Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Oct 1.
Published in final edited form as: Psychol Health. 2021 Jun 30;37(10):1253–1269. doi: 10.1080/08870446.2021.1941960

A Latent Profile Analysis to Assess Physical, Cognitive and Emotional Symptom Clusters in Women with Breast Cancer

Ruth G St Fleur 1,*, Sara M St George 1, Molly Ream 2, Michael H Antoni 2
PMCID: PMC10068690  NIHMSID: NIHMS1761140  PMID: 34187253

Abstract

Objective:

Current research on the physical and psychological functioning of breast cancer survivors often takes an approach where symptoms are studied independently even though they often occur in clusters This paper aims to identify physical and psychological symptom clusters among breast cancer survivors while assessing clinical, psychosocial and demographic characteristics that predict subgroup membership.

Design:

Using post-surgical data collected from 240 women with stage 0-III breast cancer, symptom clusters were identified using latent profile analysis of patient-reported symptoms.

Main Outcome Measures:

Baseline measures included the Pittsburg Sleep Quality Index, the Fatigue Symptom Inventory, the Hamilton Rating Scales for depression and anxiety, and the Impact of Event Scale.

Results:

Three distinct classes were identified: 1) mild physical, cognitive and emotional symptoms, 2) moderate across all domains, and 3) high levels of all symptoms. Lower socio-economic status, minority ethnicity, younger age, advanced disease stage along with lower self-efficacy and less internal locus of control were significantly associated with a higher likelihood of class 3 membership.

Conclusion:

By identifying those most at risk for severe physical and psychological symptoms in the post-surgical period, our results can guide the development of tailored interventions to optimize quality of life during breast cancer treatment.

Keywords: Breast cancer, survivors, symptom clusters, disparities

Introduction

Breast cancer is the most common type of cancer among women in the U.S. (White-Means & Osmani, 2018), and results in long-term medical and non-medical problems (Loh & Quek, 2011). The diagnosis itself along with corresponding medical procedures and treatment side effects contribute an array of stressors for breast cancer survivors (Urcuyo, Boyers, Carver, & Antoni, 2005) who often experience a decline in physical and psychological functioning (Pudkasam, Tangalakis, Chinlumprasert, Apostolopoulos, & Stojanovska, 2017). Due to the advances in cancer detection and treatment, the number of breast cancer survivors has increased steadily. It is estimated that more than 3.8 million women in the U.S. are currently living with a history of breast cancer (Miller et al., 2019). Therefore, understanding the factors that affect quality of life in breast cancer survivors represents a significant public health priority.

Across the cancer care continuum, previous research indicates that quality of life is affected by symptoms such as pain, fatigue, sleep disturbance, emotional distress and cognitive disturbance (Berger, Farr, Kuhn, Fischer, & Agrawal, 2007; DeSantis, Ma, Bryan, & Jemal, 2014; Kehlet, Jensen, & Woolf, 2006; Schreier, Johnson, Vohra, Muzaffar, & Kyle, 2019). A systematic review of 21 studies conducted to assess long term psychosocial symptoms associated with cancer survivorship found that a considerable proportion of breast cancer survivors experienced symptoms such as pain (26%−47%), fatigue (16%−49%), and sleep disturbance (59%) (Harrington, Hansen, Moskowitz, Todd, & Feuerstein, 2010). Another systematic review of the long-term symptoms of depression and anxiety in breast cancer survivors indicated that up to two thirds of survivors reported experiencing depressive symptoms, with the highest prevalence of depression reported across studies that included patients in the first year after diagnosis (Hack et al., 2003; Maass, Roorda, Berendsen, Verhaak, & de Bock, 2015). Breast cancer survivors additionally suffer from impaired cognitive processing and post-traumatic stress disorder symptoms which can manifest as cancer-related intrusive thoughts and cognitive avoidance (Chu, Wu, & Lu, 2020; French-Rosas, Moye, & Naik, 2011). Previous studies have shown that those symptoms tend to co-occur with depression and anxiety in cancer survivors (Jacobsen et al., 1998; Kulpa, Kosowicz, Stypuła-Ciuba, & Kazalska, 2014; Matsuoka et al., 2002).

Research on quality of life in breast cancer survivors has started shifting away from studying the aforementioned symptoms in isolation, given their tendency to occur in clusters (Dodd, Miaskowski, & Paul, 2001; Given & Given, 2013; Lee, Ross, Griffith, Jensen, & Wallen, 2020; Miaskowski et al., 2017; Miaskowski, Dodd, & Lee, 2004). Using a “de novo” approach (Miaskowski, Aouizerat, Dodd, & Cooper, 2007), previous studies have focused on the identification of the number and types (e.g., physical, psychological, gastrointestinal) of clusters that might have negative synergistic effects on quality of life among cancer survivors (Chow et al., 2019; Khan, Ahmad, & Biswas, 2018; Roiland & Heidrich, 2011). This variable-centered approach highlights groupings of symptoms that can be simultaneously managed with similar treatment protocols or targeted with tailored intervention components. For instance, a study conducted in a sample of breast cancer survivors (n=219) identified two viable symptom clusters: the first one consisted of sleep disturbance, concentration and anxiety while the second one consisted of fatigue, pain, bowel issues and nausea (Berger, Kumar, LeVan, & Meza, 2020). Research on aggregated and co-occurring symptoms has also used a person-centered approach where pre-specified symptoms or clusters are used to identify subgroups of patients based on their distinct experiences with those symptoms (Crane et al., 2020; Lee et al., 2020; Miaskowski et al., 2017). Through this approach, Lee et al (Lee et al., 2020) identified four subgroups of breast cancer survivors with different levels (i.e., normal, mild or moderate) of pain, fatigue, depression and sleep disturbance symptoms. Despite the abundance of symptom clusters literature, however, most studies are conducted in racially/ethnically homogeneous samples of survivors with heterogeneous cancer diagnoses, during or following adjuvant therapy (Berger et al., 2020; Crane et al., 2020; Miaskowski et al., 2017).

