MedPac Report to Congress on Medicare and Medicare Advantage

Rewrite of important points on Medicare and the over payments to commercial MA plans.

This is not the entire Chapter 12 MedPac report to Congress. Although, you may think it is as this a rather long post even after I boiled it down. Why Chapter 12? Simple . . . It talks about the rip-off. . . Medicare Advantage plans. This year’s report is no different than last year’s plan report. In 2024, we are looking at a “Uncorrected MA Coding Intensity increasing Payments to Plans by an estimated $124 Billion through 2022 and is projected to generate nearly $94 billion more in 2023 and 2024.”

This is no holds barred commercial healthcare with no limits as to what is spent in care which can be achieved for far less.

There is more going on with Medicare Advantage Plans than what I reveal here. There are more and more people going back to original Medicare after being with MA plans and the increasing costs. Original Medicare is less costly to patients than Medicare Advantage. It can be used anywhere in the US with no special plans, etc.

Mar24_Ch12, MedPAC_Report_To_Congress_SEC-1.pdf

Chapter 12 reflects revisions the method of estimating coding intensity. I am not going to discuss Chapter 13. For a matter of reference and a deeper dive; it describes the revisions, the research leading to them, and the impact of estimating coding intensity using the revised method relative to the method used in prior MA status reports.

Up to Date Information for 2024 Impact of payment to MA Plans, The Medicare Advantage program: Report to the Congress: “Medicare Payment Policy” MedPac Report Chapter 12 March 2024

“MA rebates for conventional plans have more than doubled since 2018”

For 2024, rebates for conventional MA plans (excluding employer plans and Special Needs Plans [SNPs]) to insurance companies were an average $194 per enrollee per month or $2,329 annually per enrollee. The total is reduced to $2,142 after subtracting plan projections for administrative costs and profit. A slight decrease from the record high $196 per enrollee per month in 2023 (Figure 12-2, p. 370).

When including SNPs, rebates reached a record high of $209 per enrollee per month in 2024. This is a slight increase from $206 per enrollee per month in 2023. These rebates account for 17 percent of plan payments, unchanged from 2023. The average MA rebate among conventional plans has more than doubled since 2018. See Figure 12-2.

“MA rebates in 2024 remain at nearly record levels”

In 2024, the share of plan rebates allocated toward cost-sharing reductions is projected to remain about the same as 2023 levels (Table 12-3). Plans project that $75 per enrollee per month in rebates (39 percent of rebate dollars, unchanged from 2023) will go toward reductions in cost sharing for Medicare services, 1 percent lower relative to 2023.20,21 However, plans reported allocating a slightly higher share of plan rebates to non-Medicare-covered supplemental benefits.

In 2024, plans project that 27 percent of rebates (averaging $53 per enrollee per month) will be used for non-Medicare-covered supplemental benefits. The Medicare Payment Advisory Commission (MedPac) previously reported these benefits often include coverage for vision, hearing, or dental services. However, the non-Medicare supplemental benefits are not necessarily tailored toward populations that have the greatest social or medical needs (Medicare Payment Advisory Commission 2021a). The lack of information about enrollees’ use of supplemental benefits makes it difficult to determine whether the benefits improve beneficiaries’ health (Government Accountability Office 2023).

“Conventional MA Plans project that rebates will be used to reduce cost sharing, reduce Part B and Part D premiums, and offer non-Medicare benefits in 2024.”

Aggregate Medicare payments to MA plans have always been substantially higher than what estimated spending would have been in Fee for Service (FFS) Medicare. Our review of private plan payments suggests that over a 39-year history, the many iterations of full risk contracting with private plans have never yielded aggregate savings for the Medicare program.

Throughout the history of Medicare managed care Medicare Advantage), the program has paid more than it would have paid if beneficiaries had been in FFS Medicare. Evaluations of private plan payment rates under Medicare demonstrations occurring before 1985 found payment rates were 15 percent to 33 percent higher than FFS Medicare spending (Langwell and Hadley 1990).

