New pet therapy guidance

A new Expert Guidance Document has been released by the Society for Healthcare Epidemiology (SHEA) and published in Infection Control and Hospital Epidemiology.

R Murthy, G Bearman, S Brown, K Bryant, R Chinn, A Hewlett, BG George, EJC Goldstein, G Holzmann-Pazgal, ME Rupp, T Wiemken, JS Weese, DJ Weber. Animals in Healthcare Facilities: Recommendations to Minimize Potential Risks. Infect Control Hosp Epidemiol 2015

 The guidance document covers 4 main areas

  • Animal assisted activities (aka pet therapy, hospital visitation)
  • Service animals
  • Research animals in healthcare facilities
  • Personal pet visitation

The document provides recommendations for practices to reduce the risk of animal-associated diseases from these activities. There’s also some introductory survey information that highlights a few reasons why this guidance is important.

  • Pet therapy was allowed at 89% of surveyed US hospitals.
  • Personal pet visitation was permitted at 40%.
  • Research animals were present in 35%.
  • Many hospitals that allowed animals had no policy covering their activities.
  • All facilities that allowed pet therapy permitted dogs. 21% also allowed cats, 5% miniature horses and 2% primates (ugh!).
  • A few US facilities, including a Veteran’s Hospital, indicated that they did not allow service animals, something that contravenes the Americans with Disabilities Act.

This should be required reading for anyone involved in hospital administration, hospital infection control or pet therapy programs.


Exotic animals in hospitals

What zoonotic microorganisms do pangolins commonly carry?

Are some pangolins higher risk than others?

What infection control measures should be used?

I don’t know, and from a cursory review of the literature, I don’t think anyone really knows.

So, do we really want to be exposing them to some of our most susceptible individuals - kids who are patients in children's hospitals? posted an article earlier this year about exotic animals and hospitals (something that would be contrary to international pet therapy program guidelines published in the American Journal of Infection Control, and various other policies). The cover picture is of a child, presumably quite immunocompromised, touching a pangolin. She’s wearing a mask and gloves, but it looks like she’s also wearing her pajamas (which she is touching with her glove, that maybe just touched the animal).

In reality, the title and pictures don’t have a lot to do with the actual article. It focuses on a great initiative by a group of zoos to create videos about animals and animal care, and to make a TV network that will be provided to children’s hospitals across the US. That sounds like a great idea. The pictures of exotic animals in hospitals are not-so-great.

Pet therapy’s a wonderful thing. I work with pet therapy programs and have been involved in a few initiatives to produce guidelines to make these programs as safe and effective as possible. Part of that involves knowing what you’re dealing with.

Yes, pangolins (sometimes referred to as "walking artichokes") are cool looking critters. I’m sure the kids get a kick out of seeing them. But, is there any additional benefit beyond what the children would get from interacting with a well trained dog? Beyond the cool factor, there’s not a lot of personal connection with a pangolin, I suspect. A happy, friendly dog might actually provide a greater benefit, and since we know a lot about dogs, we are able to manage the risks much more effectively.

The zoo TV initiative sounds great.

Well-run pet therapy programs are similarly great.

Bringing pangolins and other exotic animals into hospitals… not so much.

Photo: Tree pangolin (Manis tricuspis) in central Democratic Republic of the Congo (credit: Valerius Tygart)(click image for source)

Hospital animal visitation in the news

The New York Times has a nice article on hospitals that allow patient's pets to visit.  This is a controversial area, with policies (when they are actually present) that range from wide-open access to complete prohibition. Like most things in life, there's a middle ground that's the most reasonable.

The positive aspects of people being allowed to have their animals come visit are pretty obvious, since people may have close bonds with their own pets and having a chance to see their pets might make a big difference to their mental state/well-being, particularly for someone who is chronically ill.

The negative aspects are less clear. We certainly know that pathogens can be transmitted from animals to people (and in the other direction too), and people in hospitals are often at higher risk of infection and complications thereof. There's a list of pathogens we worry about, but there's a serious lack of data to help determine the severity of the risk - and how to reduce it. Organized pet therapy programs are great because they are structured and there can be a lot of scrutiny and training of the pet and handler. Visits by patients' own animals are inherently less controlled, since the individual animal and handler don't undergo the same degree of assessment and training.

I get asked to review visitation guidelines frequently, and a reasonable middle-ground can usually be found. These are some snippets from the NYT article that highlight common points.

A doctor’s order allowing the family pet to visit is typically necessary...

That's a good approach, although it's not often used. This lets the doctor decide if it is reasonably safe to have the pet visit, i.e. the patient is not at a very high risk of infection. The weak link here is sometimes the doctor, because sometimes the doctor doesn't understand the risks associated with pet contact and may not identify a concern. Other times, the doctor may not understand the relatively low risk and the potential benefits, and therefore default to banning visitation without giving it much thought. I think that's less common these days but still an issue. is an attestation from a veterinarian that the animal is healthy and up to date on all its shots.

The first part is great: making sure there are no health or behavioural issues with the animal that would pose an increased risk. The second part is very common but largely represents a lack of understanding of the issues. Too often, "has he had his shots?" is the main question that's asked about the animal, despite the fact that it's largely irrelevant from a zoonotic disease standpoint. Yes, we want to make sure the pet's rabies vaccination is up-to-date, but the other core vaccines are irrelevant from a human health standpoint (although they're very important for keeping the animal healthy overall).

Most institutions require that dogs — the most common visitors, by far — be groomed within a day or so of a visit and on a leash when they walk through hospital corridors.

Standard and logical policy. Grooming might help reduce the burden of bacteria, fungi and parasites on the haircoat, as well as a lot of loose fur and dander that could otherwise contaminate the hospital environment.

Cats must be taken in and out of the institution in a carrier.

Logical. Some cats do well on leashes but it's better to have a cat in a carrier when taking it through a strange area. It also helps prevent other people from coming into contact with it.

If a dog or cat wants to get up on a patient’s bed, a covering is laid down first.

Good policy, and it protects both the patient and the animal.

If an animal seems agitated or distressed when it comes into the hospital, staff members who meet the family and escort them to the patient’s room have the right to turn it away.

This has two important components. One is that the visitation is supervised, which is great. The other is that staff are given the ability to intervene in the unlikely even that there are problems.

If the patient shares a room with someone, that person must agree before a pet may visit.

This is often overlooked. Roommates might be afraid or allergic, and in those situations, visitation shouldn't happen in a shared room. There might be a way to do the visit in another room, so it doesn't necessarily preclude the visit from happening. This has to be broached in advance and in a manner  that the roommate doesn't feel pressured into consenting. It's best done by having the patient's healthcare providers approach the other patient and/or the other patient's family.

There's always some risk with animal hospital visitation. That's never going to be eliminated, but a lot of common sense practices can reduce the risk to a very low level, hopefully to the point that the risk is overwhelmed by the benefits. A little structure and a lot of common sense go a long way.