Beyond the identification of subgroups with greater risk for high symptom burden, it is important to consider clinical, psychosocial and demographic characteristics that might drive the differences in symptom experiences (Miaskowski et al., 2017). Results from past research in breast cancer survivors in the U.S. indicate that an increased sense of internal locus of control over breast cancer-related outcomes is associated with improved physical (Sharif & Khanekharab, 2017) and psychological outcomes (C. S. Carver et al., 2000). Previous psycho-oncological studies have also reported that increased self-efficacy is associated with less psychological distress and better quality of life (Heitzmann et al., 2011). Additionally, past studies have identified specific demographic (e.g., age, race/ethnicity (Hulbert‐Williams, Neal, Morrison, Hood, & Wilkinson, 2012)) and cancer-related factors (e.g., stage, type of surgery (Sohl et al., 2012)) that may hinder or promote psychological and physical adjustment throughout survivorship. Beyond their association with the occurrence of physical and psychological symptoms in breast cancer survivors (Ademuyiwa, Cyr, Ivanovich, & Thomas, 2016; Perkins et al., 2007), these factors could characterize persons who are among the subgroup experiencing greater vulnerabilities.

Overall, there is a need to identify subgroups of breast cancer survivors whose quality of life is disproportionately affected by higher burden of physical and psychological symptoms. Beyond identifying those subgroups, it is also important to assess the determining clinical, psychosocial and demographic characteristics in order to develop tailored interventions that target multiple (psychological and physical) symptoms. The current state of the literature on symptom clusters in oncology patients highlights the need for studies that focus on homogeneous cancer diagnoses and demographically heterogeneous samples prior to the start of adjuvant therapy (Crane et al., 2020; Lee et al., 2020; Miaskowski, 2016; Miaskowski et al., 2015; Skerman, Yates, & Battistutta, 2012). Using an ethnically diverse sample, this paper aims to identify physical and psychological symptom clusters among patients, in the period following surgery for breast cancer, based on ratings for physical (i.e., sleep, fatigue), emotional (i.e., anxiety, depression), and cognitive indicators (i.e., intrusive thoughts, cognitive avoidance). We hypothesize that there will be at least two subgroups of women who differ based on their experiences with those sets of symptoms and that certain clinical (e.g., disease stage), psychosocial (e.g., self-efficacy) and demographic (e.g., income) characteristics will be significantly associated with the likelihood of group membership.

Methods

Study Design and Participants

This secondary data analysis used baseline data collected from a stress management study conducted among 240 women with stage 0-III breast cancer. Women aged 18–75 years old within 2–8 weeks of surgery completion were recruited from cancer treatment centers in South Florida between 1998–2005 (Wang et al., 2018). The original study excluded participants who were not fluent in English, who had initiated chemotherapy or radiation treatment, had a history of prior cancer, had severe psychiatric illness, an acute or chronic comorbid condition, or were unwilling to be randomized to study conditions. The study was originally approved by the Institutional Review Board (University of Miami, Coral Gables IRB# 93/536), and written informed consent was obtained from all participants. As part of the intervention, participants were either randomized to a 10-week Cognitive Behavioral Stress Management (CBSM) intervention or 1-day psycho-educational seminar. Additional details on the parent trial are reported elsewhere (Wang et al., 2018).

Measures

Demographics:

The study team collected additional data on socio-demographic factors (i.e. age, income, education, partnered status) as well as medical information regarding stage of disease, type of surgical procedure undergone (lumpectomy vs mastectomy). Medical information was collected through self-report and verified with medical chart review.

Fatigue:

The fatigue intensity subscale of the Fatigue Symptom Inventory (FSI) was used to measure the most, least, and average levels of fatigue in the past week, and the level of fatigue at the time of questionnaire completion, on a 9-point Likert scale (Cronbach’s alpha = 0.85). Higher scores indicate greater levels of fatigue. The FSI was developed and validated with breast cancer patients (Hann et al., 1998).

Sleep:

The Pittsburgh Sleep Quality Index (PSQI) was used to retrospectively measure quality of sleep and specific sleep disturbances within the past 30 days (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). The PSQI has seven subscales, with scores ranging from 0–3. The scale has previously demonstrated good psychometric properties and a Cronbach’s alpha of .72 among breast cancer survivors (Carpenter & Andrykowski, 1998).

Clinical rating of depression:

Baseline depression symptoms experienced were measured by the Hamilton Rating Scale for Depression (HRSD) (Hamilton, 1960). The 17 items of the scale pertain to symptoms of depression experienced over the past week. Assessors who coded the ratings were trained by a clinical psychologist with extensive training in the use of this measure.

Clinical rating of anxiety:

Anxiety Symptoms were measured using the Hamilton Anxiety Rating Scale (HRSA)(Hamilton, 1959). The scale contains 14 items which are scored on a scale of 0 (not present) to 4 (severe). The items of the scale assess both psychic (mental agitation and psychological distress) and somatic anxiety (physical complaints related to anxiety). Previous studies have reported high interrater reliability, internal consistency, and discriminant validity for this scale(Schwab, Bialow, Clemmons, & Holzer, 1967).

Intrusive thoughts and cognitive avoidance:

The Impact of Event Scale was used to assess intrusive thoughts about breast cancer as well as attempts to suppress those thoughts. Both subscales are indicators of event-related distress. While the intrusion subscale measures unwanted thoughts and images related to the stressor (e.g., “I had trouble falling asleep or staying awake because pictures or thoughts about it came into my mind”), the avoidance subscale assesses attempts to prevent oneself from thinking about the situation (e.g., “I tried not to think about it”). This 15-item self-report instrument has response options coded 0 (not at all), 1 (rarely), 3 (sometimes), and 5 (often). Both subscales have demonstrated good psychometric properties in past research (Antoni, Wimberly, et al., 2006).

Self-efficacy:

The Measure of Current Status (MOCS) was used to measure cognitive coping self-efficacy (C. Carver, 2006). The MOCS has two parts. Items in Part A assess various skills such as relaxation, awareness of tension, assertiveness, and self-efficacy that can be used in challenges and demands of everyday life. Self-efficacy is represented by five items, including “It’s easy for me to decide how to cope with whatever problems arise” and “I can come up with emotionally balanced thoughts even during negative times.” Items are on a 5-point Likert scale ranging from 0 (“I cannot do this at all”) to 4 (“I can do this extremely well”).

Locus of control:

Locus of control was measured using one item which assesses an individual’s perception regarding their control over breast cancer. Responses were provided on a 9-point scale from 1 (“The outcome will be determined entirely by things that are under my personal control”) to 9 (“ The outcome will be determined entirely by things outside of my control”. A low score indicates high sense of internal control while a high score indicates high sense of external control.

