CWD and baiting - Risk Analysis

[wadegarrett]

Hi all,

TL;DR - This model shows no statistically significant link between baiting and CWD.

I have always been curious about the relationship between baiting cervids and the prevalence of CWD. When cruising around for information, I noticed there wasn't a tonne of research done on the subject aside from mostly logical conclusions (ie baiting = more proximity = spread = more infection). So, for a masters project, I attempted to crunch some numbers to see if there is a statistically significant link between baiting and CWD in certain areas.

The model defines a region as a central state/province (area) with directly adjacent states/provinces (areas). Regions were sorted in groups by number of areas within said region which allowed baiting for the purpose of hunting. Each region was then treated as a sample based on how many total areas had hits of CWD. So, for example, the Alabama "region" is in group 4 (4 areas), has 1 baiting area and 2 CWD areas. That is now a sample in the analysis. The hypothesis here is that in a group, regions with increased baiting areas should see increased areas positive for CWD when compared with the population norm. For this model, there was no such clear increase so I could not conclude with certainty that there is a link.

The model isn't the best for several reasons that I go into in the report - but taking into account various factors like state/province reporting granularity, manmade and natural borders, etc. I managed to put something general enough that could yield somewhat credible results - albeit as more of a stepping stone for more work to be done on this subject. Note that these results are ONLY based on this model, and that the results are only as good as the model. Also, full disclosure that I am no stats whiz by any means - I just thought this could be of interest to some people. My recommendation to states/provinces would be to run a similar analysis, but on a country to county scale to get much more accurate results. Also note that I do not participate in baiting while my province allows it, so I don't really have a dog in this race aside from not wanting CWD to spread.

Thanks,

WG

 

[Clarence]

Not sure if its relevant to the topic, but my mind tends to think of the spread of CWD through the spread of captive cervids. It seems to be a major culprit in my home state of Pa. Thank you for looking at this critical topic. I have never been to Texas, but it sounds like there is a large culture of baiting deer, and also importing genetics, into large high fence ranches. From the outside looking in this seems like a hot button topic I'm sure you will hear some strong opinions on. Thanks for sharing your work!

Sent from my SM-G965U using Tapatalk

 

[mlgc20]

Hey Wade. Thanks for posting. Very interesting. I did read your report. And like you, I have no dog in the race, other than wanting to slow the spread of CWD. If I understand your methodology, it looks like you included states that do not have any cases of CWD. It seems to me that states without CWD should be excluded from the analysis. I don't think anyone is theorizing that baiting creates CWD. The hypothesis is that it hastens the spread. So, if a state doesn't have CWD, then it can't spread it. An analogy could be COVID-19. If you wanted to understand how to slow the spread of COVID-19, you wouldn't want to look at a country that never had any cases. You might look to that country for other info on how to not get it to begin with. But, their information would not be useful on how to control the spread. They didn't have anything to spread to begin with. If you are just looking at whether baiting leads to CWD, I don't think that's the thing people are primarily concerned with. It seems like it would be very interesting to look at the spread of CWD in states that have CWD and then comparing the areas where baiting is and is not allowed.

Of course, just my opinion after a quick read of your paper. And it is just meant as feedback to think about. Not criticism of your work. This is an important topic and I'm glad you took it on and are furthering the conversation.

 

[Billy Goat]

We have master baiters around here, it's not legal. But very prevalent. I would imagine that can skew the results in areas where baiting isn't allowed.

 

[wadegarrett]

Hi all,

Thanks for your replies so far. I'll touch on a couple notes:

-Regarding captive vs wild cervids, this is an excellent point. I only looked at wild cervids in this model. I would be interested to see the parallels between wild and captive populations, in addition to transitivity between those populations (ie can there be transmission through a fence from captives to wild?).

-As for the model, my first cut only included infected states but there were a few things that were holding me back. The big variables I wanted to control, or at least normalize, were 1. reporting consistency and 2. animal movement. Since all states have varying degrees of reporting and cervids kind of do whatever they want and don't care about man-made borders, I wanted to capture a little bit of bleeding from one region to the next. For example, if Quebec had no reported hits but shares a border with Ontario, it's conceivable that 1. Ontario's testing isn't as robust and 2. animal movement between Ontario and Quebec isn't captured. This kind of reasoning worked well in small, dense states like the North East and regions with similar geographies (ie sharing plains), but I totally agree that it's not very refined. In a few years there will hopefully be enough data to do a transient analysis. In that case, the hypothesis would look something like "region X had no hits in 2020 but allowed baiting with adjacent area Y having positives. 10 years later, the disease spread to region X but not region Z, who was adjacent but didn't allow baiting".

-Reporting is a big variable as well, it goes in a similar vein as data collection. Sure, you can account for some variability in hunter reporting, but how can you account for purposefully unreported (read: unlawful) activities? Baiting, carcass transport where banned, those things are tough to get a good read on and even harder to integrate in the current data. If there is causality between baiting and CWD, then unreported baiting should skew the results.

You guys have raised similar questions that I had while designing this analysis. To accurately answer all these and more, we need more data! However, just having conversations about this is a great start.