Lesbian, Gay, Bisexual, and Transgender Inpatient Satisfaction Survey: Results and Implications.

BACKGROUND: Disparities in health care among lesbian, gay, bisexual, or transgender (LGBT) communities have been researched in primary care, but few have examined these phenomena among LGBT inpatients.
METHODS: Descriptive correlational study. A questionnaire was emailed to an LGBT population for a total of 508 participants. Questionnaires addressed level of satisfaction with hospital employees and whether this was related to their sexual orientation or gender identity (SOGI). Descriptive statistics describe participants and hospitals. χ2 was used to examine satisfaction and belief based on SOGI, recommendation to LGBT friends, and the relationship between location and likelihood of recommendation.
RESULTS: Those reporting dissatisfaction were likely to indicate it was related to their SOGI. They were less likely to recommend the hospital to LGBT friends or family. Analysis between location and recommendation identified rural hospitals as less likely to be recommended. DISCUSSION: Lesbian, gay, bisexual, or transgender patients perceived lack of courtesy by hospital employees to be related to their SOGI.

Introduction

The importance of culturally competent care for people identifying as lesbian, gay, bisexual, or transgender (LGBT) has been receiving increased attention internationally. The US Department of Health and Human Services has recently committed itself to ensuring equity in health care for LGBT individuals. Health risks specific to LGBT individuals have been identified, and efforts to raise awareness of a patient’s sexual orientation and gender identity among health-care employees continue. The need for education regarding culturally competent care that responds to the particular needs of LGBT patients remains imperative in providing informed and inclusive clinical care (1).

Research has shown that homophobia poses a health risk to LGBT populations. The minority stress model suggests that stressors related to stigmatized sexual orientation adversely impact the health of LGBT people, including external stressors, such as anti-LGBT discrimination or violence, and internal stressors, such as concealing one’s sexual orientation. A growing body of research has indicated that minority stressors often predict health disparities between LGBT-identified individuals and the general population (2,3).

Thus, one aspect of health risk concerns an LGBT individual’s level of comfort in disclosing sexual identity to medical professionals. Indeed, LGBT individuals are at risk for being denied specific, appropriate health care based on demonstrated homophobia (4 –10). Disparities in the provision and quality of health care faced by LGBT communities have been researched in relation to primary care, but comparatively few studies have examined these phenomena as they relate to LGBT-identified patients in the hospital setting. In its annual Healthcare Equality Index (HEI), the Human Rights Campaign (11) has acknowledged the possibility that there are health-care disparities among LGBT patients in the hospital setting. The purpose of the HEI is to provide a “national LGBTQ benchmarking tool that evaluates health-care facilities’ policies and practices related to the equity and inclusion of their LGBTQ patient.” (11) (p1) In spite of the recognized importance of LGBT supportive and inclusive policies, research that investigates any effect a patient’s sexual identity might have on overall inpatient satisfaction scores is scarce. This might be of even greater concern, given the stressful and vulnerable nature of medical issues that necessitate inpatient status, compounded by facing the possibility of discrimination or inadequate care based on sexual orientation or gender identity (SOGI). The ability to choose from LGBT inclusive providers (which might be a possibility in choosing a primary care provider) is perhaps not always an option for patients requiring hospitalization. The aim of this study was to describe patient satisfaction levels among LGBT individuals who have accessed inpatient care.

Methods

The University of New Hampshire Human Subjects Review Committee approved this project in March 2015. A questionnaire was electronically mailed to an LGBT-identified population recruited from SurveyMonkey’s online survey services. This service provides a sample to researchers based on specified demographics, and participants were recruited throughout the United States. A cover letter described the survey; the respondent’s consent to participate was required to access the survey. Participants were required to be at least 18 years old, to identify as LBGT, and to have been an inpatient at any hospital in the United States. Of 626 total responses, 118 were excluded from the survey, leaving 508 participants who fulfilled the inclusion criteria. Due to the small number of transgender participants (n = 6), the authors were unable to include them as a separate category in the statistical analysis.

