A 50-State Review of Access to State Medicaid Program Information for People with Limited English Proficiency and/or Disabilities Ahead of the PHE Unwinding

Authors: MaryBeth Musumeci, Sweta Haldar, Emma Childress, Samantha Artiga, and Jennifer Tolbert
Published: Aug 26, 2022

Issue Brief

Executive Summary

State Medicaid websites are a key source of information and an avenue for enrollment in or renewal of coverage for many applicants and existing enrollees. Medicaid households include a disproportionate share of nonelderly adults with limited English proficiency (LEP), and three in ten nonelderly Medicaid adults report having a disability. Moreover, some individuals have LEP as well as a disability. When the continuous enrollment requirement related to the COVID-19 public health emergency (PHE) ends and states resume regularly scheduled redeterminations and renewals, individuals with LEP and/or disabilities may be at increased risk of losing coverage, despite remaining eligible, due to barriers in accessing eligibility and renewal information. Several federal laws require state Medicaid agencies to provide information in a way that is accessible to people with LEP and people with disabilities (Appendix).

This issue brief reviews accessibility of information for people with LEP and people with disabilities provided through state Medicaid websites, online application landing pages, and PDF applications and call center automated phone trees as of June 16, 2022. The analysis shows that while states have taken some steps to support access to information and applications for people with LEP and people with disabilities, gaps in accessibility remain. Given the potential challenges people with LEP and people with disabilities may face maintaining coverage once Medicaid renewal and redeterminations resume following the end of the public health emergency, specific steps to minimize barriers and ensure easy access to information, applications, and assistance could help prevent coverage losses among those who remain eligible.

Most states have taken steps to support access to information and applications for people with LEP and people with disabilities.  Such steps include the following:

  • Nearly every state (49) translates information on their homepage, their online application landing page, and/or their PDF Medicaid application into languages other than English. However, there is variation across and within states in terms of which resources are translated and how many translated languages are available. Just under two-thirds of states (32) translate information on their Medicaid homepage into a language other than English, 36 states translate their online application landing page into at least one other language, and 39 states make a PDF application available in at least one other language than English. The number of languages other than English ranges from 1 to over 100 in states using automated translation systems such as Google Translate, although as noted below, there may be quality issues associated with automatically translated information. The languages most commonly available include Spanish, Chinese, Vietnamese, and Tagalog. Online application landing pages and PDF applications are typically available in fewer languages than homepages, with translations of these resources often limited to only Spanish.
  • A total of 46 states include multilingual tagline notices about how to obtain language assistance services on or within one click of their homepages online application landing pages and/or on their PDF applications. In these states, taglines are provided in between 1 and 55 languages other than English, with most of these states including taglines in at least 15 languages in addition to English. All states with taglines include taglines in Spanish. In addition to Spanish, the most common other languages in which taglines are offered are Chinese and Vietnamese.
  • Forty-five states provide general information about the availability of reasonable modifications (also known as reasonable accommodations) for people with disabilities on or within one click of their homepage or online application landing page and/or on their PDF application. Examples of alternative formats and services to increase access for people with disabilities include teletypewriter (TTY) call center numbers to assist people who are deaf, hard-of-hearing, or have a severe speech impairment, which 41 states provide on their website, online application landing page, and/or PDF application. Fewer states specify how to access large print or Braille materials and/or applications (19 states) or how to request an American Sign Language or other appropriate interpreter (26 states).
  • State Medicaid homepages and online application landing pages on average have fewer accessibility errors for people who are blind or have low vision compared to webpages in general. State homepages and online application landing pages were evaluated using WAVE, a suite of automated web accessibility measurement tools that includes 110 elements that assess potential accessibility errors for people who are blind or have low vision. WAVE detected an average of 11.4 errors per page across the 101 webpages assessed, compared to the average of 50.8 errors per page found on the top one million web homepages generally. About half of the assessed webpages showed five or fewer detectible errors, and 11 pages showed no detectible errors. These findings suggest that states have given notable attention to ensuring accessibility on these pages overall.

While states have taken steps to support access to information and applications for people with LEP and people with disabilities, continuing gaps in accessibility remain. Some of these gaps include:

  • Multilingual information is often available in a limited number of other languages. In several states, Spanish is the only language other than English in which translated information is available on Medicaid homepages. Moreover, as noted above, translated online application landing pages and PDF applications are typically available in fewer languages and often limited to Spanish. For example, 26 of the 39 states with a translated PDF application only translate it into Spanish. Of the 40 states that offer language options through call center automated menus, 31 states offer options only in English and Spanish. While Spanish is the most common language spoken among nonelderly adults in Medicaid households who have LEP, Chinese and Vietnamese are each spoken by 3% of these adults, and there are a diverse set of languages spoken overall by this group.
  • Translated information on Medicaid homepages may be less reliable than the information available on English versions of homepages and, in some cases, translations are incomplete. Over half of states providing translated homepages use automated translations through Google Translate, which may lead to quality issues associated with the translations. Google has indicated that it is not intended to replace human translators or to be used in public health contexts without having translations verified. A few states include disclaimers that the English version of the homepage is the most reliable. Additionally, in some states, some content in the headers, footers, and/or menus of the translated versions of the homepage remains in English and/or clicking links in translated versions of the homepage leads to English content.
  • Most states do not make information and applications available in multiple formats to improve accessibility for people with disabilities. A total of 19 states include information on their homepage, online application landing page, or PDF application about how to obtain materials in large print or Braille, but only two states post a large print PDF application form online. The remaining states do not mention how to access alternative formats. Making information and applications available in multiple formats is an important means of increasing access for people with disabilities. People with disabilities may access information in different ways, depending on their type of disability and type of assistive technology they use.
  • Most of the webpages (in 47 states) assessed through WAVE showed at least one incidence of low contrast text, which is difficult for people with color blindness or low vision to read. Connecticut is the only state that allows users to change the homepage to a “high contrast” mode to improve readability for users. Other commonly detected accessibility issues included missing labels to describe the various fields of content in a form (24 states) and images missing alternative text (23 states).
  • Nearly all state call centers (49 states) answer with an automated phone tree, which can increase accessibility challenges for people with LEP and/or intellectual, developmental, or mental health disabilities. In 12 states with automated phone trees, the first set of menu options include the option to speak to a live person. As of the time of our data collection, the call center wait time to speak with a live person was less than 15 minutes in 34 states, while in 8 states the wait time was more than 15 minutes, and in the remaining 7 states, we could not reach a live person.1  Access to a live person can improve accessibility for people with LEP and/or intellectual or developmental disabilities and people with mental health disabilities. 

Introduction

State Medicaid websites are a key source of information and an avenue for enrollment in or renewal of coverage for many applicants and existing enrollees. Medicaid households include a disproportionate share of nonelderly adults with limited English proficiency (LEP), and three in ten nonelderly Medicaid adults report having a disability. While Spanish is the most common language spoken among nonelderly adults in Medicaid households who have LEP, Chinese and Vietnamese are each spoken by 3% of these adults, and there are a diverse set of languages spoken overall by this group. Medicaid enrollees have a variety of disabilities, including blindness or low vision; deafness or hard of hearing; intellectual and developmental disabilities such as autism or Down’s syndrome; traumatic brain or spinal cord injuries; and mental illness. Some people may have both LEP and a disability.

When the continuous enrollment requirement related to the COVID-19 public health emergency (PHE) ends, and states resume regularly scheduled redeterminations and renewals, individuals with LEP and/or disabilities may be at increased risk of losing coverage, despite remaining eligible, due to barriers in accessing eligibility and enrollment information. The Centers for Medicare & Medicaid Services (CMS) guidance about the PHE unwinding reminds states about these obligations and prompts states to review their communications with people with LEP and people with disabilities as a strategy to mitigate inappropriate coverage loss. Several federal laws require state Medicaid agencies to provide information in a way that is accessible to people with LEP and people with disabilities. The Appendix provides background information on these laws.

This issue brief reviews accessibility of information for people with LEP and people with disabilities provided through state Medicaid websites and call center automated phone trees, as of June 16, 2022. We assessed homepages and call center automated menu options in all 50 states and DC and online application landing pages and PDF applications in the states that provide these resources2  (Appendix Table 1). See the Methods box for more detail.

Accessibility of Information for People with LEP

Translation of Website Information and Applications

Overall, 32 states translate information on their Medicaid program homepage into languages other than English, and 36 states provide translations of their online application landing page into other languages (Appendix Table 2 and Figure 1). The number of languages other than English range from 1 to over 100 in 18 states using automated translation systems such as Google Translate. Additional states also use Google Translate to offer translations in a more limited number of languages, while other states do not specify if the translation is based on automated software. All states that offer translation on their homepages or online application landing pages offer information in Spanish. The other languages most commonly available include Chinese, Vietnamese, and Tagalog. In general, states offer fewer translated languages on their online application landing pages than on the homepages. For example, 23 of the 36 states with translated online application landing pages only translate into Spanish.

Most states with online PDF applications (39 of 48 states) offer the application in languages other than English (Appendix Table 2 and Figure 1). In 26 of these states, Spanish is the only other language in which the PDF application is available. In the remaining 13 states, PDF applications are offered in additional languages, including in 10 or more languages in California, New York, North Carolina, Oregon and Washington. The most common other languages available are Vietnamese, Chinese and Korean. Most PDF applications ask about the applicant’s preferred spoken language, which may facilitate access to linguistically accessible communications in the future.

There may be potential quality issues with translations in states using automated translations, such as Google Translate. Google has indicated that it is not intended to replace human translators or to be used in public health contexts without having translations verified and other research has found Google Translate to be unreliable in medical contexts.3 ,4  A few states include disclaimers that the English version of the homepage is the most reliable. In some states, some content in the headers, footers, and/or menus of the translated versions of the homepage remains in English and/or clicking links in translated versions of the leads to English content.

Homepage Available in Languages Other than English
Figure 2: Google Translate Translation Menu from California’s Medicaid Homepage

Access to Language Assistance and Interpretation Services

Thirty-five states include multilingual taglines on or within one click of their homepages or online application landing pages with information on how to access language assistance services (Appendix Table 2). Most of these states provide taglines in at least 15 languages other than English. All 35 states include taglines in Spanish, and the other most common languages are Chinese and Vietnamese.

Most states with PDF applications (35 of 48 states) include multilingual tagline notices regarding the availability of language assistance services on the PDF application (Appendix Table 2). These taglines are generally present on the first or second page of the application or as a footer running across every page. In 26 of these states, taglines are only provided in Spanish and English, while in others, they are provided in between 2 and 15 other languages. In addition to Spanish, the most common other languages in which taglines are offered are Chinese and Vietnamese.

Figure 3: Spanish Tagline from Iowa’s Medicaid Website

In 49 states, call centers answered with automated phone trees, and 40 of these states offered menu options in languages other than English. However, in 31 of these states the only other language offered through the menu is Spanish (Appendix Table 2 and Figure 1).5  In the remaining states, phone tree menus offer between two and six languages other than English. In some states, other language options are only offered after the applicant listens to a long set of options in English.

Accessibility of Information for People with Disabilities

Availability of Reasonable Modifications

Most states (43) provide general information about the availability of reasonable modifications (also known as reasonable accommodations) for people with disabilities on or within one click of their homepage or online application landing page, while 32 states provide this information on their online PDF application (Appendix Table 4 and Figure 4).6  States often provide information about reasonable modifications for people with disabilities on a separate webpage titled “Nondiscrimination policy” or “Accessibility” rather than directly on their homepage. Users can usually find these separate webpages by clicking on a “Nondiscrimination policy” link at the bottom of the homepage. States often use these separate webpages to outline their general commitment “to serving the needs of people with disabilities” (CT). Most of these separate webpages include information about free “auxiliary aids and services necessary to afford an individual with a disability an equal opportunity to participate in all services, programs and activities” (KY). Additionally, states often include information about reasonable accommodations such as alternative formats including Braille or large print materials, teletypewriter (TTY) numbers, and how to request qualified American Sign Language (ASL) interpreters on these separate webpages. The comprehensiveness of the reasonable modification statement varies across states—some states only include a general statement while other states describe the types of aids and alternative formats that are available and provide instructions on how to access them. The availability of information in plain language also is an important means of providing access for people with cognitive disabilities, though our website review did not assess content for plain language.

Statement on Reasonable Modifications for Individuals with Disabilities Available Online

Alternative Written Formats for People with Disabilities

A total of 19 states provide information about how to obtain materials in large print or Braille (Appendix Table 4 and Figure 4). Most of these states list a phone number (and/or a TTY number) for individuals to call to request materials in alternative formats, such as large print, or advise users to contact their local office. Two states (CA, CO) post a large print PDF application form online,, and several states, including DE, KS, MS, NH, NY, TN, and VA, allow users to increase or decrease the font size of text on the webpage. Connecticut allows users to change the homepage to a “high contrast” mode to improve readability for users. Making information and applications available in multiple formats is an important means of increasing access for people with disabilities. People with disabilities may access information in different ways, depending on their type of disability and type of assistive technology they use.

