Mental Health Effects of the COVID-19 Pandemic on Aging Lesbian, Gay, Bisexual, Transgender, and Queer Populations

Abstract

Introduction

The COVID-19 pandemic affected numerous subpopulations of people in unique ways. This study evaluated the impact of the COVID-19 pandemic on the mental health status of Broward County’s aging lesbian, gay, bisexual, transgender, and queer (LGBTQ) seniors and identified strengths, weaknesses, and opportunities for interventions to enhance positive mental health outcomes. The study was performed from April 2021 through July 2021.

Methods

Online surveys measuring anxiety, depression, social support, and resilience were delivered by email to 47 residents of Wilton Manors, Florida. Study data were collected and collated using research electronic data capture tools with bivariate analysis using the Mann–Whitney test and generalized linear regression.

Results

Bivariate analysis showed that males scored significantly higher on the social interactions subscale of the Duke Social Support Index, indicating greater social support, yet there was not a significant difference with sexual orientation or living situation. Multivariate analysis revealed differences for several dependent variables including anxiety and depression reporting being lower in LGBTQ individuals.

Discussion

This study provides important insight into specific mental health challenges faced by Broward County’s LGBTQ seniors, which can be applied to other LGBTQ populations across the globe.

Conclusion

Using the results, targeted interventions can be developed to help improve mental health outcomes during periods of isolation for all individuals.

Introduction

Broward County’s diverse population has a substantial proportion of aging adults over 65 years old who are at a higher risk for negative outcomes of COVID-19.^1,2 Various studies have established that the adverse effects of health pandemics are more pronounced in the aging population than among their younger counterparts. COVID-19, in particular, had a higher severity and fatality rate in the elderly populations, with 20% of the deaths from COVID-19 in the first wave of the outbreak in China being persons over the age of 60.^1,3 In addition to the direct effects of the COVID-19 virus, older adults also have a high incidence of anxiety and mood disorders, including depression.⁴ Older adults also have a high incidence of suicide, with males over 65 accounting for the highest overall suicide rate in the United States.^5,6 Of even more concern is that a largely underserved population resides within Broward County’s aging adult community—adults who identify as lesbian, gay, bisexual, transgender, or queer (LGBTQ). A substantial portion of the 1.7 to 4.0 million LGBTQ identifying adults over age 65 in the United States call Broward County home.⁷

Older LGBTQ individuals are more prone to comorbidities such as cardiovascular disease, cancer, and HIV/AIDS, which increases their susceptibility to the adverse health impacts of COVID-19 in comparison to their heterosexual counterparts. Lesbian and bisexual women are more likely than heterosexual women to be overweight or obese.⁸ Additionally, LGBTQ people are more likely to smoke, vape, and suffer from alcohol and drug abuse than their heterosexual counterparts, which may be attributed to a concept called “minority stress.”⁹ Furthermore, LGBTQ older adults experience greater rates of disability than heterosexual, cisgender elders.¹⁰ All of these conditions and risk factors may very well indicate a higher susceptibility to negative physical health consequences of COVID-19 but may not account for the mental health impacts. In terms of mental health, LGB adults are more than twice as likely to be affected by mental health conditions than heterosexuals.¹¹ Additionally, transgender individuals are more than 4 times as likely to report mental health conditions than cisgender individuals.¹² Not only are LGBTQ individuals more likely to have a mental health condition, but their likeliness of attempting to commit suicide is 4 times higher than heterosexuals.¹³

Evidence shows strong mental health and social networks are linked to better adaptability and fewer adverse health events in the aging adult population, and that isolation from a social/support network can have negative impacts on the mental health of the aging adult population and can also cause impairment of activities of daily living.^14–16 Therefore, the potentially detrimental effects of the COVID-19 pandemic on the physical and mental health of Broward County’s elderly LGBTQ population requires heightened attention and research.