Puppy ban

Following outbreaks of campylobacteriosis in a Canberra, Australia nursing home, health officials have recommended banning puppies from aged care facilities. Two outbreaks that involved at least 15 people occurred in one such facility last year, and a healthy puppy was identified as the cause. Unlike many reports in which people try to blame an animal source without any evidence, they isolated Campylobacter jejuni from the puppy and people. That, along with ample previous evidence of a role of puppies in this disease, is pretty strong evidence that the puppy was the problem.

They concluded that puppies shouldn't be aged care companions because of "high rates of Campylobacter carriage and shedding, their social immaturity, susceptibility of elderly residents to infection and poor outcomes." Such a conclusion is not really that surprising or novel, actually. The 2008 international guidelines for animal visitation in hospitals recommend that only adult dogs and cats should be used for these activities, for several good reasons:

  • Puppies are biohazardous. It's just biology. Young animals are at much greater risk of shedding various bacteria that can cause disease in people, such as Campylobacter.
  • Contact with puppies and kittens has been clearly demonstrated as a risk factor for diseases like campylobacteriosis.
  • Compared to adult animals, puppies and kittens are more likely to poop on the floor.
  • Puppies and kittens are also more likely to nip or scratch through playful behaviour.

This is not to say that everyone should avoid puppies and kittens, after all they are cute and entertaining, and a great pet in many situations. The risk is higher in certain populations, such as people living in nursing homes, and while puppies are fun, similar positive impacts can be obtained by well-run visitation programs using older animals. That's the approach that being taken in Canberra, as trained adult dogs will still be allowed to visit aged care homes (hopefully as part of a structured program).

Dogs in the delivery room?

Don't get me wrong. I'm all for pet therapy and animal visitation in hospitals - when it's done logically. I've been involved in research in the area, helped develop international guidelines and am chair of the medical advisory board of one of the largest pet therapy groups in the US. Animals can do great things in hospitals and we need to support good visitation programs. But that doesn't mean I check my brain at the door and think that all animals in all hospital situations are a good idea.

A colleague sent me a link to a Medscape News article entitled Woof! Does Fido Belong in the Hospital Delivery Room? 

  • My first thought was... not a chance. (My second, third and fourth thoughts were no better.)

The situation in the article isn't that clear cut though, since the English woman who wanted her dog in the delivery room had a trained therapy dog that helped her with an anxiety disorder. So, if this was truly a trained therapy dog (some people unfortunately make that claim just as an excuse to take their dog everywhere, and compromise people that truly need these animals), it would be justifiable since this is a service dog, not a companion, and we need to support access of service dogs.

However, it raises questions about whether this will open the door to requests for pets to join in the birthing process, now that we've moved from the era when dad paced outside the room to a time when half the family may be present, live-streaming the event to the internet and posting on Twitter.

What are some issues here?

A delivery room is a busy environment. Things can be nice and happy and relaxed. There can also be yelling (personal experience there), lots of activity and other things that might scare or upset the dog. I'm not worried about the dog's feelings here, but what a startled or upset dog might do (e.g jump, bark, bite, pee, try to run away).

Not all deliveries are smooth and things can go from good to bad quickly. The last thing that's needed is another distraction (e.g. the aforementioned dog jumping, growling, barking, peeing, etc.) when medical personnel are dealing with a life-threatening delivery complication.

A newborn is a high risk person for infectious diseases. Every dog is shedding multiple microorganisms that can cause disease. Usually, the risk is low. However, when you have a highly susceptible person (or persons, including the mother here to a lesser degree since post-partum infections are a concern), we don't want them exposed to pathogenic bacteria if we can avoid it. Yes, it's an ever-present risk in a hospital, but why add to the potential risks? It would seem illogical to have delivery personnel in full protective gear (e.g. gowns and gloves) with a dog potentially aerosolizing bacteria nearby through breathing, coughing, barking, shaking, and tail-wagging. People would also likely contaminate their hands often by touching the dog. Yes, medical staff can be to told to avoid contact with the animal and wash their hands, but we know from previous research that hand hygiene by medical personnel after animal contact is very (very!) uncommon.

Also, we know that a baby's first bacterial encounters have a major impact on its developing bacterial microbiome (that is, the composition of normal bacterial populations at various body sites, something that's important for good health and development). Babies born by C-section have much different microbiomes for a long period of time compared to those born by vaginal delivery. Do we really want to confuse the picture more by having some of the first bacteria encountered being Fido's bacterial flora? It's not going to make the baby start barking, but I'd rather the baby not be exposed to various bacteria from a dog seconds after it's born.

In my opinion, visitation is more important the longer the person is in hospital, the more lonely they are and the more upsetting or depressing the situation is. Delivery is typically a short-term, happy hospital stay. What's the real benefit here for your average dog owner?

Personal pets in any hospital situation is a controversial area. Unlike dogs that are part of proper visitation programs, these dogs tend not to have any health screening, behaviour screening or other type of assessment. There's also no handler training. You might say "well, the dog's just visiting its owner so that's not a big deal." However, the dog has to go from the parking lot to the room and back again. What are the odds that the dog's not going to encounter lots of other people in the process, let alone potentially distracting or scary situations. Do you want your elderly immunocompromised relative to ride (or be stuck in!) a hospital elevator with an aggressive or otherwise high risk dog? Or to have you child that just had surgery step on a pile of dog poop? There are clear screening, training and supervision criteria for hospital visitation dogs, and they are there for a reason.

Back to the article.  Dr. Arthur Kaplan, the author, sums things up nicely:

"I think there are risks, and I think the risks are pretty significant. I am not sure that we should open the door to every barnyard creature we could think of to be present at birth, even when the mom-to-be says that she would like to have her pet there. But at the same time, I think there are arguments that, for some people, such as the woman in England who has a special relationship with her pet, or perhaps a woman who is blind, a case can be made."

Rudolph inquiry in Scotland

Santa gives the reindeer a few months off every year, so inevitably they're going to cause trouble.

A health board inquiry is under way after a young reindeer was taken into a Glasgow Children's Hospital and allowed to interact with patients. The reindeer fawn, from a local reindeer farm, was paraded around the hospital grounds as part of an organized event. That's great. The kids could see something unique. However, the problem occurred when a staff member decided to take the fawn inside so more young patients could see it. By doing that, the fawn was turned into a "therapy animal," meaning all the various recommended precautions for a therapy pet should apply (including a prohibition on bringing farm animals into a hospital).

It essence, good intentions + lack of critical thought + lack of clear local guidelines lead to this situation, which has caused an outcry amongst some groups (and probably a similar "so what?" amongst others).

Here are some interesting bits from the article:

"It had been checked by a vet..."

  • Okay, good start, but for what was is checked? You can't tell what infectious agents an animal carries by looking at it. We know that young animals are more likely to shed various potentially harmful microorganisms, and deer can be sources of very harmful bacteria like E. coli O157.

 "It is understood that the patients who did pet the fawn were later given antiseptic wipes with which to clean their hands."

  • I wonder what "later" means. I suspect it wasn't right after animal contact.

