Fear of cancer recurrence as a pathway from fatigue to psychological distress in mothers who are breast cancer survivors

Abstract Fatigue is prevalent and pervasive among breast cancer survivors. Mothers are particularly susceptible to fatigue due to the ongoing demands of their caring role. While fatigue has been associated with psychological distress in prior research, the pathway by which fatigue translates into psychological distress is unclear. Given the theoretical and empirical links between fatigue, fear of cancer recurrence (FCR) and psychological distress, the role of FCR in mediating the relationship between fatigue and psychological distress in mothers who are breast cancer survivors was investigated. Ninety‐two mothers who were breast cancer survivors completed the Depression, Anxiety and Stress Scale, PROMIS‐Cancer Fatigue Short Form and Concerns About Cancer Recurrence scale in an online survey. Mediation analysis via PROCESS was used to examine whether fatigue predicted depression, anxiety or stress through FCR. Fear of cancer recurrence mediated the relationships between fatigue and anxiety and fatigue and stress, while fatigue directly predicted depression. This study highlights FCR as a potential pathway to anxiety and stress in response to ongoing fatigue, and as a mechanism of action to reduce psychological distress among mothers who are breast cancer survivors. Future research examining this pathway from fatigue to psychological distress should also explore the nature of mothers' fears about their cancer recurring.


| INTRODUCTION
Cancer-related fatigue is a distressing, persistent, subjective sense of physical, emotional and/or cognitive tiredness or exhaustion related to cancer or cancer treatment, which often interferes with day-today functioning (Hilarius et al., 2011;Stone & Minton, 2008;Yavuzsen et al., 2009). Fatigue is reported as the greatest contributor to symptom burden in mothers with cancer 1 year after cancer therapy has been completed (Gotze et al., 2015), and may continue to be prevalent 5-10 years after cancer diagnosis (Bower et al., 2006;Hofman et al., 2007). Mothers are often presented with the struggle of balancing their family's needs and their own needs, in addition to the perceived societal demands to resume parenting responsibility and housekeeping despite their illness (Fisher & O'Connor, 2012;Kuswanto et al., 2018;Semple & McCance, 2010). Arguably these demands make mothers in particular more vulnerable to high levels of fatigue. Physical limitations due to cancer such as fatigue can also greatly interfere with mothers' daily parenting activities (Gotze et al., 2015;Hofman et al., 2007;Khoshknabi & Davis, 2006;Mackenzie, 2014) and their perceived ability to fulfil parenting demands (Kuswanto et al., 2018;Tavares et al., 2017), with the effects also felt by their caregivers and family members (Hofman et al., 2007).
Fatigue has also been consistently linked to mood disturbance and psychological distress, such as anxiety and depression, among cancer patients and survivors (Costanzo et al., 2007;Hofman et al., 2007;Smets et al., 1998;Stone et al., 2000;Stone & Minton, 2008). Nevertheless, the pathway by which fatigue translates into psychological distress is unclear (Hofman et al., 2007;Smets et al., 1998). A mounting body of research highlights a strong link between cancer survivors' physical symptoms, including fatigue, and their fears about cancer recurring (Crist & Grunfeld, 2013;Kelada et al., 2019;Lebel et al., 2018;Simard et al., 2013), indicating that cancer survivors may interpret their fatigue as a possible sign of cancer recurrence or progression. Accordingly, a case can be made for heightened fear of cancer recurrence (FCR) as a possible mechanism in the pathway from fatigue to psychological distress.
Leventhal's Common Sense Model (CSM; Lebel et al., 2018;Lee-Jones et al., 1997), suggests that the perception of internal cues (e.g., physical symptoms and fatigue) may trigger anxious preoccupation and rumination about cancer recurring and may manifest in repetitive checking behaviour and seeking reassurance from family, doctors and other healthcare practitioners Lee-Jones et al., 1997;McGinty et al., 2016;Vickberg, 2003). The CSM illustrates how somatic experiences that represent cancer may trigger a series of cognitive processes, whereby somatic symptoms such as pain and fatigue are interpreted as symptoms of cancer recurring. In addition, Mishel's Uncertainty in Illness theory proposes that the uncertain nature of cancer may heighten FCR Mishel, 1988). Under conditions of uncertainty and unpredictability, somatic triggers such as pain and fatigue are more likely to be interpreted as signs of recurrence, which in turn contribute to FCR . Both theories postulate subsequent emotional processes, for example, anxiety or distress, in response to such interpretations of somatic experiences (Lee-Jones et al., 1997). A more comprehensive, contemporary cognitive processing model of FCR (Fardell et al., 2016) also recognises the centrality of self-focussed sensitivity or hypervigilance to signs that cancer may have returned, as well as the links between FCR and distress such as anxiety or depression.
Accordingly, FCR is one of the strongest predictors of psychological distress symptoms in long term cancer survivors (Deimling et al., 2006;Koch et al., 2014). Furthermore, cancer survivors with at least one child also report higher FCR than those without children (Arès et al., 2014;Koch et al., 2013;Kuswanto et al., 2018;Lebel et al., 2012;Mehnert et al., 2009;Tavares et al., 2017;Wan et al., 2018), which is consistent with the cognitive processing model of FCR that identifies life circumstances, such as the demands of a caring role, as a vulnerability factor for heightened FCR (Fardell et al., 2016). Parents, particularly mothers, who have previously been diagnosed with cancer are also likely to worry about the negative impact that cancer recurrence would have on their children's physical and psychological wellbeing (Arès et al., 2014;Kuswanto et al., 2018;Tavares et al., 2017;Visser et al., 2005;Wan et al., 2018). It is thus not surprising that FCR has been associated with higher psychological distress in mothers who are breast cancer survivors (Arès et al., 2014;Wan et al., 2018).
Taken together, it is therefore plausible that fatigue provokes FCR in cancer survivors, and FCR is in turn associated with symptoms of psychological distress; indicating the potential role of FCR in mediating the relationship between fatigue and psychological distress. Moreover, a recent study demonstrated FCR mediated between fatigue and health outcomes among haematological cancer survivors, representing a pathway from fatigue to physical functioning and quality of life (Esser et al., 2019). These findings highlight the possible clinical benefits of targeting FCR to minimise the detrimental effects of fatigue on cancer survivors' quality of life without needing to directly reduce fatigue. Despite the potential clinical benefits, the associations between fatigue, FCR and psychological distress have not yet been explored in mothers who are breast cancer survivors, even though fatigue poses considerable parenting and psychological issues.
In view of the potential clinical benefits of better understanding psychological distress in mothers who are breast cancer survivors, it is pertinent to examine the complex relationship between fatigue, FCR and psychological distress. The current study therefore investigated the extent to which FCR mediates the relationship between fatigue and symptoms of psychological distress among mothers who are breast cancer survivors. It was hypothesised that fatigue would be associated with symptoms of psychological distress both directly, and indirectly via FCR, in our sample of mothers who are breast cancer survivors (Figure 1). Acknowledging that depression, anxiety and stress are distinct symptoms of psychological distress (Lovibond, 1998;Oathes et al., 2015;Stafford et al., 2016), the current study examined the mediating role of FCR for depression, anxiety and stress separately.