Website Accessibility for People Who are Blind or Have Low Vision

State homepages and online application landing pages were evaluated using WAVE, a suite of automated web accessibility measurement tools that includes 110 elements that assess potential accessibility errors for people who are blind or have low vision. WAVE is developed and made available as a free community service by WebAIM (Web Accessibility in Mind) at Utah State University. WAVE assesses web accessibility, including compatibility with screen reader software, which can be used by people who are blind to convert web content to synthesized speech, and screen magnifiers or zoom, which can be used by people with low vision. Because WAVE is automated, it does not assess all aspects of accessibility that may be encountered by website users. However, WAVE does identify the accessibility errors that are most frequently encountered and that tend to have the greatest impact on users who are blind or have low vision.7  Additionally, errors identified by WAVE have been found to correlate with the existence of other accessibility issues that WAVE does not detect but which can be encountered by a website user.8  The WAVE analysis was applied to a total of 101 webpages, including the homepages for all 50 states and DC and the online application landing pages for 49 states and DC.9 

The most common accessibility issue detected by WAVE is very low contrast text, with 67% of the assessed webpages (in 47 states) showing at least one incidence of this error (Appendix Table 3 and Figure 5). Overall, WAVE detected an average of seven instances of very low contrast text errors per page across the 101 webpages assessed. Low contrast text refers to the difference in brightness between text or a graphic and its background colors and is difficult for people with color blindness or low vision to read.  Other commonly detected accessibility issues include missing labels to describe the various fields of content in a form (27% of pages, 24 states) and images missing alternative text (25% of pages, 23 states) (Appendix Table 3). Alternative text is used to describe the content of an image. Images that lack alternative text and forms without text labels cannot be properly identified by screen readers, making this content very difficult to access for people who rely on that technology.

The average number of accessibility errors detected by WAVE across the 101 assessed state homepages and online application landing pages is substantially lower than the average number of errors found on webpages in general. Specifically, WAVE detected an average of 11.4 errors per page across the 101 webpages assessed, compared to the average of 50.8 errors per page found on the top one million web homepages generally. The number of web accessibility errors detected on the assessed webpages varies widely across states (Appendix Table 3). Over half of the states have 15 or fewer errors across their homepages and online application pages and in seven states, there are less than five errors (Figure 5). These findings suggest that states have given notable attention to ensuring accessibility on these pages overall, although some pages may have errors that could pose difficulty for users with disabilities.

Number of Web Accessibility Errors Detected by WAVE, 2022

Effective Communication for People with Disabilities

Forty-one states list a TTY call center number on their homepage or online application landing page, within one click from these pages, or on their PDF application (Appendix Table 4 and Figure 6).10  These numbers are often found in different places on states websites. Some states require users to click on a “contact us” tab to find a TTY number, while others require users to click on a different link to a “nondiscrimination” or “access for users with special needs” webpage to find a TTY number. In some states, the TTY number is only available on the language assistance page. TTY numbers allow individuals who are deaf or hard of hearing to communicate by sending typed messages over the phone line. If states communicate with applicants or the public by phone, they also must handle calls via TTY or similar technology.11  While a number of states include the TTY number in their tagline notices for language assistance, a smaller number of states, such as North Carolina, provide a Spanish language TTY (Relay Service).

Twenty-six states provide information about how to request an ASL interpreter on their website and/or PDF application (Appendix Table 4 and Figure 6). One of these states (WA) directly lists this information on their homepage, while the other states require users to click on another link such as a “nondiscrimination” webpage to find information about how to request an ASL interpreter. Both TTY numbers and ASL interpreters can help promote effective communication for people who are deaf or hard of hearing that could be essential for someone to understand how to enroll in or retain Medicaid coverage.

The large majority (49 of 51) of call centers answer with an automated phone tree and only 12 of the automated phone trees include an option to speak to a live person in the first set of menu options (Appendix Table 4 and Figure 6). Two call centers (DC, SD) answer with a live person instead of an automated phone tree. Access to a live person can improve accessibility for people with LEP and/or intellectual or developmental disabilities and people with mental health disabilities. As of the time of our data collection, the call center wait time to speak with a live person was less than 15 minutes in 34 states, while in six states, the wait time was more than 15 minutes. In the remaining seven states, we were unable to reach a live person.12  The end of the COVID-19 PHE may increase these wait times, as states resume processing redeterminations and renewals.

TTY Number Available on Medicaid Homepage

Looking Ahead

As the COVID-19 PHE ends and states resume regularly scheduled Medicaid redeterminations and renewals, people with LEP and/or disabilities may face increased challenges to maintaining coverage despite remaining eligible due to barriers in accessing eligibility and enrollment information. Under Title VI of the Civil Rights Act, Section 1557 of the Affordable Care Act, and the Rehabilitation Act, state Medicaid agency program information must be accessible to people with LEP and people with disabilities. The Biden Administration has issued a proposed rule revising implementing regulation for Section 1557, taking steps to reverse Trump Administration policy and regulations that significantly narrowed the implementation and administrative enforcement of the regulations. The proposed rule reinstates the requirement that Medicaid agencies provide notice of the availability of language assistance services and auxiliary aids and services in both physical locations and on their websites. Additionally, the rule, for the first time, requires state Medicaid agencies to give staff clear guidance and training on the provision of language assistance services and effective communication and reasonable modifications to policies and procedures for people with disabilities in order to improve compliance. Regardless of the status of the regulations, the underlying statutory protections of Section 1557, which require meaningful access to federal programs for people with LEP and/or disabilities, remain as well as protections under other federal laws and federal Medicaid regulations that require public programs be accessible to people with LEP and/or disabilities.

As states prepare to resume Medicaid redeterminations and renewals, they can take steps to prevent and minimize potential administrative barriers to maintaining coverage, particularly for people who have LEP and/or disabilities. CMS guidance outlines specific steps states can take, including ensuring accessibility of forms and notices for people with LEP and people with disabilities and reviewing communications strategies to ensure accessibility of information. Increasing access to a live person through call centers and/or providing a dedicated call center line for people with LEP may also help improve accessibility for people with LEP and people with disabilities. Ensuring accessibility of information, forms, and assistance will be key for preventing coverage losses and gaps among these individuals.

Methods

State Medicaid websites were identified by using the first result from a Google search for the state name and Medicaid (e.g., Alabama Medicaid), with the exception of DC, Illinois, and Nevada, for which the first results were not specific to Medicaid or did not include applicant or member-facing information. For DC and these states, the second result, which was more relevant, was used. Online Medicaid application landing pages and PDF applications were identified through State Medicaid websites. In some cases, online application landing pages are housed in separate websites from the Medicaid website, for example, if the online application allows individuals to apply for multiple benefits (such as health care, food, and cash assistance) or if the application is integrated with Marketplace coverage. The analysis included examination of information available on and within one click of the Medicaid homepage and online application page as well as PDF applications, where available.  An abbreviated list of indicators used to determine the accessibility and accommodation information is reproduced in Appendix Tables 1 through 4.Call center data was obtained by searching the Medicaid state homepage for the customer support call center number. Calls were made to these numbers (listed in Appendix table 1) from June 2 through June 16, 2022, between 9 AM and 5 PM EST.

KFF appreciates the assistance of Jared Smith, program manager at Utah State University Institute for Disability Research, Policy and Practice, with analyzing and interpreting the WAVE results.

Appendix

Appendix

Summary of Relevant Federal Nondiscrimination Laws

State Medicaid agency program information must be accessible to people with LEP and people with disabilities. States must provide information about Medicaid eligibility requirements, covered services, and applicant/enrollee rights and responsibilities via a website as well as orally and in writing.13  States also must allow individuals the opportunity to apply for Medicaid without delay.14  Individuals must be able to apply online, by telephone, by mail, in person, and through other commonly available electronic means.15  State Medicaid agencies must provide assistance to individuals seeking help with the application or renewal process in person, over the telephone, and online.16 

Federal law requires state Medicaid agencies to provide meaningful access for people with LEP. As recipients of federal funds, state Medicaid agencies are prohibited from discriminating on the basis of national origin, pursuant to Title VI of the Civil Rights Act of 1964 and Section 1557 of the Affordable Care Act (ACA). Current regulations require state Medicaid agencies to “ensure meaningful access to… programs or activities by limited English proficient individuals.”17  People who are LEP may include those who do not speak English as their primary language and who have a limited ability to read, write, speak, or understand English. The current Section 1557 regulations, issued by the Trump Administration, reduced the regulatory standards originally adopted by the Obama Administration. The Biden Administration has issued a proposed rule taking steps to reverse Trump Administration policy and strengthen and reinstate Section 1557 regulatory standards.

The two main types of language assistance services are oral interpreting and written translation. In addition to the requirements of Title VI and Section 1557, federal Medicaid regulations specifically require state Medicaid agencies to provide information about eligibility requirements, covered services, and applicant/enrollee rights and responsibilities as well as application assistance in a timely manner and in a way that is accessible to people with LEP, though the provision of language assistance services at no cost to the individual.18  Individuals must be informed about the availability of these language assistance services and how to access them.19  At minimum, state Medicaid agencies must provide taglines in non-English languages that indicate the availability of language assistance services.20 

Federal law requires state Medicaid agencies to make programs and services accessible to people with disabilities. Title II of the Americans with Disabilities Act (ADA) applies to state and local governmental entities, while Section 504 of the Rehabilitation Act and Section 1557 of the ACA apply to recipients of federal funds. All of these laws prohibit state Medicaid agencies from discriminating against people with disabilities. This means that people with disabilities cannot be excluded from participation in or be denied the benefits of the services, programs, or activities offered by a state Medicaid agency.21  People with disabilities also must have an equal opportunity and ability to participate in or benefit from the services, programs, and activities of state Medicaid agencies.22  Additionally, state Medicaid agencies must afford people with disabilities the opportunity to request, and must make, reasonable modifications in policies, practices, or procedures when necessary to avoid disability-based discrimination, unless doing so would result in a fundamental program alteration  or result in an undue burden.23  The definition of disability is to be construed broadly in favor of expansive coverage to the maximum extent permitted.24 

State Medicaid agencies must take appropriate steps to ensure that communication with applicants, participants, and members of the public with disabilities is as effective as communication with those without disabilities.25  State Medicaid agencies also must timely furnish appropriate auxiliary aids & services where necessary to afford people with disabilities an equal opportunity to participate in and enjoy the benefits of a program or activity.26  The type of auxiliary aid or service to be provided varies based on the individual’s communication method; the nature, length and complexity of the communication; and the context. Governmental entities like state Medicaid agencies must give primary consideration to the individual’s requested type of auxiliary aid or service. Auxiliary aids and services may include qualified onsite or video remote interpreters. A teletypewriter (TTY) or equally effective telecommunication system must be used when an agency communicates with applicants or enrollees by phone, and agencies must respond to relay calls in the same manner as other phone calls.27  Any automated attendant system (voicemail or interactive voice response system) to receive or direct incoming calls must provide effective real time communication.28 

In addition to the requirements of the ADA, Section 504, and Section 1557, federal Medicaid regulations specifically require state Medicaid agencies to provide information about eligibility requirements, covered services, and applicant/enrollee rights and responsibilities as well as application assistance in plain language, in a timely manner, and in a way that is accessible to people with disabilities, though the provision of auxiliary aids and services at no cost to the individual.29  Individuals must be informed about the availability of accessible information and how to obtain it.30 

State Medicaid Websites, Call Center Numbers, and Availability of PDF and Online Applications
Availability of Online Program Information, PDF Applications, and Call Center Menu Options in Languages Other than English
Most Common Web Accessibility Errors Detected by WAVE, by Error Type
Availability of Reasonable Modifications for Individuals with Disabilities and Live Call Center Assistance

 