It is imperative that investigations into the mental health of aging adults within this population be performed to identify specific, population-centered interventions. Feelings of loneliness and self-isolation can exacerbate underlying mental health issues such as suicidal ideation and substance use.¹⁰ This research project is crucial to better serving Broward’s aging LGBTQ communities. By circumventing a “one-size-fits-all” solution that may be ineffective for LGBTQ elders, targeted interventions to preserve and protect the mental health in this population may be developed. This study partnered with the major social services organization (The Pride Center) that serves the senior LGBTQ community in Broward County to facilitate distribution of the survey in the Wilton Manors community, which is the No. 2 city in the United States for same-sex couples.

The goal of this study was to investigate COVID-19 and social distancing influences on the mental health of Broward’s LGBTQ aging adult population, with a special focus on loneliness, anxiety, depression, and social support. Understanding the psychological impact of COVID-19 on older LGBTQ people in Broward County will allow the research team to envision unique solutions and interventions that can improve the overall mental and physical health of this population. In addition, mitigation strategies are being developed with community collaboration with The Pride Center to prevent further health challenges and improve social support in future pandemics.

Methods

A 72-question anonymous survey was distributed to 47 adults via email through collaboration with The Pride Center in Wilton Manors, Florida, and a Facebook social media promotion was aimed at collecting responses from aging adults and full-time residents in Broward County. The survey was distributed from April 2021 through July 2021. The study was approved as exempt by the Nova Southeastern University Institutional Review Board (IRB # 2020-312-Web) due to it being an anonymous survey with no identifying information being collected. Electronic informed consent was obtained prior to participants’ continuance with the survey.

The survey screened for residency in Broward County, age over 64, and used a smiley-face mood assessment scale to avoid exposing high-risk participants to triggering questions regarding mental health, suicide, and depression.

Survey instruments used

The validated instruments used in the survey consisted of the following with minor modifications where necessary: The Revised UCLA Loneliness Scale measured subjective loneliness using a scale that included never, rarely, sometimes, and always as answer choices; the Hospital Anxiety and Depression Scale (HADS) assessed anxiety and depression using a scale of multiple choices to assess lengths of time; the Short Form-8 Health Survey (SF-8) assessed physical and mental health over the previous 4-week period using a scale of multiple choice responses; and the Duke Social Support Index evaluated an individual’s level of social support using a scale of multiple choice responses.^17–20 In addition to the validated surveys, the survey included additional questions assessing access to basic needs and community resources.

Data Collection

Study data were collected and managed using the Research Electronic Data Capture (REDCap, Vanderbilt University) online data collection platform hosted at Nova Southeastern University.^21,22 REDCap is a secure, web-based software platform designed to support data capture for research studies, providing 1) an intuitive interface for validated data capture; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for data integration and interoperability with external sources.^21,22

Data Analysis

Summary statistics were calculated for all study variables. This includes the mean and standard deviation, median, minimum, and maximum values. The first step in the data analysis process was to conduct bivariate analysis using the Mann–Whitney test for the dependent variables HADS anxiety, HADS depression, UCLA 8, Duke social interaction, Duke satisfaction, and Duke total. The independent variables were gender, sexual orientation, living situation, and income. The next step was to model differences in the dependent variables for all independent variables using generalized linear regression. The authors included total number of health conditions to account for any variation in health. The software R 4.2.3 (Free Software Foundation) was used in all the data analyses, and statistical significance was found at p < 0.05.

Results

Demographics

Of the 47 total participants, 12 identified as female (25.5%) and 35 identified as male (74.5%), with 0 identifying as transgender man, transgender woman, or gender nonconforming. Thirteen identified as heterosexual (27.7%) and 34 as LGBTQ (72.3%); 19 (40.4%) were living alone and 27 (59.6%) were living with family or others, and 21 were making $50,000 or less (44.7%) and 26 were making $51,000 or more (55.3%). Summary statistics can be found in Supplemental Tables S1–S4.

