Hi Gustav,
As per yr OP, I get the impression you are mainly proposing yr methodology as an alternative to ATP rather than as a competitor to current, lengthy, culture based counts. Clearly time is the critical factor for many current users of ATP methodology.
From yr OP -
For (a,b) swabbing error would presumably be comparable. And typically (relatively) high? (obviously it may relate to the specific surface type/condition)
It is true that the operator is one of the most important factors when you consider swabbing errors. To my knowledge it is very common that it is the same operator who performs the procedure at different production sites as it reduces the variance introduced by the swabbing method.
TVC afaik has typically a Low level of accuracy as reflected in, existing, decision criteria. IMEX this is an intrinsic limitation.
This is not my impression from 19 customer interviews. Of course this is a limited data set, but it is my impression that most customers prefer to use TVC for accurate measurements, but use ATP as they need to the speed to react. I know of some companies that only monitor with TVC, because they did not trust the ATP results when they tested the method. If this is due to lack of education or training I do not know. Using TVC only allows them to do trend analysis, and they are not able to react immediately on hygiene issues.
ATP measurement error no idea (not a user) ? i deduce you consider it to be << than TVC. Any particular reason ?
This is where we quickly end up comparing apples and oranges. The ATP instrument is developed to measure and detect organic material, which may indicate hygiene issues, but correlating the presence of organic material to actual risk is a stretch, especially when you compare correlating TVC measurements to actual risk. This is exemplified in Ryan M.'s case where they can not use ATP, because there will always be some plant residue after cleaning, but that does not necessarily mean there is a risk in the production. I would argue that we do hygiene and cleaning validation to determine and reduce risk in the production, and while ATP is a good tool, there are better parameters (such as bacteria) to use for risk assessment. The problem today is that it just simply takes too long to measure bacteria. That being said, you can not do a 1:1 correlation between bacteria count and risk either, but I would argue the correlation is stronger than for ATP.
I would have thought yr (static) counting method should have a much higher accuracy than TVC ? So what is the limitation ?
I would argue that it has, but we are up against a 100 year old method and a science (microbiology) that heavily relies on routine. We are working a lot on how to tell "the story", in the sense that we often get questions like "why don't we only care about the specific bacteria that can grow?" and "Are there really that much bacteria on surfaces? We don't see that using the methods we have for a long time." The answer to question 1 is that whatever result you get with a TVC will change if you change the temperature or the non-specific agar media, so you are not measuring which bacteria can grow, you are only measuring which bacteria can grow under the provided conditions. Your product might be subjected to an entirely different environment throughout the supply chain, so there is a risk some of the bacteria you did not see with a TVC will start growing and pose a risk. The answer to question 2 can be validated by looking at literature and reminding people of the great plate count anomaly, which is text book stuff, but easily forgotten.
Everything about "the story" has to be backed by data and reference cases to really make an impact on the industry. Right now it is only early technology adopters who engage with us, as they recognize our train of thought.
