The Relationship Between a Hospital's Magnet Status and LGBTQ+ Inclusivity in Policies and Practices in US Hospitals

ABSTRACT

LGBTQ+ individuals face discrimination in healthcare settings. Magnet hospitals have been associated with positive patient outcomes, yet it remains uncertain whether Magnet designation is associated with hospitals' LGBTQ+ inclusivity in policies and practices. This study examined 801 American hospitals across 47 states that participated in the Healthcare Equality Index (HEI) in 2021. Multilevel modeling was utilized to investigate the association between Magnet status and HEI scores, adjusting for hospital type and state‐level covariates, including LGBTQ+ inclusiveness in laws, political climate, racial/ethnic minority population, and Medicaid expansion status. Among the 801 hospitals, 32.1% (257 hospitals) held Magnet status. Magnet hospitals demonstrated higher HEI scores compared to non‐Magnet hospitals (γ = 2.13, p = 0.022), despite significant variations across states (intraclass correlation = 0.22). No significant cross‐level interactions were found. Overall, Magnet designation is independently associated with improved LGBTQ+ inclusivity in hospitals regardless of the state in which the hospital is located. Policymakers and healthcare leaders should consider leveraging the Magnet Recognition Program as a benchmark for promoting LGBTQ+ inclusivity within hospitals. Additionally, all healthcare institutions should prioritize comprehensive evaluations and improvements to their policies and practices to ensure inclusivity for LGBTQ+ patients.

1. Introduction

Lesbian, gay, bisexual, transgender, queer or questioning, and other sexual and gender diverse (LGBTQ+) individuals constitute medically underserved populations facing persistent health disparities, including discrimination in healthcare settings (National Academies of Sciences Engineering and Medicine ²⁰²⁰; National Institute of Minority Health and Health Disparities ²⁰¹⁶). A recent US‐based national survey revealed that 24% of LGBTQ+ adults reported discrimination during healthcare encounters (Kaiser Family Foundation ²⁰²⁴). This discrimination in healthcare against LGBTQ+ individuals stems from social and structural inequities, leading to healthcare avoidance and poorer health outcomes compared to non‐LGBTQ+ peers (Casey et al. 2019; James et al. ²⁰²⁴; Kaiser Family Foundation ²⁰²⁴).

Since 2007, the Healthcare Equality Index (HEI), administered by the Human Rights Campaign Foundation (HRCF), has assessed multiple facets of healthcare policies, practices, and services within healthcare organizations, providing insights into the level of LGBTQ+ inclusivity (Human Rights Campaign Foundation ²⁰²²). LGBTQ+ inclusive healthcare is characterized by its support for the unique needs of LGBTQ+ individuals (Roman Laporte and De Santis ²⁰²³). In such an environment, healthcare providers are educated on the specific health concerns of this population, show sensitivity to preferred pronouns, partner(s), and nontraditional family structures, and prioritize open communication while delivering individualized care plans. This type of care ensures that the healthcare setting reflects and acknowledges the sexual orientation and gender identity of each individual, fostering an inclusive and representative atmosphere (Roman Laporte and De Santis ²⁰²³). The HRCF assigns scores to voluntarily participating US institutions based on nondiscrimination policies, LGBTQ+ patient services, employee benefits, and community engagement. Healthcare facilities that achieve a top score of 100 points are designated as LGBTQ+ Healthcare Equality Leaders (Human Rights Campaign Foundation ²⁰²²).

Magnet Recognition status signifies excellence in nursing practices and patient care (American Nurses Credentialing Center ^2023a). Magnet hospitals have demonstrated various positive outcomes, including lower rates of pressure ulcers and hospital‐acquired infections, fewer patient falls, and lower 30‐day inpatient mortality when compared to non‐Magnet hospitals (Connor et al. 2023; Dit Dariel and Regnaux ²⁰¹⁵; McCaughey et al. ²⁰²⁰; Rodríguez‐García et al. ²⁰²⁰). Hospitals with Magnet status receive higher patient ratings for their willingness to recommend the facility, show improved patient satisfaction—including better patient–nurse communication—and deliver more cost‐effective care compared to non‐Magnet hospitals (Connor et al. 2023; Dit Dariel and Regnaux ²⁰¹⁵; McCaughey et al. ²⁰²⁰; Rodríguez‐García et al. ²⁰²⁰). However, the relationship between Magnet status and inclusive policies, practices, and services to prevent discrimination against LGBTQ+ populations remains underexplored. In three previous studies, researchers found a positive association between Magnet status and HEI scores using bivariate analyses only (Blackwell ^2020a; Blackwell, Diaz‐Cruz, and Yan ^2020b; Blackwell et al. ²⁰²⁴). However, they did not consider hospital or state‐level factors that may have influenced the results.

