Paging Dr. Mario

Over at The Health Care Blog, Douglas Goldstein wrote a post earlier this month on “health eGames,” a category of video games “that deliver measurable health benefits” to patients who play them. iConecto, a developer of such games, recently reported that there are already over 300  health eGames available—and that the size of this market over the next year will be $7 billion+. Consumers definitely seem open to the idea of healthy gaming: Wii Fit, Nintendo’s fitness video game, is poised to become the best selling title of the year, having already sold 8.7 million units.

This got me thinking: what about video games for doctors?

Kevin M.D. began to answer my question on Tuesday when he posted this YouTube clip on his blog:

This video simulates emergency room situations; surgeons can use it to train themselves in particular operations. This is a pretty cool idea: test the skills of a surgeon, but in a context where his slip-ups won’t cost lives.

What’s more surprising is that even conventional video games can play a role in training surgeons. In January, the BBC reported that a British hospital asked “eight trainee surgeons to spend an hour playing [non-medical video games] before performing ‘virtual reality’ surgery” through the program in the YouTube clip above. The hospital found that “game players scored nearly 50% higher on tool control and overall performance than other trainees.” The game that was most effective at improving their skills was Marble Mania, in which the player rolls a marble through a maze and obstacle course.

This isn’t all that weird when you consider how the technical skills you need for most video games—spatial awareness, fast reflexes, dexterity, and precision—are also vital to successful surgery. If a video game can help a surgeon brush up on these skills, it doesn’t matter whether it’s about a marble or a monkey.

This brings us to Super Monkey Ball, a video game in which you play soccer as a kooky little monkey.
The game is a favorite at Beth Israel Medical Center in New York City,
where it’s regularly played by surgeons including the chief of
minimally invasive surgery. In 2005, the New York Times noted
that Beth Israel surgeons kept “an Xbox, along with PlayStation 2 and
GameCube consoles, just a few strides from the operating room…[in order
to] warm up…just before surgery.”

Super Monkey Ball’s popularity at Beth Israel stems from a study
that the hospital did in 2004:  researchers looked at 33 doctors and
found that those who “spent at least three hours a week playing video
games made about 37 percent fewer mistakes in laparoscopic surgery and
performed the task 27 percent faster than their counterparts who did
not play video games.” That’s because the skill set relevant to Super
Monkey Ball and laparoscopic surgery are strikingly similar: an AP report
on the study quotes one surgeon as sitting down for a game and
exclaiming “yes, here we go!…I need the same kind of skill to go into
a body and sew two pieces of intestine together."

There’s a clear novelty factor in the fact that surgeons are playing
games in order to prepare for surgery. But there are also more serious
considerations to reflect on here, and one of them is
cost-effectiveness.

A high-end medical training device of the type currently available
to most hospitals can be prohibitively expensive. In 2005, Rosser told
the Times that he “had been using one” of these machines to
practice for laparoscopic surgery, but that the device cost $200,000
(about $224,000 in 2008 money), making it too expensive to serve as a
widespread tool for medical training outside of wealthy medical
centers. But you can buy an Xbox from Wal-Mart for a couple of hundred
dollars, giving poorer hospitals a much cheaper option for keeping
surgeons on their game.

Not all of the virtual training programs are kids’ stuff. More
serious fare also can replace expensive training devices. Consider the
work of the Nationwide Children’s Hospital, Ohio State University’s
Department of Otolaryngology and the Ohio Supercomputer Center. The
three organizations have come together to develop the Virtual Temporal
Bone Project, a computer program that provides medical students with a
3-D model of the temporal bone, which is located near the ear canal..
Through the program, students can practice the drilling that they would
perform during ear surgery in the real world. 

According
to the Ohio Supercomputer Center, the ear diseases addressed by these
doctors-to-be account for more than $8 billion in annual health care
costs in the U.S. Further, “training the health professional charged
with treatment of this significant disease process requires five to
seven years at an annual cost of more than $76,000.” By contrast, “the
simulator’s hardware costs about $6,500” and “its software is open
source because it is a publicly funded project [thanks in part to
grants from the National Institutes of Health].” In other words,
because the money and the code are both public, developing the
simulator once means that it can be more easily picked up by other
hospitals and programs. 

But not all virtual medical training is cheap. The University of Wisconsin-Madison is developing a program called Medical Cyberworlds,
a “massive multiplayer online game” meant to immerse doctors in a
virtual reality hospital setting for training purposes. The price tag
for its development: between $20 and $60 million.

These huge numbers are something of a red flag. One of the
big—indeed, perhaps the biggest—problem with our health care system is
its unfettered proliferation of high-tech, high-cost medical devices
that then are used in cases where we really don’t need them. America
has a tendency to go way overboard when it comes to hi-tech medical
gadgets. With that in mind, it’s not too difficult to imagine a world
where we spend millions of dollars on sophisticated virtual reality
programs long after we have reached a point of diminishing returns.
Indeed, one gets the sense that software development companies are
eager to make such a world a reality: already, iConecto is talking up the size and future growth of the medical video game market as an exciting business opportunity.

Still, it would be unfair to paint Cyberworlds as a harbinger of
American excess without noting that the program is also an example of
how medical video games can train and educate doctors in the more
interactive dimensions of care. One of the key elements of Cyberworlds
is that it’s a multi-player game: instead of just having the computer
spit out an established medical condition to direct the player’s
actions, Cyberworlds lets other players control patients or medical
professionals in the online universe. The idea is to incorporate a
degree of human unpredictability and active cooperation into the
virtual training.

