Studies by Cigna and Humana challenge some long-held assumptions about high-deductible plans
When the University of Minnesota offered Steve Parente, PhD, the choice between a traditional health plan and a high-deductible plan linked to a health savings account, he went the HSA route.
It wasn't an easy choice. A family member has a chronic illness, and as a health care economist, Parente was painfully aware of the routine medical costs that the family would be expected to pay out of the family budget.
"This is probably not a slam-dunk rational thing to do," he says. "It would be more rational to go to one of the plans that covered more of the expenses. We have a PPO option with low deductibles, and the premium wasn't that much more."
However, Parente was willing to take the risk so he could see for himself just what kind of challenges people face in a high-deductible plan. Could he figure out how to shop for health care? Would the cost prompt the Parentes to deny themselves the therapies they needed? And was the upside in accumulating tax-free savings enough to keep him in a health savings account?
While Parente, associate professor in the department of finance in the Carlson School of Management at the University of Minnesota, was putting his own money on the line, he also continued his academic studies to see how people behaved in a high-deductible account. His own moment of truth came when he was looking at paying for a trip to the ER. He called a nurse hotline and then headed for the door.
"If you're relatively wealthy and know that if you don't go, you're risking your family's livelihood if something is wrong, you generally go," says the economist.
Of course, if you're not relatively wealthy, cost could pose a big problem, he adds. And that could make the decision a lot harder.
Parente's field of study has grown hot as health plans have rolled out new high-deductible health plans. The underlying rationale has been that consumers, when presented with the prospect of paying for their own care, would carefully consider what they were getting for their money.
Cutting out unnecessary expenses would help rein in galloping health care inflation. But if the plan design exposed people to too much financial risk, the great fear was that they would go without needed care — and wind up in much worse physical condition that would demand far higher costs to treat.
For a number of analysts, the data already clearly suggest that the greater the cost of care, the more likely that people will go without. New research at Rand has found a direct link between drug costs and avoidance of therapies, raising the prospect that outcomes have worsened and overall costs have risen.
But now health plans have been gathering data of their own from thousands of members who faced exactly those decisions, and they've come up with a significantly different tale.
Until recently, most of the hard data on the subject dated back a full generation.
Rand's long shadow
The Rand Health Insurance Experiment concluded after inspecting eight years' worth of data (1974–1982) that people reduced their spending on necessary as well as unnecessary care when they were responsible for paying the tab.
But Rand also found in this randomized study that the decision to opt out came without a negative effect on people's health.
More recently, a team of researchers at Rand reviewed 132 studies of cost-containment measures — such as higher drug copayments — and found a direct link between higher out-of-pocket costs and lower utilization.
Every 10 percent increase in cost-sharing led to a 2 percent to 6 percent drop in drug spending. In other words: The more that members were required to pay for drugs, the less they used, with falling adherence rates and a rising likelihood that they would discontinue usage.
But co-author Geoffrey Joyce, a senior economist at Rand, says it's hard to find a link between lower utilization and worse outcomes: "A lot of drugs improve quality of life, but they don't keep people out of the hospital."
But when you narrow the focus to serious chronic ailments like diabetes, heart disease, asthma, and schizophrenia, there is direct evidence that as utilization drops, outcomes worsen. An increase of cost-sharing in those categories, said the report published in JAMA in early July, triggered more hospitalizations and greater use of emergency room visits.
For insurers, it's a sign that they are in turn picking up a much higher tab for at least some of their members.
"You can't just push cost-sharing across the board," says Joyce. "You have to be smart about it."
For certain disease categories, he adds, the smarter route would cut drug costs of patients in order to lower overall expense. And that's exactly what companies like Pitney Bowes have been doing.
Melinda Beeuwkes Buntin, PhD, a health care economist at Rand Health, has also been attempting to determine whether anything has changed over a generation.
The most positive initial conclusion, says Buntin, is that "people are likely to substitute a generic for a brand drug. That would be a wise way to save money."
But after that first simple step, says Buntin, the choices grow more complex, and wisdom harder to find. Consumers still don't have the decision-support programs that are needed to understand how to separate the necessary from the unnecessary.
"I feel safe in saying that the tools that people need to be discerning consumers are not all ready for prime time," says Buntin. "The information out there to compare providers on price and quality is sparse, especially when you get outside the hospital setting."
