What "tiered networks" will mean to you

More health plans are selling narrow networks based on cost and quality. Here's how individual physicians have been affected so far.

One of the things that doctors disliked about '90s-style managed care were narrow networks that excluded physicians who wouldn't give hefty discounts. Those exclusive networks largely faded away as a result of the backlash against managed care. But now, as employers again seek to contain skyrocketing costs, limited networks are making a comeback. Instead of basing their choice of physicians on a willingness to accept deep discounts, however, health plans are selecting them on the basis of cost-effectiveness and, in some cases, clinical quality indicators.

While insurers are willing to sell just these limited networks to employers, the networks are usually part of a package of options that includes a broader panel as well. The combination is commonly referred to as a "tiered" network. Like tiered pharmaceutical benefits, which set different copayments for generic, brand name, and nonformulary drugs, tiered networks give patients financial incentives for choosing more cost-effective doctors and hospitals. They also cut insurance rates for employers, typically from 5 to 20 percent.

This latest health plan scheme was pioneered a few years ago on the West Coast by Premera Blue Cross. Now a number of other plans, including Aetna, have adopted tiering across the country. According to the Center for Studying Health System Change (CSHSC), nine health plans in six of the 12 communities it studied were experimenting with tiered networks in 2003. But five of the nine were tiering hospitals only.

Among those that do tier physicians, their preferred panels aren't very narrow. Premera's includes 80 percent of the doctors it contracts with; PacifiCare's Value Network embraces 70 percent. And Aetna is tiering only specialists in six fields (soon to be expanded to 12) in three markets. While about half of the contracted specialists in those fields belong to Aetna's top-rated Aexcel network, all of Aetna's primary care doctors belong.

Nevertheless, some observers think that plans might offer more selective networks down the line. Blue Cross of California has already done so, placing just one-third of its doctors in its Power Select HMO network.

Will tiered networks catch on? It's too early to tell, because most of these insurance products are fairly new. Aetna's Aexcel network, for instance, was introduced in Dallas/Fort Worth, Jacksonville, FL, and Seattle on Jan. 1, 2004 (six more markets will be included next year). So far, it has 64,000 enrollees from nine companies. Premera has moved about half of its 1.2 million members into its tiered-network plan. But they have a choice between the broad and slightly less broad networks, and it isn't clear how many have opted for the latter.

Only 5 percent of large employers offered narrow networks to their employees in 2003, and another 1 percent planned to in 2004, according to Hewitt & Associates, a benefit-consulting firm. But 28 percent were considering the move in the future.

Internist Peter R. Kongstvedt, a managed care expert and a partner in Capgemini, a large benefits consulting firm, says that employers are still hesitant to offer tiered networks, even if they save money. Others believe that the success of tiered networks depends on whether plans can supply good data to employers. "If health plans are able to identify meaningful differences in both cost and quality across providers, and set up tiers that identify those differences, I think employers could well be attracted to this kind of design," says Glen Mays, associate professor of health policy at the University of Arkansas for Medical Sciences and a co-author of the CSHSC study.

How are physicians chosen for networks?

Some plans, like Premera and Blue Cross of California, are comparing doctors' relative cost-effectiveness. Others, including Aetna and PacifiCare, are also factoring in clinical quality. But even when they do that, cost remains king.

PacifiCare, for example, culls the most efficient physicians for its Value Network before it evaluates their clinical performance. "We need to be able to offer it on a discount basis compared to a standard network," explains FP Sam Ho, PacifiCare's chief medical officer. "If you're not going to be cost efficient, there won't be much attraction for this product."

Ho observes that some doctors who score high on quality are less cost-efficient than other physicians. "If consumers want those providers, they can select them in the standard tier, but they have to pay more. For instance, they have to pay 30 percent coinsurance instead of 10 percent. Or, in the standard network, maybe they'll have to pay 10 to 15 percent more in premiums than they would in the Value Network.

"What we're trying to do is find the highest-quality, lowest-cost providers. If consumers want to choose other providers, one of the attractions of this program is that they can still choose them."

Similarly, Aetna's approach to tiering focuses on cost, tempered by the reality that consumers still prefer broad networks. According to internist Don Liss, senior medical director of Aetna, the big insurer decided to tier specialists rather than all physicians because it saw that "specialty care is the cost driver."

Since specialty care is episodic in nature, Aetna also felt this approach would disrupt care less than a system that limited access to primary care doctors, says Liss. And, because Aetna members can still choose any contracted primary physician, the tiering is invisible to most members.

Currently, the plan's Aexcel network includes cardiology, cardiothoracic surgery, gastroenterology, general surgery, ob/gyn, and orthopedics. Starting next January, ENT, neurology, neurosurgery, plastic surgery, urology, and vascular surgery will be added.

