Problems in living among breast cancer survivors

Purpose: Breast cancer survivors may experience worse social, physical, and emotional function compared to the general population, although symptoms often improve over time. Data on problems in living can help to improve interventions and supportive care for breast cancer survivors. Symptoms such as fatigue, pain, difficulties with sleep, and sexual problems may have an adverse effect on the quality of life of breast cancer survivors. Methods: We examined problems in living using data from a survey of 164 breast cancer survivors who had completed primary therapy for the disease. Results: A total of 164 women completed the study questions (response rate 16.4%). The mean age of the women was 67 years. Among all participants, 66.7% were white, 29.5% were African-American, and the remainder were of other races. Almost all of the symptoms were more likely to be reported by participants who were < 55 years of age. Other important correlates of symptoms included non-white race, marital status, and having a household income of less than $50,000 per year. Conclusion: The results of this study highlight the need for caregivers to emphasize screening for and discussion of symptoms, including sleep difficulties, fatigue, loss of strength, aches and pains, and muscle or joint stiffness. Of particular concern are younger survivors and those who are African American or low-income.


Introduction
In the United States, an estimated 3.8 million women have a history of invasive breast cancer [1]. Improvements in survival rates due to earlier detection and improvements in therapy have led to new challenges related to the quality of survivorship [2]. Previous studies indicate that breast cancer survivors may experience worse social, physical, and emotional function compared to the general population, although symptoms often improve over time [3][4][5]. Commonly reported symptoms include sleep difficulties, fatigue, pain, cognitive problems, sexual issues, and hot flashes [6]. Such symptoms may have an adverse effect on the quality of life of breast cancer survivors [2]. Frequently, they are due to the therapy received. For example, arthralgias are a frequent side-effect of aromatase inhibitors, which are frequently used for the treatment of postmenopausal women with hormone-receptorpositive breast cancer [6]. Patients who underwent mastectomy or axillary dissection may experience persistent pain at the surgical site or lymphedema in the affected extremity [2]. Data on problems in living can help to raise awareness among patients and physicians, to improve detection, and to design interventions and supportive care for breast cancer survivors.
We thus examined problems in living using data from a survey of 164 breast cancer survivors who had completed primary therapy for their breast cancer. The overall objective was to determine the prevalence and correlates of symptoms and other problems in living.

Methods
The Cardiovascular Disease outcomes among Breast Cancer Survivors Study (CVDBCS) was a postal survey of a multiethnic cohort of breast cancer survivors who reside in Augusta, GA and who had been treated at Augusta University Health and/or the Georgia Cancer Center. Non-institutionalized women were eligible to take part in the study if they resided in Augusta-Richmond County and Columbia County, GA, or in Aiken County, SC, had been diagnosed with stage I-IV breast cancer, and had completed primary therapy for the disease other than hormonals.
Data were collected using postal survey questionnaires and via abstraction of electronic medical records. The mailings were sent to 1,000 potential research participants who were randomly sampled. A sequential mailing protocol was followed using a modified Dillman method [7]. An advance letter was mailed to the women by the study principal investigator (S.S.C.). The letter provided information about the study (purpose, potential benefits, and risks) and clearly informed patients that they could opt out and not receive further mailings about the study. Three weeks later, a survey consent letter was mailed to women who had not opted out along with a copy of the survey questionnaire and a pre-addressed, stamped return envelope. Women who had not opted out but had not returned a completed questionnaire after 4 weeks of the initial mailing were sent a reminder postcard. Survey responses were checked for completeness and then coded and entered into an electronic database. Questions about demographic factors and breast cancer diagnosis were obtained from a previous study of breast cancer survivors. Respondents were asked about their physical, psychological, and social functioning using the Cancer Problems in Living Scale (CPILS) [8]. The version of the CPILS used in this study was expanded from 31 items to include 50 CPILS items [9]. All of the CPILS items use a three-point Likert-like response scale, with response options being: 0 = not a problem, 1 = somewhat of a problem, and 2 = a severe problem.
After crosstabulations and exploratory analyses of the survey data were completed, logistic regression methods were used to compare groups of breast cancer survivors who did or did not report individual symptoms according to age, race, marital status, and household income. The dependent variable in these analyses was whether or not the respondent indicated that the symptom was "somewhat of a problem" or "a severe problem." Ninety-five percent confidence intervals (CIs) were obtained for adjusted odds ratios (ORs). Levels of statistical significance were determined using Wald chi-square tests. The goodness-of-fit of each model was examined using the Log-likelihood ratio test.
Another measure of interest we considered is the total number of problems that participants considered as somewhat of a problem or a severe problem. Across the 40 questions asked, the total number of responses with a value of 1 or 2 were recorded. A multivariate Poisson regression model was then fit to identify the association of this score with various factors such as age, race, household income, marital status, etc.