The survey collected participants’ basic sociodemographic data, information regarding the location of the hospitals to which they were admitted, and inquired if and how they disclosed their SOGI to anyone on the associated health-care team. The authors developed the survey based on the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey. The present survey was developed because no known surveys specific to capturing patient satisfaction scores as they relate to SOGI were identified in the literature. During the development of the survey, the authors presented drafts to a panel of experts, including nursing academics, hospital administrators, and LGBT advocacy organization professionals, for input and development. The completed survey consisted of 23 questions.

Patient satisfaction was assessed based on participant response to the following 3 questions taken directly from HCAHPS: “How often did nurses treat you with courtesy and respect?” followed by “How often did doctors treat you with courtesy and respect?” and lastly, “How often did nonmedical staff treat you with courtesy and respect?” Participants responded to these questions by indicating either always, sometimes, or never. The authors developed 3 additional questions, which asked participants: (1) “Do you believe the level of courtesy and respect provided to you by the by nurses was related to your sexual orientation or gender identity?” (2) “Do you believe the level of courtesy and respect provided to you by the by doctors was related to your sexual orientation or gender identity?” and (3) “Do you believe the level of courtesy and respect provided to you by the nonmedical staff was related to your sexual orientation or gender identity?” Participants responded to these questions by indicating either yes, uncertain, or no. The instruments for both patient satisfaction and belief that courtesy and respect was related to SOGI demonstrated high reliability, with Cronbach α of .84 and .86, respectively.

Statistical analyses were conducted using IBM SPSS Statistics for Macintosh, version 23.0. An a priori α level of .05 was set to reduce the risk of a type 1 error. Descriptive statistics were used to describe characteristics of all participants and associated hospitals. The following relationships were studied using a χ2 analysis: the relation between participants’ reported frequency of courtesy and respect shown by medical and nonmedical staff, and the belief that said courtesy and respect was based on the SOGI of the participant; the relation between the reported frequency of courtesy and respect shown by medical and nonmedical staff and the participants’ recommendation of the admitting hospital to LGBT friends or family; and the relation between the location and geographical region of the hospital and recommendation of the hospital to LGBT friends or family. Logistic regression was conducted to test if courtesy and respect shown by nurses, doctors, and staff, as well as hospital location and region, significantly predicted participants’ recommendation of the admitting hospital to LGBT friends or family.

Results

Surveys were completed by 508 participants. Participants’ personal demographic analysis revealed that of the 508 who reported their age, 15.9% indicated they were between the ages of 18 to 29 years, and 21.3% indicated they were between the ages of 30 and 44 years. Most (38.2%) participants reported their ages to be between 45 to 60 years, with 24.6% indicating their age as older than 60 years. Of the 508 participants, 51.8% indicated their sex as female, and 47% indicated their sex as male. The majority were white at 87.2%, followed by Hispanic at 4.9%, African-American at 2.6%, multiracial at 2.4%, Asian at 2.2%, and Native American at 0.8%. Approximately 23% identified as lesbian, 35.6% identified as gay, 39% identified as bisexual, and 1.2% identified as transgender; 0.4% participants specified their LGBT identity as “queer” within a provided fill-in category.

The majority (59.4%) of participants indicated that they did not disclose SOGI to the health-care team; 24.8% indicated that they voluntarily disclosed this information as a way of correcting a heterosexual assumption, and 10.6% indicated that the admission form gave them the opportunity to disclose this information or that they were asked the question directly (5.1%) during admission/registration (see Table 1).

Table 1.

Personal Demographics and Disclosure and Institutional Demographics.