Endnotes

  1. Two states (AZ, FL) require callers to enter identifying information in order to access live assistance and were excluded from this assessment. ↩︎
  2. If the PDF application could not be found on the homepage or online application landing page, this indicator was coded as “no” even if a PDF application was available elsewhere online. PDF applications vary widely in length, ranging from 9 to 62 pages. ↩︎
  3. Salame, R. (2021, April 30). Limited English Skills Can Mean Limited Access to the COVID-19 Vaccine. Slate. https://slate.com/news-and-politics/2021/04/covid-19-vaccine-for-non-english-speakers.html. ↩︎
  4. Taira, et al. (2021). A Pragmatic Assessment of Google Translate for Emergency Department Instructions. Journal of General Internal Medicine, 36(11), 3361-3365. doi: 10.1007/s11606-021-06666-z. ↩︎
  5. We did not assess the availability of live multilingual call center assistance. ↩︎
  6. We were unable to assess this indicator for online application forms because most states require users to create an account before accessing the contents of the online application. ↩︎
  7. Conversation with Jared Smith, Program Manager, Institute for Disability Research, Policy and Practice, Utah State University (June 7, 2022). ↩︎
  8. Id. ↩︎
  9. AK lacks an online application page. ↩︎
  10. State websites often include a brief description of TTY services explaining how the service is “designed for people who are Deaf or Hard or Hearing” (NC). For example, North Carolina’s website describes how the “relay operator reads aloud the message [the caller] typed to the other party and types the other party’s voiced message and background sounds, if any, to [the caller].” To see more about how TTY numbers can be used, see North Carolina’s infographic. ↩︎
  11. If the TTY number could not be found on the homepage or within one click from the homepage, this indicator was coded as “no” even if the number was available elsewhere on the website. ↩︎
  12. Two states (AZ, FL) require callers to enter identifying information in order to access live assistance. ↩︎
  13. 42 C.F.R. u00a7 435.905 (a). ↩︎
  14. 42 C.F.R. u00a7 435.906. ↩︎
  15. 42 C.F.R. u00a7 435.907 (a). ↩︎
  16. 42 C.F.R. u00a7 435.908 (a). ↩︎
  17. Nondiscrimination in Health and Health Education Programs or Activities, Delegation of Authority, 85u00a0Fed. Reg. 37,160u00a0(June 19, 2020) (to be codified at 42 CFR Parts 438, 440, and 460). ↩︎
  18. 42 C.F.R. u00a7 u00a7 435.905 (b)(1); 435.908 (a). ↩︎
  19. 42 C.F.R. u00a7 435.905 (3). ↩︎
  20. Id. ↩︎
  21. 42 U.S.C. u00a7 12131; 28 C.F.R. u00a7 35.130 (accord Rehab. Act u00a7u00a0 504 and ACA u00a7 1557). ↩︎
  22. Id. ↩︎
  23. 28 C.F.R. u00a7 u00a7 35.130; 35.164. ↩︎
  24. 28 C.F.R. u00a7 35.101. ↩︎
  25. 28 C.F.R. u00a7 35.160. ↩︎
  26. 28 CFR u00a7 u00a7 35.104; 35.160. ↩︎
  27. 28 C.F.R. u00a7 35.161. ↩︎
  28. Id. ↩︎
  29. 42 C.F.R. u00a7 u00a7 435.905 (b)(2); 435.908 (a). ↩︎
  30. 42 C.F.R. u00a7 435.905 (3). ↩︎

The Inflation Reduction Act is a Foot in the Door for Containing Health Care Costs

Author: Larry Levitt
Published: Aug 25, 2022

In this JAMA Forum column, KFF’s Larry Levitt examines the Medicare drug-price negotiation provisions and other drug-price reforms in the Inflation Reduction Act and their potential to spur others efforts to address health care costs.

News Release

Medicare Advantage Is Close to Becoming the Predominant Way That Medicare Beneficiaries Get Their Health Coverage and Care

Published: Aug 25, 2022

As Medicare Advantage continues to grow, a gradual but significant reshaping of the Medicare program is taking place.

A new KFF analysis finds that nearly half of eligible Medicare beneficiaries – 28.4 million out of 58.6 million Medicare beneficiaries overall – are now enrolled in Medicare Advantage plans. That represents a more than doubling of the share of the eligible Medicare population enrolled in such plans from 2007 to 2022 (19% to 48%). Enrollment is projected to cross the 50 percent threshold as soon as next year, making Medicare Advantage the predominant way that Medicare beneficiaries with Parts A and B get their coverage and care.

The rise of Medicare Advantage signals the transformation of Medicare to a program in which a majority of people receive benefits by enrolling in plans offered by private health insurance companies.

The new analysis is one of three released today by KFF in which researchers examine various aspects of Medicare Advantage. It provides the latest data on Medicare Advantage enrollment, including the types of plans in which Medicare beneficiaries are enrolled, and how enrollment varies across geographic areas. A companion analysis describes Medicare Advantage premiums, out-of-pocket limits, cost sharing, extra benefits offered, and prior authorization requirements. A third examines trends in bonus payments to Medicare Advantage plans, enrollment in plans in bonus status, and how these measures vary across plan types and firms.

Among other key findings:

  • Enrollment in private plans is highly concentrated among a small number of firms, with UnitedHealthcare and Humana together accounting for 46 percent of all Medicare Advantage enrollees nationwide. In nearly a third of counties across the U.S., these two firms account for at least 75 percent of Medicare Advantage enrollment.
  • In 2022, nearly 7 in 10 Medicare Advantage enrollees (69%) are in plans with prescription drug coverage (MA-PDs) that require no premium other than the Medicare Part B premium ($170.10 in 2022).
  • Nearly all enrollees in individual Medicare Advantage plans open for general enrollment have access to some benefits not covered by traditional Medicare, including eye exams and/or glasses (99%), hearing exams and/or aids (98%), and a fitness benefit (98%).
  • Virtually all Medicare Advantage enrollees (99%) are in plans that require prior authorization for some services. Prior authorization is most often required for relatively expensive services, such as prescription drugs administered by a physician (Part B drugs; 99%), skilled nursing facility stays (98%), and inpatient hospital stays (acute: 98%; psychiatric: 94%), but it is rarely required for preventive services (6%).
  • Federal spending on Medicare Advantage bonus payments has increased every year since 2015 and will reach at least $10 billion in 2022. Payments vary across firms, with UnitedHealthcare receiving the largest total payments ($2.8 billion) and Kaiser Permanente receiving the highest payment per enrollee ($521). (KFF is an independent, nonprofit organization that analyzes national health issues and is not affiliated with Kaiser Permanente.) Recently, the Medicare Payment Advisory Commission (MedPAC) and others have raised questions about whether the bonus program includes too many measures, does not adequately account for social risk factors, and may not be a useful indicator of quality for beneficiaries.

The full analyses are available online and include:

(KFF has adjusted its methodology from previous years to calculate the share of eligible Medicare beneficiaries enrolled in Medicare Advantage, meaning they must have both Part A and B coverage. This aligns with how the Medicare Payment Advisory Commission (MedPAC) and others describe this population. The share enrolled in Medicare Advantage would be somewhat smaller using the old method. See the methods section of the analysis.)

For more data and analyses about Medicare Advantage, visit kff.org

Monkeypox (MPX) Cases and Vaccinations by Race/Ethnicity

Authors: Nambi Ndugga, Sweta Haldar, Drishti Pillai, Latoya Hill, and Samantha Artiga
Published: Aug 24, 2022

On August 4, 2022, the U.S. Department of Health and Human Services declared the U.S. monkeypox (MPX) outbreak to be a public health emergency. The Centers for Disease Control and Prevention (CDC)  released data on national-level MPX cases reported in the U.S through July 22, including demographic information where available. These data show that, in addition to nearly all MPX cases reported being among men (99%) reporting recent male-to-male sexual or close intimate contact (94%), there are also racial and ethnic disparities, with Black and Hispanic people bearing a disproportionate burden of cases to date. While the national data by race/ethnicity have not been updated since the CDC release, a small number of states, as well as some local jurisdictions, are reporting race/ethnicity data on MPX cases and vaccinations and these data show a similar disproportionate impact. Moreover, the very limited data available to date on vaccinations also suggest that Black and Hispanic people are receiving smaller shares of vaccinations despite accounting for larger shares of cases.

National Data on MPX Cases by Race/Ethnicity

Data from 43 states, DC, and Puerto Rico show that Black people made up 26% of MPX cases compared to 12% of the population, and Hispanic people accounted for 28% of cases versus 19% of the population. Data were not separately reported for American Indian and Alaska Native (AIAN) or Native Hawaiian or Other Pacific Islander (NHOPI) people. CDC notes that areas with high numbers of cases that did not submit case reports are more racially and ethnically diverse. As such, the reported data may understate disparities. Moreover, the share of cases among Black people has risen in recent weeks, suggesting widening disparities for this group.

Racial/Ethnic Distribution of Monkeypox Cases in the U.S., May 17- July 22, 2022

To date, race and ethnicity data are missing for the majority of MPX cases at the federal level. CDC reports a total of 2,891 cases through July 22 in the 43 reporting states and DC and Puerto Rico. Case reports with at least some demographic data were available for 1,195 or 41% of these cases, but only 1,095 or 38% of cases had race/ethnicity reported. Cases have continued to climb since the report, with CDC reporting over 14,000 cases as of August 18, 2022.

State and Local Data on MPX Cases and Vaccinations by Race/ Ethnicity

Similar to national data, the limited state data available to date suggest Black and Hispanic people are accounting for disproportionate shares of MPX cases. As of August 22nd, 2022, nine states (California, Colorado, Georgia, Louisiana, New Jersey, New York, North Carolina, Oregon, and Tennessee), and Washington D.C are reporting MPX cases by race/ethnicity. While these states report race/ethnicity of MPX cases and vaccines, they vary in their racial/ethnic categorizations, making it challenging to compare data across jurisdictions. Still, these data raise early concerns about disparities for Black and Hispanic people. The proportion of MPX cases among Black and Hispanic people is higher than their share of the total population in most reporting states (Table1). For example, in Georgia, Black people account for over 80% of cases but less than a third of the population, and in Oregon, Hispanic people account for 28% of cases versus 13% of the population. White people account for a smaller share of cases than their share of the population in most of the states reporting data. Asian people make up a smaller share of cases than their share of the total population in all reporting states.

Four states (Colorado, Georgia, New Jersey, and North Carolina) and Washington D.C. are reporting MPX vaccination data by race and/or ethnicity, which show that Black people are receiving smaller shares of vaccinations compared to their share of cases. For example, in DC, Black people have received 22% vaccines, while they account for 36% of cases. In contrast, White people have received 63% of vaccinations but account for 42% of cases. In Georgia, Black people have received less than half of vaccinations, while they make up 80% of cases. Conversely, White people have received 44% of vaccinations and make up 14% of cases. And, in North Carolina, a quarter (25%) of vaccinations have gone to Black people although they account for nearly three-quarters of cases (73%), while over two-thirds (69%) of vaccinations have been received by White people who account for 20% of cases. Data are mixed for Hispanic people with them receiving a smaller share of vaccinations compared to their share of cases in Colorado (11% of vaccinations vs. 37% of cases) and New Jersey (25% of vaccinations vs. 40% of cases), but a similar or higher share in Georgia (8% of both vaccinations and cases) and North Carolina (8% of vaccinations vs. 4% of cases).

Racial/Ethnic Distribution of MPX Cases and Vaccinations by State, as of August 22, 2022

While this analysis did not include a comprehensive assessment of data being reported at the local level, we did assess data from five cities (Chicago, Houston, Los Angeles, New York City, and Philadelphia) for which the federal government is providing direct MPX vaccine allotments. All five of these cities, except Houston, were reporting MPX cases data by race and/or ethnicity, with Philadelphia and New York City also reporting MPX vaccination data by race/ethnicity as of August 19, 2022 (Table 2). Similar to the state data, for cities reporting both MPX vaccination and cases data, Black people are receiving a smaller share of vaccinations compared to their shares of cases while White people are receiving similar to, or higher shares of vaccinations compared to their share of cases. The city data also show Hispanic people receiving a lower share of vaccinations compared to their share of cases. For example, New York City reports that Black and Hispanic people account for 29% and 34% of MPX cases, respectively, while they have received 12%, and 23% of vaccines. Conversely, White people in New York City account for 32% of MPX cases yet make up nearly half (46%) of MPX vaccine recipients. Similarly, in Philadelphia, Black people make up more than half of MPX cases (57%), yet only account for 23% of MPX vaccine recipients and Hispanic people make up 15% of cases but represent 12% of vaccine recipients. In contrast, White people account for 28% of MPX cases and make up 57% of vaccine recipients.

Racial/Ethnic Distribution of MPX Cases by City and/or County, as of August 19, 2022

Discussion

Overall, it remains challenging to draw strong conclusions about racial equity in MPX cases and vaccinations due to the dearth of comprehensive data, inconsistency in reporting, and the lack of disaggregated data for smaller racial/ethnic groups, particularly NHOPI people. As the COVID-19 pandemic highlighted, having comprehensive data disaggregated by race/ethnicity is fundamental for identifying and addressing disparities.

Although limited, these early data indicate significant and potential growing racial and ethnic disparities in MPX cases and vaccination uptake, highlighting the importance of centering intersectional equity in MPX response efforts, including prevention, testing, and treatment, from the outset. Moreover, addressing challenges that include homophobia, stigma, and discrimination will be key given the disproportionate impacts among men who have sex with other men. Underlying structural inequities place people of color at increased risk for public health threats, as was seen in COVID-19 and as is beginning to be observed amid the MPX outbreak. Early and intentional efforts will be key to minimizing and preventing disparities going forward amid the MPX outbreak and for future public health threats.