Analysis

Using multiple observed variables, the latent class statistical method allows for the identification of unobserved subgroups or classes of women with breast cancer who share similar physiological and psychological profiles. When the observed variables are continuous, as in the present study, a latent profile analysis (LPA) (Lee et al., 2020; J. Vermunt & Magidson) is conducted. The present study conducted an LPA to identify symptom clusters in breast cancer survivors using indicators of fatigue, sleep, depression, anxiety, intrusive thoughts, and cognitive avoidance. To evaluate model fit and determine final number of latent classes, a series of statistical indices were used: (a) the Akaike information criterion (AIC) (Akaike, 1974), (b) the Bayesian information criterion (BIC)(Schwarz, 1978), (c) the Vuong–Lo–Mendell–Rubin likelihood ratio test (VLMR), (d) parametric bootstrapped likelihood ratio test (BLRT), and (e) entropy of 0.8 or greater.

Descriptive statistics for the psychosocial, demographic and medical covariates across classes were obtained using 1) the Bolck, Croon, and Hagenaars (BCH) approach (Bakk, Tekle, & Vermunt, 2013), which estimates the mean of continuous variables across the different classes and 2) the DCAT approach (Lanza, Tan, & Bray, 2013), which is preferable for categorical outcomes.

To determine factors that predict class membership, unadjusted logistic regression models used psychosocial (i.e. self-efficacy), sociodemographic (age income, education, race), and medical (i.e. type of procedure, disease stage) predictors. Variables with significant univariate associations with class formation were retained in a multinomial logistic regression to predict class formation, using the R3STEP method (J. K. Vermunt, 2010). Mplus, version 8.4, was used for the latent profile analyses. Additional descriptive analyses were performed using IBM SPSS Statistics, version 26.0.

Results

Sample Characteristics

The sample (N=240) was mostly middle aged (M=50.3 +/− 9.0, range= 23–70), White women (63.1%) with a mean of 15.6 +/−2.4 years of education. Approximately 62.2% of the participants were married or otherwise partnered, and 73.9% of them were employed full time. Mean annual household income (in thousands) was 79.6 +/− 67.1. See Table 1 for additional descriptive statistics.

Table 1.

Baseline Sociodemographic Characteristics

Variables
Total; n (%) 240 (100%)
Age(years); M (SD) 50.3 (9.0)
Race/Ethnicity; n (%)
 White 152 (63.1%)
 Black 21 (8.7%)
 Hispanic 61 (25.3%)
 Asian †
Marital Status; n (%)
 Married (or partnered) 150 (62.2%)
 Separated †
 Divorced 51 (21.2%)
 Widowed 11 (4.6%)
 Single 23 (9.5%)
Employed Full-Time; n (%) 178 (73.9%)
Years of Education; M (SD) 15.6 (2.4)
Annual Household Income(thousands); M (SD) 79.6 (67.1)
Breast Cancer Stage
Stages 0 and 1 128 (53.1%)
Stages 2 and 3 113(46.9%)
Surgical Procedure
Lumpectomy 122(50.6%)
Mastectomy 118 (49.0%)
Open in a new tab

Abbreviations: M=Mean; SD=Standard Deviation

†

Not reported, observations fewer than 10

Description of Symptom Clusters

Three subgroups of breast cancer survivors were identified based on ratings for physical (i.e., sleep, fatigue), emotional (i.e., anxiety, depression), and cognitive symptom indicators (i.e., intrusive thoughts, cognitive avoidance). A three-class solution was chosen based on the fit indices (shown in Table 2). Smaller AIC and BIC in the three-class model showed that it fits the model better than the two-class model. Even with a smaller AIC and BIC than the three-class model, the smaller entropy and non-significance fit index for the VLMR of the four-class model corroborated that the three-class model fit the data better than a four-class model.

Table 2.

Model Fit Information for Latent Profile Analysis for the Study Sample

Class AIC BIC Entropy VLMR LMRa
2 8121.7 8187.9 0.90 P<0.0001 P<0.0001
3 8039.8 8130.3 0.85 P<0.05 P<0.05
4 7994.2 8109.0 0.79 P=0.09 P= 0.10
Open in a new tab

AIC—Akaike information criterion; BIC—Bayesian information criterion; LPA—latent profile analysis; VLMR—Vuong–Lo–Mendell–Rubin likelihood ratio test; LMRa—Lo-Mendell-Rubin Adjusted

In all three classes, mean scores for sleep and fatigue met cut-offs for clinical significance, which are five (Curcio et al., 2013) and three (Donovan, Jacobsen, Small, Munster, & Andrykowski, 2008) respectively, with increasing severity across classes. Per established clinical cut-offs, participants in class 2 had a mean that indicated mild depression while the mean in class 3 indicated moderate depression (Sharp, 2015). A similar pattern was observed for anxiety based on clinical cut-offs (Hamilton, 1959). There are currently no established cut-off scores for cognitive avoidance and intrusive thoughts (Motlagh, 2010). Each class was labeled by its overall level of symptoms compared to the other classes. As such, class 1 (61%) is labeled as low physical, cognitive and emotional symptom levels. Class 2 (28%) is labeled as moderate across all domains while class 3 (11%) is labeled as high symptom levels. Differences in severity of symptoms between latent classes are displayed in Table 3.

Table 3.

Differences in Severity of Symptoms among the Latent Classes using the Three-Class Solution

Class 1 (n= 147) Class 2 (n=67) Class 3 (n=26)
Mean SE Mean SE Mean SE
Depression 4.1 0.4 10.5 0.9 18.4 0.9
Anxiety 4.2 0.4 10.5 0.8 18.5 1.4
Intrusive Thoughts 11.4 0.7 15.1 2.0 22.3 1.3
Cognitive Avoidance 8.8 0.6 12.1 1.5 16.9 1.6
Sleep 7.0 0.4 9.8 0.6 12.6 0.6
Fatigue 3.9 0.2 4.7 0.2 5.8 0.3
Open in a new tab

Abbreviation: SE = Standard Error

Descriptive Statistics Across Symptom Clusters

Using the BCH procedure, chi square tests were conducted to assess the equality of means of continuous variables across classes. The overall chi-square tests, with 2 degrees of freedom, were significant for locus of control (χ2= 6.823, p<0.05), self-efficacy (χ2= 15.099, p<0.01), income (χ2= 9.451, p<0.009), and age (χ2=8.547, p<0.05). Further tests comparing pairs of clusters for those significant variables revealed that means for locus of control only differed significantly between classes 2 (moderate symptom levels) (M=5.062 +/− 0.267) and 3 (high symptom levels) (M=6.216 +/−0.349), with participants in class 3 having greater external locus of control on average (p<0.05). Income and age only differed significantly between classes 1 (low symptom levels) and 3 (high symptom levels).