Bivariate results

The bivariate results indicate that for the Duke Social Interaction Subscale, males have a higher score (median = 8.0, range = 25) than females (median = 4.0, range = 16), p = 0.030. No differences were found by sexual orientation or living situation. As for income, individuals making $50,000 or less scored higher on the HADS anxiety than individuals earning $51,000 or more (median = 8.0, range = 20 vs median = 4.0, range = 17), p = 0.007; and on the HADS depression (median = 6.0, range = 11 vs median = 3.5, range = 10), p = 0.016. Bivariate results can be found with the summary statistics in Supplemental Tables S1–S4.

Multivariate results

Multivariate analysis revealed differences in the following dependent variables shown in Tables 1–5:

Table 1:

Regression model for Hospital Anxiety and Depression Scale: anxiety

Predictors	HADS anxiety
	Estimates	CI	p
(Intercept)	5.57	2.88–8.25	< 0.001
Female		Reference
Male	2.52	−0.49 to 5.53	0.098
Heterosexual or straight		Reference
Gay, lesbian, bisexual, other	−4.01	−6.93 to −1.08	0.008
Alone		Reference
Family/other	0.81	−1.21 to 2.84	0.42
$50K or less		Reference
$50K or more	−2.73	−4.72 to −0.74	0.008
Total health conditions	0.72	−0.29 to 1.74	0.159
Observations	45
R2 / R2 adjusted	0.308/0.221

CI, confidence interval; p, probability value.

Table 2:

Regression model for Hospital Anxiety and Depression Scale: depression

Predictors	HADS depression
	Estimates	CI	p
(Intercept)	2.17	0.34–4.00	0.021
Female		Reference
Male	2.15	0.10–4.20	0.040
Heterosexual or straight		Reference
Gay, lesbian, bisexual, other	−2.89	−4.88 to −0.90	0.006
Alone		Reference
Family/other	−0.10	−1.48 to 1.28	0.885
$50K or less		Reference
$50K or more	−1.63	−2.98 to −0.27	0.020
Total health conditions	1.38	0.69 to 2.07	< 0.001
Observations	45
R2 / R2 adjusted	0.460/0.393

CI, confidence interval; p, probability value.

Table 3:

Regression model for UCLA 8

Predictors	UCLA 8
	Estimates	CI	p
(Intercept)	16.39	12.82–19.97	< 0.001
Female		Reference
Male	4.42	0.41–8.43	0.032
Heterosexual or straight		Reference
Gay, lesbian, bisexual, other	−5.00	−8.90 to −1.11	0.013
Alone		Reference
Family/other	−0.40	−3.09 to 2.30	0.769
$50K or less		Reference
$50K or more	−1.91	−4.57 to 0.74	0.153
Total health conditions	1.09	−0.27 to 2.44	0.112
Observations	45
R2 / R2 adjusted	0.326/0.158

CI, confidence interval; p, probability value.

Table 4:

Regression model for Duke social interaction subscale

Predictors	Duke social interaction subscale
	Estimates	CI	p
(Intercept)	8.92	5.74–12.10	< 0.001
Female		Reference
Male	0.74	−2.82 to 4.31	0.675
Heterosexual or straight		Reference
Gay, lesbian, bisexual, other	1.87	−1.60 to 5.33	0.283
Alone		Reference
Family/other	0.10	−2.30 to 2.50	0.932
$50K or less		Reference
$50K or more	1.04	−1.32 to 3.41	0.377
Total health conditions	−1.10	−2.30 to 0.11	0.073
Observations	45
R2 / R2 adjusted	0.195/0.094

CI, confidence interval; p, probability value.

Table 5:

Regression model for Duke satisfaction scale

Predictors	Duke Satisfaction
	Estimates	CI	p
(Intercept)	3.67	1.62–5.73	0.001
Female		Reference
Male	1.32	−3.62 to 0.99	0.256
Heterosexual or straight		Reference
Gay, lesbian, bisexual, other	0.73	−1.51 to 2.97	0.515
Alone		Reference
Family/other	−0.63	−2.19 to 0.92	0.413
$50K or less		Reference
$50K or more	−0.50	−2.03 to 1.03	0.512
Total health conditions	0.16	−0.61 to 0.94	0.674
Observations	45
R2 / R2 adjusted	0.195/0.094

CI, confidence interval; p, probability value.

