scorecardresearch Skip to main content

How health plans with high deductibles became the new normal

Peter and Maria Hoey for the Boston Globe/one-time use by Peter and Maria Hoey

Health plans with high deductibles are the fastest growing type of health insurance in the United States. More than a quarter of employers provide this kind of plan for their employees, and, last year, they made up 85 percent of all plans sold on the Affordable Care Act’s health insurance exchanges. These plans often save insurers and employers money — and not only by shifting more costs onto consumers. They actually result in less medical care used and, in total, cost less than more generous plans.

From a pure cost-containment standpoint, this is a victory. Generous health insurance insulates consumers from the true price of their care, encouraging overuse and driving up costs without meaningful benefit to health. That adds up — US health care spending reached $2.9 trillion in 2013, the highest in the world per capita. Unnecessary medical services cost everyone.

The Affordable Care Act was designed to improve health care access and, ultimately, health by expanding coverage to most Americans. Yet today’s most popular health plan designs do not go far enough to meet these goals. Stories abound of newly insured patients who face medical bills they cannot pay. At the same time, the overuse of medical services remains a threat to the sustainability of the overall health care system.


What could a smarter way forward look like? Many insurance providers and employers are beginning to experiment.

We know that cost-sharing through high deductibles or other benefit design features has a strong influence on our behavior. That idea goes back to one of the most famous health insurance experiments ever conducted.

The Rand Health Insurance Experiment began in 1971 and randomly assigned 2,750 families from across the United States to a health insurance plan. Most were assigned to one of four levels of cost sharing, paying between zero and 95 percent of all their medical expenses, much as deductibles and copays are used today. (They capped everyone’s payments at $1,000, about $5,700 in today’s dollars.) The researchers followed these families for three to five years and tracked the health services they used, measuring every prescription, doctor visit, and hospitalization.

Prior to this experiment, no one was sure whether medical care was similar to other things we buy. In other areas of life, a rise in price typically decreases how much we purchase. The Rand experiment showed that this happens in medical care as well. But, perhaps more importantly for the participants who had to pay more, their general health at the end of the study was no worse than those who got free care. The results from the Rand study fueled the growth of deductibles and other cost sharing in health insurance.

For many people, high-deductible plans work well, and they are happy to have additional money in their pocket to spend on other things. Of course, that is not always the case. In particular, studies have found people with low incomes and those who have chronic illnesses are more likely to delay or forgo health care and to report difficulty paying medical bills if they have a high deductible.


There’s another problem that cuts across the socioeconomic spectrum. It turns out we just aren’t good at discerning which medical care we need and which we don’t. Indeed, in the Rand experiment, those who had to pay more out of pocket cut back on all medical care. It wasn’t as if they were filling prescriptions for their heart disease medications and skimping on the prescription acne cream. They bought less of everything. The result was they were using too much of what didn’t improve health and not enough of those things that could.

That’s why some of the latest and most innovative insurance designs are trying to take consumer’s mistakes about care into account along with the idea that some medical services are more valuable than others. Employers and insurers are crafting benefits that do more than ever to address the misuse of health services. While no panacea, these new plans may get us closer to the holy grail of health reform — improving health while constraining costs.

The state of Connecticut is one of the employers leading the way. About four years ago, the state faced a budget shortfall and decided to try to improve employee health as a way of controlling long-term costs. It turned to insurance benefit design to influence employee behavior, particularly to encourage health services that have been shown to be valuable in preventing later disease.

Known as a value-based insurance design, the new plan required certain preventive care services, such as regular checkups, cervical cancer screening for women older than 21, and colonoscopies for everyone older than 50. For those with certain chronic conditions, the plan eliminated copays for medication to prevent progression of the disease. Those who didn’t comply with the new requirements paid $100 more per month for their premiums.


A University of Michigan team led by professor Richard Hirth and Dr. Mark Fendrick, and the author, is evaluating the effects of this change. We’re finding that employees used more cervical cancer screening, colonoscopies, and other preventive services. Those with chronic disease used services that suggested better control of their conditions; diabetics were more likely to have eye exams, for example, and those with heart disease were more likely to have cholesterol tests. We have just one year of results, but so far the plan appears successful in changing people’s behavior and steering them toward valuable medical care.

Helping consumers make healthy choices is the focus of an employer-based health insurance plan offered by Blue Cross Blue Shield of Michigan. The plan lowers deductibles and copays for medical services for enrollees who meet certain targets, such as low blood pressure and remission of depression. Additionally, it requires beneficiaries who smoke or have a high body-mass index to enroll in programs that address those issues. Those who don’t meet their health goals face higher costs.

The state of Michigan is now using these same principles in its Medicaid expansion effort. New enrollees can cut their required cost share in half if they complete a health risk assessment and agree to work on healthy behaviors like smoking cessation and getting an annual flu shot. Blue Cross says it has seen an improvement in health for enrollees; no word yet on how the state plan is going, though the state is currently seeking to expand some of its cost-sharing provisions.

California’s public employee retirement system sought to influence consumer behavior in a different way. After noticing that prices for hip and knee replacements varied fivefold with no difference in hospital quality, the plan decided to cap the amount it would pay for the surgeries. Consumers were free to get surgery at a more expensive hospital, but they would be on the hook costs that exceeded the cap. This design, known as reference-based pricing, led to the choice of lower-cost facilities; one study found those enrolled in a reference-based pricing plan paid 20 percent less for procedures than those in traditional plans. They’ve now expanded the program to cover more services including arthroscopy, colonoscopy, and elective cataract surgeries.


Another twist on the idea of using price information is websites that allow patients to look up the cost of their procedure or medication. While these price transparency tools are primarily developed by insurers or employers so that people can shop among services, another important use might be to help people decide whether they want a service at all. So far, there’s little evidence of this, but it is very possible that a person could see a price and decide that the potential benefit was not worth the cost.

To be sure, there are issues with the new approaches. They rely on consumers to understand their benefit design and evaluate different costs for various services. That kind of consumer education isn’t routine. Moreover, it’s unclear how much money these plans can actually save. Evidence for cost savings from value-based insurance design programs is mixed. Reference pricing can only be used with discrete elective procedures, which do not make up the bulk of medical costs. And, since hospitals may coalesce around the reference price, choosing a high price can improve consumers’ provider choices but actually raise overall expenses.


Though all of these benefit designs have their detractors, they all represent an effort to move beyond the blunt cost sharing in a plan that includes a uniform high deductible. For consumers, they could lead to more confusion balanced against potentially more coverage for valuable health services. It will surely demand a better understanding of how health insurance works. Whether we’re willing and able to accomplish that is still an open question.

Betsy Q. Cliff is a doctoral student in health policy and economics at the University of Michigan’s School of Public Health. She can be reached at