Coping strategies and anxiety in young breast cancer survivors

We sought to describe coping strategies reported by young breast cancer survivors and evaluate the relationship between utilization of specific coping strategies and anxiety in survivorship. Participants enrolled in The Young Women’s Breast Cancer Study, a multi-center, cohort of women diagnosed with breast cancer at age ≤ 40 years, completed surveys that assessed demographics, coping strategies (reported at 6-month post-enrollment and 18-month post-diagnosis), and anxiety (2 years post-diagnosis). We used univariable and multivariable logistic regression to examine the relationship between coping strategies and anxiety. A total of 833 women with stage 0–3 breast cancer were included in the analysis; median age at diagnosis was 37 (range: 17–40) years. Social supports were the most commonly reported coping strategies, with the majority reporting moderate or greater use of emotional support from a partner (90%), parents (78%), other family (79%), and reliance on friends (88%) at both 6 and 18 months. In multivariable analyses, those with moderate or greater reliance on emotional support from other family (odds ratio (OR): 0.37, 95% confidence ratio (CI): 0.22–0.63) at 18 months were less likely to have anxiety at 2 years, while those with moderate or greater reliance on alcohol/drug use (OR: 1.83, 95%CI: 1.12–3.00) and taking care of others (OR: 1.90, 95%CI: 1.04–3.45) to cope were more likely to have anxiety. Young breast cancer survivors rely heavily on support from family and friends. Our findings underscore the importance of considering patients’ social networks when developing interventions targeting coping in survivorship. NCT01468246 (first posted November 9, 2011).


Background
In the USA, breast cancer is the most common cancer among young adults, with greater than 12,000 new diagnoses each year among women younger than 40 [1,2]. Young adult breast cancer survivors may face specific issues related to their age and life stage. These include interruptions to educational or career trajectories and experiencing financial hardship [3]. Commonly used therapies may result in premature ovarian failure, sexual dysfunction, and infertility, all of which can impact quality of life [1,4]. Additionally, development and progression of their peer and romantic relationships might be affected. They may be pregnant or parenting young children, which is difficult to navigate with a cancer diagnosis [5]. Furthermore, young women are prone to feeling isolated from other breast cancer patients due to their relative young age, which can exacerbate their psychosocial difficulties [6].
Consequently, young women are at increased risk of emotional and psychological sequelae both during and after treatment compared to other age groups, with one study suggesting adolescent and young adult (AYA) cancer survivors are nearly twice as likely (11.5% vs. 5.8%) to have psychological distress than adults without a cancer history [7]. Anxiety is a common psychiatric sequelae in young breast cancer patients [8]. In addition to negatively impacting quality of life and resumption of one's normal activities, anxiety can be a barrier to engagement in survivorship care [9].
Coping has been described as emotional, behavioral, and cognitive reactions to manage event-related distress [10]. While multiple studies have examined coping strategies used by women with breast cancer, including a recent study that showed an inverse association between social support and anxiety [11], most have had limited sample sizes [10,12,13] and have included largely post-menopausal populations. With few studies exploring coping strategies employed by young breast cancer survivors, we sought to describe coping strategies reported by young women diagnosed with breast cancer at age 40 and younger in the first two years following diagnosis and to evaluate the relationship between coping strategies and anxiety.

Participants
Helping Ourselves, Helping Others: The Young Women's Breast Cancer Study (YWS) is a longitudinal cohort study of women 40 years old or younger when diagnosed with breast cancer (ClinicalTrials.gov Identifier: NCT01468246). It was established to examine biological, medical, and quality of life issues specific to young women with breast cancer. From 2006 to 2016, women with newly diagnosed breast cancer were systematically identified at several community and academic hospital sites in Massachusetts as well as academic sites in Colorado, Minnesota, and Toronto, Canada. After informed consent and enrollment, women were sent surveys every 6 months for the first 3 years after diagnosis. Participants completed the baseline survey a median of 5 months after their breast cancer diagnosis. This study was approved by Institutional Review Boards at the Dana-Farber/Harvard Cancer Center and other study sites.
In total, 1,302 women enrolled in the cohort. For the current analysis, we excluded women deemed ineligible postenrollment (n = 4) and one woman who withdrew consent. We also excluded women who completed modified versions of the surveys (n = 91), including all women enrolled at the Toronto site, women who did not complete the 6and 18-month surveys, which included measures of coping (n = 328), and women with metastatic disease at diagnosis (n = 45). The final analytic cohort included 833 YWS participants ( Fig. 1).

