Microsoft Word - BCBSA MHPAEA Technical Release Comments

From: Silver, Dana

To: EBSA MHPAEA Request for Comments

Cc: Jones, Jennifer

Subject: submission of BCBSA comments on MHPAEA Technical Release 2023-01P

Date: Tuesday, October 17, 2023 2:16:42 PM

Attachments: BCBSA MHPAEA Technical Release Comments_10.17.23.pdf

CAUTION: This email originated from outside of the Department of Labor. Do

not click (select) links or open attachments unless you recognize the sender

and know the content is safe. Report suspicious emails through the "Report

Phishing" button on your email toolbar.

To whom it may concern:

Attached are the Blue Cross Blue Shield Association’s (BCBSA) comments on the

“Request for Comment on Proposed Relevant Data Requirements for Nonquantitative

Treatment Limitations (NQTLs) Related to Network Composition and Enforcement Safe

Harbor for Group Health Plans and Health Insurance Issuers Subject to the Mental Health

Parity and Addiction Equity Act” (Technical Release).

We look forward to continuing to work with the Departments on this issue as well as

additional ways to ensure all Americans have affordable access to high-quality MH/SUD

services. If you have questions, please contact Jennifer Jones at 202.942.1269 or

[email protected]

Best regards,

Dana Silver

Manager, Legislative and Regulatory Policy

Blue Cross Blue Shield Association

(202) 649-1777

[email protected]

October 17, 2023

The Honorable Julie Su

Secretary of Labor

200 Constitution Avenue, N.W.

Washington, D.C. 20210

The Honorable Janet Yellen

Secretary of the Treasury

1500 Pennsylvania Avenue, N.W.

Washington, D.C. 20220

The Honorable Xavier Becerra

Secretary of Health and Human Services

200 Independence Avenue, S.W.

Washington, D.C. 20201

Submitted via email to mhpaea.rfc.ebsa@dol.gov

RE: Request for Comment on Proposed Relevant Data Requirements for Nonquantitative

Treatment Limitations (NQTLs) Related to Network Composition and Enforcement Safe

Harbor for Group Health Plans and Health Insurance Issuers Subject to the Mental Health

Parity and Addiction Equity Act

Dear Secretaries Su, Becerra and Yellen,

The Blue Cross Blue Shield Association (BCBSA) appreciates the opportunity to provide

comments on the Proposed Rule on “Request for Comment on Proposed Relevant Data

Requirements for Nonquantitative Treatment Limitations (NQTLs) Related to Network

Composition and Enforcement Safe Harbor for Group Health Plans and Health Insurance

Issuers Subject to the Mental Health Parity and Addiction Equity Act” (Technical Release).

BCBSA is a national federation of 34 independent, community-based and locally operated

BCBS companies (Plans) that collectively cover, serve, and support 1 in 3 Americans in every

ZIP code across all 50 states and Puerto Rico. BCBS Plans contract with 96% of hospitals and

95% of doctors across the country and serve those who are covered through Medicare,

Medicaid, an employer, or purchase coverage on their own.

We are committed to robust mental health and substance use disorder (MH/SUD) access.

BCBS Plans have made considerable progress toward and continue to execute multifaceted

strategies to build robust MH/SUD benefits. However, we have concerns that the proposed

network composition NQTL is attempting to solve for variables that are not wholly within health

1310 G Street, N.W.

Washington, D.C. 20005

202.626.4800

www.BCBS.com

plans’ control and, as a result, will either yield persistent industry-wide results that the

Departments may view as indicators of noncompliance or force changes to be compliant that

are not in the best interests of patients.

Addressing access primarily through the network composition outcomes proposed in the

“Requirements Related to the Mental Health Parity and Addiction Equity Act Proposed Rule”

assumes the primary driver of the access challenges is health plans not contracting with

MH/SUD providers. This does not account for a number of factors, but two in particular. First,

primary care is a consistent, and growing, source of care for MH/SUD needs which is not

captured by a comparison of medical/surgical (M/S) to MH/SUD services. Second, the dearth of

available MH/SUD providers is well documented. The Health Resources and Services

Administration reports that there are 6,767 Mental Health Professional Shortage Areas

containing 166 million people.

This shortage of workers is not evenly distributed as 55 percent

of U.S. counties, all rural, have no practicing psychiatrists, psychologists or social workers.

Another study found that 77 percent of counties had a severe shortage of mental health

workers, both prescribers and non-prescribers, and 96 percent of counties had some unmet

need for mental health prescribers.

In addition, the workforce was heavily impacted by the

COVID-19 pandemic along with many other medical specialties. As a result, there is a

misconception that there is a broad universe of providers who are interested in joining health

plan networks but are being prevented from doing so due to administrative barriers and

reimbursement rates.

As BCBSA notes in its response to the Proposed Rule, health plans have done significant work

to bolster reimbursement rates and to reduce administrative burdens. Yet, qualified MH/SUD

providers still have unique incentives not to participate. The best example is the difference in

average provider practice size between M/S and MH/SUD. Many M/S practices are large groups

of providers and are increasingly becoming more consolidated, whereas MH/SUD providers

tend to comprise more single-proprietor practices where the infrastructure costs required to

contract with, bill and receive payment from third-party payers may be difficult to justify. In

addition, many of these providers are accustomed to billing their patients directly and are not

working towards learning different mechanisms for reimbursement, particularly as their practices

are full regardless of whether they accept insurance. These factors and others discourage

network participation for the limited number of non-contracted providers in the country today.

We are concerned that the Departments are relying on outdated information as justification for

the proposals in the Technical Release. In particular, the Departments cite a 2019 Milliman

analysis that purported to find widening disparities in network use and provider reimbursement

for MH/SUD and M/S to justify the proposed outcomes-based comparisons. The Milliman report

was published in 2019 and the most recent data it cites is from 2017. More recent data tells a

different story. From 2017 to 2021, commercial health insurance payment rates for MH/SUD

providers are higher than Medicare payment rates and generally growing more rapidly (see

Figure 1).

https://data.hrsa.gov/topics/health-workforce/shortage-areas

Substance Abuse and Mental Health Services Administration, Report to Congress on the Nation’s Substance Abuse

and Mental Health Workforce Issues, January 24, 2013.

Source: unpublished estimates using Markets can claims data by the Employee Benefit Research Institute, 2023.

