This Article Abstract Submit a response to this article Services E-mail this article to a friend Alert me to new issues of the journal

I N T E R V I E W :
R O W E  &  R O B I N S O N
 
13 December 2005  

Consumer-Directed Health Insurance:
The Next Generation

The chairman and CEO of Aetna describes his
company’s experience as a “first mover” on the
health insurance landscape.


by James C. Robinson


ABSTRACT:

Aetna has led the transition of the commercial health insurance industry from a focus on managed care to a focus on “consumer-driven” products that offer more information while demanding higher cost sharing from enrollees. But most health care costs are incurred by very sick patients who have spent beyond their cost-sharing limits and are not directly affected by these financial incentives. Aetna CEO Jack Rowe discusses his firm’s initiatives to moderate the cost of care for ill as well as healthy patients, including providing information on provider prices, contracting for narrow physician specialty networks, and promoting disease management and case management programs.

James Robinson: Aetna has positioned itself as a first mover in the trend toward consumer-driven health insurance products. We are interested in your experience to date with this focus and what you are thinking with respect to follow-on programs. A second generation of consumerism might extend the current focus on primary care services to encompass specialized and expensive services used by very sick patients who have passed their annual cost-sharing limits. So, first of all, how central to your strategy are Aetna’s HealthFund products, which combine a high-deductible PPO [preferred provider organization] with a savings or reimbursement account to help pay for services below the deductible?

John Rowe: The importance of this suite of products is underestimated by the proportion of our membership that is in it. Consumer-directed products, by themselves, give employees more financial responsibility but do not necessarily make them more prudent purchasers of health care services. Consumers are motivated, but they need more information. These products have accelerated the development within Aetna of a variety of approaches to enhance transparency. Examples include bringing the payments that a physician gets for services into focus for members, which we recently announced in Cincinnati; providing members with more information about pharmaceutical options, such as generic versus brand-name products; and developing information about the performance of physicians and hospitals. HealthFund also has been an important factor in our growth, because employers who are interested in offering cutting-edge products may come to Aetna and contract with us, even though only a small portion of their employees will actually choose a consumer-directed product. We will get the other employees even if they are not choosing that product. We are winning accounts in the marketplace with our profile of being innovative.

Robinson: What is the current enrollment in the health savings account (HSA) and health reimbursement arrangement (HRA) HealthFund products?

Rowe: We’re in the range of 400,000–500,000 members. I think the most interesting data will come in January 2006. As you’ll recall, the Treasury Department didn’t come out with the HSA guidelines in 2004 until it was too late for most of the large employers to offer these products for 2005. We are seeing 50 percent of the RFPs [requests for proposals] from our large national accounts requesting a consumer-driven option in the suite of insurance products for 2006. January will be the point at which we expect to see a very substantial increase in the numbers of plan sponsors offering these products to their employees.

Consumerism And Employment-Based Insurance

Robinson: It’s perhaps ironic that some of the original proponents promoted consumer-driven product designs as a way of transitioning away from employment-based to individual-based insurance in the United States. Yet much of the enrollment has been with large firms, which are continuing to provide health insurance as a fringe benefit rather than dropping coverage and expecting employees to purchase individual, nongroup coverage. Aetna, for example, is a company that has traditionally been strong with large corporate accounts, as distinct from individual, self-purchased insurance. How do you see consumer-driven products fitting in with the future of employment-based insurance?

Rowe: I just learned the other day that you wrote about this in the New England Journal of Medicine not too long ago. I think that the concentration of these products in the large national accounts is a reflection of the fact that the consultant community serving those accounts has been particularly focused on this innovation and has studied it more than the brokers who serve the lower part of the middle market and the individual accounts.

Robinson: The brokers are concerned that they’re not going to get much of a commission off of these products.

Rowe: I would not use that differential in uptake as evidence against the hypothesis that this could ultimately facilitate a trend toward more individual-based and away from employment-based coverage.

Robinson: If there were such a trend toward more individual purchasing of insurance, how would Aetna play in that new environment, given its historical strengths at the other end of the spectrum, in the large corporate accounts?

Rowe: We think such a trend would be advantageous for us. If you look at Aetna’s profitable growth—in 2005 we’ll be growing more than a million members—it is greater in the small-business market and in the lower end of the middle market than in prior years. We have segmented our membership and focused very intently on growing in the small-business and individual markets. That is the fully insured portion of our book of business. At this juncture, we’re about two-thirds self-insured and one-third fully insured, and we expect that the fully insured side will grow faster than the self-insured side. We also find, in certain markets, Jamie, that of the many companies that offer products in the small-business market, we often are the only one that offers an HSA. So there again, it’s been fueling our growth, because of the first-mover effect.

