COVID-19 Experiences Predicting High Anxiety and Depression Among a Sample of BRCA1/BRCA2-positive Women in the US

Purpose. During the COVID-19 pandemic, breast and ovarian cancer survivors experienced more anxiety and depression than before the pandemic. Studies have not investigated the similarities of this trend among BRCA1/2-positive women who are considered high risk for these cancers. The current study examines the impact of COVID-19 experiences on anxiety and depression in a sample of BRCA1/2-positive women in the U.S. Methods. 211 BRCA1/2-positive women from medically underserved backgrounds completed an online survey. Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were estimated using multivariable logistic regression for associations between COVID-19 experiences and self-reported anxiety and depression stratified by demographic factors. Results. Overall, women who reported quarantining/isolation (aOR, 0.46, 95% CI, 0.24–0.88) experienced significantly fewer depressive symptoms than women who did not report this experience. Racial/ethnic minority women caring for someone at home during COVID-19 were 3.78 times more likely (95% CI, 1.04–13.6) to report high anxiety while non-Hispanic white women were less likely (aOR, 0.36, 95% CI, 0.10–1.33, p-interaction=0.011). Conclusions. To date, this is the first study to analyze anxiety and depression considering several COVID-19 predictors among BRCA1/2-positive women. Our findings can be used to inform future research and advise COVID-19-related mental health resources specific to these women.

Participants were recruited through national, online support groups: BRCA1 BRCA2 Genetic Ovarian & Breast Cancer Gene (~ 11,000 members), BRCA Genetic Sisters Support Group (~ 6,000 members), BRCA1 & BRCA2 Support Group (~ 3,300 members), BRCA Strong (~ 2,500 members), BRCA Sisterhood of Hope (~ 1,400 members), Facing Our Risk for Cancer Empowered (FORCE) message boards, Understanding BRCA (~ 1,500 members), BRCA Advanced & Other Hereditary Cancers Journal Club (~ 3,200 members), and BRCA Preventive Mastectomy & Hysterectomy Support Group (~ 900 members) from December 2020 to April 2021. One study recruitment post was posted per day within each group (BRCA Strong only allowed one post per week), with written permission obtained from group moderators prior to posting. The post consisted of a brief announcement introducing the study, eligibility criteria, and a link to an anonymous survey. Participants were eligible if they were 18 years or older, female, lived in the US, could read/speak in English, have undergone and tested positive for either (or both) BRCA1 and/or BRCA2 genetic mutations within the past ve years, and identify with at least one medically underserved population (i.e., racial, ethnic, and/or sexual minority, person with a physical disability, those with low income, rst-generation immigrant, and/or those who are chronically ill). By clicking the brief study announcement, potential participants were rerouted to an anonymous screener survey to determine eligibility, and those tting criteria were rerouted to the full online survey via REDCap (Research Electronic Data Capture) hosted at the Johns Hopkins Bloomberg School of Public Health (JHSPH) 31,32 . Survey questions prompted participants to rate anxiety, depression, COVID-19 impact, demographic characteristics, clinical cancer and genetic testing information, prophylactic surgery and ongoing surveillance history, body satisfaction, perceived worry of cancer, cancer empowerment, health-related quality of life, discrimination, and healthcare access. Participants who completed the online survey were compensated with a $20 Amazon e-gift card. This study was approved and conducted according to the ethical standards of the JHSPH Institutional Review Board (IRB) and informed consent was obtained from all participants.