Compared to those in the high symptom level class, participants in the low symptom level class had higher average income (M1= 85.044 +/− 5.894 vs M3= 55.685 +/− 7.641, p<0.01) and were older on average (M1= 51.621 +/− 0.797 vs M3 = 46.308 +/− 1.792, p<0.01). Significant differences for self-efficacy were observed between classes 1 and 2 (χ2 = 6.476, p<0.05) as well as between classes 1 and 3 (χ2 = 11.621, p<0.01). Individuals in the low symptom level class reported greater self-efficacy on average (M=16.859 +/− 0.357) compared to both individuals in the moderate level class (M=15.014 +/− 0.581) and those in the high level class (M= 14.023 +/− 0.753).

The DCAT approach was used to test for equality of probabilities across classes. The only significant chi-square tests were found for stage (χ2= 6.373, p=0.041) and Hispanic ethnicity (χ2= 6.168, p=0.046). For early and advanced stage diagnoses, the only significant differences in probabilities were observed between classes 1 (low symptom level) and 3 (high symptom level). Individuals in class 1 were more likely to have advanced stage diagnoses (58.8%) while those in class 3 were more likely to have advanced stage diagnoses (68.5%). The chi-square test comparing the two classes was significant (χ2 = 5.395, p<0.05). Approximately 20% of individuals in class 1 reported Hispanic ethnicity while 43.2% of those in class 3 were Hispanic. This difference was significant (χ2 = 4.642, p<0.05).

Multinomial Logistic Regressions

Comparison of class 1 (low symptom levels) versus class 3 (high symptom levels):

Stage at diagnosis and minority race/ethnicity (being Black or Hispanic) significantly predicted the likelihood of reporting high levels of physical, cognitive and emotional impairment (class 3). Compared to participants diagnosed at stages 0 or I, participants with disease stage II or III had a lower likelihood of membership in class 1 versus class 3 (OR=0.362, 95% CI [0.071, 1.8510], p<0.05). Black participants had a lower likelihood of membership in class 1 compared to class 3 (OR=0.075, 95% CI [0.010, 0.536], p<0.001). Similarly, Hispanic participants were less likely to be in class 1 versus class 3 (OR= 0.101, 95% CI [0.022, 0.462], p<0.001). Higher self-efficacy was associated with greater likelihood of class 1 membership versus class 3 (OR= 1.377, 95% CI [1.088, 1.743], p <0.05).

Comparison of class 2 (moderate symptom levels) versus class 3 (high symptom levels):

External locus of control and minority ethnicity significantly predicted likelihood of being in class 3 vs. class 2. Women with greater external locus of control were less likely to be in class 2 vs. class 3 (OR= 0.664, 95% CI [0.449, 0.922], p<0.05). Even though race/ethnicity appeared to differ significantly between classes (significant p-values), the significance was not maintained in the confidence intervals of the odds ratio. With non-significant confidence intervals for the odds ratios, Black women had seemingly a smaller (non-significant) likelihood of being in class 2 compared to class 3 (OR= 0.314, 95% CI [0.052, 1.907], p<0.05) as did Hispanic women (OR=0.328, 95% CI [0.076, 1.594], p<0.05).

Discussion

This study contributes to a body of research on clusters of symptoms (i.e., psychological, physical) in breast cancer survivors. Beyond identifying symptom clusters, this study aimed to assess the relationship between latent class membership and certain covariates in an ethnically diverse sample of breast cancer survivors using an LPA approach. Three distinct classes were identified. The classes were labeled based on the mean of symptom scores across domains indicative of physical, emotional and cognitive functioning: class 1 reported consistently lower symptom scores on average, class 2, moderate symptom scores, and class 3 reported higher symptom scores on average. All classes met clinical cut offs for sleep disturbance and fatigue. On average, participants in class 2 reported mild depression and anxiety while the severity increased to moderate in class 3 participants. Aside from the identification of variables that predict class membership, this study assessed descriptive characteristics of class members in order to find variables that interrelate in their influence on symptom development in breast cancer survivors. Our findings indicate that younger, Black and Hispanic survivors with low income, who are also more likely to be diagnosed with advanced stage breast cancer, are more likely to experience high levels of impairments in physical, cognitive and emotional functioning.

Similar to previous research on the co-occurrence of symptoms in breast cancer survivors (Berger et al., 2020; Khan et al., 2018; Lee et al., 2020), our findings indicate that there is a class or subgroup of survivors who experience higher levels of co-occurring physical, emotional and cognitive symptoms. With all three classes meeting clinical cut-offs, our results also point to the high prevalence of fatigue and sleep disturbances among cancer survivors. A recent review of studies focused on cancer-related fatigue in breast cancer survivors found that fatigue was the most commonly reported symptom in breast cancer survivors, with 60% indicating some degree of fatigue (Ruiz-Casado, Álvarez-Bustos, de Pedro, Méndez-Otero, & Romero-Elías, 2020). Findings from past research have also shown that, compared to controls, women with breast cancer are more likely to experience poor sleep quality from diagnosis to several years into survivorship (Trudel-Fitzgerald et al., 2017).

Results from the present study align with previous findings regarding the socio-economic and demographic factors associated with symptom severity in breast cancer survivors. Descriptive analyses across clusters revealed that survivors in the cluster with high levels of physical, cognitive and emotional symptoms tended to be younger. These results are corroborated by past studies which have found that younger patients with breast cancer tend to experience greater levels of physical and psychological symptoms (Avis, Levine, Case, Naftalis, & Van Zee, 2015; Lee et al., 2020; Roiland & Heidrich, 2011). Previous studies have attributed greater levels of symptom burden in younger survivors to more invasive tumor types needing more aggressive treatment options (Ademuyiwa et al., 2016). Furthermore, younger patients often deal with more extensive treatment regimens and additional treatment-related challenges such as preservation of fertility and social issues related to career activities and caregiving roles (Radecka & Litwiniuk, 2016).