“I don’t suppose any animal, no matter how well shampooed and clean it was, should be allowed into a hospital without prior knowledge and the correct arrangements made,” [Dr. Jean Turner of the Scotland Patients' Association] said.

  • A reasonable statement. She's not saying "no animals," she's saying "no animals without a proper plan."

"I think it was well-intentioned, but I don’t think anyone was thinking about the consequences of taking a live animal like that to a hospital.”

  • That sums it up nicely.

Every animal (and person) is carrying multiple microorganisms that could make someone else sick. Usually that doesn't happen, and we need to live life, not stay locked up in our bedrooms. However, some animals are at higher risk of shedding pathogens (e.g. young animals, farm animals), some situations make it more likely that an individual animal will contaminate the environment of patients (e.g. interacting with a farm animal, taking a non-house-trained animal inside) and some people are at much higher risk of serious disease when they encounter various bugs (e.g. hospitalized kids).

I'm sure some kids had a great time, and the overall risk was probably low. However, was there really any benefit here beyond a properly run pet visitation program with appropriate animals, established protocols, good hygiene practices and proper handler training?

Pet therapy programs are too important to be compromised by illogical events like this that sometimes cause a knee-jerk "no animals in the hospital" response. That's why there are good international pet visitation guidelines and why people need to follow them.

Clostridium difficile and hospital visitation dogs

A couple questions that I received about C. difficile and hospital visitation dogs:

Is there any concern about cross contamination from dogs/handlers that visit health facilities and get contaminated by C. difficile? I see on this site a concern about dogs being contaminated by visitation and I wonder if the Delta Society has considered this to be an acceptable risk.

Dogs (and handlers) could be sources of C. difficile in hospitals. We have shown clearly that dogs that visit hospitals are at increased risk of shedding C. difficile in their feces, presumably because they ingest C. difficile spores from the hospital environment and/or patients' hands. All dogs that go into hospitals are at risk, with dogs that lick patients and that are allowed up on beds at increased risk (Lefebvre et al 2009). We have also shown that the dog's body can become contaminated with C. difficile after visiting a hospital.

Should there be a period of time mandated between visits to account for possible contamination?

That's hard to say, but probably not. Contamination is a potential problem and certainly occurs, but we don't know how long it lasts nor whether providing a "rest" period actually does anything. Clostridium difficile spores, the form of the bacterium that would be present on a dog's coat, are very environmentally tolerant and can survive for years. Therefore, giving a few days break in between visitation would not result in the bacteria dying. However, some spores would presumably be physically removed over time, through shedding of hair, grooming and other activities. Based on that, it's plausible that the longer the time between visits, the less likelihood of contamination still being present. Whether this actually helps, we don't know.

What to do?

Among other things, we need to consider whether there is actually any evidence of risk. We don't know whether dogs are able to transmit C. difficile in hospitals. My assumption is that there is some degree of risk, but it's limited and can be controlled with good hygiene and the use of basic visitation practices, as highlighted in the "Guidelines for animal-assisted interventions in healthcare facilities" (American Journal of Infection Control, 2008).

Another thing to consider is whether there are any measures that can be taken to reduce potential risks, while maintaining a practical and effective visitation program. A key component of this is knowing that there are factors that make it more likely that a dog will be exposed to C. difficile during visitation. If a specific subset of dogs is at increased risk, then you have a clearer way to approach it. In this case, dogs that are allowed to lick patients and that are allowed on beds are at increased risk. These activities are modifiable - you can prohibit them without having a significant impact on the visits. Licking can be prohibited. Dogs can be kept off beds unless it is required, and when that's the case, they can be placed on a towel or some other barrier to reduce their exposure to C. difficile from the bed. Additionally, we know that if a dog is being treated with antibiotics, it's more likely to shed C. difficile, so dogs that are being (or have recently been) treated with antibiotics should be excluded from visitation.

For C. difficile to be a problem, it has to go from human or animal feces to a patient's hands and then to a patient's mouth. There are multiple potential interventions to interrupt this chain of transmission. At the end of the day, however, hand hygiene is the key. If people wash their hands before and after touching the dog, there should be much lower risk of disease transmission. A problem is that when I say "hand washing," I mean hand washing - not use of an alcohol-based hand sanitizer. Clostridium difficile spores are resistant to alcohol. That creates a conundrum because the use of alcohol hand sanitizers, a common and recommended hand hygiene method, doesn't have the ability to kill C. difficile. Since not all visitation patients are able to get up and go to a sink, that complicates C. difficile control.

(click image for source)

Pet therapy: What about the risks to the animals?

There's an interesting article in today's Pittsburgh Post-Gazette about Scooter, a paralyzed cat in a custom-made cart, that visits patients at HealthSouth Harmarville Rehabilitation Hospital. It's a nice story and it's easy to see the potential appeal of a paralyzed animal whipping around a rehab hospital as an inspiration to patients.

Obviously, people like having this cat in the hospital, and he's helped some patients. That's not surprising because we know that pet therapy can be beneficial to many. My concern in this case is for the cat's health, and the heart of the issue is whether the benefits to patients are because Scooter's a paralyzed cat or because he's a cat, and whether the benefit to patients justifies the risk to the cat.

Paralyzed animals are at increased risk for certain infections. In particular, they are at very high risk for urinary tract infections. These animals tend to get recurrent urinary tract infections and enter a downward spiral of infection / treatment / infection / treatment / resistant infection / treatment / more resistant infection... and in some cases end up with infections that are very difficult or impossible to eliminate. In some cases, urinary tract infections in paralyzed individuals can result in infection spreading to the rest of the body, which can be fatal.

Back to my concerns for Scooter: We know that the hospital environment is contaminated with various drug-resistant bacteria. We know that patients in hospitals are often carrying drug-resistant bacteria. We know that dogs that participate in visitation programs are at increased risk of acquiring drug-resistant bacteria. So, do we really want to be exposing a high-risk animal to such an environment, and potentially speed up the cycle of infection that could ultimately cause severe illness or even death in the animal?

There's no clear answer, but we need to consider the risks to visitation animals, and whether the novelty of having a paralyzed cat (instead of a normal, healthy cat) visiting patients is really a significant enough benefit to justify the potential risk to the cat.

Image: A paralyzed cat using a mobility cart (source:

Robotic pet therapy?

In 2005, "Paro," an interactive robotic pet, was introduced in Japan. It looks like a baby seal, and has 12 tactile sensors in its fur, touch-sensitive whiskers and a system of motors that silently move its parts. While this "pet" didn't really catch on in Japan, it's made its way to the US and ended up in some nursing homes.

Paro has been described as a low-maintenance alternative to dogs and cats for pet therapy. It's an intriguing idea. There are certainly some appealing aspects, in that a robotic pet won't bite or scratch, doesn't poop, doesn't need to be fed, can't be injured, doesn't have a large population of resident bacteria in and on its body and can't become infected with various microorganisms from patients. Those are appealing from an infection control standpoint. However, despite this, you can't approach Paro as a way to eliminate infectious disease risks, since the robo-pet could easily become contaminated by someone, then spread  infection from person-to-person. For example, if someone has a bacterium on their hands and they touch the robot, they could transfer the bacterium to its "coat." It could then spread the bacterium to the next person that touches it. Infectious disease risks would be lower, but not zero.