Between 1985 and 2004, risk adjustment was inadequate and researchers estimated private plan payments were 5 percent to 7 percent higher than FFS Medicare spending in the late 1980s through the mid-1990s (Brown et al. 1993, Medicare Payment Advisory Commission 1998, Newhouse 2002, Riley et al. 1996). Since the introduction of bids and benchmarks in MA payment policy, the Commission started using a prospective method to compare plan benchmarks, plan bids, and the resulting payment to MA plans relative to CMS’s projected FFS spending (standardizing differences in risk scores).

From 2004 through 2006, the Commission found that payments to MA plans were 7 percent to 12 percent higher than FFS Medicare spending. These estimates used FFS projections that included beneficiaries who were not eligible for MA, did not account for differences in diagnostic coding, and did not account for the favorable selection that plans experience with beneficiaries who choose to enroll in MA (see Chapter 13, text box on “MA plan and beneficiary incentives may produce a favorable selection of enrollees,” for more detail on why plans may experience favorable selection).

“Aggregate Medicare payments to MA plans have always been substantially higher than what estimated spending would have been in FFS Medicare”

To account for these differences, the Commission now applies three adjustments on an ongoing basis:

• First, starting with 2016 data, the Commission retrospectively compares actual MA payments with actual FFS spending for beneficiaries who are eligible to enroll in an MA plan (i.e., beneficiaries with both Part A and Part B coverage).

• Second, we have updated our method for estimating coding intensity and have retrospectively applied our estimate beginning in 2007.

• Third, we now account for favorable selection of beneficiaries into MA whereby the risk standardized spending of MA enrollees would be lower than the local FFS average without any intervention from MA plans. We estimated the cumulative annual effect of selection, including the effects of attrition and regression to the mean from 2017 through 2021 (the most recent year of available data).

Figure 12-3 (page 375) shows, since 2007, payments to MA plans have been substantially above the amount FFS Medicare would have spent for the same beneficiaries. Throughout the 18-year period from 2007 through 2024, we estimate that MA payments were at least 9 percent more than FFS spending for comparable beneficiaries in each year. Between 2011 and 2017, relative MA payments decreased from 23 percent above FFS spending to 10 percent above FFS spending (declining benchmarks and favorable selection).

“Coding and Selection have Increased MA Payments above what ‘Spending would have been in FFS'”

Payments to MA (Medicare Advantage) plans are risk adjusted to account for differences in health status. Higher risk scores increase payments to plans for enrollees with higher expected Medicare spending. Risk scores are based on demographic information and diagnoses that plans submit to CMS. Documenting additional diagnosis
codes raises plan enrollees’ risk scores, generating two distinct benefits for MA plans: (1) increasing the monthly payments MA plans receive from Medicare and (2) increasing the rebates plans use to provide extra benefits to enrollees. Plans that document relatively more diagnosis codes have a competitive advantage over other plans.

Figure 12-4 shows the higher payments to MA relative to what spending would have been in dollar terms if enrollees were in FFS. In estimating the payment amount above FFS spending (MA payments for beneficiaries with end-stage renal disease, whom we exclude from all of our analyses and for whom there is no evidence of selection or coding intensity).

Since 2007, we estimate that Medicare has paid $507 billion and will pay $83 billion more for MA enrollees in 2024 than if those beneficiaries had instead been in FFS—a total of $591 billion. Over half (an estimated $338 billion) of the MA payments above FFS spending will have occurred in the last five years—from 2020 through 2024. The higher payments are increasingly driven by coding intensity, which we estimate accounted for the largest share of payments above FFS spending from 2022 through 2024.

“Coding differences increase payments to MA plans in 2024 by $50 billion and continue to generate inequity across plans”

Figure 12-5 shows the impact (2007 through 2022) of differences in coding intensity on MA risk scores relative to FFS and the size of the coding intensity adjustment (the CMS reduced MA risk scores to account for coding intensity).