Results
A total of 164 women completed the study questions (response rate 16.4%). The mean age of the women was 67 years (Table 1). Among all participants, 66.7% were white, 29.5% were African-American, and the remainder were of other races. More than half (58.4%) of the women were insured through Medicare and 29.2% held private insurance. The remainder had Medicaid or were uninsured. With respect to breast cancer stage at diagnosis, 19.8% of the women had ductal carcinoma in situ, 26.8% had stage I disease, 21.0% had stage II disease, 8.9% had stage II disease, and 5.1% had stage IV disease. The mean number of years since diagnosis was 9.4 years. About 54.9% of the women reported receiving chemotherapy and only 4.9% reported biologic/targeted therapy. Problems in living reported by the study participants are shown in Table 2. Symptoms reported by at least 50% of the participants as "somewhat a problem" or "a severe problem" combined included sleep difficulties (53%), fatigue (56%), loss of strength (59%), forgetfulness (57%), aches and pains (71%), and muscle or joint stiffness (67%). Several symptoms were reported as a severe problem by at least 15% of the participants including fatigue (16%), loss of strength (18%), aches and pains (21%), muscle or joint stiffness (17%), dryness in vagina (15%), less sexual desire (23%), and hot flashes (17%).
Almost all of the symptoms were more likely to be reported by participants who were less than 55 years of age (Table 3). Other important correlates of symptoms included non-white race (Table 4), marital status (Table 5), and having a household income of less than $50,000 per year (Table 6). Compared to those who were not married/living with a partner, participants who were married or living with a partner were more likely to report sexual problems, but less likely to report signs of financial distress (Table 5). Lower income participants were more likely to report having major problems with their health and signs of financial distress (Table 6).
For the total score on problems reported as somewhat of a problem/a severe problem, there was a significant relationship between the total score on issues reported as somewhat/severe problem and age (decreasing with age), race (higher for non-whites vs. whites), marital status (higher for married women vs. non-married women), total household income (higher for low income families), stage of cancer (higher for stages II, III and IV versus ductal insitu carcinoma) and treatment type (higher for radiation and biologic/targeted therapy vs. notreatment). Coefficients and p-values for the Poisson regression model are presented in Table  7.

Discussion
The results of this study indicate that a substantial proportion of breast cancer survivors seen at an academic medical center in the southern United States reported having problems in living such as sleep difficulties, fatigue, loss of strength, aches and pains, and muscle or joint stiffness. Important correlates of having reported symptoms included age, race, marital status, and household income. Madelblatt et al. [10] noted that older breast cancer survivors may have a high symptom burden due to comorbidities and aging. However, in the current study, almost all of the symptoms were more likely to be reported by participants who were less than 55 years of age. In a study of problems of living among men and women who were adult cancer survivors, Baker et al. [8] found that more problems were reported by younger survivors (ages 18-54 years), women, non-whites, those who were not married, and those with a household income < $20,000 per year. Younger cancer survivors are more likely to still be employed and to have dependent family members [8].
Many cancer survivors identify fatigue as one of the most frequent and distressing cancerrelated symptom [11,12]. In the current study, 56% of the participants reported having fatigue. In a population-based study by Meeske et al. [11], 41% of breast cancer survivors who were 2 to 5 years following diagnosis were fatigued. These differences in the frequency of fatigue may be due to differences in study design or patient population. A subset of breast cancer survivors experience moderate to severe symptoms years after cancer treatment has ended [11,13,14]. Factors associated with fatigue in breast cancer survivors include pain, sleep problems, physical inactivity, and depression [13][14][15][16]. Persistent fatigue following cancer treatment affects survivors' physical well-being and quality of life [11]. Physical activity has been shown to be an effective non-pharmacologic intervention for fatigue in cancer survivors [11,[17][18][19].
Previous studies have showed that 24% to 84% of breast cancer patients report persistent pain following cancer treatment [2,[20][21][22][23][24]. In the current study, 71% of participants reported aches and pains, which may be partially due to comorbid conditions such as arthritis. About 67% reported muscle or joint stiffness. Report of pain symptoms has been associated with poorer quality of life among breast cancer patients [2]. Results from crosssectional and longitudinal studies suggest that younger age, more invasive surgery, adjuvant therapy, and psychosocial factors have a role in the development of chronic pain [25,26]. In the current study, age < 55 years was associated with tenderness at surgical site [OR = 0.33, 95% CI (0.12, 0.88)].
Caregivers should be aware of the increased frequency of problems in living among breast cancer survivors who are non-white or low-income. In the current study, the majority of the non-white participants were African American. The finding that non-white breast cancer patients were more likely to report certain problems in living highlights the importance of considering the special needs of minority patients in their response to illness [8]. Lowincome survivors are particularly vulnerable to experiencing problems in living that are indicators of financial distress including being less able to provide for the financial needs of their family, having difficulty in meeting their medical expenses, having difficulty in obtaining adequate insurance, and not having money for the cost of medical visits.
In the current study, participants who were married or living with a partner were more likely to report sexual problems such as vaginal dryness, pain during sexual intercourse, and having less sexual desire. In prior studies, women who received adjuvant therapy experienced more severe symptoms, including vaginal problems, musculoskeletal pain, and hot flashes [27,28].
With respect to limitations, misclassification bias is a possibility due to the use of selfreported information. The results of this study may not be generalizable to other populations of breast cancer survivors. However, the sample was diverse by race, socioeconomic factors, and history of breast cancer diagnosis and treatment. A further limitation was the crosssectional design of the study. In addition, selection bias may have occurred due to the low response rate (16.4%). Taken overall, the results of this study, when combined with findings from previous reports [2,10], highlight the need for caregivers to emphasize screening for and discussion of symptoms, including sleep difficulties, fatigue, loss of strength, aches and pains, and muscle or joint stiffness. Of particular concern are younger survivors and those who are African American or low-income. As more attention is given to increasing the quality of life of breast cancer survivors, it is important to identify problems in living, which can help establish which problems should be the focus of possible prevention efforts and supportive care.       Table 7 Coefficients obtained from fitting a Poisson regression model to the total number of problems where the participants responded as having somewhat of a problem/a severe problem