Pet therapy is all about cost-benefit. We know there will never be a zero-risk pet-human interaction. However, socializing with an animal can provide significant benefits to many people, and I'm not convinced that the same degree of benefit would be provided from interaction with a robotic critter. There might be some situations where a robotic pet would be useful in a nursing home or similar environment, but I don't think they're going to replace interaction with a live animal.

Raw Q and A

While I hate to initiate another round of emails from raw diet supporters, here are a few of the more common questions that I get about these diets and some answers:

I use frozen raw food. Doesn't freezing kill harmful microorganisms?

  • No. Freezing is an effective way to eliminate most parasites (with an adequately low temperature and adequate time, which varies between parasites). Campylobacter also does not survive freezing well. However, other bacteria, including Salmonella, tolerate freezing quite well. Studies of previously frozen raw diets have found high rates of bacteria like Salmonella.

My dog doesn't defecate in the hospital, so why are we paying attention to intestinal bacteria?

  • Inadvertent exposure to fecal bacteria is common. Most gastrointestinal infections in people are from ingestion of bacteria and viruses from feces (e.g. Salmonella, Clostridium difficile, norovirus). We don't knowingly ingest feces, but we get exposed to these organisms nonetheless. Fecal bacteria can end up on pets' haircoats, people's hands and many surfaces in the general environment, and then wind up in the intestinal tract of a susceptible person.

Is there any way to eliminate Salmonella and other harmful bugs from raw meat?

  • Yes. Besides the obvious (cooking), there are a couple options. One is irradition, which is a safe and highly effective way to eliminate bacteria. The main problems are cost and consumer fears of irradiation (which is actually harmless). Another approach is high pressure pasteurization. This process uses high pressure (with a slight increase in temperature) to kill harmful organisms. The effectiveness of this for raw meat hasn't been clearly determined, but it's an option, and one company is now doing this for all of their diets.

Why don't you just go into hospitals, ask nurses whether animals visit and compare infection rates, so you can see if there is a true health risk?

  • It would be nice if it was that easy. Firstly, asking nursing staff doesn't give enough information. You need to know if animals visit, but also if they visited particular patients, and whether they visited before those patients developed infection. Just comparing infection rates between hospitals or wards that allow dogs to visit, and knowing the dietary status of the dogs, is useless. A proper study would require clear documentation of which animals visited which patients (something that is rarely recorded) and whether patients subsequently developed any infections that were not present before visitation (which is not easy to document), while concurrently investigating other possible sources of infection (similarly challenging). Ideally, bacteria causing human infections would be compared to those found in animals to provide stronger evidence of a link. Because the incidence of infections is relatively low, a large number of people would need to be enrolled. There are significant logistical issues, research ethics board issues, problems with the quality and availability of medical records and other things that make this very, very difficult. It needs to be done but it's not as simple as many people think. If it was easy, it would have been done by now.

Dogs have a short and acidic intestinal tract and are not susceptible to Salmonella.

  • This statement appears thousands of times on the internet and there's absolutely no evidence supporting it. Dogs can and do get salmonellosis. For every email I've had talking about how a raw diet has made a big difference in someone's dog's health, I get at least one email from an owner or vet whose dog got salmonellosis while eating raw meat (and sometimes people in the house also got sick). A dog that eats Salmonella can shed it in its feces. The bacterium can clearly survive passage through the intestinal tract. Most dogs that ingest Salmonella do not get sick. Some do. Sometimes their owners do as well.

Wild dogs eat raw meat. That's what they've evolved to do.

  • Wild dogs also have a much, much shorter lifespan than domestic dogs. It's obviously not all related to diet, but I don't want my dog to have the lifespan of a "natural" dog, I want her to have the longer and healthier lifespan of a modern pet dog. Take a look at older cemeteries and see the number of headstones of very young children. Raw milk played a big role in many of those.
  • Regardless, the question isn't about the health of dogs fed raw meat. That's a completely separate issue. The issue is the risk that raw-fed dogs might pose to the highly compromised people that are found in healthcare facilities. People need to think about the health of those susceptible individuals when they get involved in this debate.

(click image for source)

More raw debate

Since my post about Delta Society and raw diets, I've had multiple questions or comments about the research behind it.

Here's one question:

"Please site the research that "clearly show..." that raw-fed animals shed bacteria at a higher rate."

Here's the answer:

Lefebvre et al, Journal of the American Veterinary Medical Association, 2009

  • Study following therapy dogs over the course of a year. Diet history was recorded. Raw-fed dogs were 17 times as likely to be shedding multidrug-resistant E. coli compared to non-raw-fed dogs, and more likely to be shedding Salmonella.

Leonard et al, Zoonoses and Public Health, 2010

  • Study of healthy dogs in households. Dogs fed a commercial or homemade raw diet were greater than 5 times as likely to be shedding Salmonella than other dogs.

Lefebvre et al, Zoonoses and Public Health, 2008

  • Study investigating therapy dogs in Ontario and Alberta. Raw-fed dogs were 23 times as likely to be shedding Salmonella and 17 times as likely to be shedding multidrug resistant E. coli.

Lenz et al, Canadian Veterinary Journal, 2009

  • Campylobacter jejuni was found in the feces on 2.6% of raw-fed dogs and Salmonella was found in 14% of raw-fed dogs. Neither was found in any dogs not fed raw meat.

That's pretty clear to me.

Raw diets banned by Delta Society

Delta Society has recently announced a policy prohibiting animals fed raw meat or raw animal products from participating in their Pet Partners program. This policy was established because of research indicating dogs fed raw meat are much more likely to be shedding harmful bacteria like Salmonella and drug resistant E. coli in their feces compared to dogs fed commercial or home cooked diets, and the fact that these dogs come into close and frequent contact with people that are more susceptible to infections and at increased risk having severe infections.

Not surprisingly, internet chat sites are abuzz, and there's much condemnation and consternation from some. Some of the more vocal minority are stating that they'll just lie and say that they're not feeding raw.  I guess such dishonest actions would be based on a combination of ignorance and arrogance - feeding raw is your own decision, but blatantly flouting a policy that was put in place to reduce risks to those most susceptible is stupid and irresponsible.

One of the problems with peoples' reactions is the fact that they are confusing two separate issues. One debate is whether raw feeding is more healthy or more harmful to the pet. That's a controversial area, but this policy has nothing to do with that. This policy deals with the increased likelihood that raw-fed pets are shedding harmful bacteria. That's been very well proven in scientific studies. Do raw-fed pets cause disease in people in hospitals? We don't know. However, we have enough evidence to indicate there is the potential for increased risk to patients, and that added risk can be eliminated by not feeding raw meat products.

Hopefully, people will realize that this policy has been put in place for a good reason, and that it's focused on protection of people at high risk of serious illness. It's not a broad condemnation of raw diets, it's just a statement that it is not considered appropriate for dogs that will have contact with high risk populations - a recommendation that's far from new.

