Like most dedicated healthcare staff, Kim is up early on weekdays. After a healthy breakfast, she heads to work, ready for her shift.

She spends the day embroiled in ground-breaking disease research and is rewarded handsomely for her efforts.

Unusually for a health worker, however, she walks on all fours and wags her tail when she’s happy.

Kim is a black Labrador and her employer is Medical Detection Dogs – an organisation that looks certain to have a greater impact in neuro-rehab circles in future.

The Milton Keynes group has caught the world’s attention in recent years for exploring the canine ability to detect cancer at its earliest stages in urine, breath and swab samples.

It is now bounding into the fields of neurological conditions and spinal injury, with huge potential implications for patients and professionals.

The charity has thirty dogs working on detection of cancer and pseudomonas (common bacteria that can cause infections).

It also trains dogs to alert owners who have conditions like diabetes and are at risk of lapsing into life-threatening comas.

The dogs sense the onset of a spike or drop in blood sugar levels long before the owner can. With certain conditions where a person may be at risk of collapsing, such as Postural Tachycardia Syndrome (PoTS), the dogs are trained to give their master a few minutes’ warning and ensure they lie or sit down safely.

The group’s latest foray is a two-year study into Parkinson’s which, if successful, could pave the way for further exploration into brain conditions.

CEO and founder Dr Claire Guest says: “If this works with Parkinson’s then there is absolutely no reason why it couldn’t work with other neurological conditions. They all have a unique odour associated with a chemical change.

“Parkinson’s is a horrible disease – my own father was diagnosed with it a year or so ago. In terms of diagnosis, there are a series of symptoms for a number of years and then it starts to look more and more Parkinsonian.

“People aren’t diagnosed until they are relatively severe and damage is being done to the brain all that time.”

The study involves Manchester University, with funds from Parkinson’s UK and The Michael J. Fox Foundation for Parkinson’s Research. Dogs will be challenged in incrementally more complicated ways to detect Parkinson’s from skin swabs.

Multiple testing will be used to thoroughly establish reliability levels.

With no definitive test for Parkinson’s – which is notoriously hard to spot in its early stages – successful results could be a landmark achievement. Claire is hopeful of meaningful evidence by early next year.

“At this point we are talking about diagnostics. We could offer a service whereby clinicians send a sample of a patient they are concerned about and we could find out whether they have the Parkinson’s odour.

“We wouldn’t be screening individuals themselves, everything would have to go through a medic. In the future, if you have a younger person who may be having some mild symptoms, we could detect Parkinson’s much, much earlier.

“We would be supporting other Parkinson’s researchers and assisting in the development of early intervention.

“We’re very excited as it seems likely that the dogs are going to be able to smell it.” 

Claire’s confidence comes from her in-depth knowledge of the power of the canine nose and also existing evidence.

“Whenever we do any detection work, it always comes from anecdotes in the early stages. A Parkinson’s nurse called Joy Milne in Scotland believed she could recognise a change in odour with the patients she was working with.

“Then her husband started getting some debilitating symptoms when he was around 40. She recognised the smell immediately when she was cleaning his clothes. It took some time but he was eventually diagnosed with Parkinson’s.”

Milne worked with Edinburgh University on a subsequent study, smelling t-shirts worn by patients.

She correctly identified whether or not the patient had Parkinson’s in nine out of 10 cases.

She believed the tenth t-shirt was Parkinsonian, which was officially deemed incorrect, but that patient was diagnosed with the disease the following year. If this is what a human can detect, imagine what a dog’s sniffing power could do for Parkinson’s detection, thought Claire.

While a human may be able to smell a teaspoon of sugar in a cup of tea, a dog can sniff it out in enough water to fill two Olympic-sized swimming pools.

Claire launched her organisation in 2008 and has since focused on turning anecdotal evidence into hard boiled facts.

“There is a lot of evidence that dogs can detect cancer with 80 to 90 per cent reliability but these have been with small sample sizes,” she says.

An NHS-backed trial is currently underway to test the accuracy of prostate and other urological cancer detection by the dogs.

The study, involving Milton Keynes Hospital NHS Foundation Trust, could spare many from unnecessary biopsies.

The trial was approved following earlier studies which showed 93 per cent accuracy in detecting prostate tumours from urine samples.

A similar trial is also ongoing for breast cancer – a disease Claire has direct experience of. In 2009, her life was saved when her Labrador Daisy sniffed out a deep-seated tumour in her breast, which might otherwise have gone undetected.

A 2006 study (McCulloch et al) tested dogs’ ability to distinguish breath samples of 31 breast cancer patients from those of the 83 healthy controls.

Among breast cancer patients and controls, sensitivity and specificity was strong but there were some limitations in the design of the study and number of subjects.

The new, larger study aims to provide further supporting evidence, using breath samples from volunteers in Buckinghamshire.

Claire was previously involved in a 2004 bladder cancer study published in the British Medical Journal.

It concluded that: “Dogs can be trained to distinguish patients with bladder cancer on the basis of urine odour more successfully than would be expected by chance alone.”

Claire says: “After that study there was huge interest in this area. Cancer is one of the most difficult things to detect because you are looking at a very complicated condition and the amount of volatile in a urine sample can be very low.”

All the charity’s dogs live at home with foster families, and come in to ‘work’ for only two or three short sessions every day.

As a working environment, it sounds like the doggy equivalent of California’s Googleplex. Instead of the hammocks and beanbags of Silicon Valley ‘work spaces’, dogs chill out on comfortable beds in their own special areas.

Just like Google’s ‘playtime’, which encourages staff to stay sharp by doing their own thing and having fun, the dogs are encouraged to romp around an adjacent field between sessions.

But this is a serious business – as the wider world is finally realising after years of scepticism.

“The reaction we got 10 years ago was either ‘ooh that’s interesting’ or ‘that’s just crazy’. But things have moved on, with published papers and a growing understanding of how our dogs are assisting people with long-term health conditions and making a genuine difference to their lives.

“It is definitely filtering through at all levels that dogs can assist.”

Accuracy and reliability are two crucial factors being tested. Every morning, the dogs are “calibrated on samples” to check their noses are tuned into what they are looking for.

During testing, databases are running in the background, checking the same sample on different dogs to confirm an answer.

A new area being tested is detection of the bacteria that causes urinary tract infections (UTIs). Specifically, Medical Detection Dogs is aiming to help spinal injury patients among whom UTIs are particularly common.

After a spinal cord injury there is an increased risk of UTIs as bladder management methods such as catheterisation can make it easier for bacteria to enter the bladder.

“The dogs are able to pick up the odour of the bacteria very quickly, before the person becomes unwell. They could smell it in the urine very reliably and rapidly.

“We are also looking at a pilot in which we would place a dog with an individual with a spinal injury to see how we could assist them with detection at a very early stage UTI.

“If this works well, the dog would probably do some other task work to assist the individual in other ways. Perhaps every morning they could screen a urine sample to detect whether bacteria levels are normal.”

And there may be many more conditions, including those in the neurological field, that dogs can assist with in the future.

“We are approached by many clients with many different chronic diseases and illnesses and we are finding that dogs can help with most of these conditions.

“It seems that when we are unwell and something acute is happening, our odours do change and we believe there are a whole range of different conditions that we can help to detect.