The Future of Risk Adjustment: Supporting Equitable, Comprehensive Health Care

Executive Summary

Risk adjustment in alternative payment models for accountable care has been designed to:

• Provide financial incentives for plans and providers to attract and retain individuals with high levels of need (thereby avoiding adverse selection),
• Support fair benchmarks for assessing performance across providers and plans accountable for differing populations, and
• Ensure that accountable organizations have appropriate resources to deliver high-quality and effective care for the populations they serve.

However, current risk adjustment methods based on fee-for-service claims data face multiple limitations:

• The incentives in current methods encourage more complete health risk reporting, but the resulting additional payments are not clearly linked to better care for patients.
• Risk adjustment based on data from fee-for-service care may be decreasingly relevant and reliable for assessing financial risk differences for value-based care models.
• Current models take limited account of a person’s functional status or socioeconomic characteristics, even though these can be strong predictors of health spending and need.

Considerable interest exists in incorporating social factors into risk adjustment, given substantial evidence linking social risk factors to unmet health needs.

• However, incorporating social factors into risk adjustment based solely on historical associations with spending (the approach generally used for other risk adjusters) could reinforce existing structural inequities in access and care despite similar health needs.
• Social risk adjustment can alternatively impute higher payments to help address structural barriers to needed care, but this creates tradeoffs if risk adjustment increases must be offset for overall financial neutrality – for instance, more resources for payments to people who live in high ADI areas may mean lower payments for someone who is lower income or has multiple chronic conditions in a lower-ADI area.
• More evidence and complementary policies are needed to enable risk adjustment to be used effectively to improve care for traditionally underserved populations.

To address these issues, we have identified a set of guiding principles for risk adjustment reform, building on the traditional core goals of risk adjustment:

• Support advanced accountable care relationships for beneficiaries with diverse medical, behavioral, social and other needs, particularly those at greater risk of adverse outcomes.
• Enable resources to shift from traditional medical services to innovative care models and valuable services that are traditionally not reimbursed well, to improve outcomes, reduce costs, and increase equity.
• Encourage data collection for risk adjustment to rely on data used for tracking disease prevalence and implementing steps to reduce risk of disease incidence, progression, and costly complications (i.e., limit the need for administrative activities focused only on risk coding).
• Include reliable data on key markers of health (e.g., functional status) that provide a more complete and accurate assessment of health status and risk than diagnoses alone.
• Reflect the association between social factors (e.g., socioeconomic status, race, ethnicity, and language) and health, while recognizing that risk adjustment based on spending will not address the financial implications of inequitable access to and use of medical services that are reflected in current Traditional Medicare utilization.
• To advance accountability for outcomes and provider capabilities to achieve health equity goals, pair risk adjustment policies with complementary payment and regulatory incentives (e.g., advance payments or supplemental payments to plans and providers caring for traditionally underserved populations, payments linked to improvements in social drivers of health or equity measures).

Building a more modernized risk adjustment system described by our guiding principles will require a combination of short- and long-term actions. Short-term actions could be implemented within the next one to two years to address current challenges to build better risk adjustors and begin laying the groundwork for longer-term reforms. Potential short-term actions include:

• Refining the Medicare Advantage coding intensity factor adjustment,
• Building a foundation for risk adjustment mechanisms that go beyond reliance on administrative claims from Traditional Medicare beneficiaries,
• Reducing payments adjustments that induce changes in reporting but not in care improvement,
• Assessing viability of using risk adjusters that better capture social factors,
• Identifying additional high-priority risk adjusters capturing key dimensions of health,
• Implementing complementary payment reforms to address structural barriers to coordinated care for traditionally underserved communities, and
• Implementing steps to address under-diagnosis and advance care management in Traditional Medicare.

Potential long-term actions include:

• Linking population health management and risk adjustment through better integrated systems to adjust payments for risk and quality of care,
• Implementing risk adjustment mechanisms that rely on automated collection from data sources used for care improvement and population health management,
• Developing alternatives to measuring utilizations and resource use based on fee-for-service, Traditional Medicare claims,
• Aligning risk adjustment incentives and models, and
• Implementing an evidence-driven strategy for risk adjustment to improve equity.