Want to help keep health bills down? Maybe it's time to go shopping for that next colonoscopy.

Millions of Americans undergo the screening procedure for colon cancer each year, and costs can range from $500 to more than $8,000.

But recent experiences in California suggest that insurers and large employers could save a bundle by picking a midrange price and letting patients put the sum toward a colonoscopy from the health care provider of their choice. The catch is, patients would have to pay the difference if they pick a center that charges more than what's called the "reference price" for the procedure.

"It leads to pretty substantial changes in consumer behavior," said Christopher Whaley, a research economist at the University of California-Berkeley. "And it leads to a pretty large savings."

Whaley and colleagues have projected that such a program could shave $107, or about 7 percent, from the cost of each colonoscopy in Minneapolis, while saving $95 million per year nationally just for three large health insurance companies included in a new study.

The research published this week is the latest example of how health economists are using price data from a group of large national health insurers, including Minnetonka-based UnitedHealthcare, to study potential remedies to tame health care inflation.

Even so, not everyone is convinced about the potential for reference-based pricing, which the California Public Employees' Retirement System (CalPERS) has used for several years for colonoscopies, hip replacements and knee replacements.

In Minnesota, state law bars health care providers from billing patients for the balance of what a health plan won't pay, said Matt Anderson, senior vice president of policy and strategy at the Minnesota Hospital Association. So a hospital or surgery center would either have to match the reference price or drop out of the health plan's network. "It could limit access to care for some Minnesotans," Anderson said.

Reference pricing for prescription drugs has generated large savings for many years in certain European countries, so the concept is "innovative," said Stefan Gildemeister, state health economist with the Minnesota Department of Health.

But without much better information about prices and coverage rules, patients could get stuck with large and surprising bills simply for not understanding how the programs work, said Karen Pollitz, a senior fellow with the Kaiser Family Foundation.

The pricing program seems like a "great idea," says Dr. Randy Saliares with CentraCare Gastroenterology Clinic in St. Cloud, but it raises a key question: How do insurers factor the quality of different care providers?

Dr. Douglas Wood, medical director of the Center for Innovation at Mayo Clinic, questioned whether the program is wrongly focused on prices when the goal should be eliminating procedures that aren't needed. "Even a lower price for hip replacement, if it was not necessary, is still too high and represents wasteful spending," Wood wrote in an e-mail. "Reference pricing for colonoscopy … won't help us reduce costs."

California's experience

About 1.6 million people have health insurance through CalPERS, making it one of the nation's largest purchasers of health insurance, said Whaley, who led the new study.

When the group started looking for new ways to control costs, he said, it decided to try something different from boosting deductibles, in hopes that people would shop around for the best deals. Reference pricing seemed like a way to get patients to shop for procedures where there weren't significant differences in patient outcomes but price variations were significant.

CalPERS found in 2011 that colonoscopies performed in hospital outpatient departments in California ranged from $552 to $8,883, with a median price of $2,273. Those performed in surgery centers ranged from $500 to $6,003, with the midpoint being $878.

The group decided to apply a reference price for the procedure at $1,500 if patients sought care in hospital outpatient departments. Patients who went to a surgery center or low-cost hospital would only pay standard co-payments and deductibles, while patients at high-cost hospitals would face standard cost-sharing plus the portion of the bill above $1,500.

The program applied to a subset of about 450,000 people covered by CalPERS.

The two-year savings was $7 million — both because patients opted for lower-cost providers and because higher-priced hospitals lowered their prices.

In the new study, researchers took data from a nonprofit called the Health Care Cost Institute to project potential savings nationwide and in particular regions of the country. HCCI is backed by the insurance industry, and has removed identification information on medical claims from about 50 million individuals with coverage from national insurers Aetna, Humana and UnitedHealthcare.

Estimates and caveats

Researchers estimated that the savings would vary by location, with projected price cuts as high as $281 in Manhattan and $289 in Chicago. Prices wouldn't decline in some places, either because they're already low, or because all providers in the market charge high prices.

Reference price programs can only work, researchers noted, with "shoppable" procedures that can be scheduled in advance, and where there's no apparent link between lower prices and lower quality.

"We estimate that the same program that CalPERS implemented would be successful in lowering health care spending in nearly all markets in the United States," researchers concluded.

Christopher Snowbeck • 612-673-4744

Twitter: @chrissnowbeck