Model Variables
Predictor variables. To measure the impact of COVID-19, the Pandemic Stress Index 33 was utilized within the current study. The items involving COVID-19 experiences were as follows: changes in life due to COVID-19, diagnosed with COVID-19, fear of getting or spreading COVID-19, worrying about loved ones, quarantining or isolation, caring for someone at home, working from home, lost job, changes in healthcare services, stigma or discrimination, personal nancial loss, frustration/boredom, not having basic supplies, more depression, more anxiety, sleep issues, increased substance use, change in sexual activity, loneliness, confusion about COVID-19, giving to the greater good by following COVID-19 mandates, and getting emotional or nancial support from loved ones. The COVID-19 experiential items were entered as predictors, with one predictor in each model. Predictors were originally dichotomous with either "no" did not experience (referent) or "yes" experienced the COVID-19-related prompt during the pandemic. Items ranged from general COVID-19 occurrences (e.g., diagnosed with COVID-19, quarantining, working from home, etc.), health-related prompts (e.g., anxiety, depression, substance use, frustration/boredom, etc.), or resource reallocation (e.g., changing travel plans, nancial loss, needing nancial support, etc.).
Outcome assessments. To measure anxiety symptomology, the Generalized Anxiety Disorder 7-item (GAD-7) scale (cite). The GAD-7 is a 7-item, 4-point Likert scale prompting, "How often have you been bothered by the following over the past two weeks?" ranging from 0 (not at all sure) to 3 (nearly every day).
Responses were added to create a nal score which ranged from zero to 21 with clinical cutoffs for mild (zero to 5), moderate (six to 10), and severe anxiety (11+) 34 . The GAD-7 has a sensitivity of 89% and speci city of 82%, utilized as a screening tool to recommend further evaluation for those scoring in the moderate to severe range 34 . In the general population, the GAD-7 re ects good reliability (α = 0.89) 35 and excellent within the current sample (α = 0.93). For the purposes of this study, clinical cutoffs were dichotomized for mild (referent) and moderate/severe anxiety. Moderate/severe anxiety will be discussed as "high anxiety".
Depressive symptomology was measured using the Patient Health Questionnaire (PHQ-9) Depression Assessment 36 . The PHQ-9 is a 9-item, 4-point Likert scale asking, "Over the last 2 weeks, how often have you been bothered by the following problems?" ranging from 0 (not at all) to 3 (nearly every day).
Covariates and strati cations. The following variables were included as covariates across all models: age at survey completion, number of comorbid conditions, years since genetic testing, education, marital status, race/ethnicity, income status, and cancer survivor status (has a history of cancer versus no cancer history). Age at survey completion, number of comorbid conditions (including a past cancer diagnosis) and years since genetic testing were treated as continuous. Polynomial categorical variables were condensed into the following covariates: education (some college or less [referent], college graduate or above), marital status (married/living as married [referent], not married), survivor/control status (no cancer history [referent], cancer survivor), race/ethnicity (non-Hispanic white [referent], racial/ethnic minority), and income status (at least $40,000 USD annually per household [referent], below $40,000 USD annually per household). In separate models, strati cations by income status and racial/ethnic minority status were included, however, were entered as covariates when not in use as strati cations. education, marital status, survivor/control status, race/ethnicity, and income status. To examine the effect of experiences during COVID-19 on anxiety and depressive symptomology by income status and racial/ethnicity among BRCA1/2-positive women, an interaction term was created for COVID-19 experience combined with income status (did/did not experience during COVID-19 x income status) and race/ethnicity (did/did not experience during COVID-19 x race/ethnicity) within appropriate models. All tests were two-sided and statistical signi cance was indicated if p-values were below 0.05.