On the basis of data available today, Buntin says that when people are forced to pay more for care, they will use less. She estimates that if all nonelderly Americans migrated from low-deductible health plans to options that require greater out-of-pocket spending, there would be a one-time cost reduction of 4 percent to 15 percent. But what she and others at Rand want to continue to probe is whether the decision to reduce spending is leading people to reject needed care. And that's not so easy to determine.
A number of things have changed over the past 25 years, notes Buntin, not the least of which is that many of the consumer-directed plans that are coming out are designed to make sure that they cover all or a significant part of the cost of preventive medicine, like annual checkups. And when employers contribute to employees' health savings accounts, the effect on economizing can be blunted.
"These accounts can have different effects on different types of people," adds Buntin, but one concern remains constant: "People with cash flow constraints might cut back on care more than those without constraints."
Income and choices
Other health care economists have studied the link between income and health care choices.
It's clear that the less you make, or the less healthy you are, the more likely you are to go without recommended care, says Sara Collins, PhD., assistant vice president for the Program on the Future of Health Insurance at the Commonwealth Fund.
"We find in the EBRI/Commonwealth Fund Consumerism in Health Care Survey and the fund's Biennial Health Insurance Survey that if you look at adults with high deductibles of a thousand dollars or more, they are more likely to say they didn't fill prescriptions, skipped a test, didn't go to a doctor because of cost," says Collins. "Larger shares of people with lower incomes and high deductibles report such problems.
"In addition, people with health problems are more likely to have higher costs in high deductible plans as a share of income," adds Collins, "and they are more likely to say they didn't get care when sick."
Other research has found similar evidence about the effect of higher cost-sharing on health care use, she adds. John Hsu, MD, at Kaiser Permanente Northern California, has reported similar results in a study published in the New England Journal of Medicine, determining that a cap on prescription drug benefits reduced Medicare beneficiaries' purchase of drugs and led to adverse health consequences.
With health care costs rising faster than wages, employers will continue to be under pressure to share more costs with their employees, says Collins, with the risk that we will see more of the same results.
Giving credit to consumers
Parente isn't the only person in the health care field to put himself on the front lines of the high-deductible trend.
Beth Bierbower, the vice president for product innovation at Humana, knew she was facing some significant out-of-pocket costs when the insurer switched its employees to a high-deductible plan. As someone who has long grappled with severe migraines, she knew that the out-of-pocket expenses for what her doctor recommended for treating a full-blown episode could be high.
So she changed course. The doctor's prescription for a brand name drug, which would have cost her $95 a month, was tossed. In its place, she buys boxes of Excedrin for a fraction of that amount, and when she feels the inkling of a migraine, she makes sure to take two to prevent a full-blown attack.
"If I catch it early enough," says Bierbower, "that will cover it."
Looking at the prescription data on 1,600 Humana employees who are responsible for out-of-pocket costs that range from $1,250 to $3,000 for a single person and twice that for a family, Bierbower says that members are not letting costs stop them from getting the therapies they need.
The company studied 14 drug categories before and after high deductibles were applied and calculated that utilization patterns stayed the same for nine, increased for two, and decreased slightly for three, with no statistically significant difference overall.
Bierbower has, of course, seen the Commonwealth Fund reports and heard plenty about the original Rand study.
Her response: "I'd give more credit to consumers."
Maybe price-conscious consumers are less likely to go to the ER, she notes, but they are also more likely to call the nurse hotline. Consumers are going to the doctor more frequently (see graph below) — and paying the $100 for the visit — so they can stay out of the hospital, she says. And they are much more likely to demand generic alternatives to the branded prescription drugs that are often prescribed.
Just because you find a cheaper alternative, Bierbower notes, doesn't mean it isn't appropriate care.
Poring over the "before" and "after" data on 38,200 members who joined consumer-directed plans, Cigna HealthCare found that the change in coverage did significantly reduce the amount of money the company spent on care.
"We found that HRA and HSA members had 16 percent lower costs than members of traditional health plans, says Joe Mondy, an assistant vice president at Cigna.
Cigna also found that people in its Cigna Choice plan increased the amount of preventive care that they requested compared to members in traditional plans, and that they continued to get the care that they needed.
The company did reduce expenses, says Mondy, but in a responsible way. Examining the CPT codes for each member's care, analysts concluded that members were able to make decisions on the kind of care they needed, opting for less expensive therapies. One example: Choosing drugs to treat a cardiac illness instead of choosing open-heart surgery, a decision that offers comparable outcomes at dramatically different rates.