From 2007 through 2013, MA coding intensity increased MA risk scores by 1.15 percentage points per year more than the FFS risk-score trend. From 2027 to 2021, MA Risk Scores by 1.5 percentage points more per year for 2017 through 2021. Deviations from the typical trend occurred in 2014, 2016, and 2017. We attribute to two factors:

(1) new versions of the risk-adjustment model that were introduced in 2014, 2016, and 2017 reduced the gap in MA and FFS diagnostic coding differences; and

(2) FFS risk scores grew faster (matching or nearly matching MA risk-score growth rates) in 2016 and 2017 than in the previous or subsequent years (likely due to Medicare’s transition from using International Classification of Diseases (ICD)–9 to ICD–10 diagnosis codes in October 2015).

“Est. Impact of Coding Intensity on MA Risk Scores was larger than Coding Adjustment, 2007-2024”

Figure 12-5 shows the impact for Coding Intensity, for 2007 through 2022. The differences in coding intensity on MA risk scores relative to FFS and the size of the coding intensity adjustment (the amount by which CMS reduced MA risk scores to account for coding intensity).

For 2024, MedPac project’s MA risk scores will be about 20 percent above risk scores for comparable FFS beneficiaries. This difference is only partially offset by the coding intensity adjustment that reduced MA risk scores by 5.9 percent. The net effect is a 13 percent increase in MA risk scores due to coding intensity, leading to $50 billion in higher payments to MA plans.

“Between 2007 and 2024, we estimate MA coding intensity will have generated $217 billion in aggregate higher payments to MA plans (Figure 12-6).”

Between 2007 and 2024, we estimate that MA coding intensity will have generated $217 billion in aggregate higher payments to MA plans (Figure 12-6). Between 2007 and 2022, MA coding intensity resulted in $124 billion in increased payments to MA plans. Using our projection of MA coding intensity, we estimate that uncorrected coding intensity in 2023 and 2024 will increase program spending by another $43 billion and $50 billion.

“Uncorrected MA Coding Intensity has increased Payments to Plans by an estimated $124 Billion through 2022 and is projected to generate nearly $94 billion more in 2023 and 2024.”

Table 12-5 illustrates the relationship between coding intensity and rebate amounts using a hypothetical example of three plans covering the same set of enrollees for whom the expected cost of care is the same, at $900 per member per month.

Plans A and Z have an expected risk score of 0.97, and Plan B has an expected risk score of 1.03 due to more aggressive diagnostic coding. All three plans have bids below the risk-adjusted benchmark and provide extra benefits funded by rebates.

Because Plan B has a higher risk score, its rebate is larger than Plan A’s rebate ($52 per month vs. $15 per month). It can offer enrollees more extra benefits. Plan B’s aggressive diagnostic coding effort has therefore given it an unfair competitive advantage over Plan A.

In addition, aggressive coding can result in greater extra benefits than the effect of MA quality bonuses. The higher risk score of Plan B, which has only 3.5 stars, gives it an advantage over bonus-level Plan Z, which has 5 stars: Plan B’s rebate amount is higher than Plan Z’s ($52 per month vs. $49 per month). Thus, by inflating its risk score from 0.97 to 1.03, Plan B can offer more extra benefits than are provided through quality bonuses.

“Figure 12-7 shows coding intensity relative to FFS coding and by MA parent organization, excluding contracts in the Program of All Inclusive Care for the Elderly, special needs plans, and organizations with fewer than 2,500 enrollees.”

Coding intensity varies significantly across MA organizations. About half of organizations (covering 18 percent of MA enrollees) have coding intensity below the 2022 coding adjustment. They are penalized by the adjustment. The other half of organizations (covering 82 percent of MA enrollees) have coding intensity increasing their payment after accounting for the 2022 coding adjustment. These differences demonstrate CMS’s across-the-board adjustment for coding intensity, which reduces all MA risk scores by the same amount, generates inequity across contracts by reducing net revenue for plans with lower coding intensity and allowing other plans to retain a significant amount of revenue from higher coding intensity.