Controlling for a person’s gender, living situation, income, and the number of health conditions, on average, LGBTQ individuals scored 4.01 points lower (95% confidence interval [CI]: 1.08–6.93) on the HADS anxiety than heterosexual individuals, p = 0.008 (Table 1). Controlling for a person’s gender, sexual orientation, living situation, and the number of health conditions, on average, individuals earning $51,000 or more scored 2.73 points lower (95% CI: 0.74–4.72) on the HADS anxiety than individuals earning $50,000 or less, p = 0.008 (Table 1). Controlling for a person’s sexual orientation, living situation, income, and the number of health conditions, on average, males scored 2.15 points higher (95% CI: 0.10–4.20) on the HADS depression than females, p = 0.040 (Table 2).

Controlling for a person’s gender, living situation, income, and the number of health conditions, on average LGBTQ individuals scored 2.89 points lower (95% CI: 0.90–4.88) on the HADS depression than heterosexual individuals, p = 0.006 (Table 2). Controlling for a person’s gender, sexual orientation, living situation, and the number of health conditions, on average individuals earning $51,000 or more scored 1.63 points lower (95% CI: 0.27–2.98) on the HADS depression than individuals earning $50,000 or less, p = 0.020 (Table 2).

Controlling for a person’s sexual orientation, living situation, income, and the number of health conditions, on average males scored 4.42 points higher (95% CI: 0.41–8.43) on the UCLA 8 Loneliness Scale than females, p = 0.032 (Table 3). Controlling for a person’s gender, living situation, income, and the number of health conditions, on average LGBTQ individuals scored 5.00 points lower (95% CI: 1.11–8.90) on the UCLA 8 Loneliness Scale than heterosexual individuals, p = 0.013 (Table 3).

Controlling for a person’s gender, living situation, income, sexual orientation, and the number of health conditions, no factor (gender, living situation, income, sexual orientation, or number of health conditions) yielded significant changes in the Duke social interaction (Table 4), satisfaction (Table 5), or combined scores.

Discussion

It is widely reported that the COVID-19 pandemic negatively impacts mental health among the aging populations in the United States and globally. However, research is lacking regarding the relationship between gender and sexual minority status and mental well-being among elderly persons during the COVID-19 crisis. This study evaluated the impact of COVID-19 and social distancing on the mental health of Broward’s LGBTQ aging adult population with specific measures including anxiety, depression, loneliness, and social support.

Depression and anxiety

In order to measure depression and anxiety, the HADS was used. HADS is frequently used as a self-rating scale developed to assess psychological distress related to depression and anxiety in nonpsychiatric patients.¹⁸ The authors’ study results indicate that individuals who identified as gay, lesbian, bisexual, or other had a much lower amount of anxiety compared to heterosexual or straight counterparts. This finding is dissimilar to Akre et al, who found that cisgender individuals had lower levels of depression, anxiety, and problem drinking during the COVID-19 pandemic time period.²³ These findings may be explained by the populations studied, with the study focusing on the aging population and the Akre study using a broader age range for participants, age 18 and older. Additionally, the Akre study was done in 5 large metropolitan areas, and this study focused on Broward County, Florida, which has known support networks for the LGBTQ community. This decrease in anxiety and depression was also seen with individuals making greater than $51,000 per year. This finding supports the already existing literature that shows an increased incidence of mood disorders (including anxiety and depression) in individuals with lower household incomes.^24–26 Interestingly, one study found no statistical difference in depression between sexual minority older adults and heterosexual older adults. The study analyzed data on older adults from the June 2020 COVID-19 module of the Health and Retirement Study.²⁷ The study, however, included 3142 heterosexual individuals and 75 sexual minorities (making up 2.3% of the total study), and with this relatively small sample size of sexual minorities, the data may represent a sampling error because it is estimated the LGBTQ+ population is 7.1% of the general population of the United States.²⁸

