The interventions for resilience of breast cancer patients: a systematic review and meta-analysis

Breast cancer has now overtaken lung cancer as the world’s mostly commonly-diagnosed cancer. With the increasing prevalence of breast cancer, the psychological problems of breast cancer patients have also received more and more attention from scholars. The aim of our study was to synthesis the available research evidence on the effectiveness of interventions designed to promote or enhance the resilience of breast cancer patients. Methods We performed a systematic and comprehensive search of 12 databases from to February and VIP. We conducted a comprehensive qualication screening of the retrieved records and analysed all the included data using Review Manager Version 5.3 and STATA/SE Version 15.1. A of 23 RCTs involving in the meta-analysis. The results showed that the effectiveness of resilience training program (SMD 95% < health via mobile devices (SMD 95% and peer support (SMD p = 0