Transition Coverage and Clarity in Self-Insured Corporate Health Insurance Benefit Plans

Introduction

Transgender people choosing between employer health benefit plans or looking for a job with gender-affirming coverage have many more inclusive options now compared to 20 years ago when total exclusions for gender-affirming coverage were the norm.¹ While this shift is important, it does not mean that transgender people's health care and insurance needs are fully met; indeed, considerable evidence shows that rates of exclusion and discrimination remain high.^2–15 Difficulties using and gaining insurance coverage remain top concerns for transgender people.^10,11,16,17

Nonetheless, 97% of plans in a sample of 1057 Affordable Care Act (ACA) silver marketplace options from 161 insurers in 38 states had removed transgender exclusions in 2020, compelled by the ACA's Section 1557 nondiscrimination requirements, which ban categorical exclusions.¹⁸ Section 1557 covers any health care entity that receives funding from the Department of Health and Human Services (HHS), every health program or activity administered by HHS, and the health insurance marketplaces and all plans by issuers who participate in those marketplaces (subject to changes discussed below).¹⁹

While many transgender people in the United States rely on programs such as Medicaid for their health care benefits because they tend to hold jobs without employer coverage and/or be lower income,⁷ transgender individuals also hold or seek jobs that offer employer-based health insurance. Eighty-six percent of respondents to the 2015 Transgender Health Survey had health insurance, and 53% of respondents reported coverage under an employer-sponsored health plan.^{17(p. 94)}

Yet, most employer-based health insurance plans' details of coverage are obscure to many consumers, and the landscape of legal protections against discrimination based on gender identity has been highly volatile in the insurance context. Scholars know relatively little about gender-affirming health insurance coverage in the private market because corporations are not required to file their plan documents at the Department of Labor and do not otherwise share them publicly.

Much research in this area has focused on insurance market segments where publication is required such as those on the ACA marketplaces, Medicare, and Medicaid, and surveys of other insurance offerings do not analyze the actual insurance contract (the consumer-facing document),^† and rely on phone calls to insurers and web searches to gather insurance company information.^7,20–22 Advocacy organizations such as the Human Rights Campaign (HRC) have been very important in filling the gap of knowledge for LGBT people with their Healthcare Equality and Workplace Equality ratings,²³ but these are self-reported surveys without scrutiny of documents.

This study examines the following: What transgender health care coverage did some of the biggest corporate employers in the United States offer in 2019, and how does coverage measure up against the international standard for transgender insurance coverage published by the World Professional Association for Transgender Health (WPATH)?²⁴ We analyzed the language of 435 insurance contracts offered by 40 major American corporations operationalized into 52 discrete elements of the WPATH standard. This study is unique in its detailed focus on the language of the insurance contract itself from a sample of self-insured firms, the least regulated by antidiscrimination laws in the United States.

Methods

We accessed AXIACI from Leverage Global Consulting, a proprietary database that contains insurance plan offerings and coverage from private and public insurance market segments. Our use of the proprietary database for the purposes of public policy research and analysis in health insurance is governed by an agreement with Leverage.

We extracted, from their database, 435 health insurance contracts for 40 corporations from the year 2019 (the total number available as of July 2019). These are self-insured plans using third party administrators (TPAs). There are no human subjects in this research and no personally identifying information involved because these are only the contract documents, not records of anyone's health insurance claims or medical information. We have chosen open access publication and data sharing to accord with research ethics in topics important to transgender people beyond the academy.^25,26 A dataset of all 435 contracts used in our analysis and the spreadsheet with our full results are publicly available at the University of Michigan's open social science data depository, OpenICPSR (Deposit number 120901).

Procedure for Coding the Contracts

We operationalized details of the current WPATH-recommended transgender health benefits into 52 discrete elements in a spreadsheet. Following a detailed written search procedure, the authors and a team of research assistants analyzed each contract for these elements. Each research assistant received training on searching the contracts, and the research team met weekly to discuss questions and resolve ambiguities. The first author checked the spreadsheet responses for accuracy by searching a sample within each corporation. Samples to check were drawn from different carriers under the same corporation, because carrier documents tend to be similar. If any error was found, the entire section of all contracts for that corporation would be redone and checked again by two different coders.