| Participants
Mothers were recruited from the Breast Cancer Network Australia (BCNA); an online network providing information and support for women affected by breast cancer in Australia. The inclusion criteria were being 18 years or older, a mother at the time of receiving the diagnosis, being able to give informed consent and to complete the online survey in English. Ethics approval was obtained from Swinburne University Human Research Ethics Committee (SHR Project 2017/008).

| Demographic and clinical information
Participants completed questions about demographics including age, relationship status, number and ages of children, highest education level, employment status (i.e., not working or working full-time/parttime); and clinical details including known disease stage, type of treatment received, and time (months) since receiving the cancer diagnosis and last treatment.

| Depression, anxiety, and stress scale (DASS-21)
The DASS-21 is a self-report questionnaire assessing three dimensions of negative emotional states: depression, anxiety and stress Page et al., 2007). There are seven items that measure each dimension, and each item is measured using a 4-point Likert-type scale, ranging from 0 (did not apply to me at all) to 3 (applied to me very much, or most of the time). Scores of at least 10 for depression, 8 for anxiety, and 15 for stress indicate mild to extremely severe distress for each subscale. The Depression, Anxiety and stress scale (DASS) full-scale scores are calculated by multiplying the DASS-21 subscale scores by two in accordance with guidelines provided by . For our study, the internal reliability for each subscale was high (α = 0.91 for depression; α = 0.81 for anxiety; α = 0.86 for stress).

| Patient reported outcome measurement information System's (PROMIS) Cancer Fatigue Short Form (PROMIS-CFSF)
Cancer related fatigue was assessed with the PROMIS-CFSF, a reliable and valid measure of fatigue in cancer populations (Cessna et al., 2016). It has seven items measuring the frequency of fatigue in the past 7 days. Each item is scored on a 5-point Likert scale, ranging from 1 (never) to 5 (always). After reverse scoring item 7, a higher sum of scores indicates greater fatigue. The internal reliability in our study was high (α = 0.81).

