On April 6, the House of Representatives added an amendment from Rep. Gary Palmer (AL) and Rep. David Schweikert (AZ) to the American Health Care Act (AHCA) before their recess. The amendment was drawn, in part, from the proposal in our previous Health Affairs Blog post.

The proposal sparked a flood of media coverage, but many stories left much to be desired. Reporters and commentators repeatedly mischaracterized the program and some quoted individuals that clearly did not understand the mechanics of the program, which only deepened confusion.

Some thoughtful discussions emerged that should influence the provision in any final bill. Our goal is to correct a few misconceptions and suggest a few changes that would improve such a program based on a comprehensive independent analysis conducted by actuarial firm Milliman.

Goal Of The Invisible Risk-Sharing Program

The Invisible Risk-Sharing Program (IRSP) will stabilize the individual insurance market and lower premiums while concurrently providing guaranteed access to coverage and protecting those with pre-existing conditions. Different than a traditional high-risk pool, no one is declined coverage, enrollees with pre-existing health conditions get the same plans at the same lower price as a healthy individual, and those with pre-existing conditions are not segregated to higher cost and limited benefit high-risk pool plans. Several questions have been raised about IRSP and the amendment. We address a number below.

Is There Enough Money Allocated?

One issue that has been raised is that $15 billion allocated with the amendment may not be enough to fund the program through 2026. This would be true only if you assume states have no responsibility to contribute, that states will not find funding in the form of an assessment (the way Maine funded a large portion of their program), or that states will not use some of the other $115 billion allocated to the over the decade as they take over the program after 2020 as the amendment lays out.

Milliman’s analysis of the most effective policy package to reduce premiums for the risk-sharing program carried a price tag at just over $3 billion per year. So at a minimum, the program could be fully federally funded for more than four years if the final bill reflects the model and before states would have to tap into the $115 billion Stability Fund or use other funding. To clear up any confusion about the devolution of such a program to states, the bill may benefit from a more explicit timeline to do so.

Is This Program Too Complex For Insurers?

The implementation of the Affordable Care Act (ACA) has been complex with a large amount of uncertainty, so we don’t believe this program is unduly complex. There are concerns that insurers may not want to participate as they have to prospectively cede individuals into the program and cede those individuals’ premiums to help pay for the risk sharing. While some insurers may be looking to game taxpayers and to do less work upfront, that’s not good for the program, patients, or the taxpayer. Of course, insurers will push to reduce their work and lobby for a straightforward taxpayer money transfer, but insurers gain with increased enrollment (up to 2 million according to Milliman depending on the program structure) and their losses are capped—but still there—for those with known pre-existing conditions or poor health at the time of enrollment.

If the policy choice is to guarantee access to coverage but do so with the lowest premium rates for everyone and in the most cost-effective way for taxpayers, then that is the balance that invisible risk sharing provides—better and cheaper—for everyone. The IRSP is about directly subsidizing a policy choice — guaranteed access to those with pre-existing conditions. It is not about reducing insurers losses generally or providing general funding.

Is Maine A Good Example To Predict A National Outcome?

The risk-sharing program was modeled after a program in Maine that was passed in 2011 and ran until the ACA kicked-in. The Maine experience is particularly relevant to this debate. Prior to this invisible risk-sharing program, Maine had very similar ACA-like regulations for decades and premiums had skyrocketed and the individual insurance market was in bad shape.

But some have raised concern that the impact will be more limited at the national level, and cite other changes in Maine that account for the dramatic premium drops. While there were a few moving parts in the Maine reform, it is important to correct a few misconceptions.

First, age band changes in Maine were gradual so it was not the primary driver behind lower premiums. In fact, the AHCA changes age bands as well, so the comparison here is apt and was modeled to reflect any difference by Milliman.

Second, while plan design changes did happen at the same time, they were not “required.” The changes serve to highlight a turning point in Maine. It signaled new plan options for the first time in many years and allowed enrollees real choices that they did not have previously. While there were changes in plan design, including plans with coinsurance and increased out-of-pocket maximums, there were also new benefits being added at the same time. For example, some plans offered lower deductibles and coverage for prescription drugs that kicked in after less up front out-of-pocket spending.

At a high level, as we pointed out in our original Health Affairs Blog post, comparing premium levels for a plan with similar deductible levels (arguably the two focal points for most consumers) the difference was dramatic. Premiums decreased for all age groups, up to 70 percent for the youngest, according to the Maine Bureau of Insurance. Even the largest Maine carrier acknowledged that the immediate impact of such a program on premiums without any plan design changes was around 20 percent in the first year.

Taking a policy from one state and broadening it to a national level can be challenging, which is why Milliman was engaged to do an independent analysis of such a policy nationwide. Milliman took into account variation across states and utilized numerous comprehensive national data sources to look at actual claims and medical status of the American population at large and their report projects the outcomes on lower premiums and increased coverage nationally.

• Individual insurance premiums would drop up to 31 percent on average for those getting coverage on their own and outside the exchange in a newly created risk pool (Scenario 1, Table 1).
• No one with pre-existing conditions would be rejected, in fact their premiums would drop as well.
• Up to 2 million uninsured individuals would voluntarily buy coverage because of the lower rate (before considering any new AHCA tax credit).
• The program would be affordable as the projected cost is about $3 billion a year nationally if set up in a targeted manner.

Milliman unpacked each of the policy options and showed how a combination of changes can maximize premium decreases while still helping those with pre-existing conditions, at an affordable price tag for taxpayers and patients.