 

COVID-19 Cases and Deaths by Race/Ethnicity: Current Data and Changes Over Time

Published: Aug 22, 2022

Over the course of the COVID-19 pandemic, analyses of federal, state, and local data have shown that people of color have experienced a disproportionate burden of cases and deaths. This brief examines racial disparities in COVID-19 cases and deaths and how they have changed over time based on KFF analysis of data on COVID-19 infections and deaths from CDC. It updates a February 2022 analysis to reflect data through mid-2022, amid the ongoing surge associated with the Omicron variant. It finds:

  • Total cumulative data show Black, Hispanic, American Indian or Alaska Native (AIAN), and Native Hawaiian or Other Pacific Islander (NHOPI) people have experienced higher rates of COVID-19 cases and deaths compared to White people when data are adjusted to account for differences in age by race and ethnicity.
  • Disparities in infections and deaths have both widened and narrowed at various times over the course of the pandemic, with disparities generally widening during periods in which the virus has surged and narrowing when overall infection rates fall. In data that has not been adjusted for age, there were some periods when death rates for White people were higher than or similar to some groups of color. However, in the age-adjusted data, White people have lower death rates than AIAN, Black, and Hispanic people over most of the course of the pandemic and disparities are larger for AIAN, Black, and Hispanic people, reflecting an older White population and higher rates of death across all age groups among people of color compared to White people.

Continuing to assess COVID-19 health impacts by race/ethnicity is important for both identifying and addressing disparities and preventing against further widening of disparities in health going forward. While disparities in cases and deaths have narrowed and widened over time, the underlying structural inequities in health and health care and social and economic factors that placed people of color at increased risk at the outset of the pandemic remain. As such, they may remain at increased risk as the pandemic continues to evolve and for future health threats, such as the Monkeypox virus, for which early data show similar disparities emerging.

Disparities in Total COVID-19 Cases and Deaths

As of August 5, 2022, the Centers for Disease Control and Prevention (CDC) reported a total of over 84 million cases, for which race/ethnicity was known for 65% or over 55 million, and a total of over 880,000 deaths, for which race/ethnicity was known for 85% or over 750,000. These estimates are based on a subset of data for which case-level demographic information has been reported to CDC by state health departments, so they differ from those reported elsewhere. For example, CDC reports a total of over 1 million deaths from COVID as of August 5, 2022. Data on cases also are likely significantly underreported as they do not reflect individuals who test positive on home tests and do not report findings to their public health agency.

Total cumulative data show that Black, Hispanic, AIAN, and NHOPI people have experienced higher rates of COVID-19 cases and deaths than White people when data are adjusted to account for differences in age by race and ethnicity. Age-standardized data show that that NHOPI, Hispanic, and AIAN people are at about one and a half times greater risk of COVID-19 infection than White people, and AIAN, Hispanic, NHOPI, and Black people are about twice as likely to die from COVID-19 as their White counterparts (Figure 1). The CDC also reports large disparities in COVID-19 hospitalizations for AIAN, Black, and Hispanic people. (CDC data does not include separate reporting for NHOPI people.) Adjusting for age when comparing groups on health measures is important because risk of infection, illness, and death can vary by age, and age distribution differs by racial and ethnic group. Age adjustment allows for direct comparison between groups on health measures independent of the age distribution differences. For example, unadjusted rates underestimate racial disparities for COVID-19 deaths, since the White population is older compared to populations of color and COVID-19 death rates have been higher among older individuals. Age adjustment has limited impact on case rates by race and ethnicity, suggesting that age plays a more limited role in risk of infection.

Cumulative COVID-19 Age-Adjusted Infection Rates by Race/Ethnicity, 2020-2022

Disparities in Cases and Deaths Over Time

Analysis of monthly data on COVID-19 infections from CDC and deaths from NCHS shows disparities in infections and deaths have both widened and narrowed over the course of the pandemic. During periods in which the virus has surged, disparities have generally widened, while they have narrowed when overall infection rates fall. However, over the course of the pandemic, specific patterns of disparities have varied by race and ethnicity and between cases and deaths.

Figure 2 shows age-adjusted monthly data on cases by race and ethnicity between April 2020 and July 2022.  There were no notable differences in patterns of disparities between unadjusted and age-adjusted data for reported cases. The data show that during periods of resurgence, disparities have generally widened for people of color compared to White people, while they have narrowed when overall infection rates fall.

  • Early in the pandemic Hispanic, AIAN, and Black people had higher rates of COVID-19 cases compared to their White counterparts, with a particularly high rate among Hispanic people. Asian people had the lowest monthly infection rate.
  • Case rates increased through Winter 2021, spiking in December 2021, and disparities further widened for Hispanic and AIAN people. Following the spike in December 2021, monthly infection rates fell across all groups. By June 2021, gaps between groups narrowed.
  • By August 2021, infection rates rose again across groups, reflecting the spread of the Delta variant, with higher rates for Black and AIAN people compared to White people during this resurgence.
  • During January 2022, infections sharply rose again across all groups amidst the spread of the Omicron variant, resulting in the highest case rates recorded since the start of the pandemic. Rates were higher for all groups of color compared to White people, with AIAN people having the highest infection rate during this surge, followed closely by Hispanic people. This surge was also the first time since early in the pandemic that the infection rate for Asian people was comparable to other groups of color. By February 2022, infection rates had fallen sharply, and disparities once again narrowed.
  • Between Winter 2022 and June 2022 case rates have fallen across groups. In April, Asian people had the highest infection rate and Black, AIAN and Hispanic people had the lowest infection rate, marking the first time these patterns were observed since the start of the pandemic. More research is needed to understand the factors that contributed to this recent shift in trends. However, by June 2022, this pattern began to reverse with White people having the lowest infection rate, and Hispanic people at highest risk of infection in June and July.
COVID-19 Monthly Age-Adjusted Cases in the United States per 100,000 by Race/Ethnicity, April 2020 to July 2022

Figure 3 presents age-adjusted monthly data on COVID-19 deaths by race and ethnicity between April 2020 and May 2022. Both the unadjusted and age-adjusted data show large disparities in death rates for people of color during surges in the pandemic, with the highest rates among AIAN people for most of the pandemic. In unadjusted data, there were some periods when death rates for White people were higher than or similar to some groups of color. However, in the age-adjusted data, White people have lower death rates than AIAN, Black, and Hispanic people over most of the course of the pandemic and disparities are larger for AIAN, Black, and Hispanic people, reflecting an older White population and higher rates of death across all age groups among people of color compared to White people.

  • During the initial surge in summer 2020, AIAN, Hispanic, and Black people experienced higher rates of death than White people. As of July 2020, Hispanic people were five times more likely to die than White people, while AIAN and Black people were roughly four and three times as likely to die than White people, respectively. Asian people were at a similar risk of death as White people.
  • Deaths peaked in December 2020 and January 2021 across groups. AIAN and Hispanic people had the highest rates of death, and all groups of color had a higher death rate than White people as of January 2021. Following that surge, death rates fell across all racial and ethnic groups and disparities narrowed by early summer 2021.
  • Moving into late summer 2021, amid the rise of the Delta variant, death rates rose and were highest among AIAN people. Black and Hispanic people had higher death rates than White people, while Asian people had the lowest rate of death Following the Delta surge, monthly deaths declined slightly across groups with higher rates of death for AIAN people persisting before climbing again across groups in late 2021 with the arrival of the Omicron variant.
  • As of January 2022, during the Omicron surge, Black, AIAN, and Hispanic people experienced the highest rates of death (36.5, 33.1, and 29.4 per 100,000 population, respectively). Following the spike in early 2022, monthly death rates fell across all groups and gaps between groups narrowed with similar rates of death across all groups (2.4 per 100,000 for AIAN and White people, 2.0 per 100,000 for Black people, 1.5 per 100,000 for Hispanic people, and 1.4 per 100,000 for Asian people.)
COVID-19 Monthly Age-Adjusted Deaths in the United States per 100,000 by Race/Ethnicity, April 2020 to May 2022

Discussion

In sum, these data show that, overall, Black, Hispanic, and AIAN people have experienced higher rates of COVID-19 infection and death compared to White people when accounting for age differences across racial and ethnic groups. The age-adjusted data also suggest that while these disparities have narrowed at times over the course of the pandemic, people of color are disproportionately impacted by surges caused by new variants, with disparities widening during these periods.

The higher rates of infection among people of color likely reflect increased exposure risk due to working, living, and transportation situations, including being more likely to work in jobs that cannot be done remotely, to live in larger households, and to rely on public transportation. Black, Hispanic, and AIAN people have experienced the highest age-adjusted death rates amid each resurgence period, reflecting higher rates of death across all age groups among people of color compared to White people and an older White population. Overall, death rates have decreased across groups over the course of the pandemic as vaccination rates have increased and vaccination – particularly with boosters – continues to be highly effective at reducing the risk of hospitalization and death. Early disparities in vaccination rates by race and ethnicity have also narrowed over time. However, data suggest potential disparities in access to COVID-19 treatments, which will be important to monitor going forward.

Looking ahead, continuing to assess COVID-19 health impacts by race/ethnicity is important for both identifying and addressing disparities and preventing against further widening of disparities in health going forward. While disparities in cases and deaths have narrowed and widened during different periods over time, the underlying structural inequities in health and health care and social and economic factors that placed people of color at increased risk at the outset of the pandemic remain. As such, they may remain at increased risk as the pandemic continues to evolve and for future health threats, such as the Monkeypox virus, for which early data show similar disparities emerging.

Methods

This analysis uses data from multiple sources including the Centers for Disease Control and Prevention (CDC) COVID Data Tracker, the Centers for Disease Control and Prevention COVID-19 Response. COVID-19 Case Surveillance Data, the National Center for Health Statistics (NCHS) Provisional COVID-19 Deaths, and the Census Bureau Annual Estimates of the Resident Population. Unless otherwise noted, race/ethnicity was categorized by non-Hispanic White (White), non-Hispanic Black (Black), Hispanic, non-Hispanic American Indian and Alaska Native (AIAN), non-Hispanic Asian (Asian), and non-Hispanic Native Hawaiian or Other Pacific Islander (NHOPI).

Data on COVID-19 infections includes reported cases as of August 1, 2022 and data on COVID-19 deaths includes provisional deaths as of July 30, 2022 where race/ethnicity was available. Age-adjusted infection and death rates were standardized to the 2019 U.S. Census Bureau Annual Estimates of the Resident Population using the direct method of standardization. The direct method of age standardization is calculated by multiplying the age specific crude rates for each population by the appropriate weight in standard population and summing them to produce an age-standardized rate. Data for monthly age adjusted NHOPI cases and deaths not shown due to small number of observations. Age standardization allows for direct comparison of groups on health measures independent of differences in age distribution that may influence the measure being examined.

News Release

Mental Health Parity at a Crossroads

Published: Aug 18, 2022

With federal agencies preparing to update parity regulations and Congress considering reforms, our new report explains existing parity requirements, including who they apply to and how their enforced, and highlights key policy issues raised by regulators and advocates that could be addressed in the future.

Those issues include:

  • Simplifying parity standards to make it easier for apples-to-apples comparisons between behavior health and medical benefits needed to assess compliance.
  • Taking a closer look at how insurers are using medical necessity criteria to make behavior health coverage decisions that potentially could affect people’s ability to access care.
  • Developing enforceable standards to determine the adequacy of provider networks across the full spectrum of behavioral health needs from screening to inpatient treatment, including the potential and limits for telehealth providers to provide coverage.
  • Evaluating enforcement tools to allow regulators to better assess and enforce parity compliance, potentially including additional data-reporting requirements or incentives for employers or other plan sponsors to monitor their behavioral health coverage.

In weighing possible changes to address such issues, other factors to consider include the potential increased costs resulting from expanded behavior health coverage as well as an existing workforce shortage and other structural problems.

The brief is part of KFF’s ongoing work examining what consumers and patients face as they navigate the health care system, including financial barriers, administrative complexity, lack of transparency, and problems accessing providers.

Mental Health Parity at a Crossroads

Author: Kaye Pestaina
Published: Aug 18, 2022

Issue Brief

More than 25 years after the first federal mental health parity protections were put in place, adequate coverage for behavioral health (BH) care – including both mental health and substance use conditions –remains elusive for many consumers with health insurance.1   Federal BH parity rules require health plans that offer BH coverage to ensure that financial requirements (such as deductibles, copayments, coinsurance, and out-of-pocket limits) and treatment limits (such as day and visit limits as well as nonquantitative limits on benefits such as prior authorization) on these benefits are no more restrictive than those on medical and surgical benefits. The COVID-19 pandemic has heightened awareness and exacerbated existing challenges in BH. Strengthening BH parity protections is just one part of a larger policy discussion that includes addressing the BH workforce shortage,  rising BH treatment needs among children and youth, an inadequate health care infrastructure to address those in crisis, and the need for improved coordination and integration of primary care and BH care in the health care delivery system.