In line with previous studies, our results indicate that individuals with higher symptom burden have lower income on average compared to those with lower symptom burden. The effect of socio-economic status on symptoms has been explained by the association between indicators of socio-economic status (i.e., education, employment) and engagement in health behaviors such as physical activity (Naik et al., 2016). Past findings have linked physical activity with decreased symptoms and increased quality of life in breast cancer survivors (Phillips & McAuley, 2014). Moreover, higher income could potentially mean greater access to high quality of care and additional services for symptom management following treatment (DiMartino, Birken, & Mayer, 2017). Women with lower socioeconomic status could also experience greater levels of distress due to inability to meet financial challenges during treatment.

Our findings also revealed that, compared to the low symptom level group, the high symptom group had a significantly higher proportion of ethnic minority survivors, specifically Hispanic women and that nearly 70% of women in that cluster were diagnosed with later stage (II or III) breast cancer. Results from studies with national data indicate that ethnic minority breast cancer survivors (i.e., Black and Hispanic women) present with more advanced stages than their White counterparts (Chatterjee, He, & Keating, 2013). In subsequent logistic regression analyses, we found that being diagnosed at a later stage as well as being an ethnic minority survivor significantly increased the odds of belonging to the high symptom level cluster, compared to the low-level symptom cluster. Those results indicate that younger, ethnic minority survivors with low income, who are more likely to be diagnosed with later stage, and more aggressive forms of breast cancer, are at higher risk of experiencing high levels of breast cancer-related physical, cognitive and emotional impairments. This subgroup should be the focus of future longitudinal studies as well as tailored interventions that promote systemic changes to facilitate symptom management in this population.

Despite considerable research on the covariates associated with symptom clusters in breast cancer survivors, this is one the few studies that considers psychological factors (i.e., self-efficacy and locus of control) as predictors of class membership (Miaskowski, 2016). The findings of this study were consistent with previous studies that have explored the overall role of locus of control in psychopathology and have found locus of control to be related with positive psychological outcomes. Specifically, increased internal locus of control following a breast cancer diagnosis is related to less emotional distress (Taylor, Lichtman, & Wood, 1984) as well as less anxiety and depression (Sharif & Khanekharab, 2017). Additionally, in breast cancer survivors, increased internal sense of control has been associated with adaptive coping mechanisms such as information seeking with regards to the illness and treatments, perhaps leading to improved symptom management (Sharif & Khanekharab, 2017).

Primarily explored in the context of behavior maintenance, coping self-efficacy was assessed as a predictor of class membership in this study. Our results indicated that higher confidence for using cognitive coping strategies (e.g., cognitive reframing) was associated with reporting less symptoms across all three domains. Previous findings suggest that self-efficacy may sometimes serve as an intrapsychic resource that protects against the negative physiological and psychological impact of breast cancer on wellbeing (Shelby et al., 2014). A longitudinal study conducted with a large sample (n=684) of breast cancer survivors found that baseline self-efficacy was associated with increased emotional wellbeing a year later (Rottmann, Dalton, Christensen, Frederiksen, & Johansen, 2010). Furthermore, previous studies have linked self-efficacy with more adaptive coping styles (Chirico et al., 2017). As a state-like expectation in one’s ability to cope with cancer-related stressors, self-efficacy might promote engagement in daily activities (e.g. Physical Activity (Dalle Grave, Calugi, Centis, El Ghoch, & Marchesini, 2010)), even when experiencing difficult symptoms, leading to greater emotional and physical wellbeing. These results imply that self-efficacy and locus of control are clinically important psychological factors that could potentially explain the effects of psycho-behavioral interventions on symptom burden in breast cancer survivors (Antoni, Lechner, et al., 2006).

This study has several limitations. Given the cross-sectional design, this study could not explore the causality or directionality between covariates and clusters. Even though some factors (e.g., lower income and self-efficacy) are more prevalent in groups of women with greater symptom burden, it is unclear whether those factors trigger the concentration of high symptom levels or are a consequence of the symptoms. Specifically, it could also be that having higher symptom burden interferes with income-generating abilities or leads to lower confidence in coping abilities. Several symptoms that are highly prevalent among breast cancer survivors during treatment (e.g., pain, sexual difficulties) were not included in the analyses. Furthermore, given that baseline data were collected shortly after surgery and before treatment initiation, our study was unable to take into consideration the association between types of treatment (e.g., chemotherapy, radiation therapy) and symptom cluster membership. A future longitudinal study including a broader range of symptoms could explore the influence of breast cancer treatments, whether symptom levels resolve over time, and whether psychosocial, sociodemographic and medical characteristics (e.g., disease stage) predict trajectories of clusters or symptom levels over time.

Despite the limitations, there were several strengths of this study, the main one being the use of an advanced analysis method, LPA. The identification of symptom clusters in breast cancer survivors have primarily been conducted using a variable-centered approach (e.g., regression and factor analysis). The use of the LPA in this study allows for the emergence of similarities between individuals and the clusters identified shed light on subgroups of survivors with higher risk of distress and whose survivorship experiences may be severely impacted by high levels of symptoms. In addition, the focus on symptoms across different dimensions (i.e., physical, emotional, cognitive) is a strength of this study because women who experience the presence of multiple symptoms often have poorer functional status overall. The identification of clusters of symptoms facilitates the development of multi-component, cost-effective interventions that target high risk groups with greater symptom burden.

Conclusion

This study identified subgroups of women in three symptom clusters (based on levels of physical, cognitive and emotional symptoms) in a sample of female breast cancer patients in South Florida and contributes to the literature on the complexity of the treatment and survivorship experience in this population. Our results revealed that lower socio-economic status and minority ethnicity were associated with clustering of high symptom levels across physical, emotional and cognitive domains. Younger women, who tend to be diagnosed at a more advanced stage of breast cancer, also reported higher levels of symptoms across all three domains. Greater confidence in coping abilities along with greater internal sense of control are psychological factors associated with decreased odds of high symptom level clustering. These findings can be used to identify those who are most at risk of experiencing the most severe symptoms and lead to the development of tailored interventions that can optimize quality of life during treatment.

Table 4.