Details about this policy can be found here.

Disclosure: I'm a member of Delta Society's Medical Advisory Board. However, the opinions expressed here are mine and do not necessarily reflect the views of Delta Society.

Rabies in a household but hopefully not in a nursing home

.A Texas couple is undergoing rabies post-exposure prophylaxis after an abandoned puppy they adopted was diagnosed with rabies. They found the puppy outside and brought it into their house. One of them was subsequently bitten and they found out about the rabies diagnosis on Christmas eve.

One of the couple is quoted as saying "The doctor said 'It was a good thing they didn't wait until Monday, because it would have been too late. We couldn't have given you the shot because it wouldn't have done any good. You would have been dead within 48 hours." I really hope they completely misinterpreted what the doctor said, otherwise the doc has no clue about rabies. Prompt treatment is the goal, and you certainly don't want to wait any longer than you have to, however rabies doesn't kill in 48 hours, and you can start post-exposure treatment any time (just the sooner the better).

The couple also have seven other pets, who may also have been bitten. There wasn't any comment about what's happening to those pets. Hopefully they are properly vaccinated so they can be given a rabies vaccine booster and only undergo a short-term "quarantine" at home with the owners. (The alternative is immediate euthanasia or strict, long-term quarantine for months).

This isn't a new scenario - adopting a stray animal then finding out it has rabies. The less you know about an animal at the time of adoption, the greater the risks. I'm certainly not saying don't adopt a stray animal. But, if you are going to do it, recognize the risk, make sure you are in a low-risk household (everyone's susceptible to rabies, but some people are at greater risk for other zoonotic diseases and stray adoptions should be avoided by them), get the animal examined by a veterinarian as soon as possible, and make sure that it gets examined by a veterinarian if it develops any signs of disease.

All this leads into another another story I read a few days ago. Basically, it was a feel-good story about someone who found some puppies, stopped by a nursing home (or similar facility) and the facility adopted one or more of the puppies. This demonstrates some good points (e.g. resident's presumably had a great time watching the pups) and bad points (e.g. disease exposure, unknown temperament, injury risks from rambunctious puppies...) of animals in long-term care facilities. What if the puppies that were adopted by the home had rabies? It's happened before, and you end up having to administer post-exposure prophylaxis to a large number of people that already have enough health issues and risks. Nursing homes and other facilities should never adopt stray animals. Hopefully we don't see a news release in the next few weeks about widespread rabies exposure in that facility. 

 Video from

The good and bad of pet therapy

.An article about a therapy dog demonstrates some good points of these programs and places to improve. The story is about "Taco", a Chihuahua involved in pet therapy at the Livingston Regional Hospital (Tennessee). The obvious benefit of the program is highlighted by the owner's comment "She creates smiles when there were none." There are definite social and emotional benefits of pet therapy. There are also some potential health benefits, although the research on that isn't the strongest. On the downside, there are disease transmission concerns. These can be greatly reduced through attention to some simple procedures, but this article describes a number of concerning yet common problems:

"(Taco) greets each patient (ones who are comfortable enough to have her in their lap) with kisses on the nose."

  • Being allowed to lick patients has been shown to be a risk factor for visitation dogs acquiring MRSA. Being allowed to like the nose is about as good of a model of MRSA transmission as you can develop, because the nose is the number-one site where this important bacterium lives. This type of licking can also transmit various other infectious agents to this compromised hospital population. Licking is an unnecessary behaviour that should not be permitted because it can be associated with infectious agent transmission.  Not permitting licking does little to decrease the value of visitation.

"(Owner Gerry) Cotnoir has had Taco since she was 9 weeks old.  She worked at Bethesda [Health Care Center) in Cookeville then and brought Taco with her to work every day. "She got used to people at an early age,""

  • Socialization of dogs is important, but a hospital is not the place to do this. Only dogs older than 1 (and ideally older than 2) years of age should be in hospitals. Young animals are more likely to bite or scratch; not necessarily from aggression but also from playful or excited behaviour. Young animals also have much higher rates of shedding of various infectious agents such as Campyobacter.
  • People in hospitals should not be bringing pets to work. Animals that are in hospitals should be there for formal, structured, short-term, properly observed and properly scrutinized visitation activities. That's not the case when someone brings a pet to work. A hospital is not a doggie day-care, although some people use them as such, with the occasional visit of a patient to explain why they are there.

The hospital's infection control personnel have approved the use of Taco in the Livingston facility, but you have to wonder how much they investigated the issues. There are clear guidelines for hospital therapy programs which aren't being followed here. Hopefully other important aspects of the guidelines, especially hand hygiene, are being followed. It's likely this is a situation where people don't understand the issues and don't realize that there are both concerns and resources to help them out.  Any facility that has, or is thinking of having, a visitation program, should be aware of these guidelines, plus other information from reputable groups such as Delta Society.

(Image source:

Antibiotics, pets and Clostridium difficile

Clostridium difficile is a high-profile bacterium, being an important cause of illness and death in people. It can also be found in various animal species, including dogs and cats. In a study we published earlier this year (Lefebvre et al, Journal of the American Veterinary Medical Association, 2009), factors associated with acquisition of Clostridium difficile by dogs involved in hospital and non-hospital therapy programs were assessed. Things that were significantly associated with a dog acquiring C. difficile were:

- Contact with human hospitals: Not too surprising since it's clear that hospitals can be highly contaminated with C. difficile and the hands of some patients petting the dogs are probably also contaminated.

- Contact with children: Most parents know that kids are biohazardous (we've getting over a round of illness in our house brought home by the kids - not an unusual event). Whether the increased risk for dogs is because kids have higher rates of C. difficile carriage, or because they have closer contact with dogs (with little hygiene) or some other factor isn't clear.

- Recent use of antibiotics: No surprise here. Antibiotic use is a well-recognized risk factor for C. difficile, since antibiotics can disrupt the normal protective bacterial population of the intestinal tract and allow C. difficile to grow.

- Recent use of antibiotics by a person in the house: I think this is a fascinating result and a great example of the close inter-relatedness of people and pets microbiologically. What presumably happens is that when someone is treated with antibiotics, they are more likely to acquire C. difficile and pass it in their feces. By doing so, there is a greater chance that their dog will be exposed to C. difficile, perhaps from the person's hands or the household environment. (The toilet would be a great source if the dog's a toilet-drinker). The implications of this, for both dogs and people, are unclear. It could be primarily an academic risk (i.e. of little practical significance), or it could be that interspecies transmission of C. difficile plays a role in disease in both species. We simply don't know at this point.

This is also a good example of why educational efforts regarding prudent antibiotic use need to be directed at both animal and human healthcare.

More information about Clostridium difficile can be found on the Worms & Germs Resources page.

Image source:

"Horse strep" in people

I received the following question the other day: "I have a friend who had chemo embolization on tumor on liver in late June. She is in hospital now, and an abscess was discovered on liver.  Pathology results said "equine strep".  Her brother visited immediately after procedure, and he works with horse full time."