The ratios fall in 2021 but that was due to a one-time adjustment in Medicare payment rates that

went into effect in 2021 during the height of the COVID-19 pandemic. (Ratios for non-MH/SUD

providers also fell in 2021.) We expect payment data in future years to show continued

improvement in commercial payment rates relative to Medicare.

Moreover, commercial payment rates for MH/SUD providers increased at a higher rate than

payment rates for non-MH/SUD providers over the 2017-2021 time period (see Figure 2).

Source: unpublished estimates using Marketscan claims data by the Employee Benefit Research Institute, 2023.

0.92

0.94

0.96

0.98

1.00

1.02

1.04

1.06

1.08

1.10

1.12

1.14

2017 2018 2019 2020 2021

Commercial Payment Rate Divided by

Medicare Payment Rate

Figure 1. Ratio of Commercial Payment Rates to Medicare

Payment Rates for MH/SUD Providers, 2017-2021

Psychiatry Average Psychology Average Supportive Therapy Average

100

105

110

115

120

125

2017 2018 2019 2020 2021

Payment Index (set to 100 in 2017)

Figure 2. Growth in Average Mental Health Payment Rates

Compared with Average Payments to Other Provider Types

Psychiatry Psychology Supportive Therapy Non-Psych

Looking at more recent data also indicates that the share of commercial premiums going

towards outpatient mental health treatment has increased. That share has increased because of

increased utilization as well as growth in payment rates. According to an analysis by the

Employee Benefit Research Institute, year-over-year increases in MH/SUD provider

reimbursement rates are outpacing medical/surgical (M/S).

Comparing common clinic visit

payment rates over the 2017-2021 period, MH/SUD payment rates increased by an average of

about 18.5%, nearly double the payment increases for non-mental health providers of 9.5%.

Given that utilization was increasing, insurers responded by increasing payment rates to attract

more providers. While overall mental health spending was relatively flat, this was because the

average price of mental health drugs fell over the period as more generics came to market. This

reduction in drug spending is important to consider in making apples-to-apples comparisons of

changes in spending for MH/SUD services.

In addition, the network composition NQTL outcomes, as proposed, imply that a patient would

only select an out-of-network (OON) provider if they had no option for an in-network provider

and that patients will always select the provider that is geographically closest to their home or

work. While not always the case, these assumptions are somewhat reasonable for M/S

services, but less so for MH/SUD services. Consumers’ preference for staying close to home or

traveling out of area varies widely by type of service, which biases comparisons of OON use for

M/S and MH/SUD services. For example, patients needing skilled nursing facility (SNF)-level

care generally prefer to be close to home. But due to the perceived or real stigma associated

with substance use, patients needing SUD treatment often prefer to go out of area for treatment,

particularly if their plan covers OON services. In addition, patients are much more likely to try to

find a provider who aligns with their personal demographics (e.g., race, gender identity,

ethnicity) than for M/S services, particularly if the individual comes from a historically

marginalized population. Unfortunately, the current MH/SUD provider workforce is less diverse

than other M/S specialties and MH/SUD providers with diverse backgrounds are not evenly

dispersed across the country. As a result, patients make the choice to travel farther to access

their preferred providers where they are available. Also, the use of telehealth for the provision of

MH/SUD services is increasingly common and, while not reflective of the patient’s true

“commute,” may increase the impression that patients are more commonly accessing providers

who are a greater distance away than seen for M/S services. While we are supportive of the

increased use of telehealth for behavioral health services, we do have concerns with the

proliferation of internet-based behavioral health companies that market directly to consumers

and often refuse to partner with health plans. These companies can appeal to consumers who

are not sure how to access care but steer them away from in-network services. Figure 3 shows

the dramatic increase in the provision of MH/SUD services via telehealth from January 2020

through November 2021.

That trend appears to have generally continued to the present day.

Paul Fronstin and M. Christopher Roebuck, “Use of Health Care Services for Mental Health Disorders and Spending

Trends”, EBRI Issue Brief, no. 569, Employee Benefit Research Institute, September 2022

Norah Mulvaney-Day, David Dean, Jr., Kay Miller, and Jessica Camacho-Cook, "Trends in Use of Telehealth for

Behavioral Health Care During the COVID-19 Pandemic: Considerations for Payers and Employers," American

Journal of Health Promotion, 2022 Sep; 36(7): 1237–1241. doi: 10.1177/08901171221112488e.

We suspect that the mental health provider community will strongly support the comparisons

proposed in the Technical Release. We note that the proposed comparisons included in the

Technical Release are largely based on the comparisons

advanced by a number of mental

health advocacy organizations as part of the Path Forward. While some employer groups have

been interested in evaluating such measures as part of voluntary efforts to evaluate mental

health coverage, we have significant concerns with using these measures as an indicator of

non-compliance because the reasons for differences in outcomes are multifactorial and cannot

be controlled by health plans. Whether raising reimbursement rates will lead to increases in the

supply of MH/SUD providers willing to accept insurance depends on their own price elasticity of

supply (more specifically on the extensive margin—their network participation decision). There

is little empirical evidence on this subject, but health plan experience suggests relatively

inelastic supply, as discussed in our comments on the Proposed Rule. This raises the potential

that increasing reimbursement may have a muted effect on some of the measures that the

Departments are considering in the Technical Release.

Based on the numerous factors that lead to differences in provider network participation for

MH/SUD and M/S benefits, differences for some of the comparisons that the Departments are

considering, such as OON utilization, are unlikely to improve substantially until there is an

adequate supply of MH/SUD providers such that the providers need volume from contracting

with health plans to support their practices. Rebalancing supply and demand for MH/SUD

services will require years and a concerted effort in terms of government and institutional

investment to improve the pipeline of providers.

See the Model Data Request Form (https://mhtari.org/Model_Data_Request_Form.pdf) produced by the Bowman

Foundation and promoted as a resource by the Path Forward (https://pathforwardcoalition.org/resources/).

Another factor that should be considered when attempting to compare MH/SUD and M/S

reimbursement are differences in practice costs. Practice expenses vary widely among different

types of health care providers. Although somewhat dated, the American Medical Association

(AMA) published practice expenditures for professional services in 2006 which found that

expenditures for clinical social workers, clinical psychologists and psychiatrists were among the

lowest of all clinical specialties.