Variable	Categories	N (%)
Age	18-29 years	81 (15.9)
	30-44 years	108 (21.3)
	45-60 years	194 (38.2)
	Greater than 60 years	125 (24.6)
Sex	Female	263 (51.8)
	Male	239 (47.0)
	Prefer not to state	6 (1.2)
Ethnicity	Caucasian	443 (87.2)
	Hispanic	25 (4.9)
	African-American	13 (2.6)
	Multiracial	12 (2.4)
	Asian	11 (2.2)
	Native American	4 (0.8)
LGBT identity	Lesbian	121 (23.8)
	Gay	181 (35.6)
	Bisexual	198 (39.0)
	Transgender	6 (1.2)
	Queer	2 (0.4)
Disclosure	Did not disclose	302 (59.4)
	Correcting a heterosexual assumption	126 (24.8)
	Admission form provided the opportunity	54 (10.6)
	Directly asked during admission	26 (5.1)
Institutional demographics
Variable	Categories	N (%)
Regional location	Western	134 (26.4)
	Northeast	107 (21.1)
	Southeast	89 (17.5)
	Midwest	88 (17.3)
	Southwest	50 (9.8)
	Central	40 (7.9)
Descriptive location	Urban	344 (67.7)
	Suburban	145 (28.5)
	Rural	19 (3.7)

Abbreviation: LGBT, lesbian, gay, bisexual, or transgender.

Participants were asked to indicate both regional location and the descriptive location of the admitting hospital. Most participants indicated the admitting hospital’s regional location was in the western United States, at 26.4%, followed by 21.1% in the Northeast, 17.5% in the Southeast, 17.3% in the Midwest, 9.8% in the Southwest, and 7.9% in the central region.

Participants were asked to describe the admitting hospital’s location as urban, suburban, or rural. In order to assist participants in providing an accurate description, definitions for urban, rural, and suburban were provided based on established definitions per the United States’ Rural Urban Commuting Area Codes. Based on definitions provided in the survey, the majority of participants described the admitting hospital’s location as urban at 67.7%, followed by suburban at 28.5% and rural at 3.7% (see Table 1).

Courtesy and Respect Related to SOGI

A χ2 test was conducted to analyze the relationship between participants’ reported frequency of courtesy and respect shown by nurses, doctors, and nonmedical staff, respectively, and the belief that the said courtesy and respect was based on their SOGI. Among the 508 participants who responded to this question, statistical significance was demonstrated when comparing frequency of courtesy and respect shown by nurses, doctors, and nonmedical staff alike, and the belief it was based on their sexual identity or gender identity. Those patients who reported the frequency of nurses, doctors, or nonmedical staff treating them with courtesy and respect as sometimes or never were more likely to believe that it was related to their sexual orientation: in the case of nurses, χ2 (4; N = 508) = 41.92 (P = .000); doctors, χ2 (4, N = 508) = 19.23 (P = .001); and nonmedical staff, χ2 (4, N = 508) = 32.66 (P = .00; see Table 2).

Table 2.

Courtesy and Respect Related to Sexual Orientation or Gender Identity and Identified Sexual Orientation and Gender Identity.

Reported Frequency of Courtesy and Respect	Related to SO/GI, N (%)			Role
	No	Uncertain	Yes	Nurse
Always	267 (85)	40 (12.7)	7 (2.2)
Sometimes	132 (71.4)	47 (25.4)	6 (3.2)
Never	1 (11.1)	6 (66.7)	2 (22.2)
χ2 (4, N = 508) = 41.92, P = .000
	No	Uncertain	Yes	Doctor
Always	221 (84.0)	37 (14.1)	5 (1.9)
Sometimes	174 (75.0)	49 (21.1)	9 (3.9)
Never	5 (38.5)	7 (53.8)	1 (7.7)
χ2 (4, N = 508) = 19.23, P = .001
	No	Uncertain	Yes	Staff
Always	243 (83.2)	38 (13.0)	11 (3.8)
Sometimes	154 (75.1)	49 (23.9)	2 (1.0)
Never	3 (27.3)	6 (54.5)	2 (18.2)
χ2 (4, N = 508) = 32.66, P = .000
Courtesy and Respect Related to Identified Sexual Orientation or Gender Identity
Reported Frequency of Courtesy and Respect	Related to SO/GI, N (%)			Role
	Bisexual	Gay	Lesbian	Nurse
Always	104 (52.5)	133 (73.5)	74 (61.2)
Sometimes	88 (44.4)	46 (25.4)	46 (38.0)
Never	6 (3.0)	2 (1.1)	> 1 (0.8)
χ2 (8, N = 508) = 23.236, P = .003
	Bisexual	Gay	Lesbian	Doctor
Always	85 (42.9)	119 (65.7)	56 (46.3)
Sometimes	105 (53.0)	59 (32.6)	63 (52.1)
Never	8 (4.0)	3 (1.7)	2 (1.7)
χ2 (8, N = 508) = 24.557, P = .002

