Relationship of Lifestyle Habits to Health-Related Quality of Life of Recently Diagnosed Breast Cancer Between Younger and Older Women in China


 Background: Breast cancer is the most common cancer among women in China. Hence, how best to live with and improve the health-related quality of life (HRQoL) of this growing population of women is thus becoming of great public health importance. The aim of the study was to evaluate the relationship of lifestyle habits to HRQoL among younger and older women who were initially diagnosed with breast cancer within the first two weeks, determine the contribution of lifestyle habits factors on HRQoL.Methods: A multi-center, hospital-based, case control study was conducted among breast cancer women from 22 hospitals in 11 provinces or municipalities in northern and eastern China from April 2012 to April 2013. The Functional Assessment of Cancer Therapy-Breast Cancer (FACT-B) simplified Chinese version 4 was used to measure HRQoL. Chi-square test, ANOVA and Multivariable generalized linear models were conducted to verify differences in HRQoL between two groups and to evaluate the contribution of lifestyle habits factors ( including smoking, passive smoking, alcohol intake, Tea, Coffee, Sleep satisfaction, Current life satisfaction, Physical activity, BMI) on HRQoL of breast cancer patients.Results: 1199 eligible breast cancer patients were used for analysis. Younger women (age < 50 years) appeared to show lower scores than older women (age ≥ 50 years) in HRQoL subscales including emotional well being (p = 0.003), functional well being (p = 0.006), breast cancer subscale (p = 0.038), and FACT-B Total scores (p = 0.028). Tea and alcohol consumption, very satisfied with sleep and current life were the strongest predictors of higher HRQoL in younger group. Meanwhile, no coffee consumption, frequent participation in physical activities, high sleep satisfaction and current life satisfaction were the key predictors of higher HRQoL in older breast cancer women.Conclusion: The relationship of the nine lifestyle habit items to HRQoL were different between younger and older women. Associated variable of low HRQoL can help clinicians take intervention early in order to improve the prognosis of breast cancer patients.


Introduction
Breast cancer is the most common cancer among women in China. The crude incidence of breast cancer is 41.82 per 100,000 women with 279,000 new cases diagnosed in 2014 (1). As evidenced by the high incidence rates and relatively low mortality rates, breast cancer is the most prevalent cancer in China (2). It has been predicted that by 2021, there will be 2.2 million cases of breast cancer in China amongst women aged 35-49 years in 2001, which is equivalent to more than 100 new cases per 100,000 women (3). Therefore, efforts to improve the health-related quality of life (HRQoL) of this growing population of women has thus become an issue of great public health importance.
In 1993, the World Health Organization (WHO) de nes quality of life (QOL) as "an individual's perception of his or her status in life in the context of the culture and value systems in which he or she lives, and in relation to his or her goals, expectations, standards and concerns" (4). The QOL re ects the overall physical and mental response and the sense of real self-worth from the heart when the individual suffers from the pain (5). This functional scale for a certain cancer can accurately evaluate the patient's condition and play an irreplaceable role (6,7), and is now considered an important endpoint in cancer clinical trials.
As well-documented in previous studies, assessing HRQoL in cancer patients could contribute to improved treatment and could even be as prognostic as medical factors could be prognostic (8)(9)(10)(11). From a clinical point of view, breast cancer patients perform poorly in terms of psychology, physiology, and sociology. Therefore, breast cancer patients need more support and help in these aspects. It is very important to evaluate and study the HRQoL of breast cancer patients, pay attention to their physical and mental condition, and take active measures and intervention methods to improve their QOL.
At present, many researchers are committed to exploring the in uencing factors affecting the HRQoL of breast cancer patients, and hope to improve the overall health of breast cancer patients by changing these factors. Psychosocial factors, sociodemographic variables, and medical variables have been identi ed as predictors of HRQoL in cancer patients (7,9), but current research on the impact of various lifestyle habits factors on the HRQoL of breast cancer patients is still controversial (12,13). Mosher and Danoff-Burg reviewed studies on age differences in psychological adjustment to breast cancer and suggested age may be best viewed as a risk factor for distress with other variables operating to produce this demographic difference (14). Furthermore, previous studies have shown that younger women have higher rates of psychological morbidity and poorer HRQoL after breast cancer diagnosis than older women (10,15). However, the differences in relationship of lifestyle habits to HRQoL between younger and older women in China has not been studied. More importantly, most studies on QOL after breast cancer have been performed in women at least 4 months to upwards of 5-10 years after cancer diagnosis, and sometimes after the completion of treatment (9,10,12). Few studies have assessed the HRQoL of women initially diagnosed with breast cancer in the rst few weeks (16).
The aim of the study was to evaluate the relationship of lifestyle habits to HRQoL among younger and older women who were initially diagnosed with breast cancer within the rst two weeks, determine the contribution of lifestyle habits factors on the HRQoL. The ndings of this study can potentially help guide the training/initiatives that are organized for shaping lifestyle habits of breast cancer patients after cancer diagnosis, and may in uence the woman's future course of breast cancer.