The idea of immersing players in virtual scenarios that test more than
their precision with a scalpel is a compelling one. In May of last
year, the American Geriatrics Society reported
that researchers had developed “3-D virtual reality video game [called]
RiskDom-Geriatrics.” The game was “designed to train medical students
to make effective home visits, simulated a patient’s home…[it] allowed
players to explore and evaluate the home for hazards that could lead to
falls and other injuries.” (You can check out an online version of the
game here).
Preliminary data found that “medical students, who were evaluated
before and after playing and had to play against time and distractions,
showed improvements in their understanding of how to make an effective
home visit.”

The geriatrics example speaks to the potential of
video games to help doctors learn to coordinating care while engaging
in “thinking medicine”—an important, but overlooked part of health
care. Anything that helps more doctors get better at  listening to and
talking to patients is a good thing—especially if it’s packaged in a
way that is broadly accessible.

While teaching doctors how to interact with patients, video games
also reveal how racial discrimination can stand in the way of good
medicine. In July, researchers from the University of Florida ran an experiment in which two dozen third-year medical students interacted with a virtual female patient. The Gainseville Sun has gives us the details: 

“For
half the students, the woman appeared to be light-skinned; the other
half interviewed a dark-skinned woman. Both women had the same voice,
animation and appearance….

“Outside observers [both medical and nonmedical] watched videos of
the students doing the interviews, but could not see the skin color of
the virtual patient. They then rated the students’ empathy toward the
woman’s medical complaints [on a scale of one to seven]. The students
interviewing the dark-skinned woman were rated consistently less
empathetic. The results [for these students] correlated with standard
psychological tests of [racial bias that the participants also
completed.]”

The importance of this study is that it introduces
the idea of “automatically detect[ing] bias and, then and there [at the
point of detection,] help[ing] medical students” improve their skills
“at interacting with people who come from different racial and ethnic
backgrounds." This is an important issue to tackle. As I’ve noted
in the past, a systemic bias against racial and ethnic minorities seems
to be built into our health care system.  (No, I’m not saying all
doctors are racist—just that, as a statistical matter, ethnic and
racial minorities are less likely to get high quality care).

Earlier this month, the University of Washington released the
results of an online survey which tested the unconscious racial
attitudes of 2,500 doctors. The study looks for subconscious signs of
bias by asking the test-taker to quickly complete
a series of questions or tasks that imply a racial preference, such as
having to quickly state whether photos of blacks and whites were
positive or negative. Reuters reports that “researchers found that most
doctors in all racial and ethnic groups showed an ‘implicit preference
for whites over  blacks,’ with the exception of black doctors who
showed no preference for either race.” As the study’s abstract
puts it, these results “lends support to speculation that provider
implicit attitudes about race may represent one pathway to unequal
treatment in health care.”

So in the end, it turns out that video games can be used for more
than just sharpening reflexes: they can also provide simulated, virtual
environments that force doctors to stretch their thinking. Just as
importantly, they allow for a more controlled, targeted approach to
medical training, which can help doctors hone specific skills in a more
targeted way than might be possible without the limitless possibilities
of a virtual world.

To be sure, the verdict is still out in terms of how useful video
games really are for training doctors. But there are a number of
interesting possibilities here. Video games can  (a) help control costs
for expensive medical training, (b) democratize access to effective
training across institutions by providing a low-cost alternative to
over-priced devices, and (c) revitalize the teaching of “thinking,”
patient-centered medicine. And you thought video games were just
time-wasters.

3 thoughts on “Paging Dr. Mario

  1. I’m a little puzzled about a rather primitive video game being likened to virtual reality simulators in widespread use. While it was quite primitive, I built a prototype, somewhere between 1973 and 1976, of a cardiac life support training tool in which I actually modeled the conductive system of the heart, the effects of some drugs, and would generate the EKG changes from the student’s responses.
    Real-world medical training devices are far beyond what you are describing. For example, many intravascular and endoscopic procedures, which are often cheaper and safer than open surgery, are controlled by vernier knobs. Current simulators have the trainee looking through the same eyepiece or at the same monitor that would be used for a real patient, and the control knobs don’t just turn the cursor; they give tactile resistance that feel like the procedure. Some 1995 work is at http://ovrt.nist.gov/projects/health/vr-envir.htm
    May I suggest you look a bit at the value of advanced flight simulators in aviation safety, before suggesting this is a “verdict is out” because some video game company put out a news release? Yes, a full-motion Boeing 7×7 cockpit simulator, to say nothing of military aircraft simulators, is expensive. It also lets the crew practice for emergency conditions that would be far too dangerous to try in an actual training flight. What is the value of having experienced taking a wide-body jet into a controlled overrun off a runway, maintaining braking control? IIRC, there was a save last week involving just such a contingency.
    Incidentally, the Laboratory of Computer Science at Mass General was doing physician-patient bias studies sometime before 1970; that’s when I ported the patient interviewing software to Georgetown.
    Am I missing something?

  2. Sorry to be so late in correcting misinformation, but Medical Cyberworlds (MCI) is not affiliated with the University of Wisconsin. Also, the technology we have developed is actually far more exciting than you indicate. Most exciting of all is the fact that Medical Cyberworlds technology will provide measurably more effective training in humanistic healthcare at far lower cost than is now possible. It comes down to having the vision for how technology can be used intelligently, to make training but less expensive and more robust. That’s where Medical Cyberworlds excels.
    Watch this space. =)