Vantage Point: Selecting a health plan: Is Definity our destiny?

BY Alain C. Enthoven and Sara J. Singer

Stanford will be offering Definity Health Plan as one of six options for health coverage to its employees this year. Definity is one of a new class of "consumer directed health plans" being offered by several companies. The common theme among them is a high deductible coupled with an employer-provided "personal care account" (PCA), out of which an employee can pay most of these expenses.

Definity offers some good ideas and some bad ideas. Fortunately, Stanford's benefits office has modified the basic concept to mitigate some of the consequences of the bad ideas.

The best thing about Definity is that it seeks to educate consumers about the cost of health services. This is an urgent necessity. Employed insured people have developed a cost-unconscious sense of entitlement to any service that might possibly benefit them, no matter how weak the evidence. Rapidly escalating health insurance premiums cannot be contained until employees realize how much their employers must take out of their total compensation for health insurance. This understanding will come only from a better recognition of the total cost of medical services.

Under the Definity plan, a Stanford employee will face a $1,500 annual deductible ($3,000 for a family) and must pay for services consumed up to that amount. The blow is softened by an employer-provided PCA of $1,000 ($2,000 for a family), refreshed by another $1,000 each year. For example, in year two, an individual who had no medical expenses in year one would have $2,000 in his or her account. Stanford has elected to make this PCA usable only for otherwise covered services, not for any other health item like glasses or hearing aids.

The aim of the high deductible is to cause consumers to seek information about the health services they consume. Definity hopes that by facing cost-conscious decisions for the first $1,500, consumers will develop cost-effective use patterns that won't change even after exceeding the deductible. Definity also believes that better educated patients will change provider behavior.

Definity will supply consumers with information tools to help them make informed purchasing decisions. How much good this will do is uncertain. Definity will provide cost and quality information about products and providers. However, it will not identify the doctor who charges less for a visit but who may cost more for diagnosis and treatment of the whole episode of illness.

In our view, while there may be some benefit from training consumers to be wise purchasers, the basic concept of relying on high deductibles to create cost consciousness is flawed. First, health expenditures are concentrated: 20 percent of consumers account for 80 percent of expenditures in a given year, and they will have exceeded their deductibles. So the cost-restraining effect of the deductible is likely to have little effect on decisions regarding 80 percent of the costs. Some patients will know from day one that they will exceed their deductible, so the deductible will provide no cost-restraining incentive for them.

Definity says they will apply "case management" and "disease management" to mitigate costs above the deductible. These efforts aim to educate patients and support them in complying with treatment plans by providing opportunities to ask questions and get support to manage their illnesses. We have some doubts about the efficacy and cost-saving potential of such "third party" case management, as opposed to case management by doctors built into their own practice patterns. However, if significant effort by Definity to manage high-cost patients is effective, it could reduce cost, improve quality and exert pressure on Stanford's other health plans to improve.

Stanford will require Definity patients to pay 10 percent of the bill for in-network services above the deductible, so they will continue to be somewhat cost sensitive until they reach their out-of-pocket maximum ($3,000 for an individual/$6,000 for a family). There is no out-of-pocket payment for services totaling more than $16,500.

Stanford's basic policy of responsible choice among several carriers and delivery systems promotes delivery system reorganization. Further, by delegating financial risk to the delivery systems with which they contract, Stanford's HMOs (health maintenance organizations) encourage providers to offer high-quality care at low cost. Unfortunately, Definity will probably not motivate physician-led reform of health care delivery and reengineering of care processes. Definity pays fee for service, so providers have little incentive to find ways to reduce patients' medical needs.

Second, high deductibles lower premiums by shifting costs to patients who use services, i.e., from the healthy to the sick. This raises a question of fairness. And there is some risk that some people who don't expect to need a doctor in the coming year will choose the high deductible/low premium and stay with it until they anticipate significant medical needs. When it is time to have a baby or a joint replacement, if there isn't enough accumulated reserve in the PCA, an employee could switch to the first dollar plan. Stanford wisely shut the door on this opportunistic switching strategy by providing that, when a consumer switches from Definity ­ short of retirement ­ he must leave behind the accumulation in his account. He may not take it in cash; at retirement he can apply it to the cost of retiree medical premiums. This will take some of the reward out of opportunistic switching to game the system.

