Michael Gentile, age 74, says the hardest part about the joint replacement surgery he had on his right knee was the grueling in-hospital rehabilitation that followed.

He’s not a fan of pain medication, and because he didn’t take much of it, therapy sessions were especially hard – except for the time he was joined by a therapy dog.

“The dog comes in there and is watching, and you pet and talk to him. Therapy is painful because you have to really push it to make the therapy work. You're trying to make yourself hurt and it makes it a lot easier when there's a dog to take your mind off of it,” Gentile says. “It's a lot easier to go through therapy when you're paying attention to the dog instead of thinking about the pain from the exercises.”

He might be on to something. 

A recent study  found that interacting with therapy dogs after joint replacement surgery could mean patients have to pop fewer pain pills.

In a study presented at the 2009 annual conference of the International Society of Anthrozoology in Kansas City, MO, researchers reported that adults recovering from total joint-replacement surgery and getting pet therapy needed 50 percent less pain medication than those not getting pet therapy.

Nurses at Edward Hospital in Naperville, IL, gathered retrospective data of 174 patients who received total joint replacement surgery in 2003. All patients received care on the same nursing unit of the hospital and could choose to receive pet therapy or not. Visits from therapy dogs lasted between five and 15 minutes and were tailored to patient preferences.

There were no differences between the groups at the beginning of the study, but by the end, patients who had received animal assisted therapy needed significantly less pain medication, about half as much as those who didn’t get the animal assistance therapy.

Researchers point out there may be differences in the patients who self-selected animal assisted therapy and those who did not. They also acknowledge that other unknown variables may account for the differences. But they say that study still represents an initial step in determining the impact of animal assisted therapy.

“I think it’s working as a distraction,” says Julia Havey, a nurse at Loyola University Health System in Chicago, who presented the study. “It gives patients an opportunity to talk about things that don’t necessarily relate to them being in the hospital.”

Havey and her colleague, Frances Vlasses, PhD, an associate professor and chair of Health Systems Management and Policy at Loyola’s School of Nursing, believe animal-assisted therapy has positive effects on a patient’s psychosocial, emotional and physical well-being by acting as a therapeutic stress reducer. They say they’ve seen this as well over the last decade since they’ve been raising puppies to become assistance dogs to people with physical and developmental disabilities through a program called Canine Companions for Independence.

“The dogs are interesting because they become a nice vehicle to diffuse a situation, talking a bit about something beyond the problem at hand,” Vlasses says.

Havey and Vlasses are now beginning their own study, comparing retrospective data from patients at Edward Hospital, which has an established animal assisted therapy program and Loyola, which does not.

Stephen Wegener, PhD, an associate professor of Physical Medicine and Rehabilitation John’s Hopkins School of Medicine in Baltimore, says the study’s authors are advancing an interesting hypothesis, but he believes more controlled studies and peer review are needed to confirm their findings.

“As the authors indicate, we need to be very cautious in interpreting the these findings as there are many explanations for what is seen in the data,” Wegener says.

For instance, he says, because therapy dogs have volunteers with them, it is important to differentiate whether the animal itself is making a difference or if the visitor alone would reduce the level of pain or need for pain medication.

“Scientifically, it is important to note that there is no active control in this study. They’re not comparing a new treatment with an old treatment. They’re comparing a treatment versus no treatment," Wegener says.  “A more interesting study would be to randomly assign patients to different types of visits where half the patients get visitors and half get visitors and animals.”

But Wegener says this study does reflect recent developments in medical treatment. “ I think it’s also fair to say that this study is part of a broader trend to make patient care more friendly, less technical and more personal as part of the patient centered care movement.”