Streptococcus is a group of bacteria that includes many different species. There are two main species in horses Streptococcus equi subsp. equi (aka S. equi, the cause of strangles) and Streptococcus equi subsp. zooepidemicus (aka S. zooepidemicus, a cause of various types of infections). As you can guess by the 'equi' name, their natural host is the horse. Strep infections are very common in people, but rarely involve these two species. Nonetheless, infections with either Streptococcus equi or S. zooepidemicus can be found in people, but S. zooepidemicus is most common. Usually, these infections develop in people who are already sick for another reason, have compromised immune systems, or in young children. Interestingly, not everyone that is infected reports direct or even indirect contact with horses.

Back to the question: it's hard to say what's going on here based on the the general term "equine strep", but presumably the person has an infection with S. equi or S. zooepidemicus. Whether horses are actually involved will be tough (or impossible) to determine. It's a tempting hypothesis that the patient's brother carried the bacterium from the farm to the hospital, but I'd be wary about making a definitive statement about the bacterium's origin solely based on that. There are ways to investigate this further, such as trying to isolate Streptococcus species from horses on the farm, typing them and comparing them to the strain that caused disease in the person, but this type of testing is very costly and almost never performed, as human infection with these species is so uncommon.

This should be a good reminder that people who are sick and in hospital are at higher risk for developing infections, and they can get infections from bacteria that rarely cause disease in healthy individuals. While there is no proof of a link to horses (at least in this case), good infection control practices should be used whenever anyone visits someone in the hospital. That would include not wearing barn clothes to the hospital and paying close attention to handwashing.

This Worms & Germs blog entry was originally posted on equIDblog on 26-Aug-09.

A need for a better definition of service animals

Because service animals are so important to the people they assit, they have much greater access to various venues than other animals. In the US, the Americans with Disabilities Act (ADA) specifically addresses service animal access issues. It was a landmark act that ensured appropriate access for these animals so that people who require them are able to take them into areas where other animals are not allowed. However, some aspects of this Act can lead to abuse of the regulations and unwanted scrutiny of "real" service animals. I was at an infection control conference recently and numerous people commented on problems they have had with people with questionable "alleged" service animals, the inability to find out whether they really are service animals, and the potential legal implications of trying to do anything to prevent them from entering certain areas.

These problems occur because of a combination of strong and vague statements in the ADA:

One problem is the definition of service animals: "Service animals are animals that are individually trained to perform tasks for people with disabilities such as guiding people who are blind, alerting people who are deaf, pulling wheelchairs, alerting and protecting a person who is having a seizure, or performing other species tasks."

  • The definition itself is fairly straightforward, but there is no clear indication of what "trained" entails, and no requirement for formal training or certification, nor restriction of any animal species. Based on this, I could say that my sheep are trained to do something for me and then take them into a restaurant with me.

Some other key points in the ADA:

Businesses may ask if an animal is a service animal or ask what tasks the animal has been trained to perform, but cannot require special ID cards for the animal or ask about the person's disability.

  • This means that while businesses can ask, all someone has to do is say "yes, this is a service animal" and the conversation is done. Some people that truly need service animals are not visibly disabled and you can't tell whether someone needs an animal by simply looking at them or talking to them. Back to my sheep example, if someone asked why I had a sheep on a leash in a restaurant, all I'd have to say if that he's my service sheep and he's trained to do something. Theoretically, I could walk into a crowded location with a Salmonella-spewing baby chick, adult cow or some other inappropriate animal and no one could do anything. Yes, those are extreme examples, but people like to test extremes.

A person with a disability cannot be asked to remove his service animal from the premises unless: (1) the animal is out of control and the animal's owner does not take effective action to control it (for example, a dog that barks repeatedly during a movie) or (2) the animal poses a direct threat to the health or safety of others.

  • The problem here is who defines "direct threat." This is an issue because it is subjective, yet people can be penalized if they ban an animal and a complaint is upheld. Think back to the recent example of the pet chimp that almost killed someone. It wasn't a service animal in this case, but some people claim their monkeys are service animals. Some probably are, since some monkeys are specially trained to help the disabled (especially people with spinal cord injuries). Monkeys can be very dangerous, yet it might be hard to look at any given monkey and say it poses a "direct threat" to another person. A properly trained and temperament-tested monkey is probably low risk and justifiable. But, proper training and temperament-testing aren't required by the ADA

Businesses that sell or prepare food must allow service animals in public areas even if state or local health codes prohibit animals on the premises.

  • Public health codes are there for a reason... to protect the health of the public. Therefore, careful consideration must be taken before breaking public health rules. The risks posed by a properly trained service dog are inconsequential, and properly trained and tested animals of appropriate species absolutely should have free access. Other species have different risks and these need to be considered. All animals are not created alike.

Violators of the ADA can be required to pay money damages and penalties.

  • This is good for true violations such as someone refusing access to someone with a trained seeing-eye dog. However, it also leads to difficulties excluding high risk situations.

I'm know I'm going to get nasty emails from people with various untested, unregulated (and probably untrained) "service animals," but I think this is an important issue. The ADA provides a great framework for ensuring proper access to and by service animals. However, I don't think it's clear enough. Vague acts create the potential for stretching the rules and violating the spirit of the law. I'd never advocate getting rid of this Act, however I think it needs to be rethought. There is a great need for a clearer definition of what constitutes a service animal. Service animals should be specially trained, temperament-tested and certified by an independent body. If someone thinks they need a service monkey or horse, the need for that should be clear and the animal should be properly trained and scrutinized. Otherwise, it's a pet and shouldn't be given the same access. Problems that occur from inappropriate "alleged" service animals risk unnecessary scrutiny of, and barriers to, real service animals.

If you disagree, please comment. However, don't just send me the typical "I have a service horse and you're an idiot" comment that comes through periodically. Tell me why you disagree with better defining species, training and certification.

Pet therapy and patients with multidrug resistant bacteria

I received this question on hospital visitation animals today...

For patients who are on contact precautions for a multidrug resistant organism, what other than hand hygiene could be suggested?  I am currently rewriting our hospital infection control policy related to pet therapy and visitation with clients with a MDRO [multidrug resistant organism]. I was thinking that visitation of clients with a MDRO should not be allowed??? Any advice?

I am surprised at how often I get asked about pets and patients with multidrug resistant infections. These patients are typically isolated and managed with "contact precautions". The goal of these precautions is to reduce the amount of contact between the infected patients and other people (staff and patients) and includes the use of barriers (e.g. gowns, gloves) to reduce the risk of disease transmission when contact must occur. If the goal is to reduce contact, we shouldn't be allowing patients with these infections to have contact with pets. Pet therapy animals usually visit a large number of patients in the hospital, and we don't want them transmitting an infection to the other patients they visit. Pets are also accompanied by people (their owners), who we don't want to  expose to infection and/or to transmit infection to other patients. So, visitation of these high-risk patients by animals should not be allowed. Animal visitation can have very positive effects and its unfortunate that people with multidrug resistant infections should be restricted from these programs, however it's for the greater good of other patients in the hospital, as well as the pet and its owners.