Effects On Use Of Care

Robinson: Have you been tracking those who have enrolled for a year, in terms of the effects of consumer-driven product designs on their patterns of health care use? What are the early returns?

Rowe: Let me back up a bit, because there have been a number of significant concerns about high-deductible health plans, whether they are HRA- or HSA-compatible or not. The issue is cost sharing and its potential adverse effects on the use of preventive services and important long-term medications. We’re very focused on understanding what the impact of the high-deductible plan design is on the behavior of our members.

It’s important to understand that not all high-deductible plans have the same design. For instance, in our HSA- and HRA-compatible products, preventive services have no copayment or deductible. We have found that the use of preventive services has increased in these products. That’s encouraging, because some people felt at the outset that members would not differentiate preventive services from other services. They would just reduce their care-seeking behavior. But that’s not been the case. They can differentiate.

Robinson: Consumers aren’t stupid.

Rowe: No, they aren’t. To reduce the likelihood that people would forgo important long-term medications, we’ve pulled out of the deductible a series of medications that go beyond the original medications indicated by the Treasury Department [as eligible for first-dollar coverage without endangering the tax benefits of the HSA]. These include antihypertensives, antidiabetic agents, lipid-lowering agents, medications for asthma and osteoporosis, as well as childhood and prenatal vitamins—a long list of medications that we have taken out of the deductible.

Robinson: I’ve got a press release from you on that.

Rowe: Another concern with respect to behavior had to do particularly with individuals who have chronic diseases. We have now followed patients with diabetes, with respect to some central elements of prevention, such as hemoglobin A1C determinations, eye exams, and the like, and we have found no measurable difference in use due to enrollment in high-deductible plans. And hence, so far, we are not seeing avoidance of important care in our commercially insured population.

With respect to hospitalizations and physician visits, we’re seeing an interesting pattern that we know is persisting, because we’re in our fourth year of our consumer-driven products. Emergency admissions to the hospital have not changed, but [ambulatory] visits to the emergency room [ER] have changed. Specialist visits have actually increased. There appear to be more outpatient consultations but less inpatient utilization.

Robinson: Right.

Rowe: What do you think, by the way?

Robinson: I think that most people are smart most of the time. My concern from an economic point of view is that most of the costs, at Aetna and everywhere else, are concentrated among a small percentage of very sick patients. These people are beyond their deductible, and they’re frequently beyond their annual out-of-pocket maximum. The consumer-driven design is not targeting these patients, although this is where the big dollars are. Rather, it’s targeting the majority of patients who are broadly healthy and use few services.

Rowe: Here are some data. Inpatient admissions for people enrolled in HealthFund, compared to their prior-year utilization, were down 5.2 percent, and then the next year they were down 6.7 percent. ER visits were down, outpatient events were down, PCP [primary care physician] visits were down 10 percent, but specialist visits were up 3.5 percent.

Robinson: How do you, as a physician, interpret these numbers? Patients are now looking at a doughnut hole in insurance coverage, with more out-of-pocket payment, and somehow they end up not going to the hospital. The usual view is that hospital admissions are heavily influenced by physicians’ advice to patients as opposed to patients walking in. Why do hospital admissions fall when enrollees have higher cost sharing for primary care services?

Rowe: I agree. I’m surprised. Regarding chronic conditions, here are some numbers: Of 833 diabetic patients continuously enrolled in the first or second year, there were no significant changes in hemoglobin levels, retinal eye exams, or urinary albumin tests. For continuously enrolled members who had integrated pharmacy [some consumer-driven benefit designs exclude coverage for drugs], there was no significant change in the percentage of hypertensives taking ACE [angiotensin-converting enzyme] inhibitors and beta-blockers. There was no significant change in the percentage of members with asthma taking inhaled steroids or antiinflammatories. In short, we’ve looked at the chronic disease patients who are in the sickest 5 percent of our enrollment and who drive 50 percent of our costs [to examine how they fare under high-deductible insurance product designs].

The Limits Of High Deductibles

Robinson: One of the criticisms of high-deductible products is that they are financially more onerous to high users of care than they are to the healthy. Healthy people don’t need much medical care, and so they won’t use much, whereas sick people do need medical care, and they spend more. High-deductible products reduce the subsidy from the healthy to the sick that operates through low-deductible health insurance.