Recruitment
Breast cancer patients were recruited from 22 hospitals in 11 provinces or municipalities in northern and eastern China from April 2012 to April 2013, as described previously (17,18). Han Chinese women newly diagnosed with primary breast cancer con rmed by histology and aged from 25 to 70 years were included in this study. Exclusion criteria were: (1) < 25 or > 70 years of age; (2) diagnosed with recurrent breast cancer; (3) diagnosed with metastatic breast cancer; (4) patients with other malignant tumors; (5) patients who refused to enroll. A self-designed structured questionnaire was used on the interview, as described previously (17,18). The questionnaire mainly includes the following contents: demographic characteristics, female physiological and reproductive factors, medical and family history, lifestyle habits, and breast cancer-related knowledge. In this study, we only analyzed the lifestyle habits: including smoking (including passive smoking), alcohol consumption, dietary habits (tea and coffee consumption), sleep satisfaction, current life satisfaction, physical activity, and body-mass index (BMI, kg/m 2 ).

QOL assessment
The Functional Assessment of Cancer Therapy-Breast Cancer (FACT-B) simpli ed Chinese version 4 instrument is administered during the baseline interview to assess QOL. The FACT-B is a 36-item questionnaire that includes both 27 items of general QOL (FACT-G) associated with cancer and another 9 items of QOL related to breast cancer, the breast cancer subscale (BCS). FACT-B consists of the following subscales: physical well-being (PWB) (seven items), social/family well-being (SWB) (seven items), emotional well-being (EWB) (six items), functional well-being (FWB) (seven items), and BCS. The simpli ed Chinese translation was performed using a standardized methodology using a series of forward and backward translations as well as review and eld testing. The FACT-B uses a ve-point scale (0 = not at all; 1 = a little bit; 2 = somewhat; 3 = quite a bit; 4 = very much) to indicate how true the statements were to the subjects over the previous 7 days. If more than half of the items that make up the subscale were answered, the missing values were calculated as an average of the observed items. Depending on the scale, higher scores may represent either a higher level of well-being or a lower level of well-being. The items that are expressed in the opposite direction were transformed before being summed up to calculate each subscale's score. Higher scores represent higher levels of well-being. The Cronbach α in this study was 0.881 for the total FACT-B, 0.821 for PWB, 0.800 for SWB, 0.757 for EWB, 0.876 for FWB, and 0.653 for BCS.
Before the start of the study, the investigators were all trained and assessed. The baseline and FACT-B questionnaires were collected in a uni ed standard and uni ed manner to reduce information bias. Faceto-face interviews were conducted to collect the basic information and QOL information from the patients. And the baseline and FACT-B questionnaires were completed within 2 weeks after the diagnosis of breast cancer.

Statistical analysis
Data entry, process and analysis methods are the same as described previously (17,18). Pearson's chisquare tests were used to compare frequency distribution differences between women aged < 50 years and ≥ 50 years. Mean and standard deviations were calculated for all QOL domains. P-values were calculated using one-way analysis of variance (ANOVA). Multivariable generalized Linear Models were used to estimate for the total FACT-B score to determine the characteristics that were most strongly associated with these QOL scores. A two-sided P-value < 0.05 was considered to be statistically signi cant.