More information about pet therapy programs can be found in our hospital visitation archives.

Image from:
(It's a cute cartoon. I just have to ignore the fact that the dog is on the bed (i.e. the patient's living space) without being on a barrier such as towel or drape.)

Dirty hands, Dirty therapy dogs?

A study published recently in the Journal of Hospital Infection (Lefebvre & Weese, 2009) looked at contamination of the haircoat of animals used in hospital visitation programs. In the study, Dr. Lefebvre petted animals that were going into a hospital and we cultured her hands.  Then when the dogs finished their visits she petting them again and we re-cultured her hands.  We tested for MRSA and Clostridium difficile, two important causes of hospital-associated infections. The goal was to see if the dogs' coats could become contaminated, presumably by patients' hands, during regular visits. The dogs' paws were also tested before and after to see what they picked up walking around the hospital.

After being in the hospital, one dog (4%) was found to have C. difficile on its feet. The strain that was recovered (ribottype 027/NAP1) is a strain of great concern, being it has caused outbreaks of illness internationally. The fact that there was C. difficile on the dog's paws, which was presumably picked up from the floor in the hospital, isn't exactly surprising. It shows that contact with dogs' feet (e.g. shaking a paw) could be a source of exposure to people in hospitals, and at home, and that dogs could be exposed to C. difficile after hospital visits by licking their paws. We previously demonstrated in another study that visitation dogs are at high risk for picking up C. difficile during visits.

The more important aspect of the study was what ended up on Dr. Lefebvre's hands after petting the dogs - a very common type of contact with dogs in hospitals and at home. She picked up MRSA from the coat of one dog (4% of dogs overall) after it had been in the hospital. The dog was not an MRSA carrier, and it did not have contact with anyone known to be carrying MRSA during its visit. Presumably, there was someone in the hospital that was an unknown carrier. If MRSA was on the person's hands, they could have transfered the MRSA to the dog's coat. The important part is that this MRSA was then able to spread to someone else's hands (Dr. Lefebvre's, in this case). This is a potential route of transmission of MRSA to patients during hospital visitations, and to other people these dogs might encounter outside of hospitals. The same thing could presumably happen with other infectious agents, including bugs like norovirus and influenza. These can't infect dogs, but dogs could potentially be short-term vectors of these viruses and spread them around a hospital.

The good thing about all this is that some basic guidelines can greatly reduce the risk of transmission:

  • Patients should wash their hands  or use an alcohol-based hand sanitizer before  (to reduce the risk of depositing something on the pet's coat) and after (to reduce the risk of picking something up from the pet's coat) touching a pet.
  • Pets should be kept off beds unless a disposal/washable barrier (e.g. drape, towel) is kept between the pet and the bed sheets, to prevent dirty feet from infecting the patient's living space/clothes/sheets.
  • People should not "shake-a-paw" with pets in hospitals.

All these precautions can be taken without compromising the benefits of hospital pet visitations. More details about guidelines for hospital visitation aniamls can be found in an earlier post. More information about Clostridium difficile and MRSA can be found on the Worms & Germs Resources page.

Service monkeys and horses...where do we stop

I was at the annual conference of the Society for Healthcare Epidemiology of America on the weekend. This is a hospital infection control organization, and one of the talks I gave was about animals in healthcare facilities. One question that came up was about unusual service animal species like monkeys. Service animals are specially trained animals that help disabled individuals with specific tasks. The most common examples are seeing-eye dogs. In the US, the American Disabilities Act protects service animals and dictates that they must be allowed to go wherever the person goes. I don't think people have a problem with that in general. However, there are concerns with respect to non-traditional species being used in these roles, and the question at the meeting was about service monkeys. Monkeys can be incredibly strong physically, and they can carry some important infectious zoonotic diseases, so there are concerns about them being allowed in hospitals. Part of the issue is what really makes an animal a service animal.  Should all animals that help someone out (in any capacity) be considered service animals?

That same topic came up in a recent ABC News article that described a seeing-eye horse in Texas, including a video of the owner riding the horse while grocery shopping. 

I have no doubt that this horse helps out its owner and provides great joy, if not increased freedom. However, I'm not convinced that a horse is necessary to fulfill this person's need for a service animal. Why use a horse when a dog could do as good (or a better) job? How was the horse trained? Was it trained under a formal program so that it is truly helpful? What types of health and behaviour screening have been used? What are the additional risks associated with using such a large farm-animal species?

Horses, even based solely on their size, can easily cause injury to members of the public without meaning to, simply by stepping on a person's foot or bumping into them, for example.  Some people might be scared of horses, especially indoors. Horses aren't litter trained, and horse manure can carry potentially infectious agents. I have a big problem with the video of this horse in a grocery store.  At end of the day, is a horse really necessary for what this person needs, and has the horse been adequately evaluated to ensure that it is low risk to the public? I don't think the answer is yes to either question, let alone both.

We certainly must do all that we can to allow full access of appropriate service animals, but we also need ensure that novelty "service" animals don't cloud the picture and potentially have a negative impact on true service animals.  The article states "...the government has begun rethinking whether the regulations should be changed to exclude some animals." That sounds like a great idea to me. Careful review of this issue, including the benefits to people, risks to the public and the need for new species over traditional options all need to be considered. 

Image: captured from video at

Pets and immunocompromised people

There was an interesting article in USA Today about the increasing recognition of the positive role that pets can play in patients recovering from serious disease, and how contact with pets can sometimes conflict with disease transmission concerns in these same patients. The attitude towards pet ownership among physicians is highly variable - some recognize the strong human-animal bond and its positive effects, while others see pets as unnecessary infectious disease risks. The infectious disease concerns are heightened in patients with compromised immune systems, to the point that sometimes people are told to get rid of their pets if they are severely immunocompromised.  However, more and more pet owners, veterinarians and physicians are beginning to question if this is truly the best approach.

The USA Today article describes the experiences of a cancer patient whose greyhounds were "banished to a caregiver on doctors’ orders". Considering she was at high risk for (potentially fatal) infectious disease because of chemotherapy, bone marrow transplant, stem cell transplant and immunosuppressive drugs, it’s not surprising that there was concern about the pets. After researching the risks, and measures she could take to reduce those risks, the patient convinced her doctors that the risks from her dogs were not as great as the benefits from having them around, and so the dogs returned home. While everyone’s relationship with their pets is different, the patient felt that the return of her dogs was an important step in her recovery, stating "There's no question that having (the dogs) with me these past few months made a huge difference in my recovery".

Infectious disease transmission from pets to people is certainly a real issue, and it is of particular concern in people with weakened immune systems. There is not, nor will there ever be, a "no-risk" pet. Every contact with a pet, just like every contact with another person, carries with it some degree of risk of disease transmission. What needs to be considered is the trade-off, the risks versus the benefits. In some people, the risks are greater than the benefits because of the severity of disease, type of pet, the person's ability (or more likely inability) to interact with the pet. In other people, especially those who have a very strong bond with the animal, the positive social and emotional benefits of pet ownership may greatly outweigh the associated disease risks. The article contains a great quote from Dr. Ray Pais, a pediatric hematology/oncology specialist, saying "Our young patients have already given up so much, I see no reason at that moment for them to also lose the dog."

