Objective To summarise the evidence on determinants of health-related quality of life (HRQL) in Asian patients with breast cancer.
Design Systematic review conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations and registered with PROSPERO (CRD42015032468).
Methods According to the PRISMA guidelines, databases of MEDLINE (PubMed), Embase and PsycINFO were systematically searched using the following terms and synonyms: breast cancer, quality of life and Asia. Articles reporting on HRQL using EORTC-QLQ-C30, EORTC-QLQ-BR23, FACT-G and FACT-B questionnaires in Asian patients with breast cancer were eligible for inclusion. The methodological quality of each article was assessed using the quality assessment scale for cross-sectional studies or the Newcastle-Ottawa Quality Assessment Scale for cohort studies.
Results Fifty-seven articles were selected for this qualitative synthesis, of which 43 (75%) were cross-sectional and 14 (25%) were longitudinal studies. Over 75 different determinants of HRQL were studied with either the EORTC or FACT questionnaires. Patients with comorbidities, treated with chemotherapy, with less social support and with more unmet needs have poorer HRQL. HRQL improves over time. Discordant results in studies were found in the association of age, marital status, household income, type of surgery, radiotherapy and hormone therapy and unmet sexuality needs with poor global health status or overall well-being.
Conclusions In Asia, patients with breast cancer, in particular those with other comorbidities and those treated with chemotherapy, with less social support and with more unmet needs, have poorer HRQL. Appropriate social support and meeting the needs of patients may improve patients’ HRQL.
- breast cancer
- health-related quality of life
- patient-reported outcomes
This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Statistics from Altmetric.com
Strengths and limitations of this study
This systematic review included over 75 determinants of health-related quality of life in Asian patients with breast cancer.
Studies included had varying patient selection criteria, which may be the reason for discordance results in certain determinants.
We were not able to conduct a meta-analysis to provide a sense of the level of association, as the choice of statistical analysis varied across studies.
In Asia, the number of breast cancer survivors is increasing, with 5-year survival rates exceeding 90% in early-stage disease.1–7 This is due to improved breast cancer treatments and early detection.8–11 As such, the number of survivors is increasing rapidly. Patient-reported outcomes on health-related quality of life (HRQL), such as physical and emotional functioning and treatment-related side effects including pain, nausea and fatigue, are increasingly important as it effects many breast cancer survivors.
Impaired HRQL is best represented as gap between an individual’s actual functional level and his or her ideal standard.12 Studies from the West reported reduced physical and emotional functioning in patients with breast cancer shortly after treatment.13–16 Breast-conserving surgery as compared with mastectomy, axillary clearance, radiotherapy and chemotherapy were associated with higher level of pain.17 Furthermore, younger patients with breast cancer reported better physical functioning but more impaired emotional functioning compared with older breast cancer patients.13–16 HRQL improves until up to 6–10 years following breast cancer diagnosis.18 In Asian population, determinants of HRQL are increasingly being studied.
So far, mainly studies from Western developed countries investigated HRQL following breast cancer.14–16 19 20 However, cultural and habitual practices such as the use of traditional medicine may limit the generalisability of results from HRQL studies in Caucasian patients with breast cancer to Asian patients with breast cancer.21 22 Drug tolerance is different across populations; paclitaxel in the Japanese population is less well tolerated than the USA.23 24 Furthermore, Asian patients with breast cancer tend to be younger at diagnosis and have more advanced stages at diagnosis than Caucasians.25 Even within Asian ethnicities, Malay patients with breast cancer were found to respond better to tamoxifen therapy than Chinese or Indian patients.26 Better understanding of risk factors for poorer HRQL in Asian patients with breast cancer would allow for targeted interventions.
As an overview of the literature on HRQL determinants in Asian breast cancer survivors is currently lacking, this review systematically summarises determinants of HRQL in breast cancer survivors from Eastern, South Central and Southeast Asia.