There is another answer to this risk selection problem, a medical-econometric process known as "risk adjustment of premiums." The electronic (but de-identified) medical records of patients with each carrier are analyzed and translated into predicted medical costs. Prospective compensatory payments are made from the carriers enrolling lower risks to carriers enrolling higher risks. This is done through the employer. The federal Medicare program is doing this for Medicare HMOs. We hope and trust Stanford will consider this carefully and phase in risk adjustment over the next few years. It is the best way to take risk selection out of the process and to create confidence in the integrity of our competitive model. Now that it is feasible, risk adjustment would be appropriate even if Stanford had not introduced the Definity product.

The university must be able to end the participation of health plans that are not serving us well. What will happen to the PCA accumulations in a few years if Stanford wants to drop Definity? Stanford views the PCA as similar to paid sick time. That is, an amount is put into a person's account for use in a specific circumstance. In the case of sick pay, the account provides income during disability, and in the case of Definity, the account pays for necessary medical care while the patient is covered by Definity. While Stanford does not view money in these accounts as owed to the person, Stanford will need to make this view explicit. One consequence of a "you can't take it with you" policy is that employees will be less likely to treat the PCA as their own money, thus limiting the cost-consciousness-raising effect of the high deductible to the $500 for which an employee is truly exposed.

Third, varying estimates suggest that 60 to 75 percent of health care costs are now associated with people with chronic conditions. In many cases, these patients' costly acute medical conditions can be prevented or postponed by effective disease management, i.e., continued coordinated interventions in the home or outpatient setting, often by allied health professionals and not physicians. High deductibles, which can be effective in deterring unnecessary elective care, can also deter patients with chronic conditions from obtaining these useful and necessary services. Definity's PCA, its emphasis on disease management and an explicit exemption from the deductible of many preventive care services are all aimed at avoiding this pitfall. Stanford's current policy for preventive care through Definity exempts from the deductible annual services recommended by the U.S. Preventive Services Task Force and other national guidelines. These do not currently include preventive services necessary for patients with chronic conditions and make it more difficult for Definity's disease management efforts to succeed.

Stanford is "self insuring" the costs of the Definity plan. That is, Stanford bears the risk if the estimated premiums turn out to be underestimates. We would feel better about introducing a new, largely untested plan if Definity bore the risk. We reviewed some examples that led us to be quite concerned that Definity is priced unrealistically low ­ that the Stanford budget may be burdened by a significant cost overrun.

Roughly half of insured people don't go to the doctor at all in a given year. Under the PPO (preferred provider option), Stanford pays nothing on their behalf. Under Definity, Stanford puts $1,000 in their PCAs. While Stanford views this money as Stanford's until employees spend it, the pricing models assumed it was spent. How is Stanford going to make up for this?

One answer might be that Definity gets a favorable selection of health risks. Experience shows that in multiple-choice models like ours, the people who switch are often healthier than average ­ for a year or two. However, there are believable scenarios in which HMO members who are sick will switch to Definity to get access to its wider selection of doctors, including Stanford faculty. So the risk selection could fall either way. The actual cost to Stanford is uncertain as long as Definity does not guarantee the cost.

Another answer, offered by Stanford's consultants, is that the Definity model will change behavior so much that Definity patients will spend enough less that their estimates will be accurate. This is conceivable, but speculative. It will probably take two or three years' experience to find out. If these savings do not materialize, Definity premiums will have to rise, potentially significantly. Employees who convert to Definity may be badly disappointed.

Definity's premium relative to the PPO raises a special concern. While the out-of-pocket exposure for individuals in Definity is $200 more than in the PPO (up to $700 more for families), the annualized employee premium is over $400 less ($1,250 less for families). This suggests that, even without changing health care consumption behavior, employees currently in the PPO would be better off financially in Definity. Substantial switching from the PPO to Definity could create instability if Definity's relative price rises in the future.

It is understandable that Stanford's benefits office should be seeking out innovation that might improve consumer satisfaction and mitigate expenditure growth. Also, they need options that permit those who so desire to obtain covered services from Stanford Hospital and Clinics. But this Definity plan and concept will need careful watching.

Alain C. Enthoven, the Marriner S. Eccles Professor of Public and Private Management, Emeritus, served as the chairman of Stanford's Committee on Faculty/Staff Benefits in the late 1980s and early 1990s and has written extensively on "managed competition" among health insurance plans. He also serves as a consultant to Kaiser Permanente. Sara Singer is executive director of the Center for Health Policy and a senior research scholar and lecturer at Stanford. She currently serves as a member of the Committee on Faculty/Staff Human Resources. The authors would like to thank Sue Cunningham for valuable criticisms of and suggestions to the initial draft of this article.