People that have compromised immune systems need to have a serious discussion with their physician, veterinarian and family about the best thing to do with their pets while they are sick. While there is very little research in this area, taking a few common sense precautions should reduce the risks of disease transmission. These include:

  • Avoiding contact with stool
  • Frequent handwashing
  • Preventing licking of the person by the pet
  • Proper training to reduce the risk of bites and scratches
  • Keeping cats indoors
  • Following a good preventive medicine program for the pet

More information about Immunocompromised Pet Owners will be available soon on the Worms & Germs Resources page. The CDC also has useful information on its website about this topic.

Thanks to Dr. Doug Powell of Barfblog for forwarding this article.

Raw meat feeding revisited: what are the risks and benefits of raw diets for dogs?

Feeding raw meat diets to dogs is a very controversial issue. Some proponents passionately advocate these diets (e.g. the BARF diet) based on vague and unproven recommendations. Opponents cite various studies showing that pets fed raw meat (not surprisingly) have high carriage rates of potentially harmful bacteria such as Salmonella, and reports of diarrhea or nutritional imbalances in these animals. However, there have been only a few good studies looking at the true health benefits and risks of feeding these diets to dogs.

A recent study in the journal Zoonoses and Public Health described the risks of therapy dogs shedding Salmonella and other potentially harmful bacteria. The authors tested 200 dogs over a 1 year period, 20% of which were fed raw meat as part of their normal diet. Dogs fed raw meat were 23 times more likely to shed Salmonella compared to other dogs. They were also 17 times as likely to be shedding extended spectrum beta-lactamase (ESBL) Escherichia coli (a highly drug-resistant form of E. coli). 

The study concluded that, because of the risk of Salmonella shedding and the high-risk nature of the patients and other people that therapy dogs interact with, dogs that are involved with hospital/patient visitation programs should not be fed raw meat.

What does this tell us about feeding raw meat to pets?

Although this study doesn't answer all of the questions about the risks of raw meat diets, it reinforces the fact that pets fed raw meat have significantly higher rates of shedding of potentially harmful bacterial such as Salmonella and antibiotic-resistant E. coli. Eating pig ear treats has also been associated with Salmonella shedding. However, there was no association between a dog being fed raw meat and the animal itself developing diarrhea. In fact, raw-fed dogs had a lower incidence of extra-intestinal infections (combination of eye, ear, skin and urinary tract infections).

In general, raw meat diets should be avoided. While adverse health effects were not reported in this study, disease (including deaths) from Salmonella has been reported in dogs in other studies. While the overall risk may be low, feeding raw meat is an avoidable risk. However, it would be inappropriate to completely ignore the finding that raw-fed dogs had lower rates of certain infections. It is possible that there can be health benefits from feeding raw meat in certain dogs, but the potential benefits must be weighed against the potential risks to the animals and people with which the has contact. Raw meat diets should never be fed to pets that have contact with immunocompromised people (in the household or as part of visitation program), infants or the elderly.

If you are going to feed raw meat to your pet, make sure you take precautions to reduce the risk of infecting yourself or someone else. We'll post more on that aspect soon.

Clostridium difficile in dogs

Clostridium difficile is a bacterium that is an important cause of disease in people and has been getting a lot of press lately in this area. Associated with that has been a lot of questions about C. difficile in dogs. Here are some facts to ponder:

  • C. difficile is thought to be a cause of diarrhea in dogs in the general population, although its true role in disease is still unclear. C. difficile infection in dogs may cause mild to moderate diarrhea that is treatable.
  • Some (<5%) healthy dogs can carry C. difficile in their intestinal tract. They may never have a problem with it.
  • Dogs that visit human hospitals have much higher rates of C. difficile carriage, but we currently don't have any evidence that they are at higher risk for disease. They probably ingest the bacterium from the hospital environment and patients' hands.
  • We are highly suspicious that C. difficile can be transmitted between people and pets (in both directions), but pets are probably not a major source of infection for people.

Certain risk factors for C. difficile carriage or infection are known, including antibiotic use (people and pets), advancing age (people) and admission to a hospital (people and pets). Some of these are not avoidable. General considerations for C. difficile prevention include:

  • Don't use antibiotics unless they are actually required. This goes for people and pets.
  • Wash your hands regularly.
  • Avoid contact with dog stool, especially diarrhea.
  • Enjoy your pet. There is always a risk of disease transmission but the risk of getting C. difficile from your pet is not something to stay up worrying about.

Prescription pets

It’s not “take 2 doses of Fido and call me in the morning”, but contact with pets has been shown to be beneficial to people in hospitals and nursing homes. However, concerns have been raised about whether Fido could be spreading more than good will; from Salmonella and ringworm to superbugs like methicillin-resistant Staphylococcus aureus (MRSA).

Recently, new international guidelines for these programs were published in the American Journal of Infection Control. A joint effort of physicians, infection control personnel, veterinarians and visitation program representatives, these guidelines provide comprehensive recommendations on how to run these programs and reduce the risk of disease transmission. Among the areas covered are appropriate animals, health care for visitation animals, appropriate human-animal interaction and the roles of visitation programs, pet owners and hospitals. “The goal of these guidelines is to keep these highly beneficial animal-assisted interventions safe for all involved.  That means protecting not only the patients but the pets, as well as other people and animals the pets may subsequently interact with”, said lead author Dr. Sandi Lefebvre.

Horse visits hospital

Recently, a story about a man who brought a horse into a hospital to visit his father was widely reported. The horse apparently made it to the man’s room, which included a trip in an elevator. The son, who appeared intoxicated, was eventually asked to leave (and take the horse with him). Said a hospital spokesperson “We do have a pet visitation policy, but it does not include a horse”. Strangely, the horse that was brought to the hospital apparently wasn’t even the father’s horse (which supports suspicions of the son’s lack of sobriety).

There are guidelines about which animals are appropriate for hospital visits, although it shouldn’t take an expert to figure out that a horse is not an appropriate candidate. Kicks, bites, and trauma from being crushed or run over are among the most obvious concerns. Horses can also carry a variety of bacteria that are potentially dangerous, especially to people in hospitals. These include Salmonella and methicillin-resistant Staphylococcus aureus (MRSA). There also aren’t that many house-trained horses out there.

So, while I can easily see how someone in a hospital would like to see his or her horse, there’s no way this should even be considered.

Some closing thoughts
-    Would you like to ride in an elevator with a horse?
-    Would you like to be stuck in an elevator with a horse?
-    Do you think the horse was house trained?
-    Do you think any of the healthcare personnel washed their hands after touching the horse?

This isn't the first time a horse has been in hospital, and some even get invited. The picture is from a story in Veterinary Practice News that described a program where horses were brought into hospitals!