This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses recommendations and was registered with PROSPERO (CRD42015032468).27
Databases of MEDLINE (PubMed), Embase and PsycINFO were systematically searched, using the terms ‘breast cancer’, ‘quality of life’ and ‘Asia’ in the search strategy (table 1). The systematic search was last updated on 12 July 2017.
Studies were included based on the following criteria: (1) the study population was on women diagnosed with breast cancer living in Eastern Asia, South Central Asia or Southeast Asia; (2) the study was on demographics, clinical, treatments or other determinants of HRQL; (3) the study measured quality of life using European Organization for Research and Treatment of Cancer – Quality of Life Questionnaire, Breast cancer module, EORTC-QLQ-C30, (with or without the breast cancer module, EORTC-QLQ-BR23), or Functional Assessment of Cancer Therapy – General (FACT-G) or Functional Assessment of Cancer Therapy – Breast (FACT-B) questionnaires; (4) the outcome was HRQL measured quality of life using EORTC-QLQ-C30 (with or without EORTC-QLQ-BR23), or FACT-G or FACT-B questionnaires; and (5) the study design was either cross-sectional or observational longitudinal studies. Studies published before 2000, in language other than English, systematic reviews, meta-analyses, pilot studies and studies with qualitative analyses, were not included in the current review.
After removal of duplicates, all titles and abstracts of the remained retrieved articles were screened. Full-text articles of potentially relevant papers were assessed for eligibility by two authors independently (PJH and SAMG). Disagreement was resolved through consensus. Data extraction was performed by two authors independently (PJH and SAMG). The following determinants were collected for each study: (1) study characteristics (year and country of publication, study design, sample size, response, median follow-up and period), (2) demographics of the study population (age, ethnicity and time since diagnosis), (3) tumour characteristics (invasive or in situ and stage) and (4) past and current treatment.
Outcome extraction included HRQL, as measured by the global health status of the EORTC-QLQ-C30 and overall well-being subscales of FACT-G or FACT-B. The EORTC-QLQ-C3028–31 and FACT-G and FACT-B32–34 are validated in different populations in different languages. Other domains of the EORTC-QLQ-C30, physical functioning, role functioning, emotional functioning, cognitive functioning, social functioning, fatigue, pain, dyspnoea, insomnia, constipation, diarrhoea and financial difficulty were extracted where available. The EORTC-QLQ-BR23, an additional breast cancer module, assesses areas that are specific to patients with breast cancer: body image, sexual functioning, sexual enjoyment, future perspectives, systemic therapy side effects, breast symptoms and arm symptoms. Similarly, determinants of other domains of FACT-G, physical well-being, social well-being, emotional well-being and functional well-being were extracted. The FACT-B, an extended version of the FACT-G, has an additional breast cancer subscale.
Critical appraisal was performed using the quality assessment scale for cross-sectional studies,35 and an adapted version of Newcastle-Ottawa Quality Assessment Scale for cohort studies.36 The maximum score attainable was 8 for each cross-sectional study and 6 for each longitudinal study. Four items on sample selection, one on comparability and three on outcome measurement, were assessed for cross-sectional studies (online supplementary table 1). Two items on sample selection, one on comparability (score of 0–2) and two on outcome measurement, were assessed for cohort studies (online supplementary table 2). Meeting all criteria in the category would confer a high score in the category. Except for the comparability criterion of cross-sectional study, studies that meet <50% of the criteria would be considered as having a low score.
Supplementary file 1
Patient and public involvement
Patients and public were not involved in the development of the research question, choice of outcome measures or the design and conduct of this systematic review.
The systematic search yielded a total of 3160 records including 2549 unique articles that were screened for title and abstract using the predefined inclusion and exclusion criteria (figure 1). After screening the full text of 182 articles, 126 articles did not meet our inclusion and exclusion criteria (figure 1). Cross-referencing identified one additional article. In total, 57 articles were included in the systematic review (43 cross-sectional studies and 14 longitudinal studies), including 24 538 women diagnosed with breast cancer from the following seven countries: Korea (n=17), China (n=14), India (n=8), Taiwan (n=6), Malaysia (n=6), Japan (n=5) and Thailand (n=1) (table 2).