Results of Multinomial Logistic Regression: Predicting Symptom Clusters

Predictor OR SE P-Value 95% CI
Class 1 vs Class 3
Age 1.10 0.06 0.08 0.99 – 1.22
Education 1.22 0.30 0.45 0.76 – 1.97
Income 1.02 0.14 0.23 0.99 – 1.05
Black Ethnicity
Yes 0.08 0.08 <0.001* 0.010– 0.54*
Hispanic Ethnicity
Yes 0.10 0.08 <0.001* 0.022– 0. 46*
Marital status
Married 0.94 0.74 0.93 0.20 – 4.41
Disease Stage
Stage II and III 0.36 0.30 <0.05* 0.071 – 1.85
Type of Procedure
Mastectomy 1.08 0.86 0.93 0.23 – 5.13
Reconstructive Surgery
Yes 0.72 0.63 0.65 0.13 – 4.03
Locus of Control 0.76 0.15 0.11 0.52 – 1.21
Self-efficacy 1.38 0.17 < 0.05* 1.09 – 1.74*
Class 2 vs Class 3
Age 1.09 0.06 0.13 0.98 – 1.20
Education 1.19 0.30 0.54 0.72 – 1.94
Income 1.02 1.02 0.25 0.99 – 1.05
Black Ethnicity
Yes 0.31 0.29 <0.05* 0.052 – 1.91
Hispanic Ethnicity
Yes 0.33 0.26 <0.05* 0.067 – 1.59
Marital Status
Married 0.41 0.35 0.09 0.077 – 2.17
Disease Stage
Stage II and III 0.48 0.43 0.22 0.083 – 2.74
Type of Procedure
Mastectomy 0.80 0.69 0.77 0.15 – 4.33
Reconstructive Surgery
Yes 1.05 0.97 0.96 0.17 – 6.46
Locus of Control 0.66 0.13 <0.05* 0.45 – 0.98*
Self-efficacy 1.17 0.14 0.22 0.93 – 1.48
Open in a new tab

Abbreviations: OR= Odds Ratio; SE= Standard Error of Odd’s Ratio; CI = Confidence Interval

Data Availability Statement:

The data that support the findings of this study are available from the corresponding author, RGSF, upon reasonable request.

