About one year after I graduated vet school, I took routine screening chest radiographs of my senior Golden, Mulan. I looked them over, frowning at a small, mottled spot near her sternum.
“She has cancer,” I thought. It’s not an unreasonable conclusion to come to with Golden Retrievers. Before I panicked, I asked my colleague to look at the x-ray, and she agreed it looked suspicious. I was devastated.
I took Mulan to the local specialty hospital, where an intern I knew from vet school patted me on the back while the resident internal medicine specialist pursed his lips sympathetically. He grabbed his ultrasound machine to prepare for a guided biopsy. Before starting, he asked the radiologist to stop by to give his thoughts as to what this strange radiographic feature might be.
“What are you looking at? That? That’s normal sternum,” he said, sipping his coffee with the mildest of eye rolls before strolling out of the now-silent room.
I knew just enough to be dangerous but not enough to actually come to the correct conclusion. Along the way I dragged two other very educated colleagues with me through sheer force of conviction. Mulan lived another 4 years, by the way.
Data and Interpretation
Lots of people have asked me about the controversial results from the Truth about Pet Food’s crowdsourced food safety study. I haven’t said anything, because I couldn’t think of anything to say. It’s the same response I have when people send me this picture over email and ask me what this lump is:
The correct answer is, “I need a lot more information before I can tell you that.” Which is about how I feel about the significance of this study.
As veterinary nutritionist Dr. Weeth points out in her excellent response, scientists kind of live to nitpick and poke holes in one another’s work. It’s necessary to allow criticism because there are so many ways one can go wrong with a project- from the way the study was designed, to the implementation, to the data interpretation. It was the persistent nagging of the science community that led to the eventual discrediting of Wakefield’s autism/vaccine research paper, the public health implications of which we are still dealing today, up to and including 19 people who were sickened with measles at The Happiest Place on Earth.
Without being allowed to evaluate the entire research process, we have no way of knowing how valid the results are. A pretty infographic does not science make. Nor does protesting “it’s not junk science” mean that it isn’t.
What We Know
I’m hopeful that the full set of data will be made public, including methodology. Until then, all we can do is go by what we have been told.
Dr. Gary Pusillo of INTI services, who has the misfortune of being out of the country while all of this debate is going down, was in charge of the testing process. Thixton writes that he is a board certified veterinary nutritionist, which in theory is fantastic because it means that he would have the background in both veterinary medicine and nutrition to not only perform the studies, but interpret the results. There’s only one problem: he’s not. (Nor does he in any way present himself as one, by the way.) A board certified veterinary nutritionist is a veterinarian who is also a diplomate of the American College of Veterinary Nutrition. You may think that’s irrelevant, it’s just semantics, but it’s not.
Credentials are a big deal, as I’m sure Dr. Pusillo himself would tell you were he around. I would really love for Dr. Pusillo and Dr. Purejav to have been available to answer questions while we’re all begging to know what the heck they did, and I’d love to hear more about how they determined “risk.” They may be the most qualified people in the world, but for right now, all I have is an infographic and a consumer advocate’s word that they’re the best.
Dr. Pusillo is a PhD who provides forensic science services, which actually sounds really cool and I would love to hear more about it. I have no reason to doubt that he is an excellent scientist. He probably knows tons and tons about how to test a food for specific substances. What he may or may not know is whether or not those substances matter clinically.
Data Collection vs. Interpretation
Let’s assume that the data collection was carried out perfectly. Data collection is only half of the equation- you still have to know what to do with it. You can have all the answers in front of you and still not know the question. The scientists Thixton contracted with are out of town at the moment, so who are we going to ask to help us interpret things?
Given who’s around right now, who could interpret the limited data we have through the filter of what matters?
A microbiologist with a background in food safety would be a good start, as someone who can tell you whether or not particular pathogens are actually of concern.
Or a board certified veterinary nutritionist, who can tell you about nutrient analyses and why dry matter comparisons without calorie content is useless. Both of them have some big reservations about this project.
They know more than I do about such things, which is why I defer to their interpretation. Little things mean a lot- for example, when you say “bacteria are present” what do you mean? Does that mean live bacteria were cultured using sterile handling procedures to eliminate environmental contamination? Or did the test just look for bacterial RNA, which could come from dead bacteria that were killed during processing and therefore prove that production works as advertised? I don’t know, but that would sure make a difference.
When the company you contract with to run your tests asks for their name to be dissociated from any press surrounding you, there’s one of two conclusions: 1. They were not happy about how their data was manipulated in the interpretation stage and didn’t want to be associated with bad science; 2. Big Pet Food Cabal. We may never know. *shrug*
A victory for food safety
I like to look at the bright side of things, and for reasons I can’t fathom, what I’ve found to be the biggest findings of the study are barely mentioned.
What are the three most common concerns I hear about pet food safety?
- pathogens of most dire human significance, specifically Salmonella and Campylobacter
- pentobarbital contamination (implying euthanized rendered carcasses in pet food.)
Why were these not mentioned in the risk report?
Because they weren’t found. They did look for all of these products. All twelve tested foods were clear of the three biggest worries in recent memory to pet food safety. That’s something, don’t you think?
I’m an optimist. Let’s look at the bright side of things, what do you say!
So let’s review here: I like asking questions. I have no problem questioning consumers, colleagues, my own professional leadership. I think concerned consumers are good consumers, and I applaud anyone who is invested enough to care about what goes into their pet, be it food, drug, or plant. I have chosen not to work in the employ of companies in the field specifically so I can feel free to say what I want without worry about my job or advertisers.
That being said, I think we also have to take the Occam’s razor approach to life and assume at some point that companies are telling the truth when they tell us they aren’t actively attempting to kill our pets. There are problems, some big and some small, and those are worthy of being addressed, but if you can’t accept at the end of the day that they are generally trying to do the right thing, then we may not ever be able to come to an understanding. As part of a profession that deals with this type of distrust on a regular basis, there comes a point where you have to say, “If you’re going to insist I’m out to harm you no matter what I say then I probably should just leave now.”
So let’s end on a high note: a toast, to those who care. I think everyone’s here arguing for that reason even if the conclusions are different. Salmonella free appetizers for all.