As illustrated in Table 1 below, practice costs were substantially

higher for the most common types of M/S providers of primary and specialty care noted in the

Milliman report.

Table 1: Hourly Practice Expense by Specialty

Specialty Total Expense per Hour

Clinical Social Workers $18.33

Clinical Psychology $21.52

Psychiatry $32.10

Internal Medicine $110.62

Pediatrics $111.31

Family Medicine $119.19

Obstetrics/Gynecology $149.02

Orthopedic Surgery $162.94

Dermatology $264.88

In establishing the Medicare Physician Fee Schedule (PFS), the Centers for Medicare &

Medicaid Services (CMS) incorporates practice costs and other factors to establish appropriate

payment rates for different types of providers in Medicare. These payment rates vary among

MH/SUD providers and M/S providers. As noted in our comments on the section of the

Technical Release concerning reimbursement rates, we recommend that the Departments

develop similar adjustments to enable appropriate comparisons between provider types.

Similar differences in the costs for inpatient services would also be evident. The costs of running

an acute care hospital are substantially different than an inpatient mental health facility in terms

of the physical space, equipment and staffing. There is significantly more equipment and

machinery for the provision of inpatient M/S services than for MH/SUD treatment, all of which

require resources to rent or purchase, space to store and use, and medical staff to operate.

To better reflect the elements of network composition, and utilization as a proxy, that health

plans have an ability to influence, we recommend the Departments:

 Not finalize outcomes reporting on OON utilization as there are many drivers of

OON utilization other than network adequacy and health plans have incomplete

https://www.ama-assn.org/system/files/practice-expense-component.pdf

information on OON usage. Use of this standard would create an unlevel playing field

across types of health plans that have access to varying levels of OON claims data and

entail significant burdens on the industry. Based on the numerous factors that lead to

differences in OON utilization for MH/SUD and M/S benefits and the long-term efforts

needed to address imbalances in the market for MH/SUD services, we ask that the

Departments not move forward with this measure at this time.

 Not require plans and issuers to collect and evaluate data on time and distance

standards for MH and SUD providers to compare against M/S providers as

equivalency across these types of services is not reflective of the availability of providers

in many of the communities Plans serve. This proposal and other components of the

network composition NQTL proposed standards could force health plans to sacrifice

provider quality and performance standards in order to ensure compliance.

 Recognize the role of telehealth in increasing access to MH/SUD care. There is no

standard to account for the role of telehealth in providing access to patients when

calculating underlying time and distance standards. This is a disservice to the critical

function telehealth providers serve in bridging gaps in care. It also means that the

calculations are skewed towards the provision of brick and mortar-based services and,

particularly given the provider shortages, establishes impossible comparisons to M/S

providers who are more evenly distributed across the country and less likely to provide

care via telemedicine.

 Develop reasonable comparators for reimbursement rates for MH/SUD providers

that can be fairly equated to M/S providers. In addition, reimbursement rates are not

monolithic – they vary by reimbursement methodology, negotiations, acuity of patient

load, training of the provider and a number of other factors can drive variation across

and within specialties. These comparators should be adjusted for differences in practice

cost across provider type.

 Provide symmetry in compliance whereby if there are not material differences in

network composition outcomes data, the plan or issuer is compliant with the NQTL

requirements. If a plan or issuer is deemed noncompliant based solely on outcomes data

then the same should be true of being deemed compliant.

 Provide adequate time for public comment on the methodology for any

comparisons or outcomes the Departments move forward. The Departments ask

dozens of important and complicated questions in the Technical Release that many

plans and issuers will not have adequate time to comment on appropriately or generate

meaningful descriptive data on to inform their recommendations. Given the highly

technical nature of these questions and the potential for comparisons to be used as

indications of noncompliance, we recommend that the Departments issue their proposed

approach for any comparisons through a white paper process that details their proposed

methodology prior to proposing standards in rulemaking. This process has been used

successfully by the Departments in recent years and would allow the public to more

meaningfully comment on these proposals.

 Create a safe harbor for plans and issuers reimbursing MH/SUD providers at or

above Medicare reimbursement rates. The rates CMS pays these providers under the

Medicare program are rigorously evaluated, subject to substantial oversight, and

objectively incorporate differences in cost of overhead. While we are not supportive of a

mandate for health plans to pay Medicare rates, we suggest this safe harbor as a

reasonable way to compare commercial payment rates to a defined and publicly

available benchmark.

We look forward to continuing to work with the Departments on this issue as well as additional

ways to ensure all Americans have affordable access to high-quality MH/SUD services. If you

have questions, please contact Jennifer Jones at 202.942.1269 or [email protected]m.

Sincerely,

Kris Haltmeyer

Vice President, Legislative and Regulatory Policy

BCBSA Detailed Comments on Technical Release 2023-01P

RELEVANT DATA TO BE COLLECTED AND EVALUATED WITH COMPARATIVE

ANALYSES FOR NQTLS RELATED TO NETWORK COMPOSITION

Issue: Out-of-Network Utilization Data

The Departments are considering requiring plans and issuers to collect and evaluate data on

the OON utilization for M/S, MH, and SUD benefits for the following types of items and

services: (1) Inpatient, hospital-based services; (2) Inpatient, non-hospital-based services,

including inpatient rehabilitation facilities and skilled nursing facilities for M/S items and

services, and non-hospital-based inpatient facilities and residential treatment facilities for

MH/SUD items and services; (3) Outpatient facility-based items and services, including

physical, occupational, speech, and cardiovascular therapy, surgeries, radiology, and

pathology, services for M/S care provided in an outpatient facility setting; and intensive

outpatient and partial hospitalization services for MH conditions or SUDs in an outpatient

facility setting; (4) Outpatient office visits; and (5) Other outpatient items and services.

Recommendation:

BCBSA recommends that the Departments not require plans and issuers to collect and

evaluate data on OON utilization for M/S, MH, and SUD benefits.

Rationale:

Plans and issuers cannot accurately report OON utilization, particularly for group health plans

that do not offer OON coverage. In addition, BCBSA does not agree that high OON utilization of

MH/SUD services necessarily signifies a network access deficiency, especially as many

exceptions are granted to treat OON usage as in-network services. Rather, it can reflect a

contracting preference on the part of behavioral health providers, as described earlier, and/or a

greater preference on the part of enrollees to see their choice of MH/SUD provider even if the

provider is OON, compared to choosing a M/S provider OON.