References

  1. Ademuyiwa FO, Cyr A, Ivanovich J, & Thomas MA (2016). Managing breast cancer in younger women: challenges and solutions. Breast Cancer: Targets and Therapy, 8, 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Antoni MH, Lechner SC, Kazi A, Wimberly SR, Sifre T, Urcuyo KR, . . . Carver CS (2006). How stress management improves quality of life after treatment for breast cancer. Journal of consulting and clinical psychology, 74(6), 1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Antoni MH, Wimberly SR, Lechner SC, Kazi A, Sifre T, Urcuyo KR, . . . Guellati S (2006). Reduction of cancer-specific thought intrusions and anxiety symptoms with a stress management intervention among women undergoing treatment for breast cancer. American Journal of Psychiatry, 163(10), 1791–1797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Avis NE, Levine BJ, Case LD, Naftalis EZ, & Van Zee KJ (2015). Trajectories of depressive symptoms following breast cancer diagnosis. Cancer Epidemiology and Prevention Biomarkers, 24(11), 1789–1795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bakk Z, Tekle FB, & Vermunt JK (2013). Estimating the association between latent class membership and external variables using bias-adjusted three-step approaches. Sociological methodology, 43(1), 272–311. [Google Scholar]
  6. Berger AM, Farr LA, Kuhn BR, Fischer P, & Agrawal S (2007). Values of sleep/wake, activity/rest, circadian rhythms, and fatigue prior to adjuvant breast cancer chemotherapy. Journal of pain and symptom management, 33(4), 398–409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Berger AM, Kumar G, LeVan TD, & Meza JL (2020). Symptom Clusters And Quality Of Life Over 1 Year In Breast Cancer Patients Receiving Adjuvant Chemotherapy. Asia-Pacific Journal of Oncology Nursing, 7(2), 134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Buysse DJ, Reynolds CF, Monk TH, Berman SR, & Kupfer DJ (1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry res, 28(2), 193–213. [DOI] [PubMed] [Google Scholar]
  9. Carpenter JS, & Andrykowski MA (1998). Psychometric evaluation of the Pittsburgh sleep quality index. Journal of psychosomatic research, 45(1), 5–13. [DOI] [PubMed] [Google Scholar]
  10. Carver C (2006). Measure of current status. Retrieved March, 12, 2012. [Google Scholar]
  11. Carver CS, Harris SD, Lehman JM, Durel LA, Antoni MH, Spencer SM, & Pozo-Kaderman C (2000). How important is the perception of personal control? Studies of early stage breast cancer patients. Personality and Social Psychology Bulletin, 26(2), 139–149. [Google Scholar]
  12. Chatterjee NA, He Y, & Keating NL (2013). Racial differences in breast cancer stage at diagnosis in the mammography era. American journal of public health, 103(1), 170–176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chirico A, D’aiuto G, Penon A, Mallia L, De Laurentiis M, Lucidi F, . . . Giordano A (2017). Self-efficacy for coping with cancer enhances the effect of reiki treatments during the pre-surgery phase of breast cancer patients. Anticancer research, 37(7), 3657–3665. [DOI] [PubMed] [Google Scholar]
  14. Chow S, Wan BA, Pidduck W, Zhang L, DeAngelis C, Chan S, . . . Sousa P (2019). Symptom clusters in patients with breast cancer receiving radiation therapy. European Journal of Oncology Nursing, 42, 14–20. [DOI] [PubMed] [Google Scholar]
  15. Chu Q, Wu IH, & Lu Q (2020). Expressive writing intervention for posttraumatic stress disorder among Chinese American breast cancer survivors: the moderating role of social constraints. Quality of Life Research, 1–9. [DOI] [PubMed] [Google Scholar]
  16. Crane TE, Badger TA, Sikorskii A, Segrin C, Hsu C-H, & Rosenfeld AG (2020). Symptom Profiles of Latina Breast Cancer Survivors: A Latent Class Analysis. Nursing research, 69(4), 264–271. [DOI] [PubMed] [Google Scholar]
  17. Curcio G, Tempesta D, Scarlata S, Marzano C, Moroni F, Rossini PM, . . . De Gennaro L (2013). Validity of the Italian version of the Pittsburgh sleep quality index (PSQI). Neurological Sciences, 34(4), 511–519. [DOI] [PubMed] [Google Scholar]
  18. Dalle Grave R, Calugi S, Centis E, El Ghoch M, & Marchesini G (2010). Cognitive-behavioral strategies to increase the adherence to exercise in the management of obesity. Journal of obesity, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. DeSantis C, Ma J, Bryan L, & Jemal A (2014). Breast cancer statistics, 2013. CA: A Cancer Journal for Clinicians, 64(1), 52–62. [DOI] [PubMed] [Google Scholar]
  20. DiMartino LD, Birken SA, & Mayer DK (2017). The relationship between cancer survivors’ socioeconomic status and reports of follow-up care discussions with providers. Journal of Cancer Education, 32(4), 749–755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Dodd MJ, Miaskowski C, & Paul SM (2001). Symptom clusters and their effect on the functional status of patients with cancer. Paper presented at the Oncology nursing forum. [PubMed] [Google Scholar]
  22. Donovan KA, Jacobsen PB, Small BJ, Munster PN, & Andrykowski MA (2008). Identifying clinically meaningful fatigue with the Fatigue Symptom Inventory. Journal of pain and symptom management, 36(5), 480–487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. French-Rosas LN, Moye J, & Naik AD (2011). Improving the recognition and treatment of cancer-related posttraumatic stress disorder. Journal of psychiatric practice, 17(4), 270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Given CW, & Given BA (2013). Symptom management and psychosocial outcomes following cancer. Paper presented at the Seminars in oncology. [DOI] [PubMed] [Google Scholar]
  25. Hack TF, Pickles T, Bultz BD, Ruether JD, Weir LM, Degner LF, & Mackey JR (2003). Impact of providing audiotapes of primary adjuvant treatment consultations to women with breast cancer: a multisite, randomized, controlled trial. Journal of clinical oncology, 21(22), 4138–4144. [DOI] [PubMed] [Google Scholar]
  26. Hamilton M (1959). The assessment of anxiety states by rating. British journal of medical psychology, 32(1), 50–55. [DOI] [PubMed] [Google Scholar]
  27. Hamilton M (1960). A rating scale for depression. Journal of neurology, neurosurgery, and psychiatry, 23(1), 56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Hann D, Jacobsen P, Azzarello L, Martin S, Curran S, Fields K, . . . Lyman G (1998). Measurement of fatigue in cancer patients: development and validation of the Fatigue Symptom Inventory. Quality of Life research, 7(4), 301–310. [DOI] [PubMed] [Google Scholar]
  29. Harrington CB, Hansen JA, Moskowitz M, Todd BL, & Feuerstein M (2010). It’s not over when it’s over: long-term symptoms in cancer survivors—a systematic review. The International Journal of Psychiatry in Medicine, 40(2), 163–181. [DOI] [PubMed] [Google Scholar]
  30. Heitzmann CA, Merluzzi TV, Jean‐Pierre P, Roscoe JA, Kirsh KL, & Passik SD (2011). Assessing self‐efficacy for coping with cancer: development and psychometric analysis of the brief version of the Cancer Behavior Inventory (CBI‐B). Psycho‐Oncology, 20(3), 302–312. [DOI] [PubMed] [Google Scholar]
  31. Hulbert‐Williams N, Neal R, Morrison V, Hood K, & Wilkinson C (2012). Anxiety, depression and quality of life after cancer diagnosis: what psychosocial variables best predict how patients adjust? Psycho‐Oncology, 21(8), 857–867. [DOI] [PubMed] [Google Scholar]
  32. Jacobsen PB, Widows MR, Hann DM, Andrykowski MA, Kronish LE, & Fields KK (1998). Posttraumatic stress disorder symptoms after bone marrow transplantation for breast cancer. Psychosomatic Medicine, 60(3), 366–371. [DOI] [PubMed] [Google Scholar]
  33. Kehlet H, Jensen TS, & Woolf CJ (2006). Persistent postsurgical pain: risk factors and prevention. The lancet, 367(9522), 1618–1625. [DOI] [PubMed] [Google Scholar]
