National APM Metric Overview

National APM Metric Overview and Methodology

APM Metric Overview

This information request involves collecting information from commercial, Medicaid, and Medicare Advantage payers to track the adoption of alternative payment models (APMs). This information will help the LAN understand general market trends and the pace of progress toward alternative payment model adoption across public and private payers. The LAN adapted the CMS payment taxonomy and expanded it by introducing refinements that describe health care payment through the stages of transition from pure fee-for-service to APMs and, ultimately, population based payments. The resulting APM Framework classifies payment models into four categories:

• Category 1—fee-for-service with no link of payment to quality;

• Category 2—fee-for-service with a link of payment to quality;

• Category 3—alternative payment models built on fee-for-service architecture;

• Category 4—population-based payment.

The LAN’s goal is to bring together private payers, providers, employers, state partners, consumer groups, individual consumers, and other stakeholders to accelerate the transition to alternative payment models, specifically shifting 30% of health care payments to APMs by 2016 and 50% by 2018. Spending through APMs will be calculated in the following manner:

Denominator: The LAN Work Group, charged with developing the APM Framework and subsequent metrics, proposed viewing the data with two different denominators. As a result, participating payers will report two denominators:

1. The total estimated in- and out-of-network health care spend (e.g. annual payment amount) made to providers in calendar year (CY) 2015 or most recent 12 months.

2. The total estimated in-network health care spend (e.g. annual payment amount) made to providers in CY 2015 or most recent 12 months.

Numerator: The numerators will generally track to the subcategories and models listed in the APM Framework. Payers will report the total estimated payment amounts of categories 1 through 4 payments made to providers in CY 2015 or most recent 12 months. (See the LAN’s recently released APM Framework White Paper to get a better sense of where these subcategories and models appear in the Framework.)

Targeted Respondents

In order for this exercise to yield a meaningful representation of the private insurance market and gauge real progress towards the Secretary’s goal of moving 30% of the total U.S. health care spending to APMs, we are targeting major insurers that, in total, represent at least 60% of each of these markets. The LAN estimates we will need to approach 200 to 400 payers, with a final respondent count of 200 to 250 most likely. We expect that greater than half of the payers we reach out to will participate, and among participants we will work to ensure greater than 90% completion of the data request. In our pilot of nine payers, all of the participants were able to complete the data request.

The LAN will continually strive to increase the number of payers participating in the LAN who have committed to 1) matching or exceeding the Secretary’s Medicare FFS goal; and 2) submitting data that when aggregated with others’ will categorize how payers are paying providers. Given payer involvement in the LAN events to date (estimated at nearly 100 organizations), CMS is confident a sufficient number of payers will participate in the national data collection effort so that the resulting data would give a strong indication of the national direction.

Neither a census nor a nationally representative sample is necessary for the purposes of this collection. Rather, we have identified a minimum proportion of the market share (i.e., 60 percent) and include as many payers as are necessary to reach that threshold. In effect, this will require recruitment of providers across a variety of regions and states, yielding a cohort of payers that covers much of the U.S. market. For instance, in addition to large national insurance companies such as Humana, Anthem, and UnitedHealthcare; we are targeting state-based insurers like the Blue Cross, Blue Shield plans in many states.

APM Data Collection & Reporting

For the data collection exercise, we will provide participating payers with Excel workbooks for each of the commercial, Medicaid Managed Care, and/or Medicare Advantage lines of business in which they are involved. Payers will respond to one or more workbooks, depending on their lines of business. The workbooks will contain specific instructions for each metric, including the number of beneficiaries involved in APM programs, total spending in these APM programs, and the payer’s overall health care spending in- and out-of-network (i.e., the ‘denominator’) for each line of business. Payers are instructed to also provide their in-network only health care spending. This will be used as an alternative denominator for comparison purposes. After the payers submit these data, a cross checking of responses will help to identify whether any data appear to be outliers and require correction. We will aggregate all data submitted which, together, will quantify the dollars flowing through the categories and subcategories specified in the Framework.