Spatial and sociopolitical contexts significantly impact resource distribution and health outcomes for residents (Bauermeister et al. 2017; Diez Roux and Mair ²⁰¹⁰; Schwartz, Susser, and Susser ¹⁹⁹⁹; Susser ¹⁹⁹⁴). Understanding population health requires examining the characteristics of the groups or contexts to which individuals belong, including state‐level sociodemographic factors, policies, and political climates (Diez Roux and Mair ²⁰¹⁰; Schwartz, Susser, and Susser ¹⁹⁹⁹; Susser ¹⁹⁹⁴). Therefore, when exploring the association between Magnet status and HEI scores, it is crucial to consider key hospital and state‐level factors.

Hospital type is critical because variations in resources, patient demographics, and organizational structures significantly influence Magnet status attainment and the implementation of LGBTQ+ inclusive policies, impacting HEI scores. For example, academic medical centers often have more resources and dedicated diversity, equity, and inclusion (DEI) departments, facilitating the implementation of inclusive policies compared to nonacademic medical centers (Shahian et al. 2012). Additionally, urban hospitals may serve more diverse populations, driving them to prioritize LGBTQ+ inclusivity compared to rural hospitals (Syrett and Sepulveda 2011). Understanding these variations is essential for developing targeted strategies to promote inclusivity and excellence in healthcare across different hospital types.

Inclusive state‐level policies provide legal protections, promote equal rights, and enhance healthcare access (Movement Advancement Project ^2020a), whereas anti‐LGBTQ+ laws increase disparities and discrimination (Goldberg 2023; Hatzenbuehler ²⁰¹⁴; Hatzenbuehler, Keyes, and Hasin ²⁰⁰⁹; Kline et al. ²⁰²²), hindering healthcare access (Human Rights Campaign ²⁰²³; Jackson, Stewart, and Fleegler ²⁰²³; Watson et al. ²⁰²¹). Additionally, state‐level laws can influence hospitals' policy and practice decisions (Adler‐Milstein, Kvedar, and Bates ²⁰¹⁴; Davidoff et al. ²⁰⁰⁰). Political dynamics, including party control, can shape policy‐making processes, affecting the adoption of LGBTQ+ inclusive policies within hospitals (Michelson and Schmitt 2020; Pomeranz et al. ²⁰¹⁷). Additionally, the demographic composition of racially and ethnically minoritized populations in regions can impact both Magnet status and HEI scores. Hospitals located in areas with higher percentages of minoritized populations may encounter unique socioeconomic challenges (Tsui et al. 2020). Medicaid expansion under the Affordable Care Act extends eligibility to all adults with incomes up to 138% of the federal poverty level, broadening coverage beyond the initial focus on low‐income individuals, families, the elderly, and the disabled (Kaiser Family Foundation ²⁰²³). States that adopted this expansion may see changes in healthcare infrastructure compared to non‐expansion states (Miller and Wherry 2017; Wherry and Miller ²⁰¹⁶). This expansion could increase insurance coverage among low‐income and uninsured individuals, potentially altering hospitals' patient populations. Additionally, it could influence hospitals' financial stability, reimbursement rates, and ability to invest in resources for Magnet designation or improving HEI scores.

Given that limited insight exists regarding the relationship between state‐level policy variations and hospital‐level LGBTQ+ inclusion efforts, understanding multilevel factors is crucial, as they may affect HEI scores and subsequently influence healthcare experiences for LGBTQ+ populations. The socioecological model of sexual and gender minority health disparities research framework (National Institutes of Health Sexual and Gender Minority Research Office ²⁰²¹) proposes that various unique influences, behaviors, and issues across multiple levels impact the health and well‐being of LGBTQ+ populations throughout their lifespan. This framework emphasizes the importance of examining how individual, community, institutional, and policy‐level factors interact to shape health inequities experienced by LGBTQ+ populations. Informed by this framework, this study explores the relationship between Magnet designation and HEI scores among hospitals across the United States, while considering hospital and state‐level factors.

2. Methods

The study design was a cross‐sectional, observational study.

2.1. Study Sample

Sample inclusion criteria were having HEI data, being a hospital, and being in a US state. Nonhospital entities were excluded due to the Magnet Recognition Program's focus on hospitals. Hospitals in US territories were excluded due to significant differences in governance, legislation, and healthcare policies compared to mainland US states (Malavet 2004; Morris ¹⁹⁹⁵). This study received an exemption from the Institutional Review Board.

2.2. Study Variables and Data Sources

2.2.1. Dependent Variable

The dependent variable was the level of LGBTQ+ inclusivity in hospitals' policies and practices, assessed using the HEI, which employs a scoring system ranging from 0 to 100. The score comprises four subcategories: Nondiscrimination and Staff Training, which assesses patient and employment nondiscrimination policies and staff training in LGBTQ+ patient‐centered care (0–40); Patient Services and Support, which evaluates the specific services and support available to meet the needs of LGBTQ+ individuals, including transgender‐specific clinical services (0–30); Employee Benefits and Policies, which focuses on the institution's policies and benefits for LGBTQ+ employees such as health insurance coverage for LGBTQ+ specific care (0–20); and Patient and Community Engagement, which measures the institution's efforts to engage with the LGBTQ+ community through outreach programs and LGBTQ+ related events (0–10). Points are deducted for discriminatory policies or recent anti‐LGBTQ+ incidents. Further details on the HRCF scoring system and methodology are available elsewhere (Human Rights Campaign Foundation ²⁰²⁰). The HRCF biennially evaluates and assigns scores to healthcare institutions that voluntarily submit the required HEI information (Human Rights Campaign Foundation ²⁰²²). We used the 2022 HEI data, which were collected from healthcare facilities in 2021.

2.2.2. Independent Variable

The presence of Magnet status in hospitals was sourced from the American Nurses Credentialing Center (American Nurses Credentialing Center ^2023b). In line with our analysis of HEI scores in 2021, this variable was a binary variable, with a value of 1 if they held Magnet status in 2021 and 0 if they did not.

2.3. Level‐1 Covariates (Hospital‐Level)

2.3.1. Hospital Type

Hospital types were classified according to Definitive Healthcare's mutually exclusive categorization (Tieche 2023). This variable was operationalized as a categorical variable, with short‐term acute care hospitals, which had the largest sample size, as the reference category. The other categories include children's hospitals, critical access hospitals, long‐term acute care hospitals, psychiatric hospitals, rehabilitation hospitals, and Veterans Affairs (VA) hospitals. Children's hospitals focus on patients under 18 years old. Critical access hospitals, designated by the Centers for Medicare and Medicaid, aim to improve healthcare access and financial stability in rural areas, with 25 or fewer beds and over 35 miles from another hospital. Long‐term acute care hospitals, providing stays of 25 days or more, offer services such as respiratory therapy and pain management. Psychiatric hospitals offer round‐the‐clock inpatient care for mental illness. Rehabilitation hospitals specialize in restoring functional abilities. Short‐term acute care hospitals provide conventional care for short‐term medical needs. VA hospitals, federally funded and administered by the Veteran's Administration, serve war veterans and retired military personnel (Tieche 2023).

2.4. Level‐2 Covariates (State‐Level)

2.4.1. LGBTQ+ Inclusiveness in Laws and Policies

The overall LGBTQ+ policy tally for each state in 2020, reported by the Movement Advancement Project (MAP) (Movement Advancement Project ^2020b), served as a level‐2 covariate. MAP assesses state‐level LGBTQ+ policies, assigning a tally based on nearly 40 laws and policies covering nondiscrimination, healthcare access, criminal justice, and other LGBTQ+ relevant areas. The possible point range for these scores is from −18.5 to 38.5 points (Movement Advancement Project ^2020a). We use the tally in 2020 to account for potential lag effects in policy implementation leading up to the HEI survey collected in 2021, recognizing that legal changes may take time to impact healthcare practices (McCourt et al. ²⁰²³; Stone et al. ²⁰²²; Tormohlen et al. ²⁰²²). The variable was standardized as a z‐score, with positive scores indicating a better‐than‐average policy environment for LGBTQ+ communities and negative scores indicating the opposite.

2.4.2. Political Party in Control of Legislative Chambers and Governorship

A state control variable, obtained from the National Conference of State Legislatures for 2021 (National Conference of State Legislatures ²⁰²³), reflects the political landscape within each state. State control is determined by the party holding power in legislative chambers and the governorship. If one party holds all positions, they have state control; otherwise, it is considered divided (National Conference of State Legislatures ²⁰²³). Nebraska has a unique nonpartisan unicameral legislature, meaning its lawmakers are elected without official party affiliations, unlike in other states with clearly partisan legislatures. Therefore, we classified Nebraska's state control solely based on the party of the governorship. In 2021, Nebraska's governor was Republican, so the state was classified as having Republican control. Washington, D.C. was categorized as having Democratic control due to Democratic Party dominance in both the city council and mayorship throughout 2021.

2.4.3. Percentage of the Racial and Ethnic Minority Population

The percentage of racial and ethnic minority populations at the state level, obtained from the 2021 American Community Survey 1‐year estimate (United States Census Bureau ²⁰²¹), encompasses individuals historically underrepresented or marginalized within a society or region in terms of racial or ethnic backgrounds (National Institute of Minority Health and Health Disparities ²⁰²⁴). This variable was calculated by subtracting the percentage of the non‐Hispanic White population in each state from 100.

2.4.4. Medicaid Expansion Status

Medicaid expansion status for each state from Kaiser Family Foundation (Kaiser Family Foundation ²⁰²³) was included as a level‐2 covariate. This variable was operationalized as binary, with 0 indicating no Medicaid expansion as of January 1, 2021, and 1 indicating Medicaid expansion as of the same date.

2.5. Statistical Analysis

We conducted descriptive statistics for both hospital and state characteristics. To examine the association between a hospital's Magnet status and HEI scores, we utilized multilevel modeling (MLM). Grounded in the socioecological model (National Institutes of Health Sexual and Gender Minority Research Office ²⁰²¹), MLM is suitable for analyzing hierarchical data structures (Kreft, Kreft, and de Leeuw ¹⁹⁹⁸; Subramanian, Jones, and Duncan ²⁰⁰³). Since our data include two levels (level‐1 = hospital level, which encompasses Magnet status and hospital type; level‐2 = state level, which includes LGBTQ+ inclusiveness in laws, political climate, racial/ethnic minority population, and Medicaid expansion status), we employed a two‐level random intercept model to investigate the association of hospital and state‐level factors on LGBTQ+ inclusivity in hospitals. Random effects in MLM allow for varying intercepts for each state, acknowledging heterogeneity across regions (Subramanian, Jones, and Duncan ²⁰⁰³).

MLM analyses were conducted using Stata BE 17.0 following established guidelines (Khine 2022; Peugh ²⁰¹⁰). The intraclass correlation coefficient (ICC), indicating the proportion of total variability in outcomes attributable to state‐level groupings, was utilized to justify the use of MLM. After justifying the adoption of a multilevel approach, we employed a multilevel linear model with the identity link function, using full information maximum likelihood estimation across two levels (hospitals nested within states).

The hypothesized multilevel linear model is presented in Figure 1. We hypothesized that a hospital's Magnet status is positively associated with LGBTQ+ inclusive policies and practices, after controlling for the level‐1 variable (hospital‐level: hospital type) and level‐2 variables (state‐level: LGBTQ+ inclusiveness in laws, political climate, racial/ethnic minority population, and Medicaid expansion status). Additionally, we hypothesized an interaction between a hospital's Magnet status and state‐level factors, suggesting that the effect of Magnet status on LGBTQ+ inclusivity in a hospital may vary depending on these state‐level conditions.

Figure 1.

The hypothesized model of multilevel approach. We hypothesized that a hospital's Magnet status is positively associated with LGBTQ+ inclusive policies and practices, after controlling for the level‐1 variable (hospital‐level: hospital type) and level‐2 variables (state‐level: LGBTQ+ inclusiveness in laws, political climate, racial/ethnic minority population, and Medicaid expansion status). Additionally, we hypothesized an interaction between a hospital's Magnet status and state‐level factors, suggesting that the effect of Magnet status on LGBTQ+ inclusivity in a hospital may vary depending on these state‐level conditions.

We initiated our analysis by specifying the null model, which included only the dependent variable (HEI scores) with no independent variables. Next, we specified the level‐1 model, which included the dependent variable and only the level‐1 variable. Finally, we specified the level‐2 model, which included the dependent variable along with both level‐1 and level‐2 variables. Before the specification of each model, we calculated generalized variance inflation factors for every independent variable to evaluate multicollinearity. All variables exhibited low‐to‐moderate generalized variance inflation factor values, allowing us to include both level‐1 and level‐2 variables in the level‐2 model. Following the specification of each model, we examined the normality of residuals and assessed homoscedasticity, confirming no violations. Additionally, we checked Cook's distance to identify influential outliers. None of the observations were deemed highly influential, as none had a Cook's distance greater than 1 (Stevens 1984). We fitted an unstructured covariance model (Gurka, Edwards, and Muller ²⁰¹¹). Last, we conducted exploratory analyses to examine cross‐level interactions to identify potential state‐level variables that could serve as moderators of the relationship between Magnet status and the HEI score. There was no missing data for any of the variables.