Patients may want to ensure that their M/S providers, spouses and other family members are

not privy to their MH/SUD treatment. As mentioned in BCBSA’s Proposed Rule response and

earlier in our Technical Response, patients needing SNF-level care generally prefer to be close

to home. But due to the perceived or real stigma associated with substance use, patients

needing SUD treatment often prefer to go out of area for treatment, particularly if their plan

covers OON services or if they live in small or geographically isolated communities.

While

some may discount the role of patient preference in OON usage, we know that the relationship

between patient and provider is especially critical in behavioral health and, unfortunately, there

is a significant shortage in the behavioral health workforce for providers from minority

communities or with otherwise diverse backgrounds. Coupled with shortages and contracting

preferences among MH/SUD providers, this can in fact promote greater OON usage.

We use this comparison as access to SUD treatment commonly surfaces in network adequacy discussions and,

while this example highlights differences between M/S and MH/SUD utilization preferences, it also highlights that

comparisons in provider types are not always “one-to-one” or as simple as it can seem on the surface – a SNF is in

many ways not comparable to a SUD treatment center, but it is the best comparison to a SUD treatment center.

As a result, any comparison between OON utilization between M/S and MH/SUD is likely a false

equivalency. In addition, as the Departments acknowledge, health plans only have visibility into

the OON claims that are submitted to the plan by the provider or patient. OON utilization data is

likely highly skewed based on the cost of the service provided (or any number of variables)

which could further complicate any comparison to in-network utilization or between M/S and

MH/SUD.

The question the Departments raise on how to measure OON claims for HMOs, EPOs, and

closed network plans points to the limitations of using this comparison as an indicator of

compliance. Such plans are unlikely to cover claims for non-participating providers. In addition,

individuals with higher deductible plans with health savings accounts will be less likely to file

claims for outpatient services and drugs than PPO plans with lower deductibles. As such, this

standard will provide an unlevel playing field across plan types with regard to compliance that

the Departments should avoid.

Recommendation:

While we ask the Departments not to finalize this measure, if the Departments do move forward

we recommend guidance clarifying the following questions related to how to calculate OON

utilization:

 Confirm that zero-dollar claims and Medicare cross-over payment should be excluded

from the reporting.

 Confirm that calculations should be based on claims with a MH/SUD diagnosis, rather

than a service provided by a MH provider (i.e., claims from M/S providers administering

MH/SUD services be included). This is critical given that health plans are investing in

integration of medical and behavioral health services with primary care providers.

 Confirm that the data should not be reported by place of service or geographic location.

More granular reporting would impose an unnecessary burden on plans and issuers.

 Clarify OON utilization should be provided in terms of percentage of claims, not number

of claims or dollar amounts. In addition, health plans should be permitted to make good

faith efforts to determine which claims or line items are for MH/SUD or M/S services.

 Confirm that plans are permitted to review all services for fraud, waste and abuse

without impacting data or parity outcomes.

 Provide adequate time for the run-out of MH/SUD claims. Submission of claims for OON

MH/SUD are often submitted in paper form by members given that OON mental health

professionals often refuse to assist consumers with filing claims. These claims

submissions typically lag behind claims submitted directly by providers.

 Align reporting to prior to the last calendar year and allow health plans to account for

OON claims submitted after the end of the applicable year for relevant dates of service.

 With regard to the issue the Departments raise about the fact that this comparison could

not be equally applied to certain plans (e.g., closed panel HMOs), consider whether

other metrics finalized by the Departments would be reasonable alternatives that would

not disadvantage PPOs or other popular products that provide out-of-network coverage.

 Given the many differences that could lead to higher OON claims for MH/SUD, consider

appropriate adjustments or thresholds if this comparison is used as an indicator of non-

compliance.

Rationale:

Guidance on these technical questions is necessary to provide health plans with critical

information on how to appropriately implement the outcomes reporting proposed by the

Departments. Otherwise, there will likely be significant inconsistencies in how plans and

issuers are interpreting the requirements. This would create uncertainty and additional

burden on all parties to understand and comply with the Departments’ expectations.

Furthermore, it would meaningfully limit any interpretation of the data as it would be difficult,

if not impossible, to fully understand what was being represented and compared.

Please also note that Plans did not have adequate time to answer the questions related to

the relevant data that plans and issuers would be required to collect and evaluate for NQTLs

related to network composition (e.g., data on the OON utilization for M/S, MH, and SUD

benefits for the indicated types of items and services). Data analysis would be required to

determine whether these items and services provide for a reasonable comparison. Please

see our recommendation later in this response regarding a white paper process to solicit

additional feedback in advance of finalizing any requirements.

Issue: Percentage of In-Network Providers Actively Submitting Claims

For this data element, the Departments contemplate requiring plans and issuers to collect and

evaluate data for different types of providers (and make comparisons between a type of

MH/SUD provider and an analogous type of M/S provider).

The potential types of providers that the Departments are considering include: (1) MH/SUD

providers including child psychiatrists and psychologists; other psychiatrists and psychologists;

psychiatric nurse practitioners; master’s level MH counselors, marriage and family therapists,

independent clinical social workers, and advanced social workers; non-master’s level MH

counselors; board certified SUD addiction medicine physicians; and other non-physician SUD

professionals; (2) M/S providers including cardiologists; neurologists; orthopedists;

pediatricians; other specialty physicians; physician primary care providers (other than

pediatricians); non-physician primary care providers; and non-physician specialty providers.

Recommendation #1:

BCBSA recommends that any definition of “relevant data” for network composition standards

not include data on providers accepting new patients.

Rationale:

Providers’ willingness to accept new patients can fluctuate according to a provider’s own

considerations, including capacity to offer a specific level of care or triaging based on what they

know about the patient (e.g., type of condition, acuteness of need) and how they were sourced

(e.g., referred by another provider, cold calling the office). Given these variables, providers

generally do not provide information to health plans on how they determine whether to accept a

new patient and issuers do not have access to this data unless a provider communicates it. As

such, any health plan reported data on this metric is a limited representation of access. This is

likely one of the reasons existing state network adequacy requirements generally do not include

a metric of this kind, but rather health plans work to provide high level information on this

variable in health plan provider directories.