Rowe: At this point, you know, we’re talking about 2 percent or so of the insurance pool being enrolled in these products, and so we haven’t reached the tipping point.

Robinson: How far, as you and your colleagues think about this, will the deductibles increase?

Rowe: Employers are still raising them. We haven’t hit the limit of cost sharing.

Robinson: Medicare is raising them, too. Medicare is starting to look like a high-deductible health plan.

Rowe: Right. Some people consider increases in the deductible to be a bad sign and believe there will be significant pushback from employees and that it won’t go much higher. But my sense is that for many employers, raising the deductible is the alternative to not offering insurance at all. And so I would not be surprised to see much higher deductibles in the small-business and middle markets. I don’t think you’ll see that in the larger corporate accounts, however.

We’re talking as if the products are uniform in the marketplace in high-deductible health plans, but they are not. In the market you see two very different types. You see plans such as the Aetna plan where the deductibles are moderate and where preventive services and medications and other things are out of the deductible and reimbursed on a first-dollar basis. But then you can also find, on the Internet, products with a $10,000 deductible—very-high-deductible health plans with nothing out of the deductible [in terms of preventive services and drugs]. Those are not the kinds of products we at Aetna think are in people’s best interest. Those are not the products that we sell.

Robinson: That’s an interesting distinction.

Rowe: They are giving high-deductible health plans a bad name.

Robinson: Do you see any discussion of regulatory obstacles being placed in front of high-deductible products, in the sense that we saw a backlash against managed care from the regulators?

Rowe: I would say not, at this point. The way I would characterize our current situation with respect to high-deductible health plans is that we are moving from the phase of concern and skepticism, perhaps even cynicism, to the phase of evidence. Aetna now has three years of experience with these products, with millions of transactions from hundreds of thousands of members. We have generated several different versions of the product—HRA-compatible, HSA-compatible, different degrees of cost sharing. We have progressively modified the products to assure that preventive services are used and medications are not forgone. We are learning how the products behave in the marketplace. The regulatory restrictions that you mentioned will come if the data that we are now generating suggest that greater regulation is appropriate. So far, in Aetna’s products, we’re not seeing it.

Robinson: That assumes that regulation is based on evidence.

Rowe: It is important to understand that the population of individuals with whom we have the most experience is the commercially insured. The subgroup that is most susceptible to adverse effects of cost sharing is the population that likely would be on Medicaid.

Robinson: That’s right.

Rowe: There may be important differences in the populations.

Consumer Access To Information On Physician Pricing

Robinson: Describe to us Aetna’s initiative in providing price information to enrollees and how far you can go in helping people understand what health care costs, without submerging them in endless microdata that they don’t want to know.

Rowe: We have a pilot in the Cincinnati area, and we’re seeing substantial interest from our customers in other areas. As the number of members in fund-based products increases, and the payments to physicians came out of the deductible and therefore come out of the fund, it has become increasingly important for individuals to know the differences in prices charged by different physicians. Previously, price differences didn’t really matter to the individual. There was no transparency. Some elements of the physician community have expressed concern about this, which I fully understand. For instance, they will say that the information is incomplete because a given physician might have contracted at a lower rate for an annual physical, but that physician might order more or more expensive laboratory tests. That physician might refer a patient to a more expensive hospital. And for an episode of care, if a patient has depression, one physician might see the patient four times during that episode of depression, and another physician’s practice pattern might be to see the patient eight times. And all we’re showing is the unit price.

Robinson: So the data that you have right now are on the price for office visits?

Rowe: Yes.

Robinson: For particular procedures, and not whole episodes of care?

Rowe: That’s right. We feel we have to start somewhere, that the information we’re providing is valid. It seems to have been very well received by physicians, patients, regulators, and elected officials as an important step forward in transparency, and if we have the data available, we would be pleased to enhance them with additional information.

Robinson: These are the Aetna-contracted prices?

Rowe: If you’re an Aetna member, you can go on the Aetna Navigator [on the Aetna members-only Web site] and see in your region the names of the physicians and the Aetna-contracted rate.

Robinson: And in the Cincinnati area, I infer that there is significant variability among physicians in what they charge for the same visit and the same procedure—

Rowe: I also would predict, to interrupt, that this type of transparency might be expected to reduce the variability—and flatten rates across the marketplace.

Narrow Networks For Specialists

Robinson: Or at least have the variability be based on something. Now, if you please, could you put this price-transparency program in the context of Aetna’s specialty network?

Rowe: What we call the Aexcel network is a network only of specialists, as opposed to primary care physicians. It is a network in which we have used our database to identify physicians’ episode costs, ETGs [episode treatment groups], and quality of care. We began it two years ago in Jacksonville, Dallas, and Seattle and for six specialties. We’ll be doing it for twelve specialties in twenty markets as of January 2006. Those specialties account for about 70 percent of costs.

We use our own data, but we use standards that have been established by professional groups, such as the American Diabetes Association, or HEDIS [Health Plan Employer Data and Information Set] measures. We eliminate from the local network a proportion, but not a majority, of the specialists in the area, and thus we encourage our members to see the remaining higher-quality specialists and specialist groups. For instance, if we reduced a network by a third, the remaining two-thirds of doctors would get a 50 percent increase in volume. We generally do not increase the payment on a unit basis, because we find that physicians are very volume-sensitive for specialty procedures. These physicians tend to have lower admission rates, lower readmission rates, shorter lengths-of-stay, and fewer complications. So by virtue of going to this higher-quality network, patients actually experience a reduction in their health care costs.

Robinson: So the Aexcel network is using episode-based costs and some quality measures to create a narrower network?

Rowe: That’s right. There’s a quality screen and a cost screen.

Robinson: Compare this specialty network initiative with the price-transparency initiative in Cincinnati, which maintains a broad [primary care] network but highlights the cost differences within the network.

Rowe: The price-transparency initiative does not reduce the scope of the network, and it does not focus only on specialists. Some other companies have developed tiered networks based only on cost differences. We don’t think that’s appropriate.

Differentiating Primary Care From Specialist Networks

Robinson: As a general rule, do you feel that narrow network designs are effective on the specialty side, while broad networks are effective on the primary care side?

Rowe: Yes, at this point I would agree with that characterization. For the time being, that’s our strategy, and we’ve been very pleased with it. At the end of 2004, we had 79,000 members in the Aexcel [specialist] network. At the end of the first quarter of 2005, we had nearly 300,000.

Robinson: In Aexcel you are using the physician group rather than the individual physician as the unit of choice, if I am not mistaken? Much of the discussion of consumer-driven health care has been having individual patients choose individual doctors based on quality and price. Yet it’s hard to get reliable quality on individual physicians because they often don’t treat enough of any one condition to have valid quality data.

Rowe: I think that’s right. And in addition, in many markets, Aetna’s market share is not large enough to generate reliable statistics, as we might only represent 10–15 percent of a physician’s experience. So you’re seeing a slice of a slice. And if patients need spine surgery, they’re not interested in how the surgeon does hip replacement. They want to know how that physician does spine surgery.

Robinson: Do consumers, once they’ve got their high-deductible incentives, make more intelligent choices?

Rowe: I see this as a different form of consumerism. If you think about high-deductible health plans, with savings funds, in those products we are trying to make the consumer a more prudent purchaser, providing the consumer with more information. In contrast, the tiered-network approach, such as Aexcel, basically makes those decisions for the consumer. The consumer can say, “This is a better network; this is where I’m going.”

Robinson: I am always amused by these consumer-driven care pundits who think we all want to spend our day on the ’Net looking up quality and price data.

Rowe: The fund-based products are subject to criticism because they might not be designed to help individuals with chronic diseases, as you said before, because chronically ill enrollees are going to be through their out-of-pocket maximum by St. Patrick’s Day.

Robinson: That’s right.

Rowe: The tiered-network products don’t have that weakness. And in fact, because they concentrate on specialists and because the individuals who are the biggest utilizers are likely to be using the services of specialists, these products may be more effective for people with chronic disease [than high-deductible health plans with HSAs].

Robinson: The underlying principles of the narrower network are really managed care principles, using performance data to develop a network, as opposed to providing information and letting consumers make their own choices. Providing comprehensive price information to consumers would be the equivalent of providing information to car purchasers on the price of spark plugs and other every component of the automobile. We don’t want to know all that.

Rowe: People vary, and there are a lot of people who will pick up a cereal box and really read about how many calories and how much sodium and how many carbohydrates are in the cereal—they want that information.

The Future Of Medical Management

Robinson: What is your view concerning Aetna’s role in medical management, including disease management and case management?

Rowe: We consider medical management to be a key part of our strategy. We believe that our commitment to disease management, case management, pharmacy management programs, and the integration of our pharmacy management programs into our fund-based [high-deductible insurance] products have been an important part of our ability to control medical costs better than most companies in the industry. We have reported, year over year, that we are at the lower end, if not the lowest, in our medical cost trend among health plans. That’s a tremendous value to us. Two-thirds of our customers are self-insured, and they’re buying that low medical cost trend. And of course it helps us compete in a marketplace for insured business with competitive pricing.