  34. Khan F, Ahmad N, & Biswas FN (2018). Cluster analysis of symptoms of Bangladeshi women with breast cancer. Indian journal of palliative care, 24(4), 397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Kulpa M, Kosowicz M, Stypuła-Ciuba BJ, & Kazalska D (2014). Anxiety and depression, cognitive coping strategies, and health locus of control in patients with digestive system cancer. Przeglad gastroenterologiczny, 9(6), 329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lanza ST, Tan X, & Bray BC (2013). Latent class analysis with distal outcomes: A flexible model-based approach. Structural equation modeling: a multidisciplinary journal, 20(1), 1–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lee L, Ross A, Griffith K, Jensen R, & Wallen G (2020). Symptom Clusters in Breast Cancer Survivors: A Latent Class Profile Analysis. Paper presented at the Oncology Nursing Forum. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Loh SY, & Quek K (2011). Cancer-behavior-coping in women with breast cancer: Effect of a cancer self-management program. Int J Appl Basic Med Res, 1(2), 84–88. doi: 10.4103/2229-516X.91150 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Maass SW, Roorda C, Berendsen AJ, Verhaak PF, & de Bock GH (2015). The prevalence of long-term symptoms of depression and anxiety after breast cancer treatment: a systematic review. Maturitas, 82(1), 100–108. [DOI] [PubMed] [Google Scholar]
  40. Matsuoka Y, Nakano T, Inagaki M, Sugawara Y, Akechi T, Imoto S, . . . Uchitomi Y (2002). Cancer-related intrusive thoughts as an indicator of poor psychological adjustment at 3 or more years after breast surgery: a preliminary study. Breast cancer research and treatment, 76(2), 117–124. [DOI] [PubMed] [Google Scholar]
  41. Miaskowski C (2016). Future directions in symptom cluster research. Paper presented at the Seminars in oncology nursing. [DOI] [PubMed] [Google Scholar]
  42. Miaskowski C, Aouizerat BE, Dodd M, & Cooper B (2007). Conceptual issues in symptom clusters research and their implications for quality-of-life assessment in patients with cancer. Journal of the National Cancer Institute Monographs, 2007(37), 39–46. [DOI] [PubMed] [Google Scholar]
  43. Miaskowski C, Barsevick A, Berger A, Casagrande R, Grady PA, Jacobsen P, . . . Xiao C (2017). Advancing symptom science through symptom cluster research: Expert panel proceedings and recommendations. JNCI: Journal of the National Cancer Institute, 109(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Miaskowski C, Dodd M, & Lee K (2004). Symptom clusters: the new frontier in symptom management research. JNCI Monographs, 2004(32), 17–21. [DOI] [PubMed] [Google Scholar]
  45. Miaskowski C, Dunn L, Ritchie C, Paul SM, Cooper B, Aouizerat BE, . . . Yates P (2015). Latent class analysis reveals distinct subgroups of patients based on symptom occurrence and demographic and clinical characteristics. Journal of pain and symptom management, 50(1), 28–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Miller KD, Nogueira L, Mariotto AB, Rowland JH, Yabroff KR, Alfano CM, . . . Siegel RL (2019). Cancer treatment and survivorship statistics, 2019. CA: a cancer journal for clinicians, 69(5), 363–385. doi: 10.3322/caac.21565 [DOI] [PubMed] [Google Scholar]
  47. Motlagh H (2010). Impact of event scale-revised. J Physiother, 56(3), 203. [DOI] [PubMed] [Google Scholar]
  48. Naik H, Qiu X, Brown M, Eng L, Pringle D, Mahler M, . . . Ho, V. (2016). Socioeconomic status and lifestyle behaviours in cancer survivors: smoking and physical activity. Current Oncology, 23(6), e546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Perkins EA, Small BJ, Balducci L, Extermann M, Robb C, & Haley WE (2007). Individual differences in well-being in older breast cancer survivors. Critical reviews in oncology/hematology, 62(1), 74–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Phillips SM, & McAuley E (2014). Physical activity and quality of life in breast cancer survivors: the role of self‐efficacy and health status. Psycho‐Oncology, 23(1), 27–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Pudkasam S, Tangalakis K, Chinlumprasert N, Apostolopoulos V, & Stojanovska L (2017). Breast cancer and exercise: The role of adiposity and immune markers. Maturitas, 105, 16–22. doi: 10.1016/j.maturitas.2017.04.022 [DOI] [PubMed] [Google Scholar]
  52. Radecka B, & Litwiniuk M (2016). Breast cancer in young women. Ginekologia polska, 87(9), 659–663. [DOI] [PubMed] [Google Scholar]
  53. Roiland RA, & Heidrich SM (2011). Symptom clusters and quality of life in older breast cancer survivors. Paper presented at the Oncology nursing forum. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Rottmann N, Dalton SO, Christensen J, Frederiksen K, & Johansen C (2010). Self-efficacy, adjustment style and well-being in breast cancer patients: a longitudinal study. Quality of Life research, 19(6), 827–836. [DOI] [PubMed] [Google Scholar]
  55. Ruiz-Casado A, Álvarez-Bustos A, de Pedro CG, Méndez-Otero M, & Romero-Elías M (2020). Cancer-related fatigue in breast cancer survivors: A review. Clinical Breast Cancer. [DOI] [PubMed] [Google Scholar]
  56. Schreier AM, Johnson LA, Vohra NA, Muzaffar M, & Kyle B (2019). Post-treatment symptoms of pain, anxiety, sleep disturbance, and fatigue in breast cancer survivors. Pain Management Nursing, 20(2), 146–151. [DOI] [PubMed] [Google Scholar]
  57. Schwab JJ, Bialow MR, Clemmons RS, & Holzer CE (1967). Hamilton rating scale for depression with medical in-patients. The British Journal of Psychiatry, 113(494), 83–88. [DOI] [PubMed] [Google Scholar]
  58. Schwarz G (1978). Estimating the dimension of a model. The annals of statistics, 6(2), 461–464. [Google Scholar]
  59. Sharif SP, & Khanekharab J (2017). External locus of control and quality of life among Malaysian breast cancer patients: The mediating role of coping strategies. Journal of psychosocial oncology, 35(6), 706–725. [DOI] [PubMed] [Google Scholar]
  60. Sharp R (2015). The Hamilton rating scale for depression. Occupational Medicine, 65(4), 340-340. [DOI] [PubMed] [Google Scholar]
  61. Shelby RA, Edmond SN, Wren AA, Keefe FJ, Peppercorn JM, Marcom PK, . . . Kimmick GG (2014). Self-efficacy for coping with symptoms moderates the relationship between physical symptoms and well-being in breast cancer survivors taking adjuvant endocrine therapy. Supportive care in cancer, 22(10), 2851–2859. [DOI] [PubMed] [Google Scholar]
  62. Skerman HM, Yates PM, & Battistutta D (2012). Cancer-related symptom clusters for symptom management in outpatients after commencing adjuvant chemotherapy, at 6 months, and 12 months. Supportive care in cancer, 20(1), 95–105. [DOI] [PubMed] [Google Scholar]
  63. Sohl SJ, Schnur JB, Sucala M, David D, Winkel G, & Montgomery GH (2012). Distress and emotional well-being in breast cancer patients prior to radiotherapy: An expectancy-based model. Psychology & health, 27(3), 347–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Taylor SE, Lichtman RR, & Wood JV (1984). Attributions, beliefs about control, and adjustment to breast cancer. Journal of personality and social psychology, 46(3), 489. [DOI] [PubMed] [Google Scholar]
  65. Trudel-Fitzgerald C, Zhou ES, Poole EM, Zhang X, Michels KB, Eliassen AH, . . . Schernhammer ES (2017). Sleep and survival among women with breast cancer: 30 years of follow-up within the Nurses’ Health Study. British journal of cancer, 116(9), 1239–1246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Urcuyo KR, Boyers AE, Carver CS, & Antoni MH (2005). Finding benefit in breast cancer: Relations with personality, coping, and concurrent well-being. Psychology & Health, 20(2), 175–192. doi: 10.1080/08870440512331317634 [DOI] [Google Scholar]
  67. Vermunt J, & Magidson J Latent Class Cluster Analysis in (2002) Hagenaars JAand McCutcheon AL. Applied latent class analysis, 89–106. [Google Scholar]
  68. Vermunt JK (2010). Latent class modeling with covariates: Two improved three-step approaches. Political analysis, 18(4), 450–469. [Google Scholar]
  69. Wang AW, Bouchard LC, Gudenkauf LM, Jutagir DR, Fisher HM, Jacobs JM, . . . Antoni MH (2018). Differential psychological effects of cognitive-behavioral stress management among breast cancer patients with high and low initial cancer-specific distress. Journal of psychosomatic research, 113, 52–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. White-Means SI, & Osmani AR (2018). Affordable Care Act and Disparities in Health Services Utilization among Ethnic Minority Breast Cancer Survivors: Evidence from Longitudinal Medical Expenditure Panel Surveys 2008⁻2015. Int J Environ Res Public Health, 15(9). doi: 10.3390/ijerph15091860 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES