Experience matters: Risk-adjusted CDHP savings

  • Print
  • Connect
  • Email
  • Facebook
  • Twitter
  • LinkedIn
  • Google+
By Jack P. Burke | 28 April 2008

April 28, 2008


Milliman recently studied actual cost savings of CDHPs.

Report coauthor Jack Burke explains what we're currently seeing from CDHPs and what we might expect in the future.

Consumer-driven health plans (CDHPs) are viewed as a potential source of major health-cost savings among reform proponents. Milliman recently completed a study on actual cost savings, the first independent risk-adjusted analysis. Report co-author Jack Burke provides some perspective on what we're seeing from CDHPs and what we might expect in the future.

Q: Where does the Milliman study focus its attention?

A: First, let me define what we mean by consumer-driven health plans (CDHP). Consumer-driven may mean different things to different people. For example, we are not talking about tiered networks. For our analysis, CDHP is a high-deductible health plan (HDHP) with a funding account. This is either a health reimbursement arrangement (HRA), funded by the employer, or a qualified HDHP, with the availability—whether or not the member chooses to use it—of an accompanying health savings account (HSA). HSAs are funded by the employer or the member and are portable, meaning that members can take their accounts along with them even if they go to other jobs. The money in these accounts is restricted to use for medical and health-related costs, with varying requirements and exclusions, but HSA dollars may often be placed in interest-bearing accounts and other investment opportunities, such as mutual funds. (The "qualified HDHP" that accompanies an HSA is one that meets criteria established in the Internal Revenue Code permitting the HSA to be tax-advantaged.)

Our study investigates the good results that are reported in the media. Many CDHP vendors show wonderful experience, but the critics can just as easily say that it is a result of signing up the young and healthy. As that debate went on, good firm experience was hard to come by.

The study is based on data from employers, which was gathered with the help of the National Business Group on Health. NBGH encouraged several of its members to participate in our study and these employers provided us with every claim associated with their covered members, including prescription drug claims, which often came from a different payer.

Q: What did the analysis consist of?

A: We compared the actual experience of the CDHP plans to that of the non-CDHP plans that were part of each employer's program. Like the experience reported in the news, the paid claims were very low compared to the much richer PPO or point-of-service plans—by about half. However, that is to be expected for a variety of reasons.

First, if you look at claims paid by the plan alone, any high-deductible plan will not have payments until after the deductible is met. For many members, that could be never if the deductible is high enough. To adjust for that, we added in the payments by the member—directly or through the side fund—and considered the total allowed claims.

Even so, the total allowed claims were still very low, about 60% of the allowed claims for non-CDHP plans.

That gets us to the other reason that may explain lower claims—the young and healthy.

We started by assigning factors that measure the expected claim cost based on each member's risk profile. First, we assigned an area factor based on the ZIP code. Then we listed the age/gender factor for each individual. Last, we assigned a risk score to each individual based upon the reported conditions in the claim detail. The risk score tells us the expected claims, on average, for a person with the conditions listed. The risk score didn't try to predict claims for every small diagnosis, such as an ear infection, but focused on chronic conditions.

Q: What were the results?

A: The risk-score adjustments were very dramatic and were consistently big for every plan. When these adjustments are made, the savings begin to appear much more modest. Instead of 41% lower than the richer plans, after these adjustments, the CDHP experience was about 5% lower (4.8%). Still meaningful, but much less than the fantastic savings seen before making these adjustments.

In terms of "results," that's the number I focus on when looking at the CDHP experience. But I want to bring in one more concept. Going back to the '80s and before, high-deductible plans have existed, and they have always encouraged a reduction in utilization. The observed savings because of the cost sharing—we call it induced utilization—has been known and measured for years, and is a part of our Health Cost Guidelines pricing. It's not new to the consumer-driven movement.

In fact, the predicted reduction in utilization due to the higher deductible is 3.3% for this group. So, in effect, the consumer-driven plans are only performing about 1.5% better than predicted for a high-deductible plan.

Q: So does that mean the CDHP concept is not working?

A: Yes and no. Let me explain.

A high-deductible plan with a side fund consisting of money the employees can potentially keep and save is only part of the consumer-driven concept. It is supposed to provide the incentive. As measured above, it is doing that—plus a percent and a half more. The other part of the concept is consumer information. That is, information about doctor and hospital cost, and doctor and hospital quality.

When we polled the employers in the study, only one in six told us that they were able to give their employees data on cost, and none of them thought they had quality information to share about the healthcare workers and facilities.

General patient education of healthy lifestyles and other health programs, support groups, and care management programs were always made available to both the CDHP members and the non-CDHP members. So any benefits that resulted from those efforts affected both populations equally.

So, "yes," the CDHP concept is delivering on the incentive part. But "no," it's not fully delivering on the consumer information part—not yet, at least—which is what will make CDHPs stand out from current plans. To date, the cost and quality information tends to lag behind the financial incentives. New cost and quality information and tools are continually emerging that may add to the results achieved through financial incentives, hopefully narrowing the information gap.

Jack Burke is a principal and consulting actuary in the Philadelphia office of Milliman.