Of the 43 studies with a cross-sectional design, none received the maximum score of the quality assessment (table 2). There were 22 articles with a low score for selection (score of 0–2) due to the use of convenience sampling and small (<300) sample size (online supplementary table 1). All cross-sectional studies described their study population, conferring a high score for comparability (figure 2). Reporting of outcome was an issue in cross-sectional studies: 20 studies did not report confidence intervals or standard errors and 27 had <70% response rate (online supplementary table 1). Nine of 14 longitudinal studies were of good quality having scores of 5–6 (max=6) (table 2). The remaining five studies of poorer quality with scores of 3 or 4, four did not have a representative sample of their target population,37–40 four had a follow-up of <70% but did not provide description of lost to follow-up and none controlled for additional determinants37–41 (online supplementary table 2).
Most determinants studied were consistent in the direction of association or were not associated with global health status and/or general well-being (table 3). In studies on global health status, marital status, household income, type of surgery, chemotherapy, radiotherapy and hormone therapy, conflicting results were found. Studies on general well-being, looking at time since diagnosis, age and unmet sexuality needs measured by short-form Supportive Care Needs Survey (SCNS) also reported conflicting results. Table 4 presents a summary of determinants which were found to be associated with global health status and/or overall well-being.
Park et al found that patients with breast cancer who were of older age had poorer overall well-being and that older age was associated with longer time since surgery.42 In patients who were at least 5-year postdiagnosis, older age was associated with poorer overall well-being in those.43 44 In patients undergoing chemotherapy or radiotherapy, So et al observed that older age was associated with better overall well-being than those aged below 60 years.45 Apart from the study by So et al,45 other studies21 46–48 on this association showed that older age was associated with poorer global health status.
Chui et al 21 and Edib et al 48 found that women who were single (as compared with ever married) and unmarried (as compared with currently married and widowed/divorced), respectively, had better global health status. However, Chang et al found that being married as compared with being single/divorced/widowed was associated with better global health status.49 The classification of widowed/divorced, which confers poorer HRQL than married, may have contributed to the difference in findings of Chui et al 21 and Chang et al,49 in addition the proportion of women who were never married (single) is small in both populations (11% unmarried and 17% unmarried/divorced/widowed, respectively).
Edib et al 48 and Huang et al 50 found that higher household income was associated with better global health status, while Chui et al 21 found the opposite. While some reported higher household income to be also associated with better overall well-being, others did not find evidence of associations.44 51 Standard of living for the population is different among the different studies, making it difficult to access if the association seen was a result of the choice of categorisation of household income. Among the six studies21 43 48–50 52 that assessed household income, Chui et al were the only ones who looked at the effect of household income during treatment, in particular during chemotherapy, and found that higher income was associated with poorer global health status.21 Lower income might have been less of a concern in Malaysia, where lower income patients have access to welfare assistance, while patients of higher income are not eligible for. In addition, Edib et al studied survivors in the post-treatment period in Malaysia and found that higher household income was associated with better global health status.48
Other demographic determinants
Shorter time since breast cancer diagnosis,39 41 46–48 50 being of Chinese or Indian ethnicity as compared with Malay ethnicity,21 48 53 lower educational level21 49 and being diagnosed at later calendar year54 were associated with poorer global health status. Shorter time since diagnosis of breast cancer43 55 56 and lower educational level43 57 were associated with poorer overall well-being.