All of these issues contribute to the access and coverage challenges in health insurance that BH parity was supposed to address. The stakes are high for coverage protection, as nearly 90% of nonelderly individuals with a BH condition have some form of health coverage. Despite having coverage, many insured adults (36%) with moderate to severe symptoms of anxiety and depression did not receive care in 2019. There have been consistent calls for more federal guidance on the specific protections in the federal BH parity law, as well as for increased enforcement. As Congress2  debates reforms to address these concerns in BH care, and as federal agencies plan to update parity regulations, this brief explains the federal BH parity requirements – including who they apply to and how they’re enforced — and sets out key policy issues.

Federal BH Parity Protections

Federal protections for BH coverage sought to correct historical differences in how health insurance covered this care when compared to medical/surgical benefits. The focal point of these protections has evolved over the years from the narrow initial federal law, the Mental Health Parity Act of 1996 (MHPA), to the broader protections in the current law, the Mental Health Parity and Addiction Equity Act of 2008 (MHPAEA) (Appendix Figure 1).

With MHPAEA, an initial focus on ensuring that consumers were not subject to higher cost sharing and more restrictive day and visit limits for BH shifted to looking at disparities in treatments limits in coverage that are not expressed numerically. These so-called “nonquantitative treatment limits” or NQTLs include plan features that limit the scope or duration of care such as prior authorization requirements and medical necessity reviews, standards for provider admission to a network, and provider reimbursement rates.

Federal regulations implementing parity for commercial plans and for Medicaid and the Children’s Health Insurance Program (CHIP) have set out substantially similar protections (See Appendix Table 1 for major differences between commercial parity and Medicaid). These are detailed and complex standards making it a challenge for consumers with coverage to know what practices violate the law. Federal agencies have issued FAQs and other guidance for both commercial and Medicaid/CHIP to explain how these standards work. The basic protections are described below.

who does the law apply to?

Federal BH parity rules apply to most health coverage, public and private, but do not apply to Medicare. Table 1 summarizes the basic rules and exceptions. Parity applies to all health coverage in the individual, small and large group insurance markets, as well as to all private employer-sponsored plans (insured and self-insured) with the exception of self-insured employer plans covering groups of no more than 50 employees, “retiree only” plans, short-term limited duration coverage and coverage considered “excepted benefits.”3  It also applies to self-insured state and local governmental plans (called nonfederal governmental plans), though these plans have the ability to opt out of MHPAEA protections, as hundreds of plans across 31 states currently do.4 

Table 1: When Do Federal BH Rules Apply?
Plan TypeApply?Exceptions
Private Employer-sponsored Group Health Plan:
Self-InsuredYesDoes not apply to self-insured small employer plans, retiree only plans and excepted benefit plans
InsuredYesDoes not apply to retiree only plans and excepted benefit plans
State & Local Government Employer-sponsored Group Health Plan (“Non-federal governmental” plans):
Self-insuredYesCan annually opt out of MHPAEA compliance
InsuredYes
Individual insurance plan (on or off exchange)YesDoes not apply to excepted benefit plans
Small group insurance plan (on or off exchange)YesDoes not apply to retiree only and excepted benefit plans
Medicaid:
MCO/PHIP/PAHP* (providing Medicaid-covered services to managed care enrollees)Yes
Alternative Benefit Plan (both fee-for-service & managed care)YesABPs services provided fee-for-service do not have to comply with MHPAEA rules regarding aggregate lifetime and annual dollar limits.
Fee-for-Service (FFS)NoApplies to state plan services provided on a FFS basis to MCO enrollees (carve-outs from MCO contract); Most parity provisions apply to ABP services provided FFS
Children’s Health Insurance Plan (CHIP)YesParity does not apply to CHIP Medicaid expansion programs that provides services on a FFS basis only, with no MCO enrollees.
Medicare Fee-for-ServiceNo
Medicare AdvantageNo
NOTES: * MCO (Medicaid managed care organization), PHIP (Prepaid inpatient health plan) and PAHP (Prepaid ambulatory health plan) are types of delivery system that provide Medicaid benefits.

Substantially similar parity rules apply to enrollees in certain Medicaid coverage. Medicaid regulations apply parity to any Medicaid service provided to Medicaid managed care (MCO) enrollees, regardless of whether those services are provided by the MCO or on a fee-for-service (FFS) basis. Parity also applies to Medicaid alternative benefit plans (ABP) – which cover adults under the Affordable Care Act’s Medicaid expansion — whether delivered through managed care or FFS.5 

Parity also generally applies to coverage under CHIP regardless of the delivery system.6 

MHPAEA does not apply to Medicare, either fee-for-service or Medicare Advantage.7   Although MHPAEA does not apply to Medicare, the Medicare Improvements for Patients and Providers Act of 2008 did lower cost sharing for outpatient mental health services to match cost sharing for Medicare outpatient medical treatment.

The MHPAEA law does include a cost exemption provision that would allow plans to apply for a temporary exemption from compliance if costs exceed a certain threshold in a given plan year.8  Medicaid parity regulations did not extend this cost exemption in the statute to Medicaid and CHIP.

are plans required to cover bH services under MHPAEA?

MHPAEA does not itself require plans to provide BH benefits, nor does it require coverage of any particular treatment or condition. Other laws requiring coverage of certain BH services apply to some plans and not others, creating a complicated landscape to navigate for consumers seeking these services. The landscape is best summarized as follows:

BH services are among “essential health benefits” that the ACA requires plans to cover in the individual and small group insurance market. The ACA required these plans to cover items and services for “mental health and substance use disorder services, including behavioral health” equal in scope to a “typical employer plan.” Covered EHB behavioral health benefits are not uniform from state to state, since EHB regulations require each state to pick a set of covered benefits from one of four “EHB benchmark plans.”9  The ACA’s essential health benefits requirement does not apply to large employer plans.

Most states have some form of BH coverage mandate that will apply to fully-insured employer plans and individual market policies. These requirements differ from state to state, and could require insurers to cover specific services such as applied behavioral analysis (ABA) for autism. These requirements might also set process rules that insurers must follow such as limits on use of prior authorization for a BH item or service.

Federal Medicaid law does not require states to cover any particular BH benefit. While Medicaid plays a major role as a source for BH coverage, covering a large percentage of adults in the U.S. with BH conditions, states have the flexibility to determine which services to offer in their Medicaid programs, subject to federal minimum requirements. Some BH services may be covered in mandatory state plan categories such as physician services or outpatient hospital services, while others are covered under optional state plan categories. As discussed earlier, instead of a state plan benefit package, states must provide ABP benefits to newly eligible adults if they choose the ACA’s Medicaid expansion. ABPs, similar to the ACA essential health benefit framework, are required to cover BH equivalent to a state’s chosen private sector benchmark. Note that Medicaid does not pay for certain inpatient BH services due to the Medicaid program’s “institutions for medical disease” (IMD) exclusion.10  CMS stated in Medicaid parity regulations that it considers this exclusion beyond the scope of the parity regulation.

Self-insured private employer plans – which are typically offered by larger employers but increasingly by smaller employers as well — have no requirement to cover any BH services, as neither state mandates nor the ACA’s essential health benefit requirements apply to these plans. Parity protections for these plans are only triggered if these employers offer BH coverage. While most employers provide some form of BH coverage, there is a growing amount of private litigation about whether the parity law requires that plans cover specific BH treatments that are either excluded or determined not medically necessary under the plan.

WHAT IS CONSIDERED A BH BENEFIT VERSUS A MEDICAL BENEFIT?

While the parity rules require plans to compare coverage for BH to medical/surgical coverage in each classification, there is no standard definition for what is a BH condition versus what is a medical condition.

The parity regulations provide that plans and states must use a “generally recognized” standard for defining what benefits are considered BH items and service but does not require use of a specific standard. A standard is considered generally recognized when it is generally accepted in the relevant medical community.11  Examples of generally accepted standards mentioned in MHPAEA regulations include the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM), the International Classification of Disease (ICD) or state guidelines. of the DSM as the baseline, for instance, would include a broad range of conditions as BH, such as sleep apnea or Alzheimer’s disease, according to the CMS Parity toolkit.12  Use of other standards might narrow what is considered BH. As a result, what diagnoses are treated as BH versus medical could differ from plan to plan. This may come up for certain conditions such as autism or nicotine addiction resulting in uneven protections.

The 21st Century Cures Act of 2016 (the Cures Act) did attempt to clarify that for the purposes of MHPAEA, treatment of eating disorders is considered a BH treatment, not medical. This requires treatments such a nutritional counseling for anorexia to be considered BH treatment under the parity law.

what are the parity standards?

Federal parity regulations include protections in three distinct areas—quantitative standards, nonquantitative treatment limits and disclosure rules– each with its own set of rules. For quantitative and nonquantitative rules, commercial plans and state Medicaid agencies must evaluate whether there is parity between medical and BH benefits separately within specific benefits classifications. BH benefits for inpatient in-network care, for example, is compared to medical care benefits covered in the same inpatient in-network classification. Commercials plans must perform an analysis comparing medical to BH benefits in each of six classifications:

  • Inpatient, in-network benefits
  • Inpatient, out-of-network benefits
  • Outpatient, in-network benefits
  • Outpatient, out-of-network benefits
  • Prescription drug benefits
  • Emergency benefits

For Medicaid and CHIP, states and plans providing this coverage must compare benefits within four of the six classifications, the two out-of-network classifications are not included.

The three main requirements of the federal BH parity regulations are discussed below:

1. Quantitative Standards

MHPAEA affects how plans and state Medicaid agencies can design their patient cost-sharing and quantitative restrictions on BH benefits. The rules fall into one of three categories:

Financial requirements and quantitative treatment limits. Plans cannot place financial requirements (copays, coinsurance) and quantitative limits (day and visit limits) on BH that are more restrictive than the predominant financial requirement or quantitative limit applied to substantially all medical benefits. The federal agencies defined these terms and developed a formula to evaluate each financial requirement and quantitative limitation. The tests are the same for private coverage and Medicaid, (although longstanding Medicaid rules already limit cost sharing, with some populations and services exempt from any cost sharing).13  These are complicated tests that typically involve looking at medical and surgical claims data to estimate the total dollars expected to be paid for medical/surgical benefits in a given year for each financial requirement and treatment limitation.

“Cumulative” financial requirements and treatment limits. The parity law prevents the application of a separate cumulative financial requirement (deductibles and out-of-pocket maximums) or cumulative quantitative treatment limit (such as caps on visits) to just BH benefits. This means that both BH and medical/surgical benefits must count toward the same deductible, out-of-pocket maximum, and visit cap. The Medicaid parity rules diverge slightly by not prohibiting cumulative quantitative treatment limits such as visit caps from accumulating separately for BH and medical/surgical care.14 

Aggregate dollar limits. A different quantitative test15  applies to annual and lifetime dollar limits, but the ACA’s prohibition on aggregate dollar limits on EHBs now largely restricts these types of limits for most commercial plans.

2. Non-Quantitative Treatment Limits

Any limitation on coverage other than numerical limits is considered an NQTL. This could bring in an endless variety of coverage design standards. MHPAEA regulations provide a non-exhaustive list of common NQTLs, such as medical management standards that limit benefits based on medical necessity or whether treatment is experimental or investigational. This would include any prior authorization requirement to obtain a BH benefit. Other NQTLs include formulary design for prescription drugs, network tier design, standards for provider admission to participate in a network (including reimbursement rates), restrictions in coverage based on geographic location, facility or provider type, standards for providing access to out-of-network providers.16  The Department of Labor (DOL) has issued several sets of FAQ guidance discussing specific types of NQTLs.

Under parity regulations, plans cannot apply a NQTL to a BH benefit unless the “processes, strategies, evidentiary standards and other factors” used to apply that restriction are “comparable to and applied no more stringently” to these benefits than they are to medical surgical benefits. This standard is challenging to implement as there are no bright line rules for determining what is “comparable” and “more stringent,” when comparing benefits. A multistep process (Appendix Figure 2), added to the MHPAEA statute as part of the 2020 Consolidated Appropriations Act, is used to compare BH NQTLs to those for medical/surgical benefits. This requires that plans evaluate the rationale for applying these restrictions to BH benefits and the evidence that backs up the rationale.

Outcomes as a red flag for possible parity violation. Federal agency NQTL guidance has consistently said that outcomes are not determinative of parity compliance. For instance, higher rates of claims denial for preauthorized BH care when compared to medical preauthorization claims, is not alone a violation of the parity law. However, the DOL has said that disparities in this type of data may be a warning sign of a possible problem that the processes used are not comparable. The DOL self-compliance tool suggests that plans evaluate average denial rates of BH and other medical/surgical claims. There is no requirement, however, as part of the NQTL comparative analysis to evaluate this or other data. At least one health care advocacy organization has developed for employers a model data request form that employers and their third-party vendors can use to evaluate compliance for certain NQTLs. Also, New York is one example of a state that has used data analysis as part of their parity evaluation of commercial insurers.

