The University of Michigan
555 South Forest Street
Third Floor
Ann Arbor, MI 48104-2531

T 734-936-9842
F 734-998-6341



Consumer-driven health care has been touted as a way to make the consumer a smarter health care shopper and a tool to control health care spending. The insurance products that drive this new behavior among consumers are high-deductible health plans, marketed as consumer-directed health plans (CDHP). Although their presence is growing, there remain a lot of unanswered questions about their utility.

Interview with Roger Feldman, Blue Cross Professor of Health Insurance and Professor of Economics at the University of Minnesota

  Roger Feldman, Blue Cross Professor of Health Insurance and Professor of Economics at the University of Minnesota, has spent his career researching the organization, financing and delivery of health care focusing on the role of health insurance. A former Marshall Scholar at the London School of Economics, Feldman has served on the senior staff of the President's Council of Economic Advisers, and directed one of four national research centers sponsored by the Centers for Medicare and Medicaid Services. Currently, he is evaluating several "consumer-driven" health plans. Feldman talked to ERIU about the growth of consumer directed health plans, who are buying these policies, how they are affecting health care purchasing decisions, and whether they hold promise for those without health insurance.

What does a typical high-deductible health plan look like today?

FELDMAN: Two versions exist. The classic consumer-driven model is exemplified by Definity Health's Health Reimbursement Account (HRA). The employer contributes money to an HRA that the worker selects. The HRA rolls over at year's end and the worker can use it to reduce future medical costs. Employees don't own them and they're not portable. The account is coupled with an insurance policy providing full coverage for preventive care, and coverage above an annual deductible. Some HRA designs have coinsurance above the deductible. You can pick from a large variety of coverage levels, with deductibles ranging from $1,000-$5,000 for an individual and $2,000 to $10,000 for a family. The newer versions, health savings accounts (HSA) are similar, except the consumer owns the account and it is portable. Employer and employee can both contribute but, the employee owns it and can keep it when they leave the company. They also have superior tax advantages.

You don't need to introduce new types of health plans specifically to achieve higher cost sharing. You can put higher cost sharing into the traditional plans, like PPOs. In fact, the trend in cost sharing in traditional health plans is fairly comparable to what we see in these CDHPs.

What is the penetration of these products?

Feldman : We don't have a complete picture. Take up rates in firms that offer CDHPs are highly variable, ranging from near-zero percent to 80 percent. On average, we are seeing take-up rates in the teens. Since 2004, the take-up rate of HSAs among individuals appears to be quite strong, at least 1.5 million and headed toward 3 million or more are being sold in the individual market. That's pretty significant because a lot of people who bought HSAs in the individual market would otherwise have chosen to forego coverage altogether.

Is that expected to increase significantly?

Feldman : I'd expect the overall penetration rate among the group market to stay low, as long as employers continue their current methods of contributing to health plan premiums. Many employers contribute an equal percentage to the premiums of all the plans they offer. As long as employers subsidize all choices on the margin, there is no real advantage to picking a lower cost plan. Employees will gravitate towards the higher cost PPO designs.

Who is enrolling in high-deductible health plans? Healthier individuals and higher-income individuals?

FELDMAN: In the group market, we clearly know these plans are favored by high-income workers. That finding is supported by several peer-reviewed studies. The plans also are chosen by highly educated workers. We do not know whether they are chosen by people with low or high medical risks.

Are CDHP enrollees more likely to put off needed care?

FELDMAN: There's no evidence that they put off needed care. There is evidence that they spend a lot more money on discretionary hospital care. Most of this money is paid by the insurance plan but some of it is paid from enrollee health care accounts and by enrollees out-of-pocket. In terms of the drug piece, data here show that pharmacy costs were down initially even though volume didn't decrease. That suggests greater use of mail-order drugs in CDHPs.

Do they tend to use more generics?

FELDMAN: Actually, we find the opposite. Enrollees of CDHPs tend to use more brand-name drugs. Why? Non-CDHPs typically use a tiered pricing structure, which steers people towards generics. The CDHP design doesn't favor generics or brands; it is neutral. So, CDHP enrollees tend to use more brand-name drugs. The reason why CDHP enrollees can use more brand-name drugs but spend less is that they use the plan's 'tools' to find low prices and buy drugs by mail order, where prescription refills generally are less expensive than at the retail pharmacy.