Characteristics of the Study Sample
Description of the study population and characteristics are shown in Table 1. A total of 211 BRCA1/2-positive women, both with and without a history of cancer meeting inclusion criteria were included in the current study. The sample ranged in age from 18 to 75 (M = 39.5, SD = 10.6) and most women did not have a history of cancer (n = 138, 65.4%). Most of the current sample was non-Hispanic white (NHW) (67.2%), completed a college degree or above (64.5%), and was married or living as married (62.6%). Some of the women did report having at least one physical disability (40.8%) and most reported having more than one comorbid condition including cancer (61.6%). Most women reported ≥$40,000/ year for their household incomes (77%) A total of 49 participants identi ed as being lesbian, gay, bisexual, transgender, queer/questioning, or something else (LGBTQ+). Study characteristics among NHW and racial/ethnic minority women differed signi cantly. Racial/ethnic minority women were more often 49 years of age or younger (p = 0.011), whereas NHW women had reported signi cantly more comorbid conditions than racial/ethnic minority women (p = 0.027). Some demographic characteristics also differed signi cantly by income status. Women with household incomes ≥$40,000/year more often reported a college degree or above (p = < 0.001) and being married or lived as married (p = < 0.001). COVID-19 experiences also differed by income status and racial/ethnic minority status, as depicted in Table 1.   Missing Missing Caring for someone at home Missing Missing Lost job due to COVID-19     Missing     Giving to the greater good by following mandates Getting nancial support from loved ones Missing LGBTQ + = lesbian, gay, bisexual, transgender, queer, or questioning, or other.
*Income status = annual household income less than $40,000 USD Disadvantaged health characteristics also include income status and racial/ethnic minority status.
Bolded font indicates signi cant p-value (< .05) Anxiety and COVID-19 experiences by income status and race/ethnicity  3.78 times more likely (95% CI, 1.04-13.6) to report high anxiety while NHW women were less likely to have high anxiety with caring for someone at home (aOR, 0.36, 95% CI, 0.10-1.33, p-interaction = 0.011). There were two additional interactions: working from home (p-interaction = 0.044) and experiencing changes in healthcare services (p-interaction = 0.026), but neither resulted in signi cant odds of high anxiety based on race/ethnicity. Table 2 Adjusted odds ratios (aOR) and 95% con dence intervals (CI) for the association between COVID-19 experiences and odds of high anxiety among BRCA1/2 women from disadvantaged health populations, overall and by income status and race/ethnicity    Table 3 Adjusted odds ratios (OR, aOR) and 95% con dence intervals (CI) for the association between COVID-19 experiences and odds of more depressive symptoms a BRCA1/2-positive women from disadvantaged health populations, overall and by income status and race/ethnicity

Discussion
Among BRCA1/2-positive women residing in the US, the current study analyzed relationships between experiencing COVID-19-related instances and odds of reporting anxiety and depression overall and strati ed by sociodemographic factors. Demographically, most of the sample was younger than age 50, consistent with past literature suggesting that women are being genetically tested at younger ages 38,39 . Most women were NHW and educated, but there was some diversity where as much as 40.8% reported a physical disability and 61.6% a chronic condition. The current study is novel in its relation to COVID-19, however research remains limited regarding the pandemic and its impact on at-risk cancer populations such as those with BRCA1/2 mutations. Commonalities existed with several COVID-19-related experiences predicting increases in anxiety and depression symptomologies. It appears reporting stigma or discrimination, sleep issues, or increased substance use during the pandemic resulted in signi cantly increased chances of having more anxiety and depression symptoms than women who did not report these instances. Although it is well-known that BRCA1/2-positive women report on average, higher levels of anxiety and depression than the general population, these increases have not been directly connected to the COVID-19 pandemic, but within past literature have focused on the stress of ongoing surveillance and prophylactic risk-reducing surgeries 21,40,41 and cancer patients generally 42 . Not surprisingly, there were differences in income status, where women with average/high income were less likely to report depressive symptoms if they quarantined due to COVID-19. As we know, individuals who have both the resources and time to seek mental healthcare are more likely to utilize such care 43 50 . The recency of the COVID-19 pandemic in the US has focused research on the general population and its mental health, while very little, to our knowledge, has been implemented among cancer patients or survivors, and none regarding BRCA1/2-positive women.
It is apparent that the COVID-19 pandemic had variable effects on certain groups such as BRCA1/2-positive racial/ethnic minority women and those with low income. While research is continuing to emerge in response to the COVID-19 pandemic in relation to cancer and cancer risk, future studies should focus on stratifying by groups who are at higher risk for cancer and those who have survived it. Larger, more inclusive nationwide studies may provide the framework necessary to distinctly analyze subgroups such as these so resources following this pandemic may be of bene t to all in the US. Longitudinal studies could be implemented to discover the impact of COVID-19 on the cancer care continuum, from screening to survivorship. Resources should be made available to individuals experiencing compounded disparities, like those mentioned in the current study, to help alleviate the adverse mental health symptoms that may arise due to COVID-19, surveillance, and surgery. The National Cancer Institute (NCI) 51 and American Cancer Society (ACS) 28,52 , and even several large hospital systems such as Johns Hopkins Medicine in collaboration with the National Comprehensive Cancer Network (NCCN) 53 have published websites to assist cancer patients and survivors navigate the COVID-19 pandemic. Clinically, mental health screening at routine healthcare appointments may be bene cial to this population in combination with available mental health resources and recommendations. However, because this is a new realm of research, additional research is needed to accurately describe the relationship between COVID-19, anxiety, and depression among at-risk cancer groups such as women with BRCA1/2 mutations.
Study strengths.
The current study has several strengths. Our study attempted to recruit from a combination of hard-to-reach populations and those with rare cancer hereditary genetic mutations not easily recruited in-person. The online nature of this study acted as a pilot to test if these populations could be recruited successfully and from areas across the US. We were able to recruit a female sample from diverse backgrounds, allowing for limited generalizability to subpopulations such as racial/ethnic minorities, those with low income, and those with cancer. Future studies can use these approaches to recruit other hard-to-reach populations for rare or stigmatized health conditions.