We coded both for the overall type of exclusion language linked to gender-affirming care as well as whether particular WPATH-recommended care was mentioned as excluded or not. At the first level of coding, we noted whether there was any exclusion language relevant to gender-affirming care at all (and of course, there would not be if there was no gender-affirming care mentioned), and then whether the exclusion was generic language about cosmetic, experimental, or investigational (CEI) exclusions (but referring to trans care specifically) or whether there were additional trans-specific exclusions listed.

Categorizing the Plans

We grouped contracts into categories based on ease of accessing and understanding the coverage: Clear, Silent, Ambiguous, and Excluded. Criteria for placing a contract in a category are listed below.

Contracts rated Clear

Contracts rated Silent

Contracts rated Ambiguous

Contracts rated Excluded have a total exclusion on all gender-affirming care.

Discussion

See Table 1 for a list of sponsoring employer companies, industry descriptions, and our ratings of their coverage for gender-affirming care clarity alongside their 2019 HRC corporate equality index rating.

Exclusions

Some of these corporate contracts continue to exclude health benefits for gender-affirming care entirely, often in outdated language that appears to have been preserved in the contracts for years (perhaps to meet “grandfathered” status under the ACA). See Table 2 for a list of corporations with at least one health contract that excludes gender-affirming care completely, the number of total contracts in our analysis compared to how many had the exclusion, the TPA for the exclusions, and the language of their exclusion.

Ten companies out of this group of 40 (25%) offered at least one health care contract with a total exclusion on coverage for gender-affirming care. These contracts represent 38 contracts out of 435 total in our analysis, or just under 9%. The companies vary based on their seeming commitment to their exclusion, with some showing the exclusion in 100% of the contracts we found, while others seem to exclude strategically, with only certain contracts—perhaps for some groups of employees or some geographic areas—including an exclusion and others lacking it. It is also important to note that the TPAs, major United States health insurance carriers, sell contracts in many different market segments with and without exclusions for gender-affirming care (adapting to company requests, but responding to other legal obligations for other market segments).

World Professional Association for Transgender Health

Few companies offer contracts that mentioned WPATH criteria (10 companies out of 40, or 25%). See Table 3 for a list of corporations with health contracts that invoke WPATH explicitly. Contracts that invoke WPATH criteria gave much more detail about coverage, but did not necessarily replicate the recommended WPATH coverage elements entirely. For example, the Rockwell Automation contract administered by United Health care lists dozens of covered procedures and exclusions, even though some of the exclusions (listed below) are WPATH recommended for coverage. None of the contracts that affirmed WPATH did so by simply stating that any WPATH-recommended procedure would be covered.

Types of exclusions other than total exclusions

Table 4 depicts the corporate contracts grouped by exclusion type. Contracts used generic language (CEI) or gave specific gender-affirming exclusions (silent contracts are not included). Insurance companies deny coverage for “cosmetic, experimental, or investigational” procedures for cisgender and transgender people, arguing that they are not medically necessary or fall too far outside accepted medical practice to cover. If a contract lists only CEI exclusions (and every contract included at least that language), we could not tell what gender-affirming procedures might be denied under it.

Table 5 shows the types of procedures singled out for exclusions beyond CEI even in contracts that otherwise covered treatments for gender dysphoria. Most exclusions are related to facial gender confirmation surgeries and hair removal, which are critically important to gender-affirming care and highly problematic to exclude.²⁷ Contracts with little detail and a general cosmetic exclusion are more difficult for employees to decipher and allow employers and TPAs to be less transparent about their decisions to reject coverage for certain procedures as medically unnecessary. Silent contracts are arguably the most confusing. Although we did not code it as an exclusion, 90 of the 435 plans specifically mentioned an age restriction of 18 for gender-affirming care (and the rest mentioned nothing about age). A diagnosis of gender dysphoria was uniformly an implicit or explicit requirement for coverage in all the contracts, and precertification or preauthorization was common for many points of access to care.

Comparison to HRC rankings

We compared our category rankings to the HRC's rankings of the same companies on the 2019 Workplace Equality Index (Table 1). Only half of the companies (11 of 21) that scored a 100 on HRC's rankings had clear coverage for all the contracts in our analysis. Three companies scoring 100 offered 88 contracts with no clear coverage (First Data, McDonalds, and Stanley Black and Decker). Seven companies scoring 100 offered 114 contracts that were silent regarding coverage for gender-affirming care. Two of these companies only offered contracts that were silent regarding coverage for gender-affirming care (First Data and McDonalds). Six companies scoring 100 offered 16 contracts that were ambiguous regarding coverage for gender-affirming care. Despite earning an HRC score of 100, five companies still offered 15 contracts with blanket exclusions for coverage for gender-affirming care (Macy's, Marsh and McLennan, McKesson Corporation, Stanley Black and Decker, and Symantec), and one company (Stanley Black and Decker) only offered contracts in our analysis with exclusions. Three companies that scored a 100 offered inconsistent coverage for employees that comprised all category rankings (clear, silent, ambiguous, and excluded): Macy's, Marsh and McLennan, and McKesson Corporation.