Courtesy and Respect Related to Identified SOGI

Participants identifying as gay indicated that they always received greater respect from nurses, doctors, and staff (73.5%, 65.7%, and 63%, respectively) than did participants identifying as lesbian (61.2%, 46.3%, and 58.7%, respectively) or bisexual (52.5%, 42.9%, and 52.5%, respectively). All groups indicated that most of the time they were shown respect and courtesy from nurses, doctors, and staff (always or some of the time). Participants identifying as bisexual indicated the highest negative choice, with 3% of nurses, 4% of doctors, and 3% of staff identified as never showing courtesy or respect, while participants identifying as gay indicated never for 1.1% of nurses, 1.7% of doctors, and 2.2% of staff. Participants identifying as lesbian indicated never for 0.8% of nurses, 1.7% of doctors, and 0.8% of staff (see Table 2). Participants indicating transgender or queer orientation were not included in the calculations, as their number was too small to be statistically stable.

Courtesy and Respect Related to Hospital Recommendation

Likewise, statistical significance was demonstrated when comparing frequency of courtesy and respect shown by medical and nonmedical staff and the likelihood that participants would recommend the hospital to LGBT-identified friends or family. Those patients who reported the frequency of medical and nonmedical staff treating them with courtesy and respect as sometimes or never were less likely to recommend the admitting hospital to their LGBT friends or family than were those who reported frequency of medical and nonmedical staff treating them with courtesy and respect as always—in the case of nurses, χ2 (2, N = 508) = 63.10 (P = .000); in the case of doctors, χ2 (2, N = 508) = 60.20 (P = .000); and in the case of nonmedical staff, χ2 (2, N = 508) = 43.28 (P = .000).

Location and Region Related to Hospital Recommendation

A χ2 analysis of the relationship between participants’ reported hospital location and the likelihood of their recommending the admitting hospital to LGBT friends or family revealed that among the 508 responding participants, statistical significance was demonstrated. Specifically, those who indicated the admitting hospital’s location as rural were less likely to recommend the admitting hospital to their LGBT friends or family, χ2 (2, N = 508) = 21.56 (P = .000). A χ2 analysis of the relationship between the region of the hospital and the likelihood of participants recommending the admitting hospital to LGBT friends or family revealed no statistical significance (see Table 3).

Table 3.

Hospital Recommendation.

Reported Hospital Location	Likelihood of Recommendation
Urban	Yes	No
	290 (84.3)	54 (15.7)
Suburban	Yes	No
	116 (80.0)	29 (20.0)
Rural	Yes	No
	8 (42.1)	11 (57.9)
χ2 (2, N = 508) = 21.56, P = .000

Lastly, a logistic regression was performed to ascertain the effects of courtesy and respect shown by nurses, doctors, and staff, as well as hospital location and region, on the likelihood of participants recommending the admitting hospital to LGBT friends or family. The logistical regression model showed significance in terms of nurse (P = .002) and physician (P = .001) demonstrated levels of courtesy and respect, as well as location (P = .013) of the admitting hospital in predicting the likelihood of participants recommending the admitting hospital to LGBT friends or family. However, hospital staff demonstrated levels of courtesy and respect (P = .149) and hospital region (P = .607) did not contribute significantly to the prediction (see Table 4).

Table 4.

Predicting Hospital Recommendations.

