Friday, October 16, 2015

Talking about shelter behavior assessments

Today I presented at APDT's 2015 conference on shelter behavior assessments. It's incredibly important to be able to identify dangerous dogs when they come into shelters so we don't put them on the adoption floor, and to be able to identify dogs who we can perhaps help improve their behavior while in the shelter.

Or is it? I talked for three hours -- well, not quite three hours; my amazing audience helped out with some really fascinating discussion -- about how shelter behavior assessments aren't really all that good at identifying dogs who are just sorta likely to be aggressive. They're great at identifying really aggressive dogs and they're great at identifying really safe dogs -- but then again, we don't really need their help at that as it isn't all that hard to do. What neither these tests nor us humans are great at is identifying the in between, hard to categorize dogs.

I argued that we should continue to perform shelter behavioral assessments on dogs because those interactions with dogs give us more information about the dogs' personalities, and that information is useful. What we really should not do is use these tests as yes-no decision making tools for deciding the dogs' fate. They are not decision making tools; they are information gathering tools. One of the other main themes of the talk was that assessing a dog's personality is something that should be done by someone with plenty of dog experience, not the shelter staff member who read the behavioral assessment guidelines once and figures that's all she needs.

After the talk I said hi to Janis Bradley of the National Canine Research Council and she basically said, Hey, fun talk, but I really think we shouldn't be doing behavior assessments on shelter dogs at all. I've asked lots of competent shelter staff if they know which of the dogs in their shelters are dangerous, and they say sure they do. I've asked if it was a behavioral assessment that helped them figure that out and they say it never has been. It's been the dog's interactions with staff and volunteers.

I replied that we really need to collect as much information as possible about shelter dogs, not to identify the easy to identify extreme cases, but to identify the harder to identify in between cases -- the dog who isn't aggressive to all dogs, just certain dogs, for example.

She said sure, but she still thinks a better way of collecting that information is through careful, possibly structured documentation of the interactions of the various shelter staff and volunteers with the dog during its time in the shelter. That's what we should be focusing on.

Now, I am absolutely down with recording as much data as possible about a shelter dog's behavior. But advocating against formal behavioral assessments, even in shelters that have the resources to do them? My heart isn't quite there yet, but it's an interesting idea. If you have opinions, feel free to weigh in in the comments or on Twitter!


11 comments:

  1. UGH, Jessica I can see how your draw dropped and heart sank when Janis said she didn't think that behavior assessments are useful. I completely concur with you that they are crucial. I also agree with her that the interactions w/ staff and dog is critical as well. The trouble is that they don't know how to read dog body language. They don't know the subtle parts of the dog that are talking so how can they access the dog correctly. I've volunteered at shelters and while their hearts are in the right place they are uneducated in how to speak dog. The obvious signs yes, but not the not so obvious signs and in different situations. It's a passion of mine to teach body language to shelters, vets, owners, dog trainers, educators and parents. That's why I developed the Dog Decoder app. It's so frustrating b/c people don't know that they really don't know how to read dogs. I'm with you on this one so don't give up. This education just needs to get out there. It's shocking to me that dog training schools curriculum doesn't have a course in body language so of course it's not on the radar as it should be. How can this be? I'm sorry, I'm ranting but it's such a pet peeve of mine. Thanks for listening and don't give up. I'm not going to. It's my mission to have the app in every shelter worldwide.

    ReplyDelete
    Replies
    1. To be clear, I wasn't criticizing Janis's idea -- I was honestly thinking it through. But I 100% agree that there should be people working in every shelter who are highly skilled in both reading dog body language and in modifying canine behavior. I do think that having a very competent trainer on board will turn out to be cost effective for all but the most cash-strapped shelters in terms of increasing adoptions, decreasing length of stay, decreasing intakes... But of course we live in the real world and it's really hard for shelters to find and keep such people with their limited resources.

      Delete
  2. Objective information gathered by competent observers, via both formal assessment and daily interaction, is only the first step of the process. The crucial step is in the correct interpretation of these observations. As our subsequent actions will be guided by this interpretative filter, we must be responsible to the animal and to the process with our education, our experience, and our empathy.

    ReplyDelete
    Replies
    1. I completely agree. In my presentation I didn't really talk much at all about how to interpret the information you get -- just that it's important to have someone with the skills to do it!

      Delete
  3. I think a formal assessment is appropriate. It should be short, repeatable, and tweaked until it does give the information that will identify ranges of problems/aggression. And if possible the same person should do it each time, or people who have trained together ... so there is comparability in the review.

    I worked as a consultant in long term care for over 25 years and can tell you that the likely-hood of maintaining careful documentation of the interactions of various shelter staff and volunteers is a pipe dream. We had licensed staff, frequent audits and hallway computerized input sites and were frequently disappointed with the disparity of what was the problem and what was in the documentation. And the staff at shelters doesn't have regulators breathing down their necks to get the documentation done. So, sure, try to get better documentation, but know that many people feel that having to write things down is a punishment for them, that many people's memories are not accurate. That they are reinforced by doing something for or with the animals, and getting praised by friends and co-workers, but not by paperwork. So, of course, there is push-back against it.