| Concerns about recurrence scale (CARS)
Fear of cancer recurrence was measured by the first four items of the Concerns about recurrence scale (CARS) (Vickberg, 2003), which assess the frequency, intensity and consistency of FCR, and potential for emotional distress caused by FCR, in the past month.
Each item is scored on a Likert scale from 1 (not at all) to 6 (all the time). Higher scores suggest greater FCR, with scores rounded to three or four indicating moderate levels and scores rounded to five or six indicating high levels of FCR (Liu et al., 2011;Vickberg, 2003). The CARS has been used in past studies with breast cancer survivors (Dumalaon-Canaria et al., 2016;Lebel et al., 2012;Liu et al., 2011;Lucas, 2013;van den Beuken-van Everdingen et al., 2008). For our study, the internal reliability was high (α = 0.93).

| Design and procedure
The study was cross-sectional with a convenience sample of volunteers from the BCNA. After providing consent, participants were F I G U R E 1 A graphical and statistical representation of the mediation model. FCR, fear of cancer recurrence KUSWANTO ET AL. requested to anonymously complete the measures online through Qualtrics software (Qualtrics, Provo, UT, USA), a secure web-based tool to conduct online surveys. The survey took participants approximately 30-45 min to complete.
Based on a minimum sample size required to detect moderate effects for a multiple regression model with 3 predictors (effect size = 0.15, power level = 0.8, p-value = 0.05), the minimum sample size required is 77 participants. Using Fritz and MacKinnon's (2007)  The indirect effect is considered significantly different from zero when zero is not included within the 95% confidence interval (CI), such that the CI is consistently either positive or negative. If paths a, b, and c are significant and c' is reduced compared to c, the criteria for partial mediation are met; if paths a, b, and c are significant and c' becomes non-significant, criteria for full mediation are met. The effect size was measured by the ratio of the indirect effect to the direct effect (Esser et al., 2019). The mediation analyses were conducted separately for depression, anxiety and stress.

| Sample characteristics
A total of 146 participants provided informed consent to participate in the online survey. Thirteen participants consented but provided no data, and data from 41 participants were excluded due to incompleteness. In total, 92 mothers were included in the study. Table 1 summarises the characteristics of the sample and the descriptive statistics for the measures.
The DASS subscale means for the sample were within the normal ranges for each symptom, though individual DASS subscale scores varied across participants from nil to extremely severe levels of distress. Notable percentages of participants (Depression: 26.1%, Anxiety: 17.4% and Stress: 27.2%) reported higher than normal levels of distress, from mild to extremely severe. The average PROMIS-CFSF score was moderate in our sample (mean = 17.68, on possible range of 7-35), though 31.5% of the participants reported moderate-to-high levels of fatigue. Also, the average CARS score in our sample was moderate (mean = 3.75, SD = 1.34) on a 1-6 scale. However, more than half of the participants (68.5%) reported moderate-to-high FCR (range 3.0-6.0); 45.7% of participants reported moderate levels and 22.8% of participants reported high levels of FCR.

| Correlations
Depression, anxiety and stress were significantly positively correlated with fatigue and FCR (Table 2)

| Mediation analyses
The mediation model predicted the indirect effect of fatigue through FCR, as well as the direct effect of fatigue on the psychological distress variables. Table 3     *p < 0.05; **p < 0.01 level (two-tailed). Abbreviations: FCR, fear of cancer recurrence; LLCI, lower limit confidence interval; ULCI, upper limit confidence intervals.