Recommendation #2:

If the Departments move forward in finalizing this proposal, BCBSA recommends that the

Departments maintain plans’ and issuers’ ability to determine the most appropriate provider

types for this reporting.

Rationale:

There is significant variation in how providers and states define provider types and

licensure/certification. Health plans maintain data, such as who is accepting new patients,

based on these localized definitions. Developing a federal standard for these definitions,

rather than deferring to existing local definitions, would force the retrofitting of data which

would likely skew some of the analyses. To ensure the most accurate data which reflects the

market under review, a plan should be allowed to defer to how a state or provider defines

their own provider type/license/certification/etc.

Recommendation #3:

If the Departments do finalize a specific list of provider types for reporting on this data,

BCBSA recommends that the list of providers for which health plans report on the proportion

of providers accepting new patients be clarified to conform with how health plans maintain

data on providers. To do so, we encourage the Departments to consult with industry experts

to ensure the taxonomy is workable.

Rationale:

A number of the proposed providers, specifically those identified in the exclusions list in this

recommendation, are not defined by current provider type or provider taxonomy, i.e.,

specialty codes, and therefore are not independently identifiable in health plan databases.

Specifically, the “provider types” proposed are either not feasible for reporting (or performing

comparisons on) or require clarification for the following reasons:

For MH/SUD providers:

 “Other” psychiatrists and psychologists: Psychiatrists and psychologists have

provider type and specialty codes which all credentialed psychiatrists and

psychologists are aligned to, respectively. Plans do not maintain a list of “other

psychiatrists” or “other psychologists” that can be reported on.

 Master’s level MH counselors: Education level (i.e., “master’s level”) is a data

attribute that is not part of the taxonomy code set. States typically require certain

education levels for licensure, but educational level is not relevant for assigning most

taxonomy codes. As such, all MH counselors are coded using the same taxonomy

codes and cannot be separated based on educational level.

 Independent clinical social workers: There is a taxonomy code for Clinical Social

Workers, but not for “independent” clinical social workers. Whether they are

independent or part of a provider group is not a factor considered in the taxonomy

code for these providers.

 Advanced social workers: BCBSA is unsure what the Departments mean by

“advanced.” However, there are applicable taxonomy codes for licensed clinical

social workers.

 Non-master’s level MH counselors: As mentioned above, educational level (i.e.,

“non-master’s level”) is outside of the taxonomy code set. States typically require

certain educational levels for licensure, but educational level is not relevant for

assigning most taxonomy codes.

 Board certified SUD addiction medicine physicians: Physicians are board

certified in Addiction Medicine, but there is not an American Board of Medical

Specialties (“ABMS”) board certification for “SUD Addiction Medicine.” Whether the

physician is board certified in their contracted specialty is a separate data element

and not one used to identify provider specialty.

 Other non-physician SUD professionals: There are taxonomy codes with SUD for

non-physicians, but it is not clear which professionals are of interest to the

Departments.

For M/S providers:

 Non-physician primary care providers: Most health plans only capture Primary

Care versus Specialists for Managed Care products, and mainly for physicians and

some limited nursing or physician assistant practitioners (i.e., HMO or POS product

type).

 Non-physician specialty providers: Most health plans only capture Primary Care

versus Specialist for Managed Care products (i.e., HMO or POS product type).

To reflect these nuances, we recommend the Departments clarify these provider types to

conform with how health plans maintain data on providers. To do so, we encourage the

Departments to consult with industry experts to ensure the taxonomy is workable.

Issue: Time and Distance Standards

The Departments are considering requiring plans and issuers to collect and evaluate data on

the percentage of participants, beneficiaries and enrollees who can access, within a specified

time and distance by county-type designation, one (or more) in-network providers within

MH/SUD provider categories (including psychiatry, inpatient care, residential treatment, mobile

crisis units, opioid treatment providers, child and adolescent providers, geriatric providers,

eating disorder providers and autism spectrum disorder providers) and one (or more) in-

network providers within certain M/S provider categories.

Recommendation #1:

BCBSA recommends that the Departments not require plans and issuers to collect and

evaluate data on time and distance standards for MH and SUD providers to compare against

M/S providers.

Rationale:

Setting an expectation that health plans meet the same time and distances outcomes as

seen for M/S providers is not reflective of the limited number of MH/SUD providers in many

communities. As the Departments note in the Proposed Rule, there is a severe shortage of

mental health providers, particularly in rural areas and communities of color. More than one

third of Americans live in areas with far fewer mental health

specialists than the minimum

needed to meet the need.

Furthermore, there is no standard to account for the role of telehealth in providing access to

patients. Telehealth has been a valuable tool across health care, but specifically has been an

invaluable tool to expanding access to MH/SUD services in underserved areas. Behavioral

health is particularly well suited to telehealth for many conditions and many patients now rely

on this technology for their care. In fact, of the Plans that reported, 50% or more of all their

behavioral health outpatient services are provided via telehealth. This statistic is reinforced

by Oliver Wyman who cites that 30-50% of MH/SUD services are provided via telemedicine

today. This is a significant, and growing, source of these services. However, these providers

are not accounted for in the calculations underlying time and distance standards. This is a

disservice to the critical function these providers serve in bridging gaps in underserved

communities and for patients looking for a provider that is not easily accessible in their

community for various reasons (e.g., specific race, ethnicity, gender identity, other

demographic). However, it also means that the calculations are skewed towards the

provision of brick and mortar-based services and, particularly given the provider shortages,

establishing impossible comparisons to M/S providers who are more evenly distributed

across the country and less likely to administer via telemedicine.

As discussed in BCBSA’s Proposed Rule response, we are concerned that this outcome and

other components of the network composition NQTL proposed standard will lead to an

impossible choice for health plans – ensure compliance with the standards by accepting

lower quality providers into networks or retain existing quality standards and run counter to

federal and state regulators. The reality that health plans are facing is that the quality

providers that Plans encourage to come in network do not express interest in doing so and

those that do join bring limited additional capacity for patients. Given that the pool of

available providers is woefully insufficient to meet the needs of patients, health plans will be

forced to consider the providers that were excluded from contracting based on quality and

performance concerns. This outcome runs counter to the goals of the Departments to

support access to drive better health for Americans.