Sociodemographic, disease, and treatment characteristics
Race, ethnicity, college education, marital status, pre-diagnosis parity (dichotomized as any children vs. no children prior to diagnosis), a single item of financial stress (enough money for special things after paying bills vs. enough money to pay bills, but little spare money to buy extra or special things vs. enough money to pay bills but have to cut back on things/difficulty paying bills no matter what) [14,15], and employment status were self-reported by participants on the baseline survey. Race and ethnicity as recorded in the medical record were used, if not available via self-report. Medical record review was used to assess stage and receptor status. Treatment was ascertained by medical record review in combination with self-report on the baseline and 6-month surveys.

Anxiety
Anxiety was assessed yearly using the Hospital Anxiety and Depression Scale anxiety 7-item subscale (HADS-A), a reliable and valid scale that has been used in other studies of young women with breast cancer [16]. For the current analysis, HADS-A data 24-month survey was used. HADS-A scores range from 0 to 21, with higher scores indicating more symptoms. Scores are typically grouped into low (0-7), mild (8)(9)(10), and moderate to high (11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21), with a moderate to high score representative of clinically meaningful anxiety and suggestive of a need for intervention [17]. Cronbach's alpha in the current study for the HADS-A was 0.87.

Statistical analysis
Frequencies and medians were reported for categorical and continuous variables, respectively. Responses of "does not apply" were tabulated for descriptive purposes but excluded from the dichotomized coping variable since this indicated a non-endorsement of any of the response options. McNemar tests were used to evaluate changes in the frequencies of coping strategies reported between 6 and 18 months. Univariable and multivariable logistic regression models were fit to examine the relationship between specific coping strategies reported at 18 months and HADS-A at 24 months (HADS-A subscale score ≥ 11 vs. ≤ 10). The models included coping strategies reported at 18 months due to the temporal proximity to the HADS-A data collected at 24 months. Due to their known or hypothesized associations with psychosocial health, particularly anxiety [18][19][20][21], the following coping strategies were included in the model: support from family, support from healthcare providers, taking care of others, alcohol/drug use, exercise, and vitamin use.
Coping strategies that were significant at the p ≤ 0.20 level in the univariable model were retained in the multivariable model, which also adjusted for age, stage, receipt of chemotherapy, and financial stress.
All analyses were conducted in SAS Studio Version 3.71 and SAS 9.4 (SAS Institute Inc., Cary, NC, USA).

Results
Study population characteristics are summarized in Table 1 Table 2 includes results from the univariable and multivariable logistic regression analyses. Compared to no reliance/ reliance to a small extent, moderate or greater reliance on emotional support from a partner/spouse/significant other (odds ratio [OR]: 0.38, 95% confidence interval [CI]: 0.20-0.72) and moderate or greater reliance on emotional support from other family members (OR: 0.44, 95% CI: 0.28, 0.70) was associated with lower odds of anxiety while moderate or greater reliance on alcohol/drug use (vs. no reliance/reliance to a small extent) was associated with higher odds of anxiety (OR: 1.98, 95% CI: 1.25, 3.13) in univariable analyses. In the multivariable model, moderate or greater reliance on alcohol/drug use (vs. no reliance/reliance to small extent, OR: 1.83, 95% CI: 1.12, 3.00) and moderate or greater reliance on taking care of others (vs. no reliance/ reliance to small extent, OR: 1.90, 95% CI: 1.04, 3.45) were significantly associated with higher odds of anxiety. Moderate or greater reliance on emotional support from other family (vs. no reliance/reliance to small extent, OR: 0.37, 95% CI: 0.22, 0.63) was associated with lower odds of anxiety.