F I G U R E 2
Notably, higher levels of fatigue also directly predicted higher levels of anxiety and stress in our study, irrespective of mothers' fear about their cancer recurring. This may be attributed to the additional challenges that mothers often endure due to fatigue. For instance, fatigue may significantly interfere with mothers' day-to-day activities in fulfiling parenting responsibilities, which subsequently threatens their sense of identity that is tightly intertwined with their role as a parent (Kuswanto et al., 2018;Mackenzie, 2014;Tavares et al., 2017). Furthermore, fatigue may necessitate that mothers rely more on family and friends for practical support with parenting and daily house chores (Coyne & Borbasi, 2007;Elmberger et al., 2005;Fisher & O'Connor, 2012;Fitch et al., 1999;Kim et al., 2012;Öhlén & Holm, 2006), though some mothers may have difficulty trusting others to care for their children (Bell & Ristovski-Slijepcevic, 2011;Billhult & Segesten, 2003;Elmberger et al., 2000;Kim et al., 2012).
Past studies have indicated that mothers perceived societal demands to continue taking responsibility for child care and housekeeping, despite their physical limitations after receiving a cancer diagnosis F I G U R E 3 Partial mediation of fatigue by fear of cancer recurrence (FCR) in predicting stress. Figure 3 depicts the mediation model tested using Model 4 of Hayes' PROCESS in predicting stress. *p < 0.05; **p < 0.01 level (two-tailed) F I G U R E 4 No mediation of fatigue by fear of cancer recurrence (FCR) in predicting depression. Figure 4 depicts the mediation model tested using Model 4 of Hayes' PROCESS in predicting depression, in which no mediation of fatigue by FCR was found in predicting depression. *p < 0.05; **p < 0.01 level (two-tailed) KUSWANTO ET AL.
-203 (Fisher & O'Connor, 2012;Kuswanto et al., 2018;Semple & McCance, 2010). Thus, mothers who continue to experience fatigue are likely to feel anxious and worry about their ability to care for their children, suggesting an important relationship between fatigue and parenting efficacy in mothers who are breast cancer survivors.
Whilst future research could elucidate this further, it is plausible that fatigue is a detrimental factor that distinguishes a mother's ability to fulfil their parenting roles before and after their cancer diagnosis, and that poor parenting efficacy might also mediate the relationship between fatigue and symptoms of anxiety or stress.
Contrary to our hypothesis, FCR did not mediate the relationship between fatigue and symptoms of depression; rather, more fatigue was directly associated with more symptoms of depression. Past research has suggested that fatigue may serve as a constant reminder of the cancer and its associated challenges in juggling between caring for their own needs and meeting their children's needs, which may lead to the rumination about the sense of loss, sadness and guilt for being the perceived less-than-ideal mother (Fisher & O'Connor, 2012;Kuswanto et al., 2018;Tavares et al., 2017). However, we are cautious in making predictive interpretations given fatigue is a hallmark of depressive symptoms (American Psychiatric Association, 2013). Whilst the DASS21 scale does not directly assess for the severity of fatigue within the depression domain, mothers may have reported depressive symptoms such as anhedonia, feeling low and pessimistic or lacking in initiative because of fatigue; likewise, mothers might report higher fatigue stemming from their depressive symptoms (Lovibond, 1998;Yavuzsen et al., 2009).

| Clinical implications
Our findings highlight the need for ongoing assessment of mothers' physical and psychological wellbeing, as well as their fears about their cancer recurring. The cognitive processing system in FCR, which involves misinterpretation of fatigue and other physical symptoms (Fardell et al., 2016), indicates that psycho-education and cognitivebased therapy is suitable in reducing distress (Lee-Jones et al., 1997). focussing on the processes of cognition (e.g., worry, rumination and attentional biases) rather than the content of it (Tauber et al., 2019).
Physical activity-based interventions may also be helpful in addressing fatigue, depression and anxiety in mothers who are breast cancer survivors. Breast cancer survivors who have been more physically active have reported experiencing lower fatigue and depressive symptomatology when compared to those who have been less physically active (Aguiñaga et al., 2018). Another study has also examined the efficacy of a combination of social-cognitive therapy and physical activity behaviour change intervention (Rogers et al., 2017 (Fitch et al., 1999).
Previous studies emphasised that many breast cancer survivors have unmet information needs, particularly with regards to the longterm physical, psychological and social sequelae of cancer and the availability of support, which can have a significant impact on their FCR and psychological distress (Cheng et al., 2014;Connell et al., 2006;Vivar & McQueen, 2005 Vickberg, 2003). These findings highlight the need for ongoing healthcare support in providing cancer-related information and monitoring for physical and psychological symptoms during longterm follow-ups in order to manage potential FCR and be informed about how to communicate effectively to their family members with regards to their long-term survivorship care. The contribution of unmet information needs to higher FCR is also identified in the cognitive processing model of FCR (Fardell et al., 2016), and future research should incorporate unmet information needs as a predictor of FCR.

| Limitations and directions for future research
The current study involved a small sample size, which increases the potential for Type II errors and restricts statistical power in our an- level of fatigue and FCR may lead to an overestimation of their typical psychological distress, particularly their anxiety level.
Most of the participants were also older and employed, currently in a relationship, and had attained a tertiary education. Understanding that age, relationship status, socioeconomic status and cancer characteristics may influence the levels of fatigue, FCR and psychological distress (Crist & Grunfeld, 2013;Gotze et al., 2015;Koch et al., 2014;Kuswanto et al., 2018;Mehnert et al., 2009), further research is required to examine whether our findings can be generalised to other mothers with different socio-economic backgrounds and cancer diagnoses. Furthermore, due to the crosssectional nature of our study, the assumed casual pathway in which fatigue influences psychological distress via FCR would need to be further investigated via a longitudinal study design.
Looking ahead, future research examining this model in which FCR represents a pathway from fatigue to psychological distress should also explore the nature of mothers' fears about their cancer recurring, in addition to FCR severity. Vickberg (2003) explains there are varying aspects of women's FCR, including fears about the possibilities of death, health deterioration and further treatment, emotional difficulties and physical limitations, and threats to identity and womanhood, body image and sexuality, and their roles and relationships. These fears have also correlated with psychological distress, intrusive thoughts and avoidance (Vickberg, 2003