Health Plans Demand Proof That DM Saves Them Money

Can disease management vendors really show a strong return on a health plan’s investment, or are the data just too malleable to be useful?

As the chief medical officer for Maryland-based Coventry Health Care, Bernard Mansheim, M.D., has had plenty of opportunity to look into the disease management companies that routinely drop by to peddle their services.

Their basic business plan is simple: Supervision of members suffering from common chronic illness will improve their health habits and reduce your emergency care, which can run into astronomical figures.

Using prenatal case management as an example, the intuitive justification for investing in the program is “if you prevent one premature delivery with its attendant morbidity and cost, which can exceed $500,000, it is easy to justify investment in the program.”

Prodding health plans involves more than just dangling the prospect of savings. Accreditation groups are pushing for MCOs to do more than claim a commitment to improving the long-term health of the country’s walking wounded. They want proof of zeal for prevention, such as a contract with a DM firm for congestive heart failure or diabetes.

“I’ve given disease management a lot of thought,” says Mansheim. But every time he does the math on the return on investment (ROI), he comes out of it feeling even more skeptical.

No financial benefit that DM companies claim, he says, can be proven. In the often heated debate about the ROI offered by DM, Mansheim has taken a stand squarely opposed to what he calls the “disease-of-the-month club.”

“It can take years, if not decades, to see if it’s effective,” says Mansheim, who has decided that DM strategies are “significantly flawed for a whole lot of different reasons.”

Mansheim is engaged now in finding ways to create an effective internal illness-prevention campaign. He disagrees with leaders in the disease management field, many of whom have committed tens of millions of dollars in the sincere belief that they can demonstrate a hard return for the money spent for their services.

Clearly, some managed care executives don’t share Mansheim’s view. To them, DM companies not only offer them hope for improving long-term outcomes, they deliver hard and verifiable returns in the form of reduced costs.

“Everybody here goes through the exercise with eyes wide open,” asserts Gifford Boyce-Smith, director of quality management services for Blue Shield of California. Boyce-Smith has inked deals on two different DM programs, as well as developed five others in-house. After two detailed pilots on asthma were reviewed, he says, Blue Shield found “absolutely dramatic” results, reducing ER visits and other medical expenses by 40 to 50 percent.

Says Boyce-Smith: “It’s important to back this up with real numbers. I’m pushing everybody to think hard about it.”

Creating a base line

The key to determining ROI, critics and proponents both agree, lies in creating a “base line” for the costs associated with the patients who need to be covered.

It’s getting to that base line that provokes some of the strongest arguments for and against disease management.

Health care data, concedes Boyce-Smith, “is just a morass. It’s a sea of ICD-9 and CPT codes and so on.” Even if the data do come from a swamp, as he puts it, “It’s still possible to do terrific reporting in this field” — if you invest in the right kind of people who understand how to mine information through multiple report analyses. “What has to happen for the data to be understood is for the business person and the technical person to be joined at the hip.”

Mansheim has taken a look at the same statistical swamp, and couldn’t lose the feeling that trying to make sense out of the numbers was a fool’s errand, no matter to whom he was joined.

“I’ve never been secure that developing a base line can be done fairly and accurately,” says Mansheim. “Probably at the end of the day, nobody knows who the loser is.”

Joel Nitzkin, M.D., M.P.H., agrees. He’s a New Orleans-based consultant who often tries to steer MCOs away from DM companies and toward creating internal systems. “Those who advocate disease management are making claims they can’t back up,” he says, adding that the problem is that everything is constantly changing in health care — thus threatening to skew the data needed to create a base line.

Claims may be delayed months, Nitzkin says by way of example. Definitions of a disease may also change, causing the numbers of people in a health plan that may be covered by any one program to fluctuate.

One of the biggest problems, many say, is the rapid churn of membership, often about 25 percent a year. With that kind of turnover, managed care groups can’t always stay focused on long-term care.

“Not to say that there isn’t good work being done,” says Nitzkin, “but the industry as a whole is in trouble because it can’t be backed up.”

For most DM companies, the difficulty of demonstrating to health plans and other buyers of DM services that there was a base line created a huge challenge to their acceptance.

Benchmarking, agrees Al Lewis, one of the champions of the field, “is not a process that inspires confidence.”

Lewis’s guarantees

To help ensure that the health plan comes out a winner, Lewis’s Disease Management Purchasing Consortium builds guarantees into contracts, with DM companies offering to return 20 to 100 percent of their fees if they fail to meet preset goals.

There are some simple figures you can rely on, says Lewis. If you’re offering a disease management program for congestive heart failure, you can add up all the claims for CHF conditions in a plan for the year before, specify the savings that are being sought, and provide a guarantee of a significant chunk of your fees that you’ll hit that mark.

However, he adds, there are some critical points that need to be figured in. For instance:

  • Adjusting for adding new patients to the plan;
  • Health plan-specific inflation for any natural increases in costs;
  • Removing patients excluded from the contract, such as end-stage renal disease patients.
  • Separating patients over and under age 65, as older patients are typically more expensive to care for regardless of preventive measures.
  • Making sure that the extraction methodology is the same before and after. The best practice is to include every eligible person in figuring out a base line and projected savings.