Another interesting finding from the study is that males were found to have higher levels of reported depression than females, contradicting the literature’s well-established reports that women have 2 times the likelihood of being diagnosed with depression.^29–31 However, other studies examined this phenomenon further and found that those gender differences may be attributed to the diagnostic criteria and differences in the expression of depression in males.^32–34 Additionally, when looking at older individuals, males have the highest incidence of suicide, which may be related to depression.^5,6

Additionally, it should be reiterated that the HADS is not a diagnostic tool for depression, and the lack of congruency between the study’s findings and other studies in the literature underscores the need for more studies to delineate the impact of the pandemic on depression and anxiety in older gender and sexual minority adults.^18,35

Loneliness

The Revised UCLA Loneliness Scale is an 8-question validated scale that enables social psychological research on subjective loneliness in an unbiased fashion.^17,36 It can assess social risk-taking as well as negative affect. The authors’ results indicated that males were significantly higher in their level of loneliness than females, supporting the current literature on loneliness.^36–38 One mechanism for this difference may be due to differences in masculinity and femininity, which were not taken into account in this study but have been found to play a role in others.³⁷

Because loneliness is pervasive among older adults and older adult sexual minorities are also more likely to report an increase in loneliness than heterosexuals, it is important to explore the more nuanced gender-based differences in the experiences of LGBTQ individuals to better meet the needs of this demographic.²⁷

Social Support

The 11-item Duke Social Support Index is a validated instrument used to determine an individual’s level of social support.²⁰ Determining social support in this vulnerable population is important to help guide future interventions to help connect the community to the support that is needed. The authors’ study indicated that males have greater social interaction than females, although other studies found no difference in the social interaction between genders.³⁹ This difference may be due to geographic differences and the social support systems in place in these areas.

Limitations

The authors’ results must be interpreted in light of the study limitations. The sample size of this study was relatively small (N = 47), and this factor alone could contribute to inexact results and conclusions. The participant pool could have been increased with further advertisement, but the length of time between the responses would have been too great to not also factor into the potential differences in the data. In addition, because this study was evaluating aging adults, it is possible that many potential participants were unaware of how to access and complete the survey due to discrepancies in technology usage. Additionally, Broward County is home to Wilton Manors, an LGBTQ+ inclusive city containing a strong support system for the community.⁴⁰ The community support may be one of the reasons that the authors did not find the LGBTQ+ population to be more affected by anxiety, depression, or loneliness.

The survey findings indicate that there was not a major impact on the population who answered the surveys related to COVID-19 in terms of anxiety, depression, and loneliness. Interventions to support the aging LGBTQ+ population should be continued to strengthen the already present support in Broward County and to increase the support in other areas.

Recommendations and future directions

The authors’ study found that in Broward County there is no significant impact of COVID-19 on the mental health of LGBTQ seniors. However, social isolation and loneliness especially are still major public health concerns that exist in this demographic. There is a lack of evidence-based interventions to reduce these conditions, hence a need to continue to envision new strategies. The AARP recently commissioned the National Academies of Sciences, Engineering, and Medicine (NAM) to study interventions for social isolation and loneliness in elderly individuals, and in this report the academies proposed recommendation 9-1: “Health care practitioners, organizations, and systems should partner with social service providers, including those serving vulnerable communities, in order to create effective team-based care (which includes services such as transportation and housing support) and to promote the use of tailored community-based services to address social isolation and loneliness in older adults.”⁴¹ The authors support this recommendation as they have seen firsthand through their interactions with The Pride Center in Broward County how social support organizations play a pivotal role in the social well-being of LGBTQ seniors.

Conclusion

Understanding the unique mental health challenges of LGBTQ seniors requires consistent research and innovation in order to maintain equitable responses to the current and future pandemics. From this study, the presence of community support can be suggested as a step in the right direction.

Supplementary Material

Table S1