At the end of each data collection and after aggregation and analysis, the LAN plans to report that X # of payers participated in the data collection for a specified market segment, X # of payers represent Y beneficiaries, which is Z% of the total covered life population of the particular market segment within a given calendar year. Among this proportion, A% of payment is in category 1, B% in category 2, and C% in category 3 and D% in category 4. A separate percentage (E%) will be reported that combines categories 3 and 4. Subsequent data collections (at least annually) will provide information on the change in the mix of payer respondents each year, and the number of new payers reporting their APMs according to the Framework.

Approach for Methodological Standardization and Improvement

The LAN has convened a cohort of payers to participate in the LAN Payer Collaborative. The purpose of this Payer Collaborative is to educate payers on the APM Framework White Paper to ensure alignment on APM definitions across payers and to increase understanding with respect to how the LAN proposes to measure progress towards APM adoption. From the initial cohort of 20 payers participating in the LAN’s Payer Collaborative, nine volunteered to participate in a data collection pilot intended to gather feedback on the proposed metrics, test the feasibility of the proposed survey instrument, develop more accurate collection burden estimates, and finalize an approach for a full-scale nationwide data collection effort of private and public payers. Among the nine pilot participants, seven reported on their commercial business, four on their Medicaid business, and six on their Medicare Advantage (MA) business lines.

The pilot began on February 18, 2016. Payers were provided draft metrics for each market segment, had the opportunity to ask questions, and received a Frequently Asked Questions (FAQ) document to address common concerns and ensure consistency in data reporting. Feedback from pilot participants so far has included discussions about the metrics, the survey format, level of effort, barriers and more. Key ‘lessons learned’ from the pilot include:

• Prioritization of resources presents the most significant barrier to completing the data request. Payers have many other projects that need staff attention, including other external requests that are mandatory and, thus, a higher priority than the LAN national data collection effort. To address this reality and lessen the payer burden, to the extent possible we are striving to align the LAN’s “asks” with other substantially similar Medicare or Medicaid data requests, state level requests, and private sector requests. For example, CMS plans to start requiring Medicare Advantage (MA) plans to report the percentage of their payments to providers that are tied to APMs in 2017. To ensure alignment around categorization of APMs and measurement methods across CMS and private health plans (to include MA plans), CMS referred MA plans to the LAN’s APM methodology in the final call letter. State Innovation Models (SIM) are also requesting for SIM Round 2 states to report on APM metrics at the state level, so SIM states are being advised to refer their health plans to the LAN’s national collection effort to increase methodological alignment. Lastly, the LAN is collaborating with health plan trade organizations who are currently conducting similar APM surveys to ensure alignment. In this way, plans will only have to go through the internal process of collecting APM data once, though the same metric can be used for multiple reporting requests. In addition to lessening the burden, this approach will enable the use of a consistent methodology.

• The alternative payment models within the categories are not linked to universal definitions. The lack of consistent definitions has led to some payer confusion, which can lead to inconsistency in payment classifications. To solve this issue for the nationwide data collection effort, we have developed a definitions document that defines APM terms and provides examples.

• We encountered methodological inconsistencies in payer reporting. For example, providing annualized (point in time) versus year to date data, classifying payments by attributed providers versus members, categorizing payments in one method when several are used (hybrid programs), counting Blues members and payments across states, etc. These points will need to be clarified prior to the national reporting process to ensure that there is consistency across payers.

• Errors were identified in the formulas in the metrics/survey instrument, and some respondents found the Excel workbooks challenging to navigate. Formula errors have been corrected, and we have developed an improved Excel file that is easier to use for the participants in the national effort.

• Plans already participating in similar data requests are better equipped to report data than the payers that have had no previous experience. Additionally, some plans are able to query data systems in an automated way, while others have to query their systems manually. Despite experienced payers being better equipped, they have had to adjust their data queries to match the LAN framework, which requires additional staff and resources. Both types of payers and their different challenges lead to the need for an extended timeline to complete the data request. We realize that all payers that participate will experience some challenges in generating data, but our extended time line for submission (8 weeks) and future LAN data requests can mitigate this process.

2