Results
The study initially recruited 1489 eligible breast cancer patients as described previously, and 1199 cases were used for analysis as appropriate. Of these women, the mean age was 47.66 years. Patients were divided into two groups in this study: women aged < 50 years and women aged ≥ 50 years. Women aged < 50 years constituted 62.0 percent of the entire dataset. Figure 1 presents the ow of patients in the study. The basic characteristics of the two groups are shown in Table 1. There were statistically signi cant differences between the two groups in education levels (χ2 = 43.845, P < 0.001), family average revenue (χ2 = 11.962, P = 0.018), and postmenopausal status (χ2 = 585.054, P < 0.001). No signi cant differences were found for location, economic status, social status, marriage, and family history of breast cancer between different age groups. In this study, no signi cant lifestyle habits differences existed based on age categories, except for BMI distribution. Women age 50 years reported signi cantly lower BMI (χ2 = 19.080, P 0.001).  As shown in Table 3, in order to more speci cally examine which items in lifestyle habits were correlated with HRQoL among younger and older women diagnosed with breast cancer within two weeks, we compared the mean scores of the FACT-B Total between two age groups. The relationships of the 9 lifestyle habit items to HRQoL were different between younger and older women. Firstly, very satis ed with sleep and current life at diagnosis were associated with higher scores in all the women with breast cancer. Secondly, tea (p = 0.009) and alcohol drinking (p = 0.001) women showed a signi cantly higher score in younger age group, while drinking coffee (p = 0.009) showed a signi cantly lower score in older age group. Thirdly, compared to the younger age group, frequent participation in physical activities in older age group was associated with higher HRQoL overall (p = 0.002). And smoking showed worse HRQoL in the older age group (p = 0.045). No other signi cant items of lifestyle habits were observed in association with FACT-B Total scores. The relationship of lifestyle habits to HRQoL was further analyzed using GLM, as shown in Tables 4 and  5. Tea and alcohol drinking, being very satis ed with sleep and with current life were again the strongest predictors of higher HRQoL in younger age group. Meanwhile, no coffee consumption, frequent participation in physical activities and very satis ed with sleep and current life were the key predictors of higher HRQoL in older age group.  Center, Chicago, United States, which is widely used to assess the HRQoL of breast cancer patients (19). It has been translated into many languages, such as (simpli ed) Chinese, Malayalam, and Korean (20).
Previous studies have demonstrated that Chinese versions of the FACT-B (version 4) are effective, sensitive, and reliable in evaluating the HRQoL of breast cancer patients in China (20)(21)(22), which was also con rmed in this study. Using this internationally consistent and effective scale to assess the quality of life of Chinese patients is important to improve our understanding of the prognosis of breast cancer.
Most of the previous studies examining the HRQoL of women with breast cancer have been conducted in women at least four months after a breast cancer diagnosis (9,10,12). By this time, women have begun their initial treatment process, and have had some time to adjust to their condition. There are few studies that have examined HRQoL within a few weeks after women are diagnosed (16). As previous studies have demonstrated, a cancer diagnosis can have direct impacts on a person's mental health immediately, and subsequent the ability to cope with the diagnosis can vary substantially (10,23). The early psychosocial adaptation to a diagnosis of breast cancer may affect important survivorship issues, such as receiving and adhering to treatment (24,25), coping mechanisms (25), and long-term prognosis (26)(27)(28). Hence, in this study, we set the time point as within two weeks of breast cancer diagnosis to analyze the early factors affecting HRQoL of breast cancer patients in China.
Moreover, the age group with the highest incidence of breast cancer in women is 45-59 years old, and some independent studies have reported that the peak age of breast cancer was between 45-55 years in China (3,17). In this study, the age was divided into < 50 years and ≥ 50 years as an approximate indicator of menopausal status, this cutoff point is used in epidemiologic literature and large breast cancer HRQoL studies as well as clinical practice. The results of this study suggested that older breast cancer patients showed better HRQoL than younger women in most of the HRQoL domains except SWB and PWB, which was supported in previous studies (29,30). Compared to older women, the younger women are more susceptible to suffer from psychosocial in uences (10,29), may receive more aggressive treatment than older groups, are more likely to receive chemotherapy, while older patients have more resources or skills to deal with breast cancer and maintain economic stability.