Several transportation companies (BNSF Railway, Union Pacific Railroad, and Hunt Transportation) only offered contracts with exclusions. Finally, three companies with HRC rankings 65 or lower offered at least some contracts with clear coverage (Avnet, Costco, and Windstream). While the HRC Equality Index provides important information about LGBT inclusivity overall, this research underscores that a perfect score does not always equate to coverage for gender-affirming care. Significant variability exists in health insurance coverage for transgender employees within the same company and among employees working at different companies that all earned a perfect score from HRC.

Legal and policy analysis

We focus here on corporate health care benefit plans and their coverage provisions for gender-affirming care because they have been previously unavailable in a centralized database for academic research. They sit in a complex web of regulation between the states and the federal government, and both Title VII of the 1964 Civil Rights Act (prohibiting sex discrimination in employee benefits) and the ACA's nondiscrimination provisions under Section 1557 apply to insurance plans, although differently, incompletely, and in shifting ways that have just been realigned yet again by the June 2020 Supreme Court ruling in Bostock v. Clayton County and the Trump administration's new Section 1557 regulation issued just days before Bostock.^19,28–31

Insurance plans can be an employee benefit or not depending on how the person obtains it, and whether the plans offered by these major corporate employers were legally permitted to exclude health care coverage for gender-affirming care in 2019 depends most importantly on how the company set up its insurance benefits (fully insured, meaning the company pays premiums to an insurer, who bears the risk of paying claims, vs. self-insured, meaning the company itself bears the risk of paying claims, but typically uses a TPA to manage its plans). The company contracts we evaluate here are all self-insured.

Insurance is normally regulated at the state level under the McCarran–Ferguson Act,³² but the Employee Retirement Income Security Act (ERISA) contains a preemption clause that allows ERISA to supersede state insurance laws, including any state insurance mandate for gender-affirming health coverage.^33,34 Specifically, ERISA preempts enforcement of state insurance mandates on self-insured health insurance plans that are part of employee benefits (but not fully insured corporate health benefit plans).^34–36 Twenty-four states plus the District of Columbia have banned gender identity discrimination in insurance,³⁷ but because these corporations' plans are self-insured, these laws do not apply.

Critically, however, ERISA does not preempt federal nondiscrimination requirements under Title VII, and employee benefits clearly fall under Title VII coverage.²⁹ The Supreme Court recently held that discriminating against a transgender employee is discrimination “on the basis of sex” and therefore covered under Title VII.²⁸ Previous case law in many jurisdictions had affirmed transgender inclusion at lower court levels, and so some of the firms analyzed in this study already had this legal obligation. However, now Title VII transgender inclusion applies to all of the plans analyzed in this study and the exclusions, if they remain, are flatly illegal. It is not clear what courts will require in detail for nondiscriminatory benefit coverage, however. The plaintiffs in the June 2020 cases were all employees who had been fired and did not address the content of employee health benefits.

Regulations issued in 2016 implementing the ACA's nondiscrimination clause, Section 1557, had explicitly included gender identity as part of sex discrimination and prohibited covered entities from having or implementing a categorical exclusion for gender-affirming care.³⁸ Section 1557 applies to health care entities that accept federal funds, but self-insured private employers were not covered because they themselves are not health care providers. Their TPA was covered under Section 1557, however, and the Office of Civil Rights policy was to determine who was responsible for discrimination (the self-insured firm or the TPA), and refer any noncovered complaints to the Equal Employment Opportunity Commission for possible Title VII liability for the firm.^{38(pp. 31431–31432)}

Just 3 days before the ruling in Bostock, however, the Trump administration issued a final rule repealing most of the Obama administration's 2016 Section 1557 rules. The new rules removed all explicit protections based on gender identity, transgender status, or gender transition.³⁰ LGBT health and advocacy groups (e.g., Whitman–Walker Health³⁹ and the HRC⁴⁰) immediately filed lawsuits challenging the regulations as arbitrary and capricious (given well-documented problems of transgender health discrimination, especially under pandemic conditions), and untenable after the Bostock ruling that transgender discrimination is sex discrimination.^41,42 Bostock is a Title VII case and Section 1557's protections are rooted in Title IX, meaning there is no direct application, although courts typically interpret definitions from Title VII as applicable to Title IX.⁴³