3. Results

Out of 906 healthcare institutions participating in the HEI in 2021, 803 were hospitals in Washington, D.C., Puerto Rico, and 46 US states. No hospitals from Alaska, Idaho, Maine, and Wyoming participated in the HEI in 2021. For this paper, Washington, D.C., was considered a state. The two Puerto Rico hospitals were excluded. Therefore, the final study sample consisted of 801 hospitals, representing 47 states. On average, there were approximately 17 hospitals per state, ranging from 2 to 129.

Table 1 provides an overview of the characteristics of the 801 hospitals that participated in the HEI in 2021. Among them, 32.1% (257 hospitals) were Magnet hospitals, while 67.9% (544 hospitals) were non‐Magnet hospitals. The geographic locations of these hospitals can be found in Figure 2. When categorized by hospital type, 73% were short‐term acute care hospitals, 14% were VA hospitals, 4.6% were children's hospitals, 3.4% were psychiatric hospitals, 3.1% were critical access hospitals, 1.4% were rehabilitation hospitals, and 0.5% were long‐term acute care hospitals. The mean HEI score for all hospitals was 93.3 with a standard deviation (SD) of 12.1. Magnet hospitals had a mean HEI score of 95.3 (SD 9.5). Non‐Magnet hospitals had a mean HEI score of 92.3 (SD 13.1).

Table 1.

Characteristics of hospitals and states.

Hospital characteristics
	Hospital (N = 801)		HEI score		Deduction
	n	(%)	M	SD	M	SD
All hospitals	801	100	93.3	12.1	−0.9	2.4
Magnet status
Magnet hospitals	257	32.1	95.3	9.5	−0.3	2.3
Non‐Magnet hospitals	544	67.9	92.3	13.1	−1.2	2.4
Hospital type
Short‐term acute care hospitals	585	73.0	94.4	12.0	−0.3	1.7
Children's hospitals	37	4.6	97.7	6.4	−0.7	4.1
Critical access hospitals	25	3.1	90	11.9	−0.2	1.0
Long‐term acute care hospital	4	0.5	100	0	0	0
Psychiatric hospitals	27	3.4	88.9	13.5	0	0
Rehabilitation hospitals	11	1.4	81.8	32.5	0	0
Veterans Affairs hospitals	112	14.0	88.7	8.2	−5	0

State characteristics
	State (N = 47)		Hospital (N = 801)		HEI Score		Deduction
	n	(%)	n	(%)	M	SD	M	SD
Political party in control
Democratic	15	31.9	403	50.3	95.7	9.6	−0.7	2.1
Divided	11	23.4	234	29.2	90.2	14.5	−0.7	1.7
Republican	21	44.7	164	20.5	91.7	12.8	−1.8	3.4
Medicaid expansion
Expanded	34	72.3	655	81.8	93.4	12.2	−0.8	2.0
Not expanded	13	27.7	146	18.2	92.4	11.5	−1.5	3.5

	M	SD
LGBTQ+ policy tally	14.2	14.3
Racial and ethnic minorities	35.4	16.0

Note: The states included 47 US states, excluding AK, ID, ME, and WY due to no HEI participation. Children's hospitals focus on patients under 18 years old. Critical access hospitals, designated by the Centers for Medicare and Medicaid, aim to improve healthcare access and financial stability in rural areas, with 25 or fewer beds and over 35 miles from another hospital. Long‐term acute care hospitals, providing stays of 25 days or more, offer services such as respiratory therapy and pain management. Psychiatric hospitals offer round‐the‐clock inpatient care for mental illness. Rehabilitation hospitals specialize in restoring functional abilities. Short‐term acute care hospitals provide conventional care for short‐term medical needs. Veterans Affairs (VA) hospitals, federally funded and administered by the Veteran's Administration, serve war veterans and retired military personnel.

Abbreviations: HEI, Healthcare Equality Index; SD, standard deviation.

[Correction added on 29 October 2024, after first online publication: Table 1 and Table 2 were updated.]

Figure 2.

Geographic locations of Magnet and non‐Magnet Hospitals that participated in the HEI in 2021. The majority of Magnet hospitals participating in the HEI in 2021 are situated in the northeastern region.