    The advantage of periodic, formal, behavioral reviews is the repeated assay, the necessity of doing and the discussion that ensues when it doesn't match what people are feeling or think they know. So that's the discussion that builds the information.

    Note: now that I'm not in the long term care consulting business, I train dogs ... I listen to what people say they know about their dog - it seldom matches well to what I see in the relationship. Their bias stains what they observe. Often the main reason they have contacted me, changes quite soon in the training process too. They miss the interactions that set the pathway and only notice the more obvious results (once this is clarified the issues change). Maybe some of this might be why so many think that formal behavioral evaluations don't reflect the total dog. To make the formal assessment more palatable perhaps key training, fun doggy games could be highlighted when scoring in any parts of the review are below the recommended levels of acceptability. Good luck.

    ReplyDelete
    Replies
    1. All true, and thank you for this thoughtful response! I guess the question is -- is it more of a pipe dream to have people recording accurate information or to have an assessment that is "short, repeatable, and ... does give the information that will identify ranges of problems/aggression" ?

      Delete
    2. Yes, it can be developed. Others have done it, the MDS (minimum data set) was developed as an assessment and implemented nationwide in long term care in 1998. Did we groan and complain, yes. Is it still in place (with some tweaking)? Yes. The issue was the variability and questionable accuracy of the assessments not driving appropriate care/interventions ... sounds familiar?

      OK, so this was for people, not dogs, but still similar issues of being able to assess and then figure out what to do based on the assessment. And the idea was to have a minimum set of data that was always gathered and based on that, gather more in areas of concern. The other idea was to have something that would be consistent and provide national data ... which the MDS does because it is computerized and collected/evaluated.

      Recently lots of trainers have been reporting much higher rates of reactivity, more aggression, more fear, more ... but there is no real data. Just anecdotes. Maybe it's time for data.

      Delete
  4. The evaluation of a dog's behavioral tendencies starts with the gathering a good history of their behavior in the home (if known) because there is no better predictor of future behavior than past behavior. From there the evaluation continues as we observe the dog’s behavioral responses to the shelter environment - how they cope with the daily stress of the shelter, how they interact with the variety of people they encounter, how they handle the confinement, how they behave when they walk past a dog, etc. This evaluation should start the moment the dog enters the facility and not end until the moment they leave.
    The benefit of having a standardized assessment that is done a few days after the dog’s entry into the facility is that it provides us with information that may take weeks to elucidate simply through daily interactions with staff. The goal of an animal shelter should not be to house the dogs for weeks or months in order to get to know them - but to house them short term and find them an adoptive home quickly. Additionally, it is important to promptly identify dogs that have dangerous propensities before those propensities “come out” and a staff or volunteer is bitten.
    The formal assessment can also elucidate the dog’s behavioral responses to situations that will most likely occur in the home but that do not occur on a daily basis in the shelter - being handled in ways the owner will, having someone interact with them when they have a desired resource, etc. This information adds to our knowledge of the dog and can be used to guide our recommendations for training and behavior modification and to improve the welfare of the dog during his stay.
    The information can also help us match the dog to the most appropriate adopter and provide that adopter with recommendations to improve the likelihood of a successful adoption. The way I see it, every bit of information we can gather about the dog is useful. That said, the people who conduct these assessments must understand dog behavior, be able to read dog body language, have good handling and observational skills and be trained how to properly conduct and interpret the assessment. The assessment is only as good as the person conducting it. That is the only way they are fair to the dog.
    There are so many benefits to a well-executed formal behavior assessment in our shelters so I will continue to help shelter staff learn how to do them properly.

    ReplyDelete
    Replies
    1. For those who don't know her, Kelley Bollen, MS, ACAAB is the author of one of the papers on behavior assessments that I covered in my talk:

      Bollen, Kelley S., and Joseph Horowitz. "Behavioral evaluation and demographic information in the assessment of aggressiveness in shelter dogs." Applied Animal Behaviour Science 112.1 (2008): 120-135.

      Thanks for the reply, Kelley. Very compelling arguments.

      Delete
  5. I attended the conference and your talk. Part time I help a no-kill shelter. Since I have a different full time job, I cannot attend all of the days the vet staff pulls dogs from the city shelters. I set up a modified assessment to make sure the vet staff was picking dogs not just based on health or the ones that the city shelter staff guilts them into taking. We do mostly handling to also make sure the vet staff doesn't get snapped at when they medicate the dogs. I'm not sure if it's common in other places to medicate or vaccinate the dogs on pull day, but it seems unsafe to me. Anyway, I've also found a lack of standardization in both pre-evals or observations. The "pullers" complain about the length, so some rush through and others care and take their time. For example of at shelter observations, one kennel tech might not think the strength of being jumped on is as forceful as another.

    ReplyDelete
    Replies
    1. Great point -- it's really hard to standardize these assessments. Which in my opinion is why it's so important to have skilled people performing them.

      I agree that it's probably behaviorally unwise to vaccinate a dog on intake, but speaking as someone with advanced training in shelter medicine, medically it is extremely important to do so. Particularly if you are pulliing from poorly managed shelters, I think your staff is right to do it (as humanely as possible, distracting dogs with treats for the poke if they are willing to eat).

      Delete