This study complements the very limited number of research studies that access the impact of lifestyle habits (i.e., smoking (including passive smoking), alcohol intake, dietary habits (tea and coffee), sleep satisfaction, current life satisfaction, physical activity, and BMI) on HRQoL in Chinese women who were diagnosed with breast cancer within two weeks. The results suggest that the breast cancer patients who adopted different lifestyle behaviors had different HRQoL between ages, as shown in Table 3 and ESM_1-5. In this study, cigarette smoking in younger women showed a better SWB, while associated with worse EWB in older age group, and alcohol drinking was signi cantly related to better SWB, EWB and BCS in younger age group. Previous studies have demonstrated that Women who drunk more alcohol daily reported fewer disturbing vasomotor symptoms (31), which are considered to be the most speci c symptom of menopause (32), while women who smoked cigarettes daily had more symptoms of depression than non-smokers, except for the menstrual symptoms domain (31,33). These observations are in agreement with the results of this study which show worse EWB in older smokers and high HRQoL (including SWB, EWB and BCS) in younger alcohol consumers.
Clinically signi cant premenstrual syndrome (PMS) affects 15-20% of premenopausal women and signi cantly reduces quality of life (34). Rossignol et al. (35,36), along with three other similar studies (37)(38)(39) found a strong positive correlation between caffeine and coffee intake and premenstrual syndrome. Women with severe premenstrual symptoms appear to be able to alter caffeine intakeincreasing caffeine intake to treat symptoms such as fatigue. These observations support our result that younger women with tea and coffee intake have higher HRQoL (87.22 and 88.81, respectively). In contrast, tea in China has thousands of years of cultural heritage, the Chinese older women prefer to drink traditional tea but refuse coffee, which causes no coffee to have a higher HRQoL. However, this explanation should be provided with discretion and needs more study. But it's worth noting that tea intake was associated with higher HRQoL in both younger and older women. Some studies have shown that tea or its constituents (40), particularly, epigallocatechin-3-gallate as the most abundant and biologically active tea catechins(41), suppress mammary tumorigenesis via effects on antioxidant activity(42), sex hormones (43), or different molecular pathways (44), which may also have a potential impact on quality of life.
It is important to note that frequency of physical activity was positively associated with higher HRQoL of breast cancer patients in older group, while there was no statistical difference in the younger group.
Angenete et al found that the preoperative physical activity is positively associated with an enhanced physical recovery after breast cancer surgery (45). More importantly, evidence from observational studies shows a statistically signi cant positive correlation between inactivity and sedentary behavior and breast cancer risk and poorer health outcomes (46,47). In breast cancer patients, higher levels of physical activity have been shown to be associated with fewer adverse treatment related side effects, higher HRQoL and improved disease-speci c prognoses including longer survival and reduced risk of recurrence and mortality (13,48,49 Finally, the inherent limitations of this study should not be neglected. The most important limitations of this study include that the study was descriptive and cross-sectional, and some factors were collected retrospectively, which may have in uenced our results; in our study, the exact stage of the disease has not been determined, which may be an important factor affecting the HRQoL of patients. Notwithstanding its limitations, the results of this large population-based study may help guide interventions to improve quality of life. It is believed that this study can provide reference and basis for future research.

Conclusion
In summary, we found that younger patients < 50 years showed signi cantly lower QOL than older patients ≥ 50 years. Tea and alcohol consumption, very satis ed with sleep and current life were the strongest predictors of higher HRQoL in Chinese women when diagnosed with breast cancer at younger age. Meanwhile, no coffee consumption, frequent participation in physical activities and high sleep satisfaction and current life satisfaction were the key predictors of higher HRQoL in older breast cancer women. Associated variable of quality of life can help clinicians identify patients at risk for low quality of life. When these characteristics or situations can be balanced, changing them through intervention can improve a patient's quality of life, and as women gradually receive treatment and then enter into their long-term survivorship period, their effects may change subsequent adjustments and functions regarding breast cancer, consequently improve the prognosis of breast cancer patients.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. ElectronicSupplementaryMaterial.docx