acetophenone? We might already have rejected this idea as the mice only reacted to the relevant smell, not the control smell. But the researchers also performed a test to see if the mice were particularly anxious over all, by examining the mice's fear of open spaces. The mice were no more afraid of open spaces than average, suggesting that they were not particularly anxious individuals in general.

Was there some social influence passed down from the fathers? The researchers had begun by fear conditioning male mice, who never had direct contact with their offspring, but did have direct contact with the mothers. It was possible that the fathers had somehow taught the mothers to fear the smell of acetophenone, and the mothers had passed this down to the offspring. To control for this, the researchers artificially inseminated some mice so that the females never interacted with the males, and had the offspring raised at an entirely different lab. They also fear conditioned mothers, and then fostered the offspring to mothers who had not been fear conditioned (and fostered offspring from normal mothers onto fear conditioned mothers). None of this changed the findings: the phenomenon appeared to be genetic, not social.

The researchers had chosen this particular smell because they knew what gene controlled the receptor for it. So they looked at the mice’s brains, and indeed the offspring of fear-conditioned mice did have more of the receptors for the relevant smell, which is why the mice were able to detect it at lower concentrations, even though they had never been exposed to the smell previously. Looking at the DNA for the gene controlling this receptor, they found epigenetic changes, specifically less methylation — basically, less stuff on the DNA, making it easier to express genes from. This is a plausible explanation for how the receptor changes happened.

Which means the story goes like this: mouse is trained to fear a smell; there are changes to the mouse’s DNA, marking a particular gene as one that should be expressed more often; these epigenetic changes are passed on to the mouse's offspring; that offspring generates more of a particular kind of smell receptor, because that gene is marked as “important, make lots!” And I am okay with that, as far as it goes. But how do we get from “make lots of receptors for this smell” to “when you smell this smell, be prepared for Bad Things to Happen”?

There is some precedent for this sort of thing, although it's limited. Primates are known to be primed to recognize snakes, although it's less clear if we are primed to fear them. Mice fed acetophenone while pregnant produced offspring who preferred the smell. Neither of these phenomena are epigenetic, which makes them inherently less freaky. It's particularly interesting to me that mice will “prefer” acetophenone if their mothers have eaten it: another case of inherited emotional content or salience, although in this case due to the in utero environement, not to epigenetics.

But in the end we don't really know how the salience of the smell was transmitted. More receptors for the smell don’t cause salience: just because you can smell it better doesn’t mean you’ll like or fear it. The researchers don't try to make a guess at how this happens, but they do comment on its importance for future research: “Such a phenomenon may contribute to the etiology and potential intergenerational transmission of risk for neuropsychiatric disorders, such as phobias, anxiety and post-traumatic stress disorder.” And dogs! Mice are actually probably a better model for dogs than for humans in this case, because dogs are so much more scent-oriented than we are.

So what does this mean for fearful dogs? We all want to know what makes a fearful dog fearful. How much of it is environment (poor socialization) versus genetics (starting life having been dealt a bad hand)? Well, first of all, remember that this was a very simple stimulus — a very specific smell and very straightforward classical conditioning. That’s why the researchers chose it. Could fear of the mailman be passed on as well? It would be harder, since there is not a single receptor to recognize the mailman, controlled by a single gene which can be expressed more frequently. (I love the idea of a mailman receptor, though.) So I wouldn't extrapolate these findings to non-scent stimuli quite yet. But that doesn't mean that this weird epigenetic force is not out there, interacting with the other poorly-understood forces of environment and genetics, in a crazy storm of things we can't separate out.