Aetna uses software to pick its networks

Aetna uses a four-step process to determine who will be in the network. First, a specialist must have at least 10 "episodes of care" in the Aetna database. Then the plan looks at clinical quality indicators. These might include a physician's 30-day readmission rate and his complication rate for hospitalized patients. Specialty-specific measures rely on how well the doctor's care conforms to specialty society guidelines. For cardiologists, for example, Aetna considers how many patients with cardiac disease are taking lipid-lowering drugs, and how many patients with congestive heart failure are on ACE inhibitors.

Some patients, of course, don't fill prescriptions—an inherent limitation of claims-based analysis. Also, many patients receive the prescriptions and other services they need from primary care doctors. Liss says Aetna is cognizant of that, but still regards specialists as the ones responsible.

To evaluate the physicians' cost-effectiveness, Aetna, like many other plans, uses a claims analysis program known as the "episode treatment grouper" or ETG. The software totals all of the medical, surgical, and pharmacy claims for a particular episode of care and compares that to the total for claims when other doctors in the same specialty manage a similar case.

As an example, Liss cites an orthopedist treating an upper arm fracture. "The software looks at the cost of the ER, the X-ray, the orthopedist's consultation, pain medication, follow-up visits, and so on." The analysis includes all cases of that type that Aetna has claims for.

Finally, Aetna looks at whether the doctors it has chosen for Aexcel are located where patients can easily reach them. If not, it will add physicians who otherwise wouldn't have made the cut. It also considers whether it has enough doctors in a specialty. For instance, nearly all cardiothoracic surgeons under contract to Aetna are included in the network.

Philosophies differ on who gets profiled

While health plans argue that physician profiling methodology has improved, there are still problems with claims analysis and with profiling in general. Claims contain many errors and are generally analyzed long after the date of service. Moreover, a doctor must have a sufficient number of cases or diagnoses of a particular type or must provide particular services often enough to make the analysis statistically valid.

Small numbers are a problem wherever a doctor's business is divided among many plans, and small practices are much harder to profile than groups are. "If you try to get to the individual doctor level in terms of either cost or quality, the data are highly variable," says cardiologist Jacque J. Sokolov, a healthcare consultant in Scottsdale, AZ.

Premera Blue Cross tiers only groups of at least 30 to 40 doctors, and only on the basis of cost-effectiveness. There are about 25 Washington groups in this category. Although about 300 practices (mostly single-specialty) have enough Premera patients and do enough dollar volume to be tiered using ETG methodology, the plan hasn't done that yet, says Richard Maturi, senior vice president of healthcare delivery systems for Premera. Small practices are automatically included in the preferred tier, he says, because Premera doesn't believe it can profile them accurately.

PacifiCare plays both sides of the street

PacifiCare is tiering both groups and physicians in small practices, using its Quality Index report card for the groups' quality component. But Sam Ho admits that profiling methods are much better for groups. That makes it easier to apply tiering in Southern California, where groups are prevalent, than elsewhere.

"As we've moved this product out to other states, we've had challenges," he admits. "We've identified five or six measures, including preventive health measures, for small practices. So instead of the 50 measures that we have in the Quality Index, we'll have five measures. It's not as robust, but it's equally credible."

When rating specialists, Ho adds, PacifiCare considers which hospitals they refer patients to, but it doesn't consider specialty-specific clinical measures. The plan has a hospital Quality Index that uses 56 measures related to complications, mortality, and other factors. It plans to use this information to tier hospitals, says Ho, and doctors will be evaluated partly on their admitting patterns.

Both PacifiCare and Premera have "pay for performance" programs that give physicians extra money for meeting quality standards. Criteria for these rewards are very similar to the quality criteria for inclusion in the Value Network, says Ho. The difference is that high-quality, high-cost doctors can get financial rewards but might not be eligible for the network.

Physicians haven't seen the impact of tiering yet

Ob/gyn Angela S. Martin of Jacksonville, FL, was very hopeful when Aetna approached her and her two partners about including them in the Aexcel network. "We were told we'd get a lot more patients and maybe more referrals," she says. In addition, the local newspaper ran a story about Aetna's program with a photo of Martin.

"But we haven't seen a big influx of patients on Aetna Aexcel," she says. "We've gotten only about five patients since the article ran in the paper. I don't know how many patients are in that tier, and I don't know whether other practices are seeing more than we are. But that seems like a small number."

Across the country in Washington state, Premera's tiering of physician groups doesn't seem to be having a big effect, either. Internist Albert W. Fisk, medical director of the 225-physician Everett Clinic, says that, even after two years, he hasn't noticed an increase in Premera patients because of the group's inclusion in the plan's narrower tier. However, he doesn't know how he could measure that.

The Everett Clinic also contracts with Aetna, and some of its specialists belong to the Aexcel network. "I'm getting early signals that we might get referrals from primary care groups that didn't used to refer to us because they're farther away," he says.

Fisk would like Premera to include quality criteria in its tiering process. But he says that the group is very happy about the insurer's pay-for-performance program, which has been awarding the Everett Clinic a quality bonus of 2 to 3 percent every year for the past few years.

 

Ken Terry. What "tiered networks" will mean to you. Medical Economics Sep. 17, 2004;81:45.