Recommendation #2:

If the proposal is finalized, BCBSA recommends that the time and distance standards not

include the following proposed MH/SUD provider types, or, at a minimum, that the

Mental health disorders “involve changes in thinking, mood, and/or behavior.” https://www.samhsa.gov/find-

help/disorders

https://data.hrsa.gov/topics/health-workforce/shortage-area

Per Oliver Wyman, Individuals seeking MH/SUD treatment are significantly more likely to utilize telehealth than

those seeking other M/S services (e.g., 30-50% of MH/SUD through telehealth vs. 5-10% of M/S).

Departments clarify these provider types: inpatient care; mobile crisis unit; opioid treatment

providers; geriatric providers; eating disorder providers; and autism spectrum disorder

providers.

Rationale:

The specified proposed provider types are not provider types BCBS Plans typically define and

do not have associated taxonomy codes. Below, we explain why certain “provider types” are

not appropriate to report on (or perform comparisons on), or require clarification from the

Departments:

 Inpatient care: This is not a provider type, but rather a site of care. Many provider

types with valid taxonomy codes provide inpatient care. Rather than defining providers

based on their sites of care, we recommend the Departments only include traditional

provider types as part of the required analyses.

 Mobile crisis unit: Generally, payers do not enroll providers as a mobile crisis unit.

Instead, the providers who staff these units are enrolled according to their traditional

provider type designation (e.g., psychiatrist, licensed clinical social worker).

 Opioid treatment providers: This is not a discrete provider type, but rather a

specialization across providers.

 Geriatric providers: BCBSA recommends specifying the provider type as Geriatric

Medicine Providers, which has an existing taxonomy code in place.

 Eating disorder providers: This is not a discrete provider type, but rather a

specialization across providers.

 Autism spectrum disorder providers: BCBSA notes that not all Behavior Analysts

are autism spectrum disorder providers. Behavior Analysis is provided for patients with

autism spectrum disorders, but also patients without autism spectrum disorders.

Recommendation:

If finalized, we urge the Departments to consider including a pathway for health plans to

explain why time and distance standards were not comparable across MH/SUD and M/S

providers.

Rationale:

A pathway for a health plan to explain why, in certain geographic regions, it is unable to meet

the outcomes standard for time and distance across MH/SUD and M/S could mitigate the

unintended consequences of lowering network participation standards in order to be

compliant. For example, this pathway could include a written explanation, supported by data

analyses, of why there are insufficient MH/SUD providers to meet the standard even if all

contracted with the health plan based on the level of participation for M/S providers.

Alternatively, it could be an explanation, supported by data, of how a significant portion of

outpatient MH/SUD services are being provided virtually and those providers are not

captured by time and distance standards.

Recommendation:

If finalized, we recommend the Departments align the calculations for time and distance with

the network adequacy standards outlined for Qualified Health Plans (QHPs).

Rationale:

This approach would create a national standard for calculating the outcomes data rather than

relying on different state-based standards and, therefore, requiring significant additional

complexity for health plans when conducting the analyses. Furthermore, as this is an existing

and current standard, it would not require the regulators to identify new metrics or standards,

reducing the burden on all parties and supporting consistency in development and

interpretation of the analyses across health plans and markets.

Issue: Reimbursement Rates

The Departments are considering specifying the relevant data that plans and issuers would be

required to collect and evaluate which would include the following data: (1) In-network payments

and billed charges for inpatient MH/SUD and M/S benefits, outpatient office visit MH/SUD and

M/S benefits, and all other outpatient MH/SUD and M/S benefits; (2) Allowed amounts for CPT

codes 99213 and 99214 as well as CPT codes 90834 and 90837 for specific types of MH/SUD

and M/S providers.

Recommendation #1:

BCBSA recommends that any definition of “relevant data” for network composition standards

not include comparing provider reimbursement rates to billed charges, but instead compare

payments by specialty under private plans to Medicare reimbursement rates.

Rationale:

Billed charges are arbitrary amounts unilaterally determined by the provider and not necessarily

tied to any independent benchmark. In addition, billed charges are often not consistent with or

reflective of the cost of the service and relying on billed charges gives MH/SUD an incentive to

hike rates artificially. This reality has been consistently recognized in the work done by

policymakers to address surprise billing. Appreciating the lack of foundation of these often

heavily inflated charges, lawmakers explicitly excluded any reference to billed charges in the

independent dispute resolution process to prevent biases in decision-making by the arbiters.

We believe the Medicare Physician Fee Schedule (PFS) is the most appropriate benchmark for

comparing the adequacy of reimbursement for private plans. CMS establishes and updates the

PFS for the Medicare program. The PFS incorporates practice costs and other factors and,

unlike billed charges, is an unbiased, evidence-based payment mechanism that is beyond the

control of both providers and issuers, and as such, is probably the most accurate and reliable

information source that could be used by the Departments for assessing the reasonableness of

commercial insurer payments to MH/SUD providers.

Recommendation #2:

BCBSA recommends that comparisons between MH/SUD and M/S provider payments be

adjusted to reflect differences in both the requirements necessary to qualify as a licensed

professional in a particular specialty and the expenses that would be incurred to operate a

practice.

Rationale:

Some types of medical professionals are subject to more extensive requirements in order to

qualify to practice medicine in that field. For example, an orthopedic surgeon must complete a

five-year residency in orthopedic surgery following graduation from medical school, and often

then take an additional year or two to complete a fellowship in an orthopedic sub-specialty,

while a psychiatrist is required to complete a four-year residency. Psychologists and licensed

clinical social workers have less demanding licensing requirements than medical doctors who

may provide a subset of the same services. With regard to the latter, mental health therapy

generally does not require any specialized equipment, supplies or furnishings, whereas most

other M/S professionals must make significant investments in specialized equipment and devote

a considerable amount of resources to maintaining their clinical facilities and stocking them with

appropriate supplies necessary for providing care.

As noted above, data from the AMA indicates that there are substantial differences in practice

costs between MH/SUD and M/S specialties. This information is an input into the methodology

used by Medicare to set the payment rates in the PFS. The mean practice expenses for the

three MH/SUD provider types are the three lowest amounts, which demonstrates the extent to

which MH/SUD care differs from the care from other specialties.