Internal Clinical Guidelines as NQTL. Agency guidance has included the use of medical necessity criteria as a common example of an NQTL. For instance, in an example included in a recent enforcement report to Congress, CMS required an issuer to remove continued stay criteria for a BH condition that required a demonstration of significant improvement in an enrollee’s condition to warrant continued coverage. There were no similar criteria applied to medical/surgical criteria in the same classification.17 

“Home grown” medical necessity standards examined. Much more complicated issues arise when plans make coverage decisions using medical management criteria based on internal clinical guidelines developed by a health plan. An open question under MHPAEA and other laws is whether the use of internal clinical guidelines unique to a specific plan violates parity by applying to BH claims medical necessity standards that are not comparable and are more stringent than those used for medical necessity decisions on medical claims. These guidelines have been criticized as narrowing the scope of covered BH treatments and as inconsistent with generally accepted standards of practice. Wit v. United Behavioral Health (UBH), a class action lawsuit challenging the use of these internal clinical criteria as medical necessity criteria, has received national attention. Box 1. Some states have insurance coverage rules that require the use of specific guidelines or include additional oversight of the clinical guidelines used by private insurers. Rhode Island and Illinois, for example, require use of criteria developed by the American Society of Addiction Medicine when reviewing relevant claims for substance use treatment. California requires plans in state’s private insurance market to cover medically necessary treatment for BH under the same terms and conditions as done for medical care. Insurers are prohibited from using their own clinical coverage criteria and must choose generally accepted guidelines developed by an independent nonprofit organization, subject to state approval.

Box 1: Summary of Wit. v. United Behavioral Health (UBH)

This ongoing class action litigation, while not a case brought under MHPAEA, challenges the use of internal clinical guidelines for BH in making coverage decisions. The named plaintiffs in the case were largely covered by employer-sponsored coverage insured or administered by UBH. Seeking treatment for their substance use conditions, they sued after their plan denied claims for residential or intensive outpatient treatment. In 2019, a California federal district court held that UBH  applied overly restrictive internal clinical guidelines to deny coverage of thousands of BH claims violating certain provisions of the Employee Retirement Income Security Act (ERISA)—the federal law that regulates private sector employer-sponsored health plans. The court concluded that the guidelines were not consistent with generally accepted standards of care because they improperly focused on addressing acute symptoms and stabilizing crises rather than providing effective treatment for the plaintiffs’ underlying BH conditions. The court also concluded that UBHs’ use of these guidelines benefited UBH’s financial bottom line in violation of ERISA. UBH was ordered to reprocess over 60,000 claims, drop its internal guidelines, and choose generally accepted guidelines develop by an independent nonprofit organization.

The Ninth Circuit federal court of appeals overturned this decision in March 2022, stating that the lower federal court should have deferred to the decision of UBH in denying benefits, because their decision was “not unreasonable.”  UBH, according to the court, was not required to cover the care involved even though the treatment may have been consistent with generally accepted standards of care. Plaintiffs have appealed the decision to the full panel of judges on the Ninth Circuit.

NQTLs that affect Network Adequacy. With many reports of narrow network for BH services, there is some potential to address these problems under the parity law. The parity regulations include several plan limitations related to network adequacy as NQTLs that plans must evaluate under the parity framework. This includes rules for provider admission to a network, including reimbursement rates, and restrictions on facility types covered in-network. As with other NQTLs, the inquiry is simply whether the process for setting out standards in these areas is comparable and not more stringent for BH providers versus medical providers. Disparate numbers of BH and medical/surgical providers in a plan’s network might be an indicator of a parity violation but is not itself a violation. Examples of violations in FAQ guidance include:

Setting a network adequacy standard only for medical providers. Setting an appointment wait time standard for admission of medical providers to a network but having no equivalent network adequacy standard for BH providers violates the parity rules, according to agency FAQs18 .

Using different methodologies for setting reimbursement rates for in network providers. Agency guidance has also stated that a methodology that reimburses in-network BH providers at the Medicare rate, while reimbursing medical health providers at two times the Medicare reimbursement violates the parity law. The agencies have also called out as not permissible under MHPAEA disparities in the method used to determine reimbursement for nonphysican practitioners who provide medical care versus nonphysican practitioners who provide BH care.19 

Blanket exclusions as NQTL. An exclusion of all benefits for a particular BH condition is not considered a treatment limit under MHPAEA regulations.20  While other state or federal laws require coverage of certain BH items or services, these types of exclusion do not explicitly violate parity.21   However, if a plan covers treatment for a specific BH condition, the plan must cover treatment for that condition in every classification in which medical benefits are covered. The reach of this provision is unclear, but the rule often comes up in practice to prohibit commercial plans from excluding all substance use disorder benefits in the inpatient, out-of-network classification, if benefits for that substance use disorder are covered in all of the other five classifications. This type of exclusion is considered an NQTL that violates parity.

Private litigation under commercial parity rules has often involved challenges to exclusions of specific treatments such as ABA for autism. The cases are based on language in the MHPAEA law that, plans must ensure that. “…no separate cost sharing requirements or separate treatment limitations apply only to mental health and substance use disorder benefits.” For example, a recent appellate decision in the First Circuit Court of Appeals, N.R. v. Raytheon, allowed a parity act challenge to an exclusion for habilitative services to move forward. That case involved a denial of a claim for nonrestorative speech therapy for a child with autism covered by Raytheon’s self-insured group health plan. Plaintiffs in this case assert that that the exclusion only applied to a BH benefit and therefore violated parity. This is just one example of litigation in this area examining whether exclusions of a specific BH treatment or condition is permitted under MHPAEA.

3. Disclosure Rules

Federal parity law requires plans to disclose to plan enrollees upon request the criteria that is used by the plan to determine the medical necessity of BH treatment.22  The MHPAEA also required that plans provide the rationale for denying a BH claim to an enrollee.

Amendments to the parity law in the 2020 Consolidated Appropriations Act  (CAA) now requires plans to prepare and provide an NQTL comparative analysis and make it available to the relevant government agency upon request.23 

who enforces MHPAEA?

Enforcement of MHPAEA is overseen by several different federal and state agencies depending on the type of plan involved. Table 2 summaries the agencies involved and their enforcement methods.

Table 2: MHPAEA Enforcement
Plan TypeWho Enforces?Enforcement Tools
Private Employer-sponsored Group Health Plan (both insured and self insured):DOL/Treasury
  • DOL can investigate/sue to enforce MHPAEA, refer cases to Treasury for excise tax
  • Treasury can assess excise tax of $100 per day per individual affected
  • Individuals have a private right of action to sue to recover benefits
Non-Federal Employer-sponsored Group Health Plan (State and local government employer plan)CMS
  • CMS can investigate and assess a penalty up to $100 per day per affected individual
  • Individuals have no private right of action under federal law
Individual insurance plan (on or off exchange)State insurance agency/ CMS as federal fallback*
  • State enforcement tools include market conduct exams, penalties and other remedies will differ from state to state
  • CMS can come in where state fails to “substantially enforce” and assess a civil monetary penalty up to $100 a day per day per affected individual; no individual private right of action under federal law
Small group insurance plan (on or off exchange) and large group insurance planState insurance agency/CMS as federal fallback
  • State enforcement tools include market conduct exams, penalties, and other remedies; will differ from state to state
  • CMS can come in where state fails to “substantially enforce” and assess a civil monetary penalty up to $100 a day per day per affected individual; no individual private right of action under federal law
Medicaid plansState Medicaid agency/CMS
  • State enforcement tools vary, could include new contract requirements on MCO/PHIPs/PHAPs
  • CMS oversees state implementation by auditing requirements, requiring states document compliance, etc.
NOTES: *”federal fallback” means CMS enforces when state does not substantially enforce the requirement.

While DOL can investigate and enforce violations of the parity law for private employer plans, it does not have independent authority to assess a civil penalty specific to parity.24   It can refer cases to Treasury, which can assess a $100 a day penalty per affected individual for violations.25  A provision in ERISA26  prevents DOL from enforcing parity and other federal health insurance protections, in certain circumstances, directly against insurers. Unless the insurer functions as a “fiduciary” under the plan, DOL has limited ability to enforce most of ERISA’s health coverage provisions directly against these entities.27 

CMS and state Medicaid agencies oversee compliance activity for federal Medicaid parity rules. States generally are responsible to determine whether the overall delivery system complies with the Medicaid/CHIP parity rule and make documentation available through a public posting and to CMS.28  States have other oversight tools as part of their contracts with managed care entities, such as performance guarantees (penalties assessed when specific contract requirements are not met).

Evidence of low parity compliance. Earlier this year, DOL, Treasury and the Department of Health and Human Services (HHS) jointly issued a report to Congress on the current status of agency parity enforcement for non-Medicaid plans. The report raised alarms in its conclusion that in 2021 none of the NQTL comparative analysis submitted by employer-sponsored plans and insurers, upon initial agency review, contained information sufficient to meet the standards required under the parity law.29  According to the agencies, some plans were completing an NQTL comparative analysis for the first time in 2021 only in response to the agency request for the analysis. These plans had imposed NQTLs on BH benefits for years without performing a compliance assessment. Other plans provided a compliance analysis with limited information on the factors and evidentiary standards supporting use of specific NQTLs. For Medicaid plans, there is concern that few substantive changes were made in response to parity implementation. A 2021 report prepared by the Medicaid and CHIP Payment and Access Commission (MACPAC) concluded that implementation of MHPAEA had done little to substantially improve access to BH care for Medicaid and CHIP beneficiaries.

DOL recommends enforcement enhancements. The DOL has requested tools to expand its ability to enforce the law. The Department’s  2022 report to Congress includes proposals for legislative changes to improve their enforcement related to private employer-sponsored ERISA plans. DOL calls on Congress to amend ERISA30  to:

  • Give DOL the authority to assess civil monetary penalties for MHPAEA violations
  • Allow DOL to pursue “all appropriate actors” when enforcing the law, not just plan sponsors and others meeting the definition of fiduciary. This would include the third parties who design and implement BH benefits for employers. Third party administrators for large, self-insured employer plans, for example, are often the same insurance companies that provide fully insured coverage not only to employer plans, but in the individual and small group insurance market and in Medicaid managed care plans.
  • Clarify that participants and beneficiaries who are harmed by a parity violation can seek relief to recover what they lost due to the violation. The parameters of this “make whole” relief is not specified in the report.

First federal litigation for parity violations. In August 2021 the DOL initiated its first litigation to enforce the parity law. In Walsh v. United Behavioral health (UBH), the agency alleged, among other issues, that UBH (acting as an ERISA fiduciary) reduced reimbursements for certain out-of-network non-physician BH providers, but did not do so for non-physician medical providers. The case was settled for $15 million, with UBH agreeing to take specific actions to address the NQTL. The settlement was done jointly by the DOL and the New York Attorney General.

Looking Ahead: Key Policy Issues

Implementation of federal BH parity protections is at an inflection point. As evidence of poor compliance increases, so have questions about whether the current standards are effective to address the issues. proposed legislation specific to BH parity is expected . Several pieces of pending legislation include parity changes, including legislation that recently passed the House that would remove the ability of self-insured state and local government plans to opt out of parity compliance and provide funding to states to enforce commercial parity rules. Other pending bills would increase DOL’s enforcement authority over parity through additional civil penalties and other enforcement features.31 

Key issues raised by both advocates and regulators include:

Simplifying standards. Parity standards are solely based on a comparison of BH benefits to what is provided for medical benefits by each plan. These comparisons are often challenging to make given the differences between BH care and medical care, where apples-to-apples comparisons are sometimes difficult. Also, since MHPAEA does not require coverage of any specific BH service, these comparisons will differ from plan to plan, making the comparison even more complicated and resource intensive. Federal regulators in their 2022 Enforcement Report to Congress suggested that the use of more objective and uniform national standards as part of MHPAEA may strengthen BH coverage and provide clearer rules that consumers can understand and that will apply regardless of the type of health coverage or the state where they live. Developing external benchmarks that are at a minimum based on nationally recognized standards of care and that change as the evidence base evolves is one possible policy direction. For example, a uniform approach to determining what covered items and services are considered BH versus medical would simplify the required comparative analysis.

Taking a closer look at medical management criteria. The Wit case and similar litigation shines a light on the medical necessity criteria used to make BH coverage decisions. Some states, such as California, now have laws that require insurers use independent criteria developed by a nonprofit to determine medical necessity for BH claims. The federal parity law, however, does not include equivalent standards or require plans to develop coverage criteria based on generally accepted standards of care. Closer scrutiny of the medical management techniques applied to BH claims may take place because of this litigation. For instance, the DOL’s first parity litigation involved use of alleged arbitrary thresholds for utilization review of certain psychotherapy treatment.