Type of surgery
Edib et al observed that women who underwent breast-conserving surgery had better global health status than women who had mastectomy.48 However, Dubashi et al 59 and Huang et al 60 found that patients who had breast-conserving surgery had poorer global health status than those who had mastectomy. This could be due to the higher levels of, pain, breast symptoms and arm symptoms experienced by patients who had breast-conserving surgery as compared with those who had mastectomy.59 60 Furthermore, other studies comparing breast-conserving surgery and mastectomy did not find associations with global health status46 47 61 62 or overall well-being.43 44 55 57 63 64
Kao et al 46 and Shi et al 47 found that at 2 years postdiagnosis, women who have had radiotherapy had better global health status as compared with those who did were not treated with radiotherapy; however, Edib et al 48 found contrary results. After adjusting for potential confounders, the association between radiotherapy with poorer global health status was no longer significant.48 Park et al 58 and Hong-Li et al 55 did not find association between having had radiotherapy and global health status or overall well-being.
Edib et al 48 found hormone therapy was associated with poorer global health status; however, Kao et al 46 and Shi et al 47 found the opposite. Kao et al 46 and Shi et al 47 obtained information on hormone therapy from medical records. Using the classification of ever or current user of hormone therapy may result in misclassifying those who had discontinued with those on active therapy. Furthermore, patients who suffer adverse events, like hot flushes, are more likely to discontinue hormone therapy, which may result in patients who are on hormone therapy to be incorrectly perceived as having better global health status.65 66 In other studies, hormone therapy was not associated with global health status58 or overall well-being.44 55 64
Other treatment determinants
Ongoing treatment (vs completed treatment),67 having received chemotherapy46 48 or not having delayed chemotherapy21 39 were associated with poorer global health status. Recent (≤30 days) postsurgery (vs presurgery)68 and having received chemotherapy43 64 were associated with poorer overall well-being.
Complementary and alternative medication
The use of complementary and alternative medication in general, including dietary supplements, prayer, exercise and/or self-help techniques, was not associated with overall well-being.21 22 However, the use of traditional Chinese medication,43 empowerment of patients with breast cancer57 and participating in self-help groups57 were independently was associated with better overall well-being.
Gong et al found that patients who had less healthy behaviour (comparing zero healthy behaviour, 2, or 3 to 1) had lower global health status and overall well-being.69 Patients with breast cancer who did not exercise (vs exercise) or with lower frequency of exercising (vs ≥5 times a week) had lower global health status and overall well-being.69 Furthermore, those who had low vegetable (vs >250 g per day) intake and did not eat fruits daily had lower global health status and overall well-being.69
Having more unmet needs, especially in the physical and daily living, were associated with poorer global health status48 70 and poorer overall well-being.44 51 So et al 51 found that women who had unmet sexuality needs (measured by SCNS) had poorer overall well-being, while Park et al 71 reported the opposite. Park et al found that higher needs was associated with better overall well-being in 52 women who experienced recurrence of breast cancer, citing that patients who have better sexual functioning are more likely to have more sexuality needs.71 Akechi et al 70 found that unmet sexuality need was associated with poorer global health status, while Edib et al 48 did not find such association.
Lack of involvement in decision making,67 72 lower self-efficacy in symptom management,54 poorer perceived overall medical care67 and having higher Charlson comorbidity index or comorbidities, including diabetes, hypertension and arthritis,46 50 73 were associated with poorer global health status. Adopting a give-in coping mode or believing that they are not in control,74–77 lower perceived social support and lower self-efficiency43 52 57 and poorer perceived overall medical care43 were associated with poorer overall well-being.
Differences in quality of life between patients with breast cancer patients and general population
Two studies both conducted in Korea studied differences in global health status between patients with breast cancer and the general population.67 78 Lee et al found that global health status was not different among patients who had completed treatment for recurrent breast cancer as compared with the general population.67 However, role functioning, cognitive functioning and social functioning were lower, and fatigue levels and financial difficulties were higher in patients treated for recurrence as compared with the general population.67 Lee et al compared patients with breast cancer to the general population at two time points, immediately after diagnosis and 1 year after diagnosis and found that the general population had higher global health status at both time points.78
In Asia, patients with breast cancer have poorer HRQL than the general population. Patients with comorbidities, with chemotherapy, lower social support and with more unmet needs have poorer quality of life. However, HRQL improves with time since diagnosis and having healthier behaviour is associated with better HRQL. Within and across the scope of each questionnaire, most associations with poor global health status or overall well-being were concordant. Discordant results in studies were found in the associations of age, marital status, household income, type of surgery, radiotherapy and hormone therapy, and unmet sexuality needs with global health status or overall well-being.