Table 1 also presents the characteristics of the 47 states in which the 801 hospitals are located. In terms of political party in control of legislative chambers and governorship, 31.9% were Democratic, 23.4% were divided, and 44.7% were Republican. Regarding Medicaid expansion, 72.3% of states had expanded Medicaid as of 2021, while 27.7% had not. The mean LGBTQ+ policy tally among the states was 14.2 (SD 14.3), with a range from −6.5 to 34.75. The mean racial and ethnic minority population was 35.4% (SD 16.0).

The multilevel linear model comprises a fixed effects component, denoted by gamma (γ) coefficients, encompassing level‐1 and level‐2 slope coefficients as well as cross‐level interaction coefficients. Additionally, it includes a random effects component represented by tau (τ) coefficients, which signifies variability across state contexts (Paek et al. 2008). Table 2 presents the results of MLM, including the null model, level‐1 model, and level‐2 model.

Table 2.

The multilevel linear model with the HEI score as an outcome.

Null model
Parameters	Estimate	Standard error	95% CI	ICC
Constant term	90.98	1.04	88.95; 93.01	0.22
Residual	121.89	10.11	110.21; 134.82
Intercept (State ID)	33.99	6.27	18.98; 60.89

Level‐1 model
Parameters	Estimate	Standard error	95% CI	p‐value
Intercept	91.36	1.11	89.14; 93.59	< 0.001
Magnet status (reference: non‐Magnet hospital)
Magnet hospital	2.18	0.93	0.36; 4.01	0.019
Hospital type (reference: short‐term acute care hospital)
Children's hospital	2.13	1.89	−1.57; 5.84	0.259
Critical access hospital	−3.66	2.29	−8.15; 0.83	0.110
Long‐term acute care hospital	3.42	5.48	−7.34; 14.19	0.532
Psychiatric hospital	−3.66	2.17	−7.92; 0.60	0.092
Rehabilitation hospital	−10.62	3.35	−17.21; −4.04	0.002
VA hospital	−3.24	1.23	−5.66; −0.82	0.009

Random‐effects parameters	Estimate	Standard error	95% CI	p‐value
Residual	116.90	6.01	105.69; 129.30	< 0.001
Intercept (State ID)	30.43	9.27	16.75; 55.30	< 0.001

Level‐2 model
Parameters	Estimate	Standard error	95% CI	p‐value
Intercept	92.18	4.45	83.22; 101.14	< 0.001
Level‐1 predictors
Magnet status (reference: non‐Magnet hospital)
Magnet hospital	2.13	0.93	0.30; 3.96	0.022
Hospital type (reference: short‐term acute care hospital)
Children's hospital	2.13	1.89	−1.57; 5.84	0.259
Critical access hospital	−3.59	2.29	−8.09; 0.92	0.118
Long‐term acute care hospital	3.42	5.49	−7.35; 14.19	0.534
Psychiatric hospital	−3.71	2.17	−7.97; 0.55	0.088
Rehabilitation hospital	−10.55	3.36	−17.14; −3.96	0.002
VA hospital	−3.23	1.25	−5.68; −0.79	0.010
Level‐2 predictors
LGBTQ+ inclusiveness in laws and policies	0.01	1.47	−2.89; 2.89	0.999
% of racial and ethnic minority population	−0.07	0.08	−0.22; 0.08	0.371
Medicaid expansion status (reference: no Medicaid expansion)
Medicaid expansion	−2.22	2.70	−7.51; 3.07	0.411
Political party in control of legislative chambers and governorship (reference: Democratic)
Divided	−5.81	3.09	−11.87; 0.25	0.060
Republican	−4.88	3.65	−12.04; 2.27	0.181

Random‐effects parameters	Estimate	Standard error	95% CI	p‐value
Residual	117.12	6.04	105.86; 129.57	< 0.001
Intercept (State ID)	25.28	8.51	13.07; 48.89	< 0.001

Note: The statistical significance of the random‐effects parameters was determined using the likelihood ratio test.

Abbreviations: CI, confidence interval; ICC, intraclass correlation coefficient.

3.1. Null Model

The constant term for the overall average HEI across hospitals, assuming no state differences, was 90.98. The ICC was 0.22, indicating substantial variability between states. Significant variability was observed among hospitals within the same state (τ = 121.89; 95% confidence interval [CI]: 110.21–134.82) and across states (τ = 33.99; 95% CI: 18.98–60.89).