Discussion
To the best of our knowledge, this analysis is the first to examine coping strategies used specifically by young adult breast cancer survivors. Social supports, particularly family,  friends, and partner support, were the most commonly relied upon coping strategies at both 6 and 18 months. In addition, we also found that a significantly greater percentage of women reported exercise as a coping strategy at 18 months compared to 6 months, suggesting that women may increase their physical activity following the completion of active treatment.
Partner support was the most commonly reported coping strategy employed by the cohort, with nearly 90% of women reporting moderate or greater support from their partner at 6 and 18 months. Young breast cancer survivors may be particularly vulnerable to a perceived lack of partner support, in part due to less life experience with hardship, with a prior study from our team finding that among young women with breast cancer who were partnered, those who felt unsupported had higher odds of experiencing anxiety than those who felt supported [22]. Young couples often have competing demands on time and emotional energy, including financial issues, career trajectory and work productivity, fertility issues, and raising young children [1,3,4]. Distress screening and early intervention to address these challenges should include exploration of individuals' sources of social support, particularly from their most important relationships.
In our study, moderate or greater reliance on alcohol and/or drugs was associated with anxiety. Alcohol use is a known coping strategy with significant morbidity. Alcohol use can co-occur with anxiety in the general population, with increased odds of anxiety ranging from 2.1 to 3.3 in several studies [23], similar to the magnitude of our findings. Furthermore, alcohol may be a risk factor for cancer recurrence [24,25], underscoring the importance of addressing We observed that moderate or greater reliance on taking care of others was associated with anxiety. Many young women have dependent children and they also may be caring for aging parents. Caretaker roles can increase feelings of stress and anxiety [26]. Breast cancer survivors with young children in particular can feel burdened by physical demands of that care as well as the need to explain their diagnosis to their children, potentially having to answer uncomfortable and upsetting questions about their mortality [27]. Additionally, this cohort of women, with a median age of 37, are in their prime earning years, with work and career potentially competing with other priorities. Given these challenges, interventions that aim to better support young patients should incorporate patients' families, particularly their partners and children. Studies have shown that holistic interventions for cancer patients, their children and the rest of their family unit are beneficial for all [28].
Our study has some limitations, including limited generalizability, as YWS participants are largely White, collegeeducated, and financially secure. Furthermore, the coping items were investigator-developed and are not part of a validated scale. Additionally, because women are enrolled into the YWS following diagnosis, we were unable to account for pre-existing psychological or physical co-morbidities, which may impact psychosocial health following diagnosis.
Young women with breast cancer are a uniquely vulnerable population, having to navigate the diagnosis of a serious illness and cancer treatment while confronting challenges specific to their stage of life. For both oncology and primary care providers who may be caring for this population, awareness of these issues is critical to improving survivorship care and outcomes. While young women in our study reported adopting a spectrum of coping strategies, identification of those patients who are not able to employ protective coping strategies on their own should be prioritized. Furthermore, there have been interventions that have been developed to successfully target coping skills. A recent randomized study of a family-focused consult-based intervention in breast cancer patients demonstrated a positive impact on coping ability [29]. Additionally, a pilot trial of a positive affect skills intervention (online vs. in-person vs. online and in-person) for women with metastatic breast cancer reported a "marginally significant" improvement in utilization of positive affect skills, mindfulness and self-compassion, in the online and in-person group [30]. As new interventions to promote healthy coping continue to be developed, attention to the needs of young breast cancer survivors should be a consideration, with the goal of optimizing psychosocial health in this population.

Acknowledgements
The authors wish to thank Valerie Hope Goldstein for editorial and submission assistance.  Table 2 Univariable and multivariable logistic regression of coping at 18 months and anxiety (HADS-A subscale score ≥ 11) at 24 months (N = 765) † † Excludes participants who did not have available HADS anxiety HADS-A subscale score ≥ 11 vs. ≤ 10. data at 2 years (n = 68); missing indicator variables used for coping variables, pre-diagnosis parity, financial stress where this data was unavailable ‡ Adjusted for age, pre-diagnosis parity, financial stress, stage, and chemotherapy