If you're considering surgery for back pain, a new study suggests you first should try physical therapy, which is both less risky and less costly.

The University of Pittsburgh study found that surgery and physical therapy were equally helpful for lumbar spinal stenosis, a common condition in older people that makes walking painful.

It is the latest of numerous studies to question whether Americans get too much back surgery.

It also raises questions about financial incentives. Surgery is far more expensive for insurers - an average of $24,000 in 2010, the study found, compared with a Medicare cost of $1,440 for a dozen therapy visits. But co-payments for physical therapy can make it the more expensive option for insured patients, and that was enough to discourage some patients, the study found.

The primary message for patients is that "they should exhaust nonsurgical options, which include physical therapy, before they consent to surgery," said lead author Tony Delitto, chair of Pitt's physical therapy department. The study team included William C. Welch, a spine surgeon who now works at the University of Pennsylvania.

An editorial published with the study in Monday's issue of Annals of Internal Medicine reached a similar conclusion, but added that "patient preferences should weigh heavily in the decision."

Welch and two other prominent spine surgeons in Philadelphia - Alexander Vaccaro, president and chair of the Rothman Institute, and F. Todd Wetzel, vice chair of the department of orthopedic surgery and sports medicine at Temple University - said they already encourage patients to try physical therapy first.

"This is pretty much the way I've always talked to [patients] about it," Wetzel said. "I'm definitely on the conservative side compared to many of my colleagues."

Welch said surgeons "have a moral and ethical obligation to do the least intervention to afford them the best results."

Previous studies found that patients felt better faster with surgery, although the difference between surgical and nonsurgical treatments diminished over time.

Vaccaro and Wetzel said patients with intense symptoms may want quicker results. "Surgery can get you back to a baseline level of activity faster," Vaccaro said.

Because of its design, the new study did not prove that either surgery or physical therapy were better than doing nothing, nor did it show that patients who did not improve with physical therapy were likely to do any better with surgery. However, Delitto said the conventional wisdom is that surgery is a sensible option if other treatments don't help.

"I think what most people are questioning is the prevalence of back surgery," he said. "I don't think there's any question that some people need back surgery."

Surgery for lumbar spinal stenosis is one of the fastest-growing procedures in the country, Delitto said, possibly because a wave of baby boomers is hitting retirement age. He said a 2010 study in the Journal of the American Medical Association found that 100 of every 100,000 Medicare enrollees were getting the procedure. It is more popular in some parts of the country than others - regions that have the most spine surgeons tend to have more surgery.

Some patients told study leaders they stopped physical therapy because their co-payments were too high. Delitto said typical co-payments are $25 to $35 per session. Total out-of-pocket expenses for patients can be higher for physical therapy than for surgery.

Delitto said some insurers are experimenting with lower co-payments for physical therapy.

He conceded that some patients may have preferred a quicker fix that did not involve working out. "It's hard to get a patient to say, 'I'm just too lazy,' " he said.

Lumbar spinal stenosis is a narrowing of the spinal canal at the point where the nerves exit the spine. It is very common in people over 50, doctors said, and does not always produce symptoms. Patients tell doctors they get severe cramps or feel like their legs are going to give out after they walk a couple blocks. They feel better after they sit down or lean over something like a shopping cart, opening the spinal canal.

The study, which was done between 2000 and 2007, involved patients who had chosen to have surgery. Many refused to join the trial. Eventually, 169 agreed, with 87 randomized to surgery and 82 to physical therapy. More than half the physical therapy group dropped out and had surgery instead.

The study found no significant differences in outcomes among the three groups, Delitto said. All improved about 20 points on a 68-point scale of physical functioning. The percentage who had what was considered a "successful" outcome ranged from 52 percent in physical therapy to 61 percent in surgery. Delitto said that difference could have been due to chance.

The surgical group received what's called a "decompression" procedure, which widens the canal. Those in physical therapy did aerobic exercises and back stretches. A typical stretch involved lying on the floor and pulling the knees to the chest.

While insurance companies could require patients to try physical therapy before surgery, Delitto hopes that better education will lead to better decisions. He doubts that either patients or doctors would react well to a mandate. "A lot of people don't like that," he said.

215-854-4944 @StaceyABurling