Patients with one or more comorbidities during the time of survey had poorer HRQL. Comorbidity occurs in 20%–30% of patients with breast cancer.79 Comorbidities may be pre-existence or developed after diagnosis; hypertension, arthritis and diabetes are common to patients with breast cancer.14 Studies outside Asia showed similar results; having less co-morbidity was also found to be associated with better HRQL in African-American and Latina breast cancer survivors.80 81 Having pre-existing diabetes was associated with poorer HRQL, in patients with early breast cancer in the USA.82 In addition, patients with pre-existing comorbidities are more likely to have treatment complications, which may lead to poorer HRQL.79
In Asian patients with breast cancer, of all treatments studied, only being on or received chemotherapy was clearly associated with poorer HRQL. This is in agreement with Wöckel et al, who found that patients who received chemotherapy had decreased HRQL, and it was more likely to remain low.83 However, patients on chemotherapy are more likely to be diagnosed with advanced stage disease which was also found to be associated with HRQL. Other treatments, like surgery, are less likely to be associated with advance stage disease, and may be the reason for the null findings. Furthermore, patients with poorer prognosis or who are undergoing chemotherapy are more likely to experience pain, fatigue and potentially other adverse events.84 85
The lack of social support and higher unmet needs were associated with poorer HRQL, in Asian countries. Having a large percentage of unmet needs is not unique to Asia.86 87 Provision of social support should be in-line with the needs of the patient, so as to not adversely impact their HRQL.88 89 In this review, social support, in areas that enable patients to be empowered with higher self-efficacy, was associated with better HRQL. The provision for the educational needs or having access to the service of a breast care nurse may help in reducing unmet needs and provide social support from an institutional effort.89 90
We acknowledge that this systematic review has some limitation. The studies included had varying patient selection criteria, which may be the reason for discordance results in certain determinants. Studies conducted in patients during the treatment period would differ from those conducted after completion of treatment. The choice of statistical analysis varies, with most reporting associations from linear models and some from correlation analysis; thus, we were not able to provide a sense of the level of association. Non-standard methods of measuring determinants were used in some studies, limiting the comparability of the studies. Furthermore, we cannot determine the direction of association from cross-sectional studies; it is possible that some determinant, such as unmet needs and use of CAM, were the result of poorer HRQL. While most of the studies of longitudinal design were of high quality, the majority of the cross-sectional design studies were of moderate or poor quality. Future cross-sectional studies should consider reporting reasons for non-response and include multiple sites if sample size is insufficient.
Patients with breast cancer in Asia have a poorer HRQL than the general population. A shorter time since diagnosis of breast cancer,39 41 43 46–48 50 55 56 having a Chinese or Indian ethnic background as compared with Malay ethnicity,21 48 53 lower educational level21 43 49 57 and advanced stage breast cancer disease46 48 50 52 58 were associated with poorer HRQL. There is some evidence that patients with comorbidities or with chemotherapy are more likely to experience poorer HRQL. The lack of social support and having unmet needs may predict poorer HRQL. Further studies into methods to provide social support in the Asian setting is needed to identify effective ways to improve patients’ HRQL.
Contributors HMV, PJH and SAMG designed the study. PJH and SAMG performed the systematic review. PJH wrote the manuscript. All authors discussed and revised the manuscript.
Funding The study was carried out with the support from the National University Hospital, Singapore, Clinician Scientist Award, National Medical Research Council R-608-000-093-511 and Asian Breast Cancer Research Fund N-176-000-023-091 awarded to MH.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement No dataset was used in this systematic review.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.