Health care delivered in the emergency room is often derided as expensive and inefficient, the source of our health spending woes. Physician Robert O'Connor has a different way to describe emergency medicine: An incredibly good deal.

O'Connor chairs the department of emergency medicine at the University of Virginia School of Medicine. As an emergency room doctor, he is not unbiased in defending the work he and his colleagues do. He's also pretty tired of all the rhetoric about emergency rooms as the health spending culprit.

He says that ERs only account for 2 percent of all health care spending—and argues that patients actually get tons of bang for their buck.

"It represents a remarkable value that we have this service available at all times, in all parts of the country, for every citizen," he told reporters during a press conference at the Association of Emergency Care Physicians' annual meeting. 

That 2 percent figure comes from the Medical Expenditures Panel Survey, a large-scale federal survey of patients and doctors' health care spending. It found that Americans spent $51.1 billion on emergency care in 2009. O'Connor estimates that's 2 percent of the $2.4 trillion spent on health care annually.

The MEPS data only covers $1.2 trillion of health care spending; it excludes any health spending that isn't a direct payment to a provider (administrative costs, for example) as well as services of dentists, optometrists and some other doctors.

To compare apples to apples, it's probably better to use the MEPS denominator for overall health spending. Even using that, emergency department spending only comes out to 4 percent of the nation's health care bill. 

Another surprising data point: Emergency room spending is pretty uniform across different types of insurance coverage. That challenges some of the assumptions that the uninsured tend to visit the emergency room the most frequently. As it turns out, 89 percent of emergency room patients have some form of public or private insurance.

"We were hearing a lot of dialogue during the health reform debate about these expensive emergency rooms and how we have to get patients out of there," O'Connor recalls. "We decided to go look at the numbers, so we got the federal data and were surprised it was as low as it was.

"There’s a conventional myth its 40 percent or 50 percent. It’s not that big in the grand scheme of things."

Government data does show that the average emergency room visit is significantly more expensive than a trip to the primary care doctor, $922 versus $199. But O'Connor pushes back against the notion that the same services are delivered in each situation.

A growing body of research has shown that relatively few visits to the emergency room are for non-urgent care, things like runny noses that could easily be seen in an outpatient setting. Just about 7 percent of emergency room visits by those with Medicaid were for "non-urgent" needs, according to research earlier this year from the Center for Health System Change. That number stood at 9.9 percent for those with private coverage.

"We're handling a third of the acute care visits and essentially 50 percent of the decision making in hospitalizations, whether someone will be admitted or not," says ACEP president Andrew Sama. "I think what were spending on that seems to be reasonable."

Sama and O'Connor are skeptical that cost control will happen in their offices. O'Connor's own research shows that repeat emergency room visitors tend to have high rates of chronic health issues, things like diabetes and high blood pressure. Management of those conditions is more likely to happen in an outpatient setting, rather than in an emergency department. 

Even if emergency room visits do drop, its unclear what savings that might yield.  As O'Connor puts it, even if you prevent 10 percent of emergency room admissions, you're still working with in a very small sliver of the country's health care spending. "It's just not a whole lot of money," he says. 

At the same time, he still does see space for emergency rooms to operate more efficiently. If he had access to medical records from nearby hospitals, for example, O'Connor thinks that would prevent repetitive tests and scans. More access to mental health resources could also improve his work.

These are all relatively small scale changes. O'Connor's bigger message seemed to be about not having the fingers pointed in his direction when it comes to driving up health care costs—instead turning them to other parts of the health care system.