Also found is significant variation in coding intensity across the largest eight MA organizations (covering 77 percent of MA enrollees) from 4.7 percent to 20 percent above FFS levels. Seven of the eight largest MA organizations had greater coding intensity than the 2022 coding adjustment and therefore received a net increase in payment due to aggressive coding practices. These differences are large enough to give MA organizations with higher coding intensity a significant competitive advantage by increasing the size of plan rebates and helping them to attract more enrollees. Our finding that coding intensity varies across MA organizations is consistent with other research assessing variation in coding intensity across or in the use of health risk assessments and chart review, which are key drivers of MA coding intensity (Geruso and Layton 2015, Kronick and Welch 2014, Office of Inspector General 2021).

MA organizations with the highest diagnostic coding relative to FFS are located in California and Florida

In the course of reviewing our coding intensity estimates by MA organization, we found that several organizations with the highest diagnostic coding (relative to FFS) are located in California and Florida. Of the 23 MA organizations offering plans in California and Florida, organizations with a majority of their enrollment in California or Florida, (excluding the 8 largest MA organizations); 10 were among the 21 organizations with the highest coding intensity, including 6 of the top 7 highest coding organizations (Figure 12-8). These six organizations had MA risk scores that ranged from 29 percent higher to 52 percent higher than scores for comparable FFS beneficiaries.

To address why these California- and Florida-focused organizations account for so many of the highest coding organizations, we considered health care plans in California and (to a somewhat lesser extent) Florida have long participated in a form of capitated payment for providers known as the “delegated model.”

Under the delegated model, the responsibility for health care delivery and associated financial risk are delegated by the plan to a medical group or independent physician association. Typically, a plan pays a medical group a risk-adjusted sum per enrollee. This is often calculated as a share of a plan’s total Medicare revenue. Because a plan’s revenue increases when more diagnoses are documented, the capitated payments to providers (determined as a percentage of the plan’s revenue) increase proportionately. In these arrangements, the financial incentive to document more diagnoses is passed on to the medical group, which has direct access to an enrollee’s medical records and diagnostic information.

It could not be confirmed the plans offered by the highest-coding California and Florida organizations use the delegated model. A review of the share of 2021 provider payments that were capitated for 9 of the top 10 such organizations (one organization did not have 2021 data) were made. Of these nine organizations, the share of provider payments capitated was above the national average (33 percent in 2021) for six organizations, including two organizations with provider payments that were almost entirely capitated. Two other organizations had some capitated provider payments but a lower share than the national average, and one organization reported no capitated provider payments (Figure 12-8).

MA plans’ use of health risk assessments to increase diagnosis coding Health risk assessments are provided to Medicare beneficiaries as part of an annual wellness visit. Also for MA enrollees, health risk assessments are often provided during a plan-initiated home visit. Health risk assessments sometimes rely on patient self-reporting of medical conditions, which may result in HCCs based on inaccurate diagnoses, diagnoses which are no longer active (and therefore not eligible for risk adjustment), or diagnoses without sufficient evidence to conform to ICD coding guidelines (Department of Justice 2022).

In 2021, about 6.9 million MA enrollees had a health risk assessment that identified at least one HCC, and a total of 15.0 million unique HCCs were identified through health risk assessments. Of those 3.2 million, the only source for at least one of the HCCs identified (and a total of 5.0 million HCCs [one-third of all HCCs
identified on health risk assessments]) were identified only on a health risk assessment. Seven HCCs each
generated more than $500 million in payments from these assessments, accounting for nearly 60 percent of
all payments generated by health risk assessments.

We found in 2022, diagnostic coding that was associated with only health risk assessments accounted for $13 billion in payments to MA plans, or a little more than 3 percent of all payments to MA plans. About 60 percent of these payments were from health risk assessments conducted as part of an annual wellness visit or initial preventive physical examination, while the rest of these payments were from in-home health risk assessments.

I stopped here as we start to get into Chart Reviews to increase diagnosis coding. More on this later. Unless you do a deep dive into the yearly MedPac report, much of what happens to Medicare resulting from Medicare Advantages goes unknown. As I post this the 2025 MedPac Report is already out.