3.2. Level‐1 Model – Hospital‐Level Effects

In the level‐1 model, Magnet hospitals had significantly higher HEI scores compared to non‐Magnet hospitals (γ = 2.18, p = 0.019). Rehabilitation hospitals (γ = −10.62, p = 0.002) and VA hospitals (γ = −3.24, p = 0.009) showed significantly lower HEI scores compared to short‐term acute care hospitals. The inclusion of level‐1 predictors reduced within‐state variability from the null model (121.89) to the model with level‐1 predictors (116.90), explaining 4.1% of the HEI score's within‐state variability. Meanwhile, the decrease in variance between states from the null model (33.99) to the model with level‐1 predictors (30.43) accounted for 10.5% of the between‐state variability in the HEI score, with an associated ICC of 0.21. Despite this, significant variability persisted within states (τ = 116.90; 95% CI: 105.69–129.30) and across states (τ = 30.43; 95% CI: 16.75–55.30), prompting the introduction of additional predictors.

3.3. Level‐2 Model – State‐Level Effects

In the level‐2 model, Magnet hospitals maintained significantly higher HEI scores compared to non‐Magnet hospitals (γ = 2.13, p = 0.022). Rehabilitation hospitals (γ = −10.55, p = 0.002) and VA hospitals (γ = −3.23, p = 0.010) exhibited significantly lower HEI scores compared to short‐term acute care hospitals. However, no level‐2 predictors were found to be significantly associated with the HEI score, and no cross‐level interaction was found to be statistically significant. The between‐state variability was reduced from the null model (33.99) to the model with level‐1 and level‐2 predictors (25.28), suggesting that the model explains approximately 25.6% of the between‐state variance in the HEI score.

4. Discussion

Consistent with previously documented positive outcomes associated with Magnet status, the results of this study indicate that Magnet designation is independently associated with increased LGBTQ+ inclusivity, regardless of geographical location. These findings underscore the potential of the Magnet Recognition Program as a robust benchmark for fostering LGBTQ+ inclusivity within hospitals.

There are several potential explanations for why Magnet hospitals exhibited higher HEI performance compared to non‐Magnet hospitals. First, the Magnet Recognition Program has assessed nurses' delivery of culturally appropriate care (American Nurses Credentialing Center ²⁰¹⁷). Consequently, professional development in Magnet hospitals may have included training and educational opportunities that promote awareness and understanding of diverse patient populations, including LGBTQ+ individuals. Second, the program has also evaluated the organizations' workplace advocacy initiatives for all staff, including diversity (American Nurses Credentialing Center ²⁰¹⁷). Leadership within Magnet hospitals may have emphasized creating an inclusive and supportive work environment where diversity is valued and celebrated. Nurses in these hospitals may have been encouraged to advocate for LGBTQ+ patient needs and engage in initiatives targeting health disparities within these communities. Last, the Magnet Recognition Program underscores healthcare organizations' commitment to patient‐centered care (American Nurses Credentialing Center ²⁰¹⁷), emphasizing the importance of tailoring care to meet individual patient needs.

Healthcare policymakers should heed this evidence and explore ways to leverage the Magnet Recognition Program's principles and evaluation criteria to bolster LGBTQ+ inclusion initiatives across healthcare settings. Additionally, non‐Magnet hospitals stand to benefit from adopting the principles and evaluation criteria inherent in the Magnet Recognition Program. By embracing these standards, non‐Magnet hospitals can enhance their organizational culture, policies, and practices to better support LGBTQ+ patients, families, and employees. This proactive approach not only aligns with ethical imperatives to provide equitable care but also contributes to improved healthcare experiences and access, health outcomes, and staff morale.

Although not specifically tailored to LGBTQ+ populations, the Magnet Recognition Program recently integrated a DEI requirement as outlined in the 2023 Magnet application manual (Bryant, Pruski, and Yarbrough ²⁰²²). Since our study utilized HEI scores predating the manual's DEI focus, our findings highlight the unique contribution of Magnet status to LGBTQ+ inclusivity. The decision to incorporate DEI criteria into Magnet status reflects societal concerns, challenges within healthcare delivery systems, and insights from subject matter experts (Bryant, Pruski, and Yarbrough ²⁰²²). Therefore, future research exploring whether the DEI component leads to further enhancements in hospitals' HEI scores is warranted. Additionally, implementing more specific and inclusive measures for LGBTQ+ individuals can systematically facilitate hospitals' efforts toward LGBTQ+ inclusion.

