Transfer Problems
Exploring the imbalance in the ACA’s risk adjustment transfer formula
Web ExclusiveRisk adjustment, utilized as a method to spread revenue across populations, is not a new concept in health care. CareSource, a managed care organization (MCO), is focused on providing health care to a membership comprised of the most vulnerable in our society—the Medicaid population. Our mission focuses on this population, even now as we expand our product scope. Risk adjustment has played an important role in supporting this mission.
Programs like Medicaid have many challenges above and beyond the unique needs of the population, such as state-specific regulations, program funding, definitions of eligibility, benefit differences and churn (or turnover) in a population, just to name a few. For years, state-managed Medicaid programs have utilized zero-sum risk adjustment models to right-size capitation payments across MCOs. Through their marketing approach, network make-up and other differentiating program elements, MCOs work to increase membership. Over time the population risk skews, and selection bias toward the MCOs that truly have differentiated themselves occurs. Risk adjustment is essential to ensure appropriate sizing of capitation to the population risk enrolled.
An Imperfect Transfer Formula
For many of these same reasons, risk adjustment also plays a key role in the individual Affordable Care Act (ACA) Marketplace. However, a significant challenge presented for the marketplace is developing a risk adjustment model when the premium (or capitation) is set by each insurer individually, based on its specific product offering. Also, similar to Medicaid, the risk adjustment program is funded as a zero-sum program, which means solely the participating insurers generate the funds. In order for the risk adjustment transfer formula to balance to a zero-sum across a market, the formula utilizes the state average monthly premiums as one of the key elements. Although the end result is financially neutral at the state level, a few unintended consequences result. To attract CareSource’s target population (low-income members), we set our premiums low enough to allow the Advanced Premium Tax Credit to cover the entire, or at least the majority of, the premium cost for our offered products. As a result, our 2015 average premium was 14.6 percent lower than the statewide average in one market in which we participate.
Why does a variance from the statewide average matter? The risk adjustment transfer formula does not account for plans that have more advanced approaches to care management or more progressive value-based contracting methods that help drive premiums lower in a market. The transfer formula penalizes plans with lower premiums and rewards those with higher premiums in relation to the statewide average premium. Unfortunately, low cost insurers working to keep the marketplace affordable for individuals are being hit with the formula imbalance. Similarly, the inclusion of administrative costs (e.g., non-claims costs) into the calculation further exacerbates the issue for highly efficient plans with low administrative costs. Figure 1 depicts a simple example of the imbalance in the transfer formula as described.
Figure 1: Sample Transfer Formula Variables | |||||||
---|---|---|---|---|---|---|---|
Membership Percentage | Actuarial Value (AV) | Plan Liability Risk Score (PLRS) | Allowable Rating Factor (ARF) | Induced Demand Factor (IDF) | Geographic Cost Factor (GCF) | Premium | |
Statewide | 100% | 0.700 | 0.967 | 1.800 | 1.050 | 1.000 | $275.00 |
Plan 1 | 33% | 0.700 | 0.900 | 1.800 | 1.050 | 1.000 | $250.00 |
Plan 2 | 33% | 0.700 | 0.900 | 1.800 | 1.050 | 1.000 | $275.00 |
Plan 3 | 33% | 0.700 | 1.100 | 1.800 | 1.050 | 1.000 | $300.00 |
Figure 2 shows that Plan 1 pays the same transfer amount as Plan 2, even though Plan 2 has a higher average premium (e.g., Plan 1 pays an additional 0.7 percent of premium). For many plans with low premiums, a 0.7 percent loss due to the formula inequity could equate to their entire margin/profit load in their rates. Using plan average premiums may help reduce this imbalance in the formula; however, additional steps should be considered, such as an additional funding source (e.g., Medical Loss Ratio dollars) or inclusion of additional factors (e.g., care management, disease management, administrative efficiency, etc.) to net to zero.
Figure 2: Sample Transfer Formula Calculations | ||||||
---|---|---|---|---|---|---|
PLRS*IDF*GCF | Normalized (PLRS*IDF*GCF) | AV*ARF*IDF*GCF | Normalized (AV*ARF*IDF*GCF) | Transfer Amount | Transfer Percentage of Plan Premium | |
Statewide | 1.015 | 1.000 | 1.323 | 1.000 | $0.00 | 0% |
Plan 1 | 0.945 | 0.931 | 1.323 | 1.000 | ($18.97) | −8% |
Plan 2 | 0.945 | 0.931 | 1.323 | 1.000 | ($18.97) | −7% |
Plan 3 | 1.155 | 1.138 | 1.323 | 1.000 | $37.93 | 13% |
Conclusion
There is no question that a risk adjustment pool is necessary for the marketplace to thrive; however, a strong risk adjustment program is essential to drive market competition and consumer choice. Improving the model will fortify insurers that currently are offering coverage in a market, as well as help to encourage expansion into new markets.