Assessing how parity can address network adequacy challenges. Evaluating what an adequate BH network of providers looks like, one that addresses the full spectrum of BH needs from screening to inpatient services, will be key to developing a standard that can be enforced. This will likely involve looking at existing networks and the disparate number of in-network BH providers, not simply the processes involved in developing the network that is the current focus of NQTL review. Phantom networks, where providers listed in the network directory are no longer taking new patients, is a particular problem for accessing BH services.

Development of a common approach to assess BH network adequacy may be useful. New regulatory standards for the federal marketplace will include time-distance access standards for BH providers, as well as appointment wait time standards for BH visits. CMS regulations do require states to set a quantitative network adequacy standard for Medicaid managed care plans and the agency has developed a toolkit for states to use in monitoring BH network adequacy. For Medicaid, network adequacy standards will differ from state to state, and a recent study found that care may still be concentrated among a small number of providers even where quantitative network adequacy standards apply. Similar network adequacy initiatives could focus on large, self-insured employer plans, who currently do not have any network adequacy requirements under parity or any other federal law.

Use of telehealth is an important part of improving access. An additional part of the inquiry is looking at access problems that cannot be addressed through telehealth alone, including treatment for those with serious mental illness. Finally, where the provider shortages in certain areas mean plans cannot meet a network adequacy standard for BH, some mechanism for patients to obtain out-of-network care at in-network rates may be important.

Evaluating enforcement tools. The limited compliance with parity standards might be improved by giving the agencies more mechanisms to hold plans accountable; and by exercising current regulatory authority in different ways.

With respect to coverage decisions, additional data reporting could inform oversight of how plans cover BH services compared to other claims. Federal transparency data reporting, enacted as part of the ACA in 2010, requires private health plans, including employer sponsored plans, to periodically report data to the federal government related to how coverage works in practice. Among other things, the law requires plans to report data on claims denials. To date this requirement has been implemented on a limited basis for nongroup plans in the federal marketplace, and not at all for employer-sponsored plans. For Medicaid plans, CMS and states might be able to use their existing oversight authority to require this type of data reporting.

Finally, enforcement enhancements that create incentives for plan sponsors to perform program oversight of their BH coverage is another option being evaluated. Employers, for example, may be limited in their ability to self-monitor compliance unless third party insurers and other entities that administer their plans share information about how they design and manage BH benefits. More emphasis on data sharing with vendors so employer plan sponsors can perform their monitoring role might improve compliance. CMS and state Medicaid agencies, by contrast, have oversight responsibilities under Medicaid parity rules, along with documentation requirements for public disclosure of the comparative analysis. They may already have the ability to access information from MCOs and other entities and can utilize this authority to enhance their parity enforcement.

These possible changes to BH parity come at the same time as concerns about a BH workforce shortage and other structural problems in the delivery of BH care that will impede progress even if BH coverage is expanded. Also, there are always tradeoffs between increased costs versus broader BH coverage that will be important to consider as existing BH medical management practices are reevaluated.

This work was supported in part by Well Being Trust. KFF maintains full editorial control over all of its policy analysis, polling, and journalism activities.

Appendix

Figure 1: Evolution of Federal Mental Health Parity
Appendix Table 1: Major Differences – Commercial MHPAEA v. Medicaid MHPAEA
TopicCommercialMedicaid
Definition of behavioral healthPlans define what is considered behavioral health versus medical care.State Medicaid agencies can define what is behavioral health versus medical care. MCOs, PHIPs and PAHPs must follow the state’s definition.
Long term care servicesLong term care services are considered “excepted benefits” that do not have to comply with parity.Long term care services generally must meet parity standards.
Out-of-network classificationsPlans must evaluate parity for out-of-network inpatient care and out-of-network outpatient care.The two out-of-network classifications do not apply to Medicaid MHPAEA.
Network tiersPlans can determine parity separately for each provider network tier in a classification.Medicaid parity regulations do not give Medicaid plans with network tiers the ability to evaluate parity within each tier. Parity is evaluated for all care within a classification.
Cumulative Quantitative Treatment LimitsQuantitative treatment limits (such as a limit on the number of visits for a service per year) cannot accumulate separately for behavioral health and medical.Quantitative treatment limits can accumulate separately for behavioral and medical services if certain standards are met.
Disclosure of reasons for denial of a claimPlans subject to ERISA must provide the denial reason in the form/manner included in  ERISA claims review rules that are separate from commercial parity rules.  NonERISA plans that follow this form and manner are deemed to comply with this parity requirement.Medicaid is subject to its own form/manner notice standards for providing the reason for a claim denial in adverse action notice regulations that are separate from Medicaid parity rules.
Cost exemptionPlans can qualify for a temporary exemption from meeting the MHPAEA standards if they meet certain requirements for cost increases.Medicaid regulations do not apply the cost exemption to Medicaid.
SOURCE: 70 Federal Register 68240-68296; 80 Federal Register 18390-18445
Figure 2: Steps in NQTL Comparative Analysis

Endnotes

  1. Behavioral health care includes any medication or service used to treat a mental health and/or substance use disorder and would include conditions and treatment for developmental disabilities such as autism. ↩︎
  2.   Development of a bipartisan package of legislative reforms on various aspect of behavioral health care concerns is underway in both chambers of Congress. The Senate Finance Committee issued a Request for Information in 2021 and in March 2022 released a report, Mental Health Care in the United States: The Case for Federal Action, that includes a discussion of mental health parity.  https://www.finance.senate.gov/chairmans-news/finance-committee-releases-bipartisan-report-on-mental-health-care-in-america- ↩︎
  3. Parity applies regardless of status as a grandfathered or nongrandfathered plan if the plan covers BH. ↩︎
  4. Plans can opt out annually using a process administered by CMS. 42 U.S.C. § 300gg-21(a)(2). According to the last CMS posted list, over 200 nonfederal plans have opted out of compliance with MHPAEA. ↩︎
  5. ABPs are an alternative to the traditional Medicaid state plan benefit package and are based on private insurance designs. States can generally choose to offer ABPs but are required to offer them to newly eligible adults if the state expands Medicaid to this population under the ACA. The Medicaid parity rules do not apply to other Medicaid covered BH services that are provided fee-for-service although CMS encourages states to apply parity protections to those services. ↩︎
  6. Variations in how MHPAEA applies to Medicaid expansion CHIP programs, separate CHIP programs and programs that are a combination of both are described in section 8 of the CMS Parity Compliance Toolkit Applying Mental health and Substance Use Disorder Parity Requirements to Medicaid and Children’s Health Insurance Programs (January 2017), pp. 57-58. https://www.medicaid.gov/sites/default/files/2019-12/parity-toolkit.pdf ↩︎
  7. This means that parity does not apply to Medicare Part A, B or D services covered by Medicaid MCOs through integrated plans for dual eligible (Medicare and Medicaid) beneficiaries. 61 Federal Register 18424 (March 30, 2016). ↩︎
  8. 29 U.S.C. § 1185a(c)(2). Federal agencies have not indicated whether any plans have applied for this exemption. ↩︎
  9. The EHB benchmark choices are (1) any of the three largest state employee plan health benefit options, (2) the largest plan in any of the three largest small group insurance market products, (3) any of the three largest national Federal Employee Health Benefit Program (FEHBP) national plan options, and (4) the coverage offered by the largest insured commercial non-Medicaid HMO in the state. 45 C.F.R. § 156.100. ↩︎
  10. The IMD exclusion prohibits Medicaid payment for services for adults aged 22 to 64 in institutions with more than 16 beds that are primarily engaged in behavioral health treatment. While this might be considered a limitation on coverage that must be compared with limits on medical benefits in the inpatient classification under MHPAEA, CMS did not include this as part of their Medicaid MHPAEA regulation. Despite the exclusion, some states use Section 1115 waivers and other options to finance IMD services. Kaiser Family Foundation, State Options for Medicaid Coverage of Inpatient Behavioral Health Services (November 2019) https://modern.kff.org/medicaid/report/state-options-for-medicaid-coverage-of-inpatient-behavioral-health-services/ ↩︎
  11. 75 Federal Register 5412 (February 2, 2010). ↩︎
  12. CMS, Parity Compliance Toolkit Applying Mental health and Substance Use Disorder Parity Requirements to Medicaid and Children’s Health Insurance Programs, pp. 10-15. ↩︎
  13. 42 C.F.R. §§ 438.910(c)(4), 440.395(b)(3)(iv). ↩︎
  14. 81 Federal Register 18398-9 (March 30, 2016). ↩︎
  15. Separate dollar limits can apply to BH and medical/surgical benefits but more restrictive limits on BH are prohibited. For example, if an annual aggregate dollar limit applies to less than two-thirds of expected payments for medical/surgical benefits in a year, plans cannot impose any annual aggregate dollar limit on BH benefits in a year. ↩︎
  16. 45 C.F.R. § 146.136(c)(4)(ii); 42 C.F.R. § 438.910(d)(2) ↩︎
  17. DOL, Treasury and HHS, 2022 MHPAEA Report to Congress (January 2022) p. 34 ↩︎
  18. DOL, Treasury and HHS, FAQs About Mental Health and Substance Use Disorder Parity Implementation & the 21st Century Cures Act Part 39 (FAQs Part 39), Question 7 (September 5, 2019). https://www.dol.gov/sites/dolgov/files/EBSA/about-ebsa/our-activities/resource-center/faqs/aca-part-39-final.pdf ↩︎
  19. FAQs Part 39, Question 6. ↩︎
  20. 45 C.F.R. § 146.136 (definition of “treatment limitation”) ↩︎
  21. FAQs Part 39, Question 4. ↩︎
  22.   A model disclosure request form was developed to facilitate enrollee requests for this information as well as information on a plans’ comparative analysis of NQTLs.  The form is optional for plans to use and was created by DOL and the other agencies who implement the non-Medicaid parity rules. ↩︎
  23. Medicaid plans are generally deemed to comply with these CAA rules if they met certain disclosure requirements in Medicaid parity regulations.  In 2017, states had to document similar parity analysis and post it publicly on a website for benefits provided to Medicaid managed care enrollees. Similar documentation rules apply to Medicaid Alternative Benefit Plans. 42 CFR § 438.920; 42 CFR § 440.396. ↩︎
  24. Separate from MHPAEA, DOL can assess civil penalties in other circumstances, such as a failure of a plan sponsor to disclose certain information to a participant upon request (such as medical necessity criteria). DOL also has other enforcement tools under ERISA to bring actions against those who are “fiduciaries” under ERISA. An ERISA fiduciary is any entity that has discretionary authority or control of a plan’s management or administration or the disposition of plan assets. Who is a fiduciary is often very fact specific, subject to evolving case law, and has generally left out the entities that design and control the key factors that determine what health benefits are covered. ↩︎
  25. It is not clear whether Treasury has ever assessed a penalty using this authority. According to a 2019 General Accountability Office (GAO) report on parity enforcement, the DOL has never referred a case to Treasury to assess a penalty. GAO, State and Federal Oversight of Compliance with Parity Requirements Varies (December 2019) (GAO Parity Enforcement Report), p. 25  https://www.gao.gov/assets/gao-20-150.pdf ↩︎
  26. 29 USC § 1132(b)(3). ↩︎
  27. Insurers are often the third-party administrators (TPA) of self-insured plans and are the entities that design and implement behavioral health benefits, build provider networks, and oversee medical management structures. Employers often adopt the TPA’s prototype plan as is with limited changes. These TPA activities are not always defined by courts as ERISA fiduciary functions. ↩︎
  28. CMS, Frequently Asked Questions: Mental Health and Substance Use Disorder Parity Final Rule for Medicaid and CHIP, Q7 and Q8 (March 29, 2016). https://www.medicaid.gov/sites/default/files/2019-12/faq-cms-2333-f.pdf ↩︎
  29. A 2019 GAO report on parity enforcement for commercial plans concluded that the extent of parity compliance was unknown and that agency reliance on targeted reviews based on complaints and other information may not be adequate to assess compliance. GAO Enforcement Parity Report, (December 2019). ↩︎
  30. These three changes are also included in President Biden’s Fiscal Year 2023 Budget. https://www.whitehouse.gov/wp-content/uploads/2022/03/budget_fy2023.pdf. The House-passed Build Back Better legislation from 2021 also included these three changes. https://www.congress.gov/bill/117th-congress/house-bill/5376/text. ↩︎
  31. The House-passed Build Back Better legislation from 2021 would have given the DOL the authority to assess civil penalties for parity violations. This is similar to legislation introduced in 2021, H.R. 1364, the Parity Enforcement Act of 2021. https://www.congress.gov/bill/117th-congress/house-bill/1364/text. H.R. 7780 would make additional changes to ERISA’s enforcement scheme. https://www.congress.gov/bill/117th-congress/house-bill/7780/cosponsors?r=1&s=1   ↩︎