Variable	B	SE	Wald	df	Significance	Exp (B)	95% CI for EXP (B)
							Lower	Upper
Step 1a
Location	0.530	0.212	6.233	1	0.013	1.699	1.121	2.576
Region	0.035	0.068	0.264	1	0.607	1.036	0.906	1.184
Nurse sat	0.896	0.290	9.514	1	0.002	2.450	1.386	4.329
Doctor sat	0.995	0.299	11.107	1	0.001	2.704	1.506	4.854
Staff sat	0.409	0.284	2.082	1	0.149	1.506	0.864	2.626

Abbreviation: SE, standard error.

a Variable(s) entered on step 1: location, region, nurse, doctor, staff sats.

Discussion

Results from this study suggest that LGBT-identified patients may perceive a lack of courtesy and respect shown to them by health-care employees to be related to their sexual orientation and/or gender identity. Although the majority of participants (59.4%) reported not disclosing their SOGI to the health-care team, it might be reasonable to suggest in some cases that a particular SOGI may have been presumed by health-care providers. In any case, sexual orientation and gender identity have yet to become an essential consensus-level variable to consider in relation to patient satisfaction. Cultural competencies attentive to race, ethnicity, and language (exclusive of SOGI) have demonstrated improved patient satisfaction scores among those specific minorities (12). In an effort to maximize improved patient satisfaction scores, it seems reasonable to suggest that all hospital education and training efforts include alternative sexual orientations and gender identities as additional markers of minority status. These would not be exclusive of other existing markers of minority status.

Efforts to implement or improve cultural competency related to LGBT patient needs and considerations would need to address specific local challenges/resistance to these efforts, particularly in rural areas. Of the few studies on LGBT health care in rural areas, most have demonstrated increased stigma, higher rates of depression, substance abuse, and lower utilization of health-care services among rural LGBT-identified people (10,13,14). Stigma is thought to have demonstrated greater influence in rural communities related to less exposure to diverse lifestyles and an intolerance of diverse lifestyles and less anonymity (13,15,16).

Although not statistically significant (P = .059), results from this survey in terms of region showed that patients admitted to hospitals located in the northeastern and western regions of the United States were more likely to recommend the hospital to LGBT friends or family. These regions also have the most states with prohibitory laws based on sexual orientation and gender identity, as well as laws and policies inclusive of transgender health care. In addition to the incorporation of established standards of care for LGBT patients, hospitals should consider both location and region in providing the most effective training and implementation strategies to make LGBT cultural competency part of the institutional culture. In addition, even in areas where health-care institutions are providing culturally competent care inclusive of LGBT patients, general patient education regarding the clinical benefits of disclosing SOGI should be provided.

Limitations

Limitations to this study are related to sample size. A larger sample size would increase the diversity of participants, particularly with regard to the transgender population. Determining and encouraging methods to increase transgender participation should be of particular concern to health disparity researchers given the disproportionate health risks, victimization, and discrimination experienced by this population. Furthermore, statistically significant findings from this study are related to sexual orientation only, due to the small sample size of those identifying as transgender. Future studies might best separate questions related to sexual orientation from gender identity in order to avoid possible misinterpretation of results. Although participants reported both sexual orientation and gender identity, the authors concluded that those who identified as transgender would be referring to their gender identity rather than sexual orientation as it related to their reported level of courtesy and respect. Furthermore, a larger sample size in general would have contributed to improved similarity in locational and regional representation. Also, recruitment may be limited in terms of participant verification through the use of an online survey provider. In addition, the survey and associated instruments would benefit from additional testing and possible refinements in a larger study. Finally, an entirely quantitative survey is perhaps inadequate to fully understand issues experienced by LGBT-identified inpatients. In spite of these limitations, the findings are an important step in documenting the experiences of LGBT patients in the hospital environment.

Conclusion

Results from this study demonstrate that LGBT-identified patients tend to believe that insensitive care shown to them by health-care employees may be related to their sexual orientation and/or gender identity. This belief also revealed a tendency among those patients not to recommend the associated hospital to LGBT friends and family.