Robinson: How does that vision of Aetna as medical manager fit with the other vision of Aetna as facilitator for consumer choice?

Rowe: We believe that they are perfectly consistent, in that they focus on different subsets of the patient population. The medical management approaches tend to be focused on that 5 percent of the population that drives 50 percent of the expenditures, whereas the enrollees in the consumer-driven products are more likely to be healthier. The exception [to the potential for differential risk selection] is when a plan sponsor decides to do a full replacement [enroll all employees in a single health plan’s products, without choice of carriers]. At that point, the sponsor tends to be very interested in purchasing our disease management and pharmacy management initiatives to supplement the effectiveness of the consumer-driven products. It becomes valuable as a synergistic effect. This commitment to medical management and the belief that there is much more that can be done was the main reason we acquired ActiveHealth Management.

Robinson: Tell us a little bit about your decision there.

Rowe: ActiveHealth is a very successful company that has a proprietary care engine, which uses computerized algorithms to monitor quality of care to identify opportunities to improve quality of care. It’s been successful in the marketplace. We use ActiveHealth as a vendor, and our customers like it very much. We decided that the leaders of ActiveHealth were cutting-edge, the best in the marketplace. By purchasing the company, we are able to let them inform the development of the next generation of our products. At the same time, we believe that access to ActiveHealth is a right, not a privilege, and therefore we continue to encourage other health plans to purchase its services for their members. But we feel that it’s an important addition to our capacity to design future products.

Robinson: The other health plans that had been using ActiveHealth—are they a little nervous about the fact that it is now a subsidiary of one of their competitors?

Rowe: That was an initial concern when we made the acquisition, so we made the decision to brand this as ActiveHealth, rather than as Aetna, and to maintain it as a freestanding subsidiary. We’ve been very pleased with the fact that the membership in ActiveHealth has actually grown, not reduced, since the acquisition and that the other health plans that had been using it continue to use it.

Consolidation Of The Health Insurance Industry

Robinson: Just one more question. Aetna has turned itself around financially, but as part of that, you were willing to accept substantial declines in enrollment. In the past year, enrollment was back on the upswing, but Aetna is now significantly smaller than the two biggest companies in the industry, WellPoint and United. How do you see your role going forward? Is this difference of scale acceptable? Are you now interested potentially in a significant acquisition, which you have not been for the past few years?

Rowe: Well, as you say, we’ve entered a growth phase, and it’s a profitable growth phase: We’ll add over a million members in 2005. Our focus with respect to growth is on a number of very specific markets. We’ve identified nineteen key markets throughout the United States in which we believe the combination of our market position, the regulatory environment, and our network provides us with the opportunity to grow significantly. To date, our growth has been largely organic, although we’ve recently acquired HMS Healthcare, a leading regional health care network. This acquisition will enable Aetna to strengthen its local market presence—particularly in Michigan and Colorado. We also acquired Chickering, which provides health insurance to college students, and SRC, which provides group health coverage for part-time workers. Although we acquired membership in both of these, like the ActiveHealth acquisition, they are really designed to enhance our capabilities so we can meet varied customer needs and grow organically going forward. We would be interested in either organic growth or acquisitions that are relevant to our key growth markets. The focus is not on national size for the sake of national size.

Robinson: In virtually every U.S. state and market, the local Blue Cross Blue Shield plan has the largest share and a good brand name. How does Aetna, as you move into the small- and midsize-firm markets, compete against such traditional brands that have such a large market share?

Rowe: We’ve had success competing with the Blue Cross plans over the past couple of years in both small-business and midsize markets. What we see in the marketplace is interest on the part of employers in one-stop shopping. The fact that we offer our own PBM [pharmacy benefit manager] integrated into our health products and our own behavioral health services are very significant advantages. The fact that we offer consumer-directed health plans is increasingly important to these customers. The fact that we have a large commitment to disease management and case management is important as well.

Jack Rowe, a physician, is chairman and chief executive officer of Aetna, one of the nation's largest insurance companies. Jamie Robinson (jamie{at}berkeley.edu) is the Kaiser Permanente Distinguished Professor of Health Economics at the University of California, Berkeley, School of Public Health.

DOI: 10.1377/hlthaff.w5.583
©2005 Project HOPE–The People-to-People Health Foundation, Inc.