While we understand that publishing a list of comparative specialties with appropriate

adjustment factors may be more complex for the Departments, there is a significant risk that

comparing costs for the broad categories of services that the Departments outline in the

Technical Release will have an inflationary impact on rates, and ultimately on costs for

businesses, workers and their families. It is important to note that the average premium for an

employer-provided family health insurance policy reached $22,221 in 2021 (one-third of the

median household income), nearly triple what it was in 2001.

And the average employee

contribution now accounts for 9% of the median household income. Rising prices are the

primary drivers of rising health care costs. In the market for medical services, acquisition of

physician practices by hospitals and private equity firms is driving up prices. The share of

physician practices owned by hospitals more than doubled from 2012 to 2018. In a study for

Physician Advocacy Institute, Avalere examined the impact of the COVID-19 pandemic on

physician practice acquisition in 2019 and 2020 and found that 48,400 additional physicians left

independent practice during the two-year study, and, by the beginning of 2021, only 30% of

physicians in the U.S. were practicing medicine independently.

According to a recent study on tax filings by physicians, the average physician in the US earned

$350,000 in 2017.

This study also found a significant difference in earnings across specialties.

Statistics are calculated from Sarah Flood et al., Integrated Public Use Microdata Series, Current Population

Survey: Version 9.0. Minneapolis, MN: IPUMS, 2021, https://doi.org/10.18128/D030.V9.0, and Annual Employer

Health Benefits Survey, Kaiser Family Foundation, for years 2001 and 2021, https://www.kff.org/wp-

content/uploads/2013/04/6458.pdf and https://www.kff.org/report-section/ehbs-2021-summary-of-findings/.

“2020 Health Care Cost and Utilization Report,” Health Care Cost Institute, May 2022,

https://healthcostinstitute.org/images//pdfs/HCCI_2020_ Health_Care_Cost_and_Utilization_Report.pdf, and BCBSA

calculations based on data from Congressional Budget Office, Consumer Price Index, Historical Data and Economic

Projections, May 2022, https://www.cbo.gov/data/budget-economic-data#4.

https://www.washingtonpost.com/business/2023/08/04/doctor-pay-shortage/

It would be hard to rationalize benchmarking the reimbursement for MH/SUD professionals to

the broad category of M/S physicians, which would include neurosurgeons that made $920,500,

orthopedic surgeons that made $788,600, or dermatologists that made $655,200 a year during

their peak earning years in 2017, according to this study.

We are concerned that the proposed comparison of reimbursement outcomes would extend the

factors leading to unsustainable prices in medical and surgical services sector to mental health

professionals. Thus, if the Departments proceed, the Departments should develop appropriate

comparisons with adjustments for reasonable provider reimbursement or benchmark to some

reasonable measure of adequate provider reimbursement, such as what is paid by Medicare.

Recommendation #3:

CMS should establish a limited list of comparative specialties for MH/SUD and M/S services for

health plans to use to compare their average reimbursement rates, based on the

recommendations outlined at a high level in Table 2.

Rationale:

If the Departments finalize the proposed categories of “relevant data” for the network

composition standards, BCBSA recommends the Departments publish analogous provider

types for the sake of comparing reimbursement that are standardized and specific enough to be

truly comparable, as recommended in Table 2. These provider comparison standards should be

made available for public comment, and should include (1) provider type, (2) common CPT

codes, (3) geographic region, and (4) a materiality standard.

We specify the relevant provider types and example CPT codes we would recommend for this

comparison below:

Table 2: Recommended Comparisons of MH/SUD to M/S Provider Types by CPT Code

M/S Specialty MH/SUD Specialty CPT Code Examples

• Internist

• Endocrinologist

• Neurologist

• Pediatrician

• Psychiatrist 90792

99203

99204

99213

• Nurse Practitioner

(M/S)

• Nurse Practitioner

(BH)

90792

99203

99204

99213

• Chiropractor

• Podiatrist

• Psychologist 90832

90791

• Occupational

Therapist

• Physical Therapist

• Licensed Clinical

Social Worker

97161 - 97168

By developing and publishing a granular list of standards to conduct provider comparisons,

health plans will be better equipped to accurately measure parity in reimbursement rates.

Recommendation #4:

BCBSA recommends that the Departments adopt a materiality standard for comparing

reimbursement rates as proposed in Table 2.

Rationale:

Application of a materiality standard will ensure plans and issuers are not reviewing provider

type and/or CPT code combinations for which there are very low claims experience as such

analyses are likely to be skewed, highly variable based on the reporting period, or otherwise

problematic, making these analyses of limited credibility and value.

Issue: Determination of Noncompliance based on Outcomes Data

If the relevant data evaluated reveals “material differences” in access to MH/SUD benefits as

compared to M/S benefits, the differences would establish that the plan or issuer is not in

compliance.

Recommendation #2:

BCBSA recommends that the Departments adopt a symmetrical rule whereby if there are not

material differences in outcomes data, the plan or issuer is deemed compliant with the NQTL

requirements for network composition.

Rationale:

As noted by the Departments, the goal of the statute and regulations is to ensure access to

MH/SUD care.

Where objective outcomes data demonstrates that NQTLs in no way impede

access to MH/SUD benefits relative to M/S benefits, the parity goal is met. In these situations, it

is reasonable that plans and issuers should be presumed to be in compliance with the

regulatory requirements and additional investigations would not be needed. This approach

would reduce administrative burden on both sides but, more importantly, would allow the

Departments and state regulators to focus their available resources on investigations that could

have material impacts for patients’ access.

Issue: Recommended Safe Harbor for Plans Paying Medicare Rates for MH/SUD

Providers

The technical appendix seeks comment on two safe harbors but does not include a safe harbor

that relates to the adequacy of payments to a publicly available benchmark that could be easily

compared across health plans.

Recommendation:

We recommend that the Departments create a safe harbor for plans and issuers reimbursing

MH/SUD providers at or above what CMS pays these providers under the Medicare program.

This comparison would be to the ratio of reimbursement levels for the plan or issuer compared

to Medicare for the MH/SUD specialists and CPT codes outlined above. Under this safe harbor,

failure to meet any other reimbursement comparison finalized by the Departments would not be

See, e.g., Prop. Treas. Reg. § 54.9812–1(a)(1), Prop. DOL Reg. § 2590.712(a)(1), Prop. HHS Reg. §

146.136(a)(1).

considered a strong indicator that the plan or issuer is not in compliance if plans paid at or

above Medicare rates during the relevant period.

