Reductions in service utilization in 2020 will impact MA payments for 2021. Specifically, lower utilization results in fewer opportunities for providers to code diagnoses. The CMS uses diagnosis data from claims to calculate MA risk scores used to adjust payments to MA plans. Because diagnoses from 2020 claims are used as an input to determine 2021 risk scores, fewer claims in 2020 could mean lower risk scores, even though the health status of enrollees has not changed. Consequently, risk scores may not fully reflect the cost of care. Using MA claims data through June 2020, Avalere estimated the impacts of the reduction in claims on risk scores and payments for 2021.

Avalere Found Large Reductions in MA Claims in 2020

The number of MA enrollees with at least 1 claim was 47% lower in April 2020 compared to April 2019, which corresponds with the declaration of the public health emergency in March 2020 (see Figure 1). Though utilization rebounded moderately in May and June, it was still substantially below 2019 levels. Overall, as shown in Figure 2, the total number of claims decreased by 66% in April 2020 compared to April 2019 and remained low in May and June compared to 2019.

Notably, without new telehealth flexibilities, utilization may have declined even further. Telehealth utilization increased to 21% of all claims in April 2020 and 17% in May compared to approximately 5% in April, May, and June of 2019 (see Figure 3).

On April 10, 2020, the CMS announced that due to the COVID-19 pandemic, MA plans and other eligible organizations may submit diagnoses from telehealth visits for risk adjustment. Diagnoses may be submitted for risk adjustment if they are from a face-to-face inpatient, outpatient, or professional service encounter. Diagnoses can meet the face-to-face requirement when services are provided through communication services that have interactive audio and video systems, allowing for real-time communication.

Note: Telehealth claims include both audio and video telehealth encounters.

2021 Payments Could Be 4–6% Lower Due to Lower Risk Scores

Each year, plans calculate an initial risk score using claims from July to June (e.g., July 2019–July 2020) and a midyear score using a calendar year of data (e.g., January 2020–December 2020). To estimate the potential impact of the pandemic on MA plan payments, Avalere first calculated the 2020 midyear score to use as a baseline, and the initial score for 2021 (i.e., claims from July 2019 to July 2020). The initial 2021 score, based on complete claims data through June 2020, only accounts for the first 4 months of the pandemic and already shows an estimated 3.5% decrease in health care utilization, leading to a 2.39% decrease in payment. In the summer of 2021, the CMS will update the risk scores for 2021 payment by using a full year of the 2020 data, which would also include the impact of the pandemic from July to December.

Avalere modeled 3 scenarios based on different levels of assumed utilization (i.e., low, medium, and high compared to the previous year) for July–December 2020 (Table 1). The low impact scenario assumes that utilization increases moderately from June 2020 levels, while the medium scenario assumes that it remains at similar levels. The high impact scenario accounts for a decrease in utilization below June levels, potentially due to the worsening of the pandemic during the fall and winter. Under these scenarios, MA plans’ risk scores could decrease between 6.3% and 9.6%, which could lower payment up to 6.1% (see Table 1).

*Note: Risk scores are standardized to equal a 1.000 for the baseline.

**Uses diagnoses from July 2019 to June 2020.

Avalere also assessed the number of diagnoses—or hierarchical condition categories (HCCs)—per enrollee under each risk-score scenario. Looking only at enrollees who had at least 1 HCC included in the midyear 2020 score, 9% did not have any HCCs recorded for the initial 2021 score, as shown in Figure 4. Avalere estimates that the percentage of enrollees with no HCCs recorded would increase further under all 3 midyear 2021 scenarios. Similarly, in comparison to the baseline, the number of enrollees with 4 or more HCCs (i.e., enrollees with multiple conditions) recorded decreases from 31% to 26% in the initial 2021 score and down to 22% under the high impact scenario.

This decrease in the number of HCCs recorded likely does not reflect that enrollees’ health improved but could mean that those who are at highest risk if they contract COVID-19 (i.e., those with multiple pre-existing conditions) may be the most likely to defer care. Consequently, their actual health status is not being accurately captured for risk-adjustment calculations.

MA Plans Must Prepare for the Long-Term Impacts of the Pandemic

Because many enrollees with chronic illnesses have not seen their providers during the pandemic, their disease states could worsen, increasing the need for more expensive interventions. At the same time, plans may be facing lower payments and challenges managing care due the ongoing spread of COVID-19. Plans will need to continue to adjust their strategies to account for the longer-term implications of deferred care and use telehealth and other flexibilities to mitigate the impacts.

Methodology

Avalere used MA claims data through June 2020 from our in-house, proprietary Inovalon MORE² Registry® dataset, a large national multi-payer health insurance claims dataset including data on 324 million patients, for enrollees who were continually enrolled in a MA plan between January 2019 and June 2020. Using demographic data (e.g., dual status, date of birth) available as of December 2019, Avalere identified appropriate CMS-HCC model software to calculate the midyear 2020 baseline risk score, the initial 2021 score, and the 3 midyear 2021 score scenarios (i.e., low, medium, and high impact). Avalere adjusted all risk scores for coding intensity and normalization and set the average risk score to 1.000 in the baseline (midyear 2020) by dividing all risk scores by the average calculated risk score for the baseline. Avalere then used its proprietary bid model for each level of risk score to project the overall average impact on payments for 2021. Individuals on dialysis and those who have received a kidney transplant were excluded from the analysis.

Funding for this analysis was provided by Better Medicare Alliance.

To receive Avalere updates, connect with us.