Limitations.
The current study's ndings should be interpreted with consideration of its limitations. Overall, while the current study provided a moderately large sample, the data is cross-sectional and self-reported, which may introduce misclassi cation or recall bias. Strati ed results should be interpreted with caution due to limited sample sizes among the subgroups of interest. Our ndings should be replicated in a larger study with a similar study population to con rm similarities. It is also possible that by using predictors that were originally dichotomous may limit the implication of detailed information, as future studies may ask about the severity of COVID-19 experiences in addition to incidence. These participants were recruited from online support groups, which may introduce bias by being more open and willing to share experiences than others not in support groups 54 . Therefore, generalization of these ndings is limited to the populations analyzed in the current sample.

Conclusion
The current study provides a unique view in beginning to understand the impact of the COVID-19 pandemic on anxiety and depression among women with BRCA1/2 mutations. This perspective allowed the identi cation of several COVID-19-related experiences in relation to mental health outcomes, strati ed by income status and race/ethnicity, showing that there are distinct disparities among both groups. Future research can target the development of anxiety and depressive symptom relief during and after the COVID-19 pandemic utilizing prospective longitudinal study designs, while interventions can focus on recurrent training for medical professionals working with this population. Clinically, medical professionals should offer referrals to mental health counseling for all patients, not only those who are visibly struggling during this pandemic. With genetic testing becoming more widely available, especially with the utilization of telemedicine, it is possible that women may require ongoing mental healthcare that are not currently widely available for those of low income and racial/ethnic minority groups to reduce the inequities among those with BRCA1/2 mutations.

Declarations
Funding: The current study was funded by the Johns Hopkins Ho-Ching Yang Memorial Faculty Award. KD received research support from the National Cancer Institute (T32CA009314) Cancer Epidemiology, Prevention, and Control training program.
Availability of data and material: The datasets generated during and/or analyzed the current study are available at the Principal Investigator (PI)'s discretion upon reasonable request.
Code availability: Syntax coding is available upon reasonable request from the corresponding author.

Author contributions
Kate E Dibble conceptualized and designed the study, was in charge of data acquisition, data analysis, and interpretation. Kate E Dibble also wrote the main manuscript text and revised the article, as well as approving the nal version. Avonne E Connor assisted with study conceptualization and design, data