Does The Past Experience Of Some Prospective Programs For Small Business Tell Us About Future Success In The Individual Market?

While a comparison about risk sharing in the individual market versus the small group market is valid, some have tried to imply that past programs in the small group market should raise doubt about targeted invisible risk sharing working in the individual market. The small group market is very different than the individual market. In addition, any comparison should unpack unique design differences to gauge effectiveness along with any local political history that might point to insurer influence or lobbying.

How Is This Different From The ACA’s Reinsurance?

Both the ACA’s reinsurance and other insurer focused funds share a similar goal of reducing premiums, but the programs are structured and run in very different ways, and have very different records when it comes to stabilizing the individual market and lowering premiums, particularly for those outside the exchange (the focus of the IRSP).

The ACA’s reinsurance program serves as an expansive partial “bailout” for insurers for all high-cost enrollees. Funds are simply being injected on a retrospective basis to an insurer (after medical claims have been paid by an insurer), to reimburse a portion of their losses. They are simply reducing a pre-defined portion of some claims paid. They represent a $1 for $1 reduction in claim dollars spent by an insurer ($1 reinsurance reimburses $1 of claims paid). The effect on premiums is directly tied to the amount of reinsurance dollars paid. For example, if a reinsurance program paid a million dollars into a $100 million insurance market, you would expect to see premiums reduced by about 1 percent. Insurers also keep premium dollars paid by enrollees putting the taxpayer on the hook for the entire cost of the program.

By contrast, the IRSP is a policy choice to guarantee access to coverage for those with pre-existing conditions and subsidize only a select group of enrollees to prevent distortions in the market. Of the entire private insurance market, the IRSP would only apply to those in the individual market buying insurance outside their employer and outside the exchange and that have been designated for the program by an insurer and only for those with actual medical claims that exceed a threshold (for example $10,000 of medical costs). This translates into an estimated 1.7 percent of the entire private insurance market.

The program utilizes health questionnaires to identify individuals with specified pre-existing and likely high-cost medical conditions at the time they are applying for an insurance policy. Insurers then pay a significant portion of the policy premiums for any individual designated to the IRSP while maintaining responsibility for a certain level of claims incurred. Incentives are aligned to both maximize the effectiveness of the program and guard against insurers gaming the system. This approach also leads to more predictable costs as it only applies to a small group of individuals. All this allows insurers to price those in the IRSP and those outside the program as if they were healthy, because the losses of those known to be high cost are capped. That’s why Milliman projected such a drop in premiums. The IRSP pulls that risk out of the market. Critics conflate the risk of providing guaranteed access to a few with the claims costs of all high cost individuals. They are two very different issues.

In talking with actuaries there is a belief that prospective models are more effective in lowering premiums when structured correctly. By having insurers cede individuals upfront it allows the insurer to have more predictability when gauging the level of impact that ceding will have on claims and therefore premiums for the remaining enrollees. Moreover, the IRSP also becomes a predictable risk-transferring entity for the insurer that covers the claims of those ceded which is different from the ACA’s reinsurance program that had a volatile element for insurers as their portion of funding lacked predictability. These two important differences enable insurers to lower premiums more under a IRSP structure.

Additional Policies To Consider

Due to a number of factors, some which are due to the limits of the reconciliation process, some more details would be beneficial to be added in the future. Some that the Senate should consider are the following:

Create A Separate Risk Pool

The Milliman report makes it clear to maximize premium reductions in the first year, creating a new risk pool for new insurance plans to be offered to those outside the exchange (Scenario 1) has the biggest impact on premium reduction for all age groups. This would likely include grandfathering in the old exchange plans and could place some requirements on insurers to continue those plans in order to sell in the new market for those buying coverage outside their employer and outside the exchange (and thus paying the full cost of the premiums).

Create Special Enrollment Period

Set in place a special enrollment period in 2018 to provide premium relief as soon as possible. Maine had new plans on the market in 13 months (after passing the law in May 2011, new plans were on the market by July 2012).

Set A Clear Plan For A Devolution & Financial Commitment By The States

A clear timetable for states to customize and run the IRSP would be appropriate, regardless of the level of federal funding, and would be beneficial and appropriate.

Medicare Rates

Milliman modeled paying risk-sharing claims at Medicare rates (or about 25-30 percent lower than commercial rates) once IRSP patients incur a certain level of claims. This policy choice of switching to Medicare rates produces a few extra percentage points of premium reduction for everyone and recognizes that taxpayer-funded programs (such as Medicaid or Medicare) typically pay less than commercial rates. This is a policy choice with implications for patients, providers, and taxpayers that should be fully discussed.

If a limited Medicare rate proposal is included in a final bill, the following context should be noted:

• It would apply to a tiny percentage of claims for a given hospital in a community (again only about 1.7 percent of all those with private health insurance are in the IRSP, and would only be paid for patients that exceed the threshold for claims in any given year).
• A few percentage points lower premiums means more currently uninsured individuals will sign up for commercial insurance which will result in higher payments for providers.

The IRSP is simple, straightforward, and effective. It achieves a very tough policy balance – guaranteed access to those with pre-existing conditions while alleviating the massive premium increases that the ACA’s guaranteed issue provisions have produced in the individual market outside the exchange. It is worthy of debate and discussion, but those discussions should be based on the full understanding and not misconceptions of the program or the legislative process.

https://healthaffairs.org/blog/2017/04/25/correcting-misconceptions-about-invisible-risk-sharing/