2TurkeysMore on risk management with farm products, as triggered by Bill Marler’s speech at FOF.

Non-Zero Risks

The thing about risk that can be surprising before you think about it is, it’s rarely possible to bring a risk level down to 0.00000%. Probabilities of risk almost always exist in some quantity, even if small and remote. The key is to do everything we can practically do to bring risk to an acceptable level, and what is considered acceptable varies, depending on the situation. It’s a risk-benefit analysis. 

If I am making a child’s toy, obviously it needs to be pretty safe, because the benefits of a kid having a fun time with a toy are small compared to the risks of injury or dying. Even still, we know that many playthings sold today come with quite measurable risk (mitigated by warning labels): think of bicycles, skateboards, scooters, roller skates, etc. When we venture into the territory of medical products and food, the risk-benefit analysis gets even more complicated.

When I first started working in the medical industry, I had a hard time getting my mind wrapped around this. If a company can make a medical product that potentially saves millions of lives, while at the same time risking an estimated two lives, and it has done all it can to reduce that risk, that may be considered a worthwhile risk. If the company can think of a technology that would convincingly reduce that two-life risk to close to zero, but it costs billions of dollars, then it can no longer profitably make the product. So, it likely wouldn’t be required to employ that particular risk mitigation; because of course the company would just say, forget it, we can’t make the product. And then society would miss the opportunity to save millions of lives.

This is a startling reality, because it’s not much different than herd management, only applied to the human “herd.”  Akin to my choice to vaccinate my ewes, possibly sacrificing the few to save the many. The really hard part with human products is considering that it could be me, or a close family member, who ends up being one of the few. It’s a little bit creepy, and not at all emotionally comforting; but it is logical.

So, the story’s twist is, every product has some risk. Every time we put fork to mouth, we take a risk that we may be eating something lethal. But we can’t stop eating, or we’d introduce a new and very high risk: starvation. Winking smile The best we can do is reduce the risk of bad food down as low as we can practicably get it, and that’s never going to be zero. But it’s actually really important to realize this and be honest about it; because our emotional tendency is to pretend small risks are zero risks, and then we can start adding to the risk by being sloppy. As can our customers. I think keeping that risk in the backs of our minds every day, in everything we do, is key: knowing that it’s always there, and that we have to be constantly working to push it back towards zero.

Ongoing Monitoring

Once we’ve assessed risk and documented everything we’re doing to reduce it to an acceptable level, there is a second piece: monitoring. We have to acknowledge that risk analyses are imperfect, they are often based on assumptions and conjecture. It’s entirely possible that we were wrong in some of our analyses, or something has changed since we originally did them.

If new evidence emerges which changes our view of a risk, then we have to go back and re-evaluate our FMEAs and risk mitigations in light of the new information. Say I find out that a neighboring farm has been identified as a salmonella outbreak location. Now all of a sudden my risk estimates are different. Where before I may have assumed that the likelihood of my small and isolated chicken flock being exposed to salmonella was pretty low, now it’s quite high, because wild birds may carry it between local farms. So I may have to change my practices, at least temporarily, to account for the change in estimated risk.

Next, I’ll talk about thoughts of the worst happening- someone gets sick, and we face investigation and liability.