A Research paper into the ability of sniffer dogs to detect the American foulbrood (AFB) causing Paenibacillus larva spores was published in December, a year after the work was completed. At that time of research the AFB Pest Management Agency called for recognition of the project – in the form of peer review and publishing – as the next step required before the use of canines would be considered in the national Pest Management Plan (PMP). Now that scientific stamp of approval has been given, is there a next step for AFB detector dogs and the PMP?

Recognition of an Odour Pattern from Paenibacillus larvae Spore Samples by Trained Detection Dogs was published in international open-access journal Animals on December 30. It capped off the efforts of lead scientist Neroli Thomson of Massey University and dog trainer Pete Gifford of K9 Search Medical Detection in Manawatu, to prove dogs can detect AFB causing bacteria.
While the duo, along with beekeeper Jason Prior of Downunder Honey in Manawatu who helped organise the project, were confident in the robustness of their work, gaining publication is a seal of approval from the scientific community. Further to that, AFB Agency national compliance manager Clifton King called on the publishing of the paper before the Agency would further consider the use of dogs.
The Agency Board is set to meet on February 17 and the research is on the agenda. However, King says that PMP AFB detector dog use is still likely some way off, as more research is required.
“Feasible research generally has a very small scope. In this instance the only research question which they set out to answer was, can dogs detect odours from AFB spores? The paper comes out with a solid ‘yes, dogs can’,” King says.
“The study is proof of concept. It proves dogs can detect AFB spore odour. It also highlights that trials to investigate the sensitivity and specificity in field conditions are required. It is not a case of leaping from proof of concept to using it across the board in the field, there is a next step to undertake.”
The national compliance manager will share his views on the research at the February Board meeting.

He is not alone in the call for more questions to be answered. The paper itself identifies the need for further research, stating, “The high success rate of two dogs in this study provides convincing evidence that dogs can recognise an odour pattern from P. larvae spore samples, but the test design was not suitable for evaluating parameters such as sensitivity and specificity. These would require additional repetitions, requiring a large number of samples, in an operational (field) setting where there may be multiple positive samples or no positive samples in a single search.”
Thomson describes the sensitivity and specificity as the “accuracy measures”.
“I.E what number of the positive hives in there can the dog find? That’s the sensitivity. Then the specificity is, how many other hives or pieces of equipment does the dog indicate on that turn out not to have AFB,” the Massey university scientist says.
Getting that work done will be costly, likely more than $100,000 and potentially around $200,000 various stakeholders estimate. Also challenging would be getting a team of researchers together to carry out more work, with there no guarantee those who have got it this far will be available to continue.
The Southern North Island Beekeeping Group have sought to raise funds to help Gifford keep the two dogs he has successfully trained thus far, but that has only raised $1700 total.

Prior says the Ministry for Primary Industries have been supportive of the potential for further funding of AFB dogs research, after their Sustainable Farming Fund contributed $50,000 to the $94,000 initial four-year project. However, like the first project where some industry funding was found to go alongside in-kind work of the research team, a substantial proportion of funding for further research would need to come from industry.
“Ultimately beekeepers should fund it, through levies or private contributions, but I don’t think there is a lot of point doing the work if we don’t figure out how we are going to use it. You need to come up with an operating model, and test towards that,” Prior says.
Thomson is also of that opinion and says “the intended use of the dogs would need to be confirmed, then a study shaped to find out how accurate they could be in that use”.

The AFB Management Agency’s input will therefore be vital, but to this point their input has been non-existent Prior says.
“What it needs is the Agency to run the project. The next step of the project needs to be driven by the ultimate, end users of the capabilities. I don’t think that is private industry,” Prior says.
“The AFB Board should be asking Clifton for the business case for using a different operating model, such as using dogs. If he does the numbers and it shows there could be a model where the dogs identify more AFB, at a lower cost, then there is your business model. From there the Board can help figure out how to get the research done to achieve it.”
However, at present it is not as simple as that according to the national compliance manager, who says an accurate cost benefit analysis (CBA) will be difficult with only the current level of information.
“To complete a CBA we need to understand the costs of training and maintaining detector dogs at an appropriate performance standard, and we need to understand detector dog performance in the field. That is the performance of detectors dogs that have been trained to appropriate standards,” King explains.
That means a greater understanding of sensitivity and specificity of the dogs is required so that the costs of missing AFB beehives and destroying healthy beehives can be factored into the analysis. Or, if a second diagnostic method is to be added to the protocol, such as visual inspection of all beehives that the dog indicated on to address specificity issues, this cost needs to be included in the CBA as well.
“With CBAs, it is not about proving your assumptions are 100 percent correct, it is a matter of making them reasonable and proving you can defend them. [For detector dogs] we don’t have enough information, at present, to make sure our CBA stands up to scrutiny,” King says.
And that scrutiny will likely be applied, especially if it came to the Agency going so far as to fund research.
“There has been a lot of support from beekeepers for that, but there is also a significant number of beekeepers who are totally against the use of dogs under the AFB PMP. That sizable proportion of beekeepers would, in all likelihood, be against their levies funding a project which runs into six figures,” he says.
For Prior though, he just wants to see support of any kind emanate from the Agency.
“The Agency are not willing to make the effort. There’s been very little participation or interest shown. What we have had is continual pushback that they have no mandate to look at this work until it’s peer reviewed. They have never been to see (dog trainer) Pete Gifford. There was some discussion about Neroli speaking at the Apiculture New Zealand conference, but that didn’t seem to happen. No one seems to give a hoot and I think that reflects poorly back to the research sector as, why do this sort of work if the stakeholders are not interested?” Prior says.
The apathy to their research will make the already potentially preventive task of forming a research team to carry out any future projects even more difficult Prior points out.
“The Project wasn’t fully funded so we are expecting the likes of Pete Gifford and I to provide in-kind contributions of our time and really, for what? This project was never about us.

“Pete has been knocking on doors for five years and he just doesn’t feel any love for his efforts. If you want people to do this work, such as Neroli did without getting paid, you need to give them some encouragement.”
And that points to a potentially larger problem for New Zealand beekeepers’ efforts to reach the ultimate goal of eliminating AFB and the improved tools and technologies that would bring that aspiration closer.
“If the Agency’s opinion is that they are not going to accept new technologies until there is peer reviewed research to lean on, then that simply won’t work. Because, no one is going to go and do research for them when they can’t even be bothered turning up to show stakeholder interest in the projects. They are just going to piss off what few research and development people there are in New Zealand,” Prior says.
It's therefore seemingly a ‘chicken and egg’ situation – what comes, or at least should come, first, more research, or more engagement from the Agency to help make that research a possibility?
February’s meeting where the AFB PMP Agency Board of Mark Dingle, Val Graham, Gabriel Torres, Jason Ward, Russell Marsh, Murray Lewis, Dennis Crowley and Jane Röllin will gather could be critical to the extension of the current detector dogs research. For Prior, with the research fresh, dogs already trained, and key personals’ motivations waning, action must be taken soon.
“I think the goodwill of the industry has reached a cap. What we need now is some stakeholder engagement to determine what we do with this capability and the Agency is the ultimate owner of the capability, so they need to take some ownership,” Prior says, adding, “it is their job to go to industry and say ‘please fund this’, but, before then, they need to run the numbers on how much money it could save levy payers if dogs could be adopted.”
Comments