The observed variations in HEI scores across different hospital types carry significant implications for the health and well‐being of diverse LGBTQ+ subpopulations. Our findings indicate that VA hospitals and rehabilitation hospitals demonstrate lower HEI performance compared to short‐term acute care hospitals. VA hospitals received the highest score deductions because the VA medical benefits plan excludes gender‐affirming surgeries. This exclusion is a policy set at the system level, beyond the control of individual providers and facilities, underscoring systemic barriers within the VA health system. Addressing these issues and improving LGBTQ+ inclusivity within these settings is crucial, especially given ongoing reports of health disparities among LGBTQ+ veterans (Carey et al. 2022; Kondo et al. ²⁰¹⁷). Targeted interventions or policy changes are necessary to rectify these disparities and ensure equitable healthcare access for LGBTQ+ individuals across all hospital types. While rehabilitation hospitals demonstrate poorer performance in the HEI compared to short‐term acute care hospitals, it is essential to note that inferences may not be robust due to their limited sample size. Moreover, out of over 6000 US hospitals (American Hospital Association ²⁰²⁴), slightly more than 10% participated in the HEI. This low rate underscores the need to develop more effective strategies to encourage greater participation in the HEI. Doing so would enable a more thorough evaluation of LGBTQ+ inclusivity in hospital policies and practices, thereby fostering improvements in these areas.

There are several reasons why state‐level policies in our study might not have reached statistical significance. First, we used an overall policy tally score that focused on the sum of nearly 40 laws across various policy areas. It is possible that specific laws and/or specific combinations of policies have a greater impact on HEI scores than others. Future research should consider disaggregating these laws and examining their individual effects on HEI scores to identify the most influential policies. Second, although we used policies from the prior year, the lag between the enactment of policies and their practical impact on hospital operations or staff training may extend beyond a single year. Future research should consider examining the effects of policies over a longer time frame, potentially looking at multiple years of data to capture the full extent of their impact. Finally, we recognize that these policies may affect patient‐level outcomes directly rather than operating via hospital‐level characteristics. Future research endeavors could benefit from exploring the impact of implementing LGBTQ+ inclusive policies and practices on patient‐level outcomes.

Despite the strengths of this study, including its theory‐based multilevel approach and national hospital sample, several limitations must be acknowledged. The voluntary nature of hospitals' participation in the HEI, with slightly more than 10% of hospitals participating in 2021, introduces potential selection bias. Hospitals that choose to participate in the HEI may systematically differ from those that do not, particularly in terms of their LGBTQ+ inclusivity practices. For example, hospitals more committed to LGBTQ+ inclusivity might be more inclined to participate in the HEI to showcase their policies, whereas those with less emphasis on inclusivity might opt out. As a result, the sample of participating hospitals may not fully represent all hospitals, potentially leading to biases in evaluating LGBTQ+ inclusivity and its relationship with other factors such as Magnet status. Another limitation of our study involves endogeneity, which may bias our estimates in the MLM analysis. Endogeneity occurs when an independent variable is correlated with the error term, often due to omitted variables, reverse causality, or measurement error (Nethery et al. 2022). In our case, hospitals' decisions to pursue Magnet status or participate in the HEI may be influenced by unobserved factors such as internal policies, leadership priorities, or resources, which we were unable to account for in the model. These unmeasured variables could lead to biased or inconsistent estimates (Huntington‐Klein ²⁰²¹). Additionally, despite including all hypothesized hospital‐level and state‐level variables, variations still existed across states, suggesting the presence of unknown and unexplained state‐level variables in our model. This indicates the necessity for future research to further explore these unexplained factors.

5. Conclusions

Our study provides valuable insights into the intricate dynamics of LGBTQ+ inclusivity within hospitals, considering both organizational and state‐level factors. With a socioecological approach, a unique aspect of our research lies in its specific focus on the association between a hospital's Magnet status and the implementation of LGBTQ+ inclusive policies and practices, disentangling the distinctive contribution of Magnet hospitals to LGBTQ+ inclusivity. Policymakers and healthcare leaders should consider leveraging the Magnet Recognition Program as a benchmark for promoting LGBTQ+ inclusivity within hospitals. Additionally, all healthcare institutions should prioritize comprehensive evaluations and improvements to their policies and practices to ensure inclusivity for LGBTQ+ patients.

Author Contributions

Hyunmin Yu, Stephen Bonett, and José A. Bauermeister made contributions to the conception and design of this article. Hyunmin Yu, Tari Hanneman, Stephen Bonett, and José A. Bauermeister contributed to the acquisition, analysis, and interpretation of data. Hyunmin Yu drafted the manuscript. Hyunmin Yu, Tari Hanneman, Dalmacio Dennis Flores, Stephen Bonett, Seul Ki Choi, Steven Meanley, and José A. Bauermeister revised the manuscript. All authors approved the final manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data supporting the findings of this study are available from the Human Rights Campaign Foundation. Data requests should be directed to the Human Rights Campaign Foundation.