The opposing language leaves gender-affirming health coverage in a confusing place for now, although the trajectory is clearly moving toward affirming transgender nondiscrimination obligations across settings. For our purposes here with 2019 plans, the trans-inclusive Obama-era Section 1557 rules applied (although without application to self-insured corporate plans) and Bostock had not yet been decided. Going forward, Bostock governs employee health plans under Title VII, but the Section 1557 rules for covered health care entities are either in litigation or, if implemented, lack explicit transgender protections.

There are thus several interlocking explanations for why we have found complete exclusions for gender-affirming care in this study of corporate health contracts: (1) the plan is not covered under Section 1557's nondiscrimination clause because the company is not a health care entity and has not accepted federal funds; (2) there is no case law in that company's legal jurisdiction finding that the exclusion is a violation of employment nondiscrimination laws, either because no one has brought a case or because someone did and lost on that point or settled with a confidentiality agreement; (3) it is a discriminatory vestige that could successfully be challenged under Title VII, but remains anyway because no plaintiffs have sued; and (4) no labor union or employee group has mobilized for their removal.

The other side of the question is why extend benefits if one has not been legally required to do so? Civil rights protections have transformed the industry toward broad inclusion,⁴⁴ and there is a strong consensus in the medical community that gender-affirming services are medically necessary.⁴⁵ Companies may have been responding to extralegal pressures such as the desire to recruit LGBT employees and earn high ratings from HRC. They may calculate that gender-affirming services will cost relatively little, but be worth the progressive image benefit and offer them even if not legally compelled to do so. Indeed, the long history of success in expanding sexual orientation nondiscrimination in private company policies without a federal law shows that internal employee organizing for benefits can be successful.^46–48 Transgender advocacy organizations such as the National Center for Transgender Equality and the Trans Health Project at the Transgender Legal Defense and Education Fund counsel clients to approach their employers directly to extend coverage since this strategy may be much easier than litigation or fighting the insurance company. After Bostock, this push will be much easier.

Conclusion

The contract language we found and the legal and policy context of a fragmented and insufficiently regulated United States health care insurance system help explain why transgender employees continue to have problems finding out about and using the coverage that their employers promise. This study shows how even under conditions of coverage in corporate America, there are still ambiguities, a lack of transparency in coverage, and a host of important exclusions for recommended procedures. Contracts vary widely in how much they discuss gender-affirming coverage when it is explicitly mentioned as covered, and many contracts are still confusing to decipher or completely silent on the question of whether and what types of gender-affirming care will be covered.

Even companies rated highly under HRC criteria for LBGT-friendly workplaces can fail to provide adequate coverage for transgender employees under closer scrutiny. Corporate health plans are more likely to completely exclude gender-affirming coverage for transgender employees (9% with total exclusions) than plans that have been more clearly governed by Section 1557 nondiscrimination provisions (only 3% with exclusions). Any insurance contract offered in the United States should provide full, nonexclusionary gender-affirming care and be easily available for public scrutiny.

Our analysis is confined to the 40 corporations made available to us in the AXIACI database as of July 2019. While we believe these represent a reasonable cross-section of industries, we make no claims of their representativeness of U.S. corporate gender-affirming coverage overall. In addition, Leverage cannot guarantee that all of a company's health plan offerings for 2019 are contained in the database. We adopted the perspective of an employee and so limit our discussion here to what is on the document of the contract itself, without extension to medical policies or provider manuals that give more of the internal perspective of the insurer and provide additional direction to health care providers. We measured coverage against the WPATH health insurance guidelines, but scholars have pointed out that the gatekeeping model of WPATH does not fit all transgender and nonbinary people's needs or self-understandings.⁴⁹ Nonbinary people's health care needs are not addressed in our analysis, and may not be met even under a transgender-inclusive policy.

Author Disclosure Statement

All the authors have no disclosures. None of the authors has any commercial relationship to or investment in Leverage Global Consulting.

Funding Information

Funding for this study was provided by the National Science Foundation (Award # 1654645), the Institute for Research on Women and Gender at the University of Michigan, and the University of California, Irvine School of Law.