So, what does the evidence suggest in terms of people in high deductible health plans altering their habits?

FELDMAN: There simply hasn't been enough peer-reviewed research to answer that question. The answer is going to depend on the design of the insurance plan. How big is the account, how big is the gap between the account and the major medical policy, and does the major medical policy have coinsurance when it kicks in?

One of the knocks against CDHP is that they are really limited in controlling health care costs, as the big expenses are incurred by insurers. Does the research show that?

FELDMAN: It is true that a large share of health care expenses is accounted for by a small percentage of the population. It is also true that the quickest way to run up a large bill is a hospital admission or a hospital outpatient surgery. That fact, however, does not mean that cost sharing won't control health care expenses. The RAND Health Insurance Experiment showed that a high-deductible health plan had a very significant effect on total spending, through reduced hospital admissions, just like a traditional HMO controls your access to high-cost hospital admissions and high-cost technology.

So is it fair to say that HMOs and PPOS can achieve some of the same objectives as a high-deductible plan?

FELDMAN: That's right. You've got to bite one bullet or the other: it's either consumer cost sharing, or supply side controls, like we see in an HMO. You do have a choice in which bullet you want to bite.

But the added tax advantages lie largely with the CDHPs?

FELDMAN: Under both HRA and HAS designs, contributions to the spending account get a tax break since they are made with tax-free dollars. The government is attempting to put out-of-pocket health expenses on the same tax footing as insured expenses, which have enjoyed a tax subsidy since the 1950s. However, a tax subsidy is not neutral to all forms of individual insurance; it is specifically tied to the purchase of a high deductible health insurance policy. Instead of letting all out-of-pocket expense be tax favored, to get this special advantage, you have to couple it with specific deductibles.

What suggestion would do you have for health care policymakers?

FELDMAN: My first suggestion to policy makers would be to either cap or get rid of the tax subsidy all together. But, that isn't going to happen. So, my second suggestion is to put insured expenses and out-of pocket expenses on the same tax footing.

What would this approach achieve?

FELDMAN: The policy of subsidizing insured expenses leads to people buying insurance policies that are overly generous on the margin, policies that don't have enough cost-sharing to control the ordinary type of expenses. So, if you apply the tax subsidy equally to out-of-pocket expenses, you would give people the right incentive to purchase policies with a better mixture of insurance and out-of-pocket spending. Specifically, they might decide to have higher coinsurance rates on discretionary services, or they might decide to have higher deductibles, but they would make these decisions without being influenced by the tax system.

So catastrophic or bare-bones like policies may be the way to go in order to increase health insurance coverage while making it more affordable for individuals as well as employers?

FELDMAN: Yes, absolutely. The bottom line is that high deductible health insurance has made insurance affordable to a much broader group. I am not going to argue that it is affordable to everyone. That is clearly still not the case. There is going to be health screening in the individual market. That is a whole other question.

Where do you see this consumer-driven health movement going five years from now?

FELDMAN: As long as employers offer a choice of plans and subsidize the more expensive ones, I don't see pure consumer-driven health plans gaining a big foothold in employment-based insurance. I see them gaining a big foothold in the individual market, especially if the subsidy for premiums is enacted.

When it comes to efforts to cover the uninsured with high deductible insurance or via traditional plans with higher cost sharing, what advice would you have for policymakers?

FELDMAN: We've done some work for the Assistant Secretary of HHS in the form of simulating the effect of additional subsidies for the individual market. Our simulations suggest that additional subsidies could substantially increase coverage for individuals, but they need to be set high enough. The design of the subsidy is important. A refundable tax credit of up to $1,000 per adult and $500 per child in low-income families could increase HSA take-up by 3.8 million people and reduce the number of uninsured by 2.9 million. More generous subsidies would increase HSA take-up even more but, if they were widely available to anyone who buys a policy in the individual market, this could be pretty expensive.

Back to top

Funded by The Robert Wood Johnson Foundation, ERIU is a five-year program shedding new light on the causes and consequences of lack of coverage, and the crucial role that health insurance plays in shaping the U.S. labor market.