That way, says Lewis, the only real risk a health plan runs is that the DM company goes bankrupt. A very cautious organization can arrange for fee insurance to protect against even that contingency.

A 15-year veteran of HMOs, Christobel Selecky, the CEO of Newport Beach, Calif.-based LifeMasters Supported SelfCare, says her company first asks for the raw numbers from health plans and then runs its own analysis.

“Data being what they are, you have to work with what you have,” she says. “It’s their data. It isn’t like they dump the data on us and we come up with a number. It’s a very collaborative process. Until both sides agree, we don’t have a base line.”

It still takes about 18 months for the programs to demonstrate a true ROI, she says. “We’re just starting to see disease management report definitive savings. We’re still in the stage when the people buying our services are the innovators.”

Selecky is also quick to acknowledge a host of problems in gathering health care numbers. Many of the institutions involved have outdated systems that are hard to work with. There isn’t any sign of that changing anytime soon.

“The MCOs are still operating on pretty thin margins and don’t have much money to invest in new systems,” says Selecky. “I’m not seeing a giant investment in information systems infrastructure being made.

That’s a problem, she adds. “For health care to improve quality, there has to be greater focus on data.”

Reluctant plans

Until the numbers begin to come in on DM, and the industry starts seeing peer-reviewed material on its effectiveness, plenty of big health plans will be reluctant to buy into her vision.

The view on Wall Street toward start-ups in disease management may have turned slightly chillier in recent months, particularly for DM companies relying on the Internet to deliver services — and success.

“A lot of the people who are paying for DM claim that there is a demonstrable savings,” says one New York-based financial analyst who specializes in health care. It’s going to take more than a few skeptics like Mansheim to change that impression.

Lewis is even harder to budge on the topic. He insists that anyone who gets a guarantee from a member of the consortium can negotiate a DM contract in good faith.

Mansheim, though, says DM advocates’ arguments are riddled with holes. “I have an inherent skepticism,” he says, “that anyone can prove that spending $100 to try to reduce morbidity costs of a patient with CHF can save $1,000 in treatments.”

Lewis agrees that’s ridiculous. “Returns average 1.6 to 1, depending on the disease. Still a 60 percent net return is pretty good, considering that it’s guaranteed and, to paraphrase Henry Kissinger, it has the additional virtue of being the right thing to do.”

However, Mansheim insists that just because you identify a population of people with a diagnosed chronic condition and reduce its medical costs in one year versus the prior year doesn’t mean the disease management program worked in creating savings. Patients may have done better entirely on their own, without any direct benefit from having a separate organization calling on them occasionally to encourage better care.

Where the “disease-of-the-month club” suffers, Mansheim says, is from its extraordinary level of fragmentation. Most disease management companies offer programs for only one or two chronic conditions, which is virtually meaningless when trying to create savings for populations that typically suffer from a high degrees of comorbidity.

“We live in a world of multiple comorbidities,” says Mansheim. “People do not have homogeneous problems. They have multiple problems.”

If a disease management company is assigned to control only one “discrete disease,” he says, the approach is so shortsighted that it is bound to fail in reducing the cost of care.

What makes sense on the medical scene is gaining traction in the financial world as well.

“You have too many people running around doing a little of this and a little of that,” says one Wall Street banker. Only when the DM industry undergoes a major consolidation will you see a few big players emerge to offer a broad cross-section of programs covering many comorbidities.

Addressed by DM industry?

Again, Lewis agrees with Mansheim’s basic point, but adds that the industry addresses it. “Everyone knows that the leading chronic disease vendors do manage comorbidities as well and have for about two years,” he says.

Mansheim, for his part, isn’t waiting for DM, or Wall Street, to sort things out.

The foremost job of the health plan, says Mansheim, “is to take care of the sickest people. It doesn’t matter if they have AIDS or cancer or MS. Those are the people we need to manage.” After that, it is appropriate to reach out to all members.

By creating a complex case management system, says Mansheim, a managed care group can begin to care for the really chronically ill, no matter what disease or diseases they have.

“With the use of computers, we have — for the first time — the opportunity to manage every one” of Coventry’s 1.5 million enrollees, he says. “By classifying each member into one of 114 categories using the Ambulatory Care Groups Case-Mix System developed at Johns Hopkins, programs can be tailored to individuals. This would account for disease management in all 14,000 ICD-9 codes, not just a select few like diabetes or asthma.”

Again, Lewis has a response. “Some of the best vendors do exactly what Bernie is saying, but they start showing returns in months instead of the years it takes for a health plan to do it.”

Still, Mansheim says that while population management has become a trendy concept, physicians have never lost their focus on the individual because you can’t easily lump people by discrete illnesses.

Adds Mansheim: “We don’t manage statistics. We manage the process of health care delivery to individuals.”