The Inflation Reduction Act (IRA) of 2022 includes a number of climate, tax, and health care provisions.  Major health care provisions include prescription drug reforms that primarily affect Medicare as well as a three year extension of enhanced subsidies to purchase Marketplace coverage that were included in the American Rescue Plan Act (ARPA).  However, a temporary federal policy to close the coverage gap – which was included in the House-passed reconciliation bill — was not included in the IRA and an amendment offered by Senator Warnock to add it back in did not garner enough votes.  The coverage gap includes roughly 2.2 million individuals living in 12 states that have not adopted the Affordable Care Act’s (ACA) Medicaid expansion. These individuals do not qualify for Medicaid and have incomes below poverty, making them ineligible for premium subsidies in the ACA Marketplace. Without a federal option to close the coverage gap, attention once again turns back to the states for now, where Medicaid expansion could be a key issue in some upcoming elections.  Public polling consistently shows that the majority of people living in non-expansion states want to see their state expand Medicaid. (more…)

Fate of Medicaid Expansion and Filling the Coverage Gap May Once Again Depend on the Outcome of State Elections

Authors: Meghana Ammula and Robin Rudowitz
Published: Aug 17, 2022

The Inflation Reduction Act (IRA) of 2022 includes a number of climate, tax, and health care provisions.  Major health care provisions include prescription drug reforms that primarily affect Medicare as well as a three year extension of enhanced subsidies to purchase Marketplace coverage that were included in the American Rescue Plan Act (ARPA).  However, a temporary federal policy to close the coverage gap – which was included in the House-passed reconciliation bill — was not included in the IRA and an amendment offered by Senator Warnock to add it back in did not garner enough votes.  The coverage gap includes roughly 2.2 million individuals living in 12 states that have not adopted the Affordable Care Act’s (ACA) Medicaid expansion. These individuals do not qualify for Medicaid and have incomes below poverty, making them ineligible for premium subsidies in the ACA Marketplace. Without a federal option to close the coverage gap, attention once again turns back to the states for now, where Medicaid expansion could be a key issue in some upcoming elections.  Public polling consistently shows that the majority of people living in non-expansion states want to see their state expand Medicaid. (more…)

The Inflation Reduction Act (IRA) of 2022 includes a number of climate, tax, and health care provisions.  Major health care provisions include prescription drug reforms that primarily affect Medicare as well as a three year extension of enhanced subsidies to purchase Marketplace coverage that were included in the American Rescue Plan Act (ARPA).  However, a temporary federal policy to close the coverage gap – which was included in the House-passed reconciliation bill — was not included in the IRA and an amendment offered by Senator Warnock to add it back in did not garner enough votes.  The coverage gap includes roughly 2.2 million individuals living in 12 states that have not adopted the Affordable Care Act’s (ACA) Medicaid expansion. These individuals do not qualify for Medicaid and have incomes below poverty, making them ineligible for premium subsidies in the ACA Marketplace. Without a federal option to close the coverage gap, attention once again turns back to the states for now, where Medicaid expansion could be a key issue in some upcoming elections.  Public polling consistently shows that the majority of people living in non-expansion states want to see their state expand Medicaid. (more…)

Key Facts About Medicare Part D Enrollment and Costs in 2022

Authors: Juliette Cubanski and Anthony Damico
Published: Aug 17, 2022

A version of this brief with information for 2023 is available: Key Facts About Medicare Part D Enrollment and Costs in 2023

The Medicare Part D program provides an outpatient prescription drug benefit to older adults and people with long-term disabilities in Medicare who enroll in private plans, including stand-alone prescription drug plans (PDPs) to supplement traditional Medicare and Medicare Advantage prescription drug plans (MA-PDs) that include drug coverage and other Medicare-covered benefits. This analysis provides the latest data about Medicare Part D enrollment and costs in 2022 and trends over time, based on data from the Centers for Medicare & Medicaid Services (CMS).

Enrollment in Medicare Advantage drug plans has surpassed stand-alone prescription drug plan enrollment

A total of 48.9 million people with Medicare are currently enrolled in plans that provide the Medicare Part D drug benefit, including plans open to everyone with Medicare (stand-alone PDPs and MA-PDs), and plans for specific populations (retirees of a former employer or union and Medicare Advantage Special Needs Plans, or SNPs).

In 2022, more than half of all Part D enrollees (53%, or 25.8 million) are enrolled in MA-PDs and 47% (23.1 million) are enrolled in stand-alone PDPs—the first year that MA-PD enrollment has surpassed PDP enrollment (Figure 1, Table 1). Over time, Part D enrollment in MA-PDs has increased, reflecting enrollment growth in Medicare Advantage plans overall, while enrollment in PDPs has decreased each year since 2019. Between 2019 and 2022, the number of MA-PD enrollees increased by 34%, while enrollment in PDPs decreased by 8%.

More Than Half of All Medicare Part D Enrollees Are Now in Medicare Advantage Drug Plans; Enrollment in Stand-alone Drug Plans Has Declined Since 2019

Part D enrollment is concentrated in 3 national firms – UnitedHealth, Humana, and CVS – which have a combined 57% of total enrollment

The top three firms – UnitedHealth, Humana, and CVS Health – cover close to 6 in 10 of all beneficiaries enrolled in Part D in 2022 (57%), while the top five firms – including Centene and Cigna – account for three-quarters of Part D enrollment (Figure 2). These shares are unchanged from 2021.

The Top 3 Firms - UnitedHealth, Humana, and CVS Health - Cover Close to 6 in 10 Medicare Part D Enrollees in 2022

Apart from Kaiser Permanente, which exclusively offers MA plans, the top Part D plan sponsors offer both stand-alone PDPs and MA-PDs. For most firms, Part D enrollment is more concentrated in one market than the other; for example, CVS Health, Centene, and Cigna have greater enrollment in PDPs than MA-PDs, while UnitedHealth and Humana have higher MA-PD enrollment than PDP enrollment (Figure 3).

Most of the Major Firms Sponsoring Medicare Part D Plans in 2022 Offer Both Stand-alone and Medicare Advantage Drug Plans

Nearly 6 in 10 beneficiaries receiving the Part D Low-Income Subsidy are enrolled in Medicare Advantage drug plans

In 2022, 12.8 million Part D enrollees (26% of all Part D enrollees) receive premium and cost-sharing assistance through the Part D Low-Income Subsidy (LIS) program. These additional financial subsidies, also called “Extra Help,” pay Part D premiums for eligible beneficiaries (as long as they enroll in stand-alone PDPs designated as premium-free “benchmark” plans) and reduce cost sharing. LIS enrollment in MA-PDs has been steadily increasing, and now totals 7.3 million, or 57% of all LIS enrollees (Figure 4, Table 1).

More Than Half of all Medicare Part D Enrollees Receiving the Low-Income Subsidy Are Now Enrolled in Medicare Advantage Drug Plans

As part of increased enrollment in MA-PDs, close to one-third of all LIS enrollees (32%) are now enrolled in Medicare Advantage Special Needs Plans (SNPs), up from only 4% in 2006. Overall, 4.6 million Medicare beneficiaries are enrolled in SNPs in 2022. SNPs limit enrollment to beneficiaries with certain characteristics, including those with certain chronic conditions (C-SNPs), those who require an institutional level of care (I-SNPs), and those who are dually enrolled in Medicare and Medicaid (D-SNPs), which account for the majority of SNP enrollees.

The average monthly premium for PDPs is substantially higher than the premium for drug coverage in MA-PDs

In 2022, the enrollment-weighted average monthly premium for PDPs is $40, a 4% increase from the weighted average PDP premium in 2021 ($38). The average monthly PDP premium is substantially higher than the enrollment-weighted average monthly portion of the premium for drug coverage in MA-PDs ($11 in 2022). Lower average premiums for Part D coverage in MA-PDs are mainly due to the ability of MA-PD sponsors to use rebate dollars from Medicare payments to lower or eliminate their Part D premiums. The total average premium for MA-PDs, including all Medicare-covered benefits, is $18 per month in 2022.

The calculation of the average monthly premium for Part D coverage in MA-PDs is relatively low because it includes the 69% of MA-PD enrollees in plans that do not charge a monthly premium in 2022. By comparison, all PDPs charge a monthly premium (although Part D enrollees who are receiving the full Part D Low-Income Subsidy and are enrolled in a benchmark PDP pay no monthly premium). The average enrollment-weighted premium for Part D coverage across the 31% of MA-PD enrollees in plans that charge a monthly premium is $35, which is within a few dollars of the average monthly premium for PDPs (Figure 5).

The Average Monthly Premium for Stand-alone Drug Plans in 2022 Is Substantially Higher than the Average Premium for Drug Coverage in Medicare Advantage Drug Plans

The average Part D deductible has increased for PDPs, while decreasing for MA-PDs; the 2022 weighted average annual drug deductible is 4.5 times larger in PDPs than in MA-PDs

In 2022, a large majority of PDP enrollees (86%) are in plans that charge a deductible, with nearly 8 in 10 (79%) in PDPs that charge the standard amount of $480 in 2022. Conversely, 1 in 5 MA-PD enrollees are in plans that charge the standard Part D deductible, and just over half (51%) are in plans that charge no drug deductible. These enrollment patterns explain the wide divergence between PDPs and MA-PDs in the enrollment-weighted average Part D deductible amount. For PDPs, the average Part D deductible in 2022 is $398, nearly 4.5 times larger than the average drug deductible in MA-PDs ($90) (Figure 6).

The Average Annual Deductible for Medicare Part D Coverage in 2022 Is Nearly 4.5 Times Higher in Stand-alone Drug Plans Than in Medicare Advantage Drug Plans

Most Part D enrollees face low copays for generic drugs, higher copays for preferred brands, cost sharing up to $100 or 50% for non-preferred drugs, and 25-33% coinsurance for specialty drugs

Most Part D enrollees pay low or no copayments for preferred generic drugs, but higher copays for generics not on the preferred tier. Over half of Part D enrollees (both PDP and MA-PD enrollees) pay $0 for preferred generics in 2022, but many pay $10 or more per prescription for generics that are not on the preferred tier (Figure 7). For these generic drugs, most PDP enrollees (76%) pay less than $10, while just over half of MA-PD enrollees (54%) pay between $10 and $20.

For preferred brands, virtually all MA-PD enrollees and two-thirds of PDP enrollees are in plans that charge copayments instead of coinsurance. Roughly two-thirds of MA-PD enrollees (68%) and 28% of PDP enrollees face copayments for preferred brands of $45 to $47, which is the maximum copayment amount permitted by CMS for preferred brand drugs. Another one third of PDP enrollees (34%) are in plans that charge a coinsurance for preferred brands of less than 25%, while only 1% of MA-PD enrollees pay a coinsurance for preferred brands.

For non-preferred drugs, most MA-PD enrollees (89%) are in plans that charge copayments while virtually all PDP enrollees are in plans that charge a coinsurance. For drugs on the non-preferred tier, which can be all brands or a mix of brands and generics, nearly half (49%) of all MA-PD enrollees are in plans that charge $100, the maximum copayment amount allowed by CMS for non-preferred drugs. More than half (57%) of PDP enrollees are in plans than charge 40% to less than 50% for non-preferred drugs and 11% of PDP enrollees are in plans that charge a 50% coinsurance rate, the maximum coinsurance allowed.

A larger share of MA-PD enrollees than PDP enrollees are in plans that charge the maximum 33% coinsurance rate for specialty tier drugs. For specialty tier drugs, defined by CMS as those that cost at least $830 per month in 2022, a much larger share of enrollees in MA-PDs (63%) than PDPs (14%)  are in a plan that charges the maximum 33% coinsurance, while a much larger share of enrollees in PDPs (79%) than MA-PDs (4%) are in a plan that charges the minimum 25% coinsurance. Plans that waive some or all of the standard deductible – which is the case for most MA-PDs – are permitted to set the specialty tier coinsurance rate above 25%.

It is important to note that for this analysis, we did not compare which specific drugs are covered on each tier in PDPs and MA-PDs, which, in addition to the cost-sharing amounts that plans charge, would also influence enrollees’ out-of-pocket costs. Given the different cost-sharing structures adopted by MA-PDs and PDPs, particularly for non-preferred drugs, it can be difficult to compare the actual out-of-pocket costs that beneficiaries would face for different types of drugs across plan types. It can also be difficult to know how a coinsurance rate will translate into actual out-of-pocket costs without knowing the underlying list price of a drug.

Most Part D Enrollees Face Low Copays for Generic Drugs but Higher Copays for Preferred Brands, Copays or Coinsurance up to $100 or 50% for Nonpreferred Drugs, and 25-33% Coinsurance for Specialty Drugs

Juliette Cubanski is with KFF. Anthony Damico is an independent consultant.

Medicare Part D and Part D Low-Income Subsidy Program Enrollment, by Plan Type, 2006-2022

The Biggest Health Care Reform in a Decade Could Lower Your Costs

Author: Larry Levitt
Published: Aug 13, 2022

Larry Levitt writes about the political and practical impact of the health care provisions in the Inflation Reduction Act  in The New York Times guest essay, “The Biggest Health Care Reform in a Decade Could Lower Your Costs.”