Rationale:

Comparing payment rates to Medicare is appropriate because Medicare’s payments objectively

incorporate differences in costs relating to overhead, such as training and licensing

requirements, equipment, supplies, and furnishings used to provide care, and other

miscellaneous office expenses. In addition to consideration of the costs of delivering MH/SUD

care, Medicare’s MH/SUD payment rates are updated by CMS on an annual basis and subject

to rigorous public comment and Congressional oversight. Moreover, the payments under the

Medicare program are evaluated annually by MedPAC, a non-partisan Congressional agency.

Direct comparison to private and Medicare reimbursement rates for a defined subset of

MH/SUD services would avoid many of the problems identified in our comment letter. In the

Technical Release, the Departments seem most concerned about disparities in payment levels

for outpatient mental health services and have cited research indicating that some plans and

issuers are paying less than Medicare. It would make sense to focus on outpatient services for

this safe harbor given the complexity of comparing payments for inpatient services.

While we are not supportive of a mandate for health plans to pay Medicare rates, we suggest

this safe harbor as a reasonable way to compare commercial payment rates to a defined and

publicly available benchmark. However, we would note that in some cases, health plans have

multi-year contracts with providers, whereas Medicare’s payment rates are updated annually.

Thus, we would recommend the Departments provide reasonable tolerances for year-to-year

fluctuations based on the update factors to Medicare fee schedules.

Issue: Aggregate Data Collection

For all four specific types of relevant data, the Departments are considering requiring relevant

data to be collected and evaluated by a third-party administrator (TPA) or other service provider

in the aggregate for all plans or policies, as applicable, that use the same network of providers

or reimbursement rates because, in many instances, plan-level or product-level data may not

reflect sufficient claims experience to provide enough data for plans and issuers to evaluate and

consider the impact of an NQTL related to network composition on access to MH/SUD benefits

as compared to M/S benefits.

Recommendation #1:

BCBSA supports the proposal that data would be in the aggregate.

Rationale:

BCBSA is concerned that it is not feasible for issuers to collect and evaluate outcomes data on

a plan-level or plan-by-plan basis as issuers administer thousands of plans. Collection and

evaluation of data at the plan level would add a significant burden on carriers in terms of time

and resources. In addition, claims experience can fluctuate significantly for large group plans,

Medicare Payment Advisory Commission, March 2023 Report to the Congress: Medicare Payment Policy, 15

March 2023, Washington DC. https://www.medpac.gov/document/march-2023-report-to-the-congress-medicare-

payment-policy/.

and BCBSA does not believe that at the individual plan level the claims data would be credible.

Issue:

The Departments’ proposed approach in the Technical Release does not provide health plans

with sufficient details, methodology or time for comment.

Recommendation:

BCBSA recommends that the Departments issue their proposed approach for any comparisons

through a white paper process for public comment that details their proposed methodology prior

to proposing standards in rulemaking.

Rationale:

The Departments ask dozens of important and complicated questions in the Technical Release

that many health plans will not have adequate time to comment on appropriately. Given the

highly technical nature of these questions and the potential for comparisons to be used as

indications of noncompliance, the Departments should provide adequate time for public

comment through a white paper process on the methodology for any comparisons the

Departments move forward with.

The Administrative Procedure Act, 5 U.S.C. § 553, gives federal agencies substantial flexibility

to conduct outreach and communications in connection with informal rulemaking. The

Administration has encouraged agencies to take additional steps to seek public input outside of

a formal comment period on several occasions, including through the 2023 Executive Office of

the President memorandum

that states the expectation agencies go beyond required notices

when communicating with the public and the 2023 Executive Order on Modernizing Regulatory

Review, which notes the need for regulatory actions to be informed by input from the affected

members of the public.

Other previous communications encouraged agencies to offer the

public increased opportunities to participate in policymaking.

The Departments have previously established similar informal or white paper processes that

have been used successfully for other data-dependent processes, such as development of

adjustment models. There are several recent instances in which the Departments have sought

additional public comment in ways similar to that which BCBSA is proposing for this Technical

Release:

 In connection with implementing the ACA risk adjustment rulemaking, HHS created a

white paper for which comments could be submitted at any time, including after the close

of the NPRM comment period.

 When HHS was developing the essential health benefits (EHB) under the ACA, the

Institute of Medicine (IOM) submitted a white paper that the HHS held listening sessions

https://www.whitehouse.gov/wp-content/uploads/2022/04/M-22-10.pdf

https://www.whitehouse.gov/briefing-room/presidential-actions/2023/04/06/executive-order-on-modernizing-

regulatory-review/

https://obamawhitehouse.archives.gov/the-press-office/transparency-and-open-government

https://www.cms.gov/cciio/resources/forms-reports-and-other-

resources/risk_adjustment_implementation_issues#:~:text=This%20white%20paper%20serves%20both,and%20Risk

%20Adjustment%20Notice%20of

for stakeholders to raise concerns. From the IOM white paper, HHS also commissioned

a study that recommended “the criteria and methods for determining and updating the

EHB.” Final Rule, 78 Fed. Reg. 12834 (Feb. 25, 2023).

 The DOL’s MHPAEA self-compliance tool serves as an example of an informal process

outside of the rulemaking comment period whereby the Departments have collected

comments and engaged stakeholders.

We believe that by engaging in a similar white paper process, the Departments would allow the

public to more meaningfully comment on these proposals. Such a process could be used to

gather more meaningful information on the technical questions that could not be sufficiently

addressed in the 75-day Technical Release comment period, including:

 Whether out-of-network utilization data be provided in terms of the percentage of claims,

number of claims, total dollar amounts of all claims, and/or something else.

 How the Departments should control for treatment received from MH/SUD providers

where no claim for benefits was made.

 The existing models or methodologies the Departments consider when specifying the

OON utilization data that plans and issuers would be required to collect and evaluate as

part of their comparative analyses for NQTLs related to network composition.

 The CPT codes that would help plans and issuers evaluate their reimbursement rate

structures.

https://www.federalregister.gov/documents/2013/02/25/2013-04084/patient-protection-and-affordable-care-act-

standards-related-to-essential-health-benefits-actuarial