"These plans are designed to take costs out," says Mondy, without reducing members' access to needed care. Anyone using the system would see in an instant that a generic drug is going to cost a fraction of a branded therapy.
"An MRI is almost a commodity," he says, "but you can see costs ranging from $600 to $2,000 in your town, depending on the facility.
"We have a star-based rating system on cost efficiency and outcome," says Mondy. "Members can drill down to some very specific cost comparisons. When you click on a hospital for a procedure, you get the cost range and how much you would pay out of pocket. That tool was introduced last year and we later extended the types of procedures you could look at."
Mondy adds that "the first thing they look at is outcome," based on a reasonable notion: "I want to come out alive."
Cigna's approach to costs is to expose members to some risk, but without simply shifting expenses from employers to members. That way, the members can reduce their own costs alongside their employer's.
"You see a lot about cost shifting being a primary problem of CDHP plans," says Mondy. "We do actuarial modeling to show the employer how he can reduce costs without shifting costs.
"People are saying that there isn't enough information out there, and they're right," says Mondy. Costs for specialists still need to be added, for example. But Cigna and other health plans are getting that information now. And in the not-too-distant future, he adds, "you'll get more information on quality and cost than in any other industry."
The big responsibility for health plans, says Humana's Bierbower, is to do everything they can to give consumers the information they need to make the right decisions.
"We know that you can educate consumers," says Bierbower. "Nine times out of 10, they will make the right decision."
In this field, many conversations about the relationship between out-of-pocket costs and refusing care quickly point back to the original Rand study.
"I think the 1982 Rand Health Insurance Experiment is relevant to today," says Steffie Woolhandler, MD, who led a recent study of cost sharing by Harvard Medical School researchers. "We do know with all the certainty that we can have with science that high deductibles will reduce needed care, as well as less necessary care."
That poses a particularly painful challenge for women, says Woolhandler. She found in reviewing the 2003 federal Medical Expenditure Panel Survey of 33,000 Americans that men under 45 in a high-deductible plan had median costs of $500 compared to $1,200 for women. Mammograms, new cancer vaccines, pap tests, and reproductivity-related medical care all put an extra financial burden on women.
Anyone trying to understand how patients decide on the care they get when confronted by costs has to start with an understanding that physicians control the decisions on 90 percent of all costs, says Mai Pham, MD, a senior researcher at the nonpartisan Center for Studying Health System Change, who led a recent study on the topic.
"Whether increased cost sharing can effectively control health care spending depends on whether patients and physicians can together consider costs during clinical decision making," she says.
As for generic drugs, says Pham, doctors are now typically quick to recognize that they can save money. But when you move to a review of various tests and other kinds of care, there are many challenges. First, there is an absence of accurate, up-to-date information on a patient's insurance coverage.
"Doctors can have all the good intentions they want, but they won't be able to execute it," Pham adds. "Another hurdle is getting patients to talk at all. It's an awkward dance. How much are you willing to pay for this versus that?
"Physicians don't generally know what prices are floating out there. There's not such great information on formularies, and that's the lowest barrier that insurers would have to overcome. Take electronic prescribing, where you would think you'd have data. It turns out often to be incomplete, inaccurate, and not updated."
Should I or shouldn't I?
"Then you run into this pernicious problem where the conversation can't take place during the visit," adds Pham. "There's no good information until patients go to get the test and find out what the copayment is. And then they have to decide: Should I or shouldn't I? And then they are on their own."
Once he went self-paid, Parente knew he'd have that escrow account built up by the end of the year. And as the family added to the HSA, it became a little like a mortgage, he says. "We were paying more up front and getting back-end value."
Even a catastrophic illness in a high-deductible plan isn't the end of the world, Parente adds.
"Ten thousand dollars is generally the worst case scenario," says Parente. "With an income of $40,000, that would be pretty high. But the alternative is not getting care, and $10,000 in debt is recoverable." That's not too high if you consider that even low-income families can still pay $6,000 for a car.
As more and more people are being forced to pay, says Parente, more risk could also have a huge effect on bad personal habits, like eating too much.
"As the world goes more high-deductible, if that's what we can afford as a society, the word might get out more that this isn't the same system we had 20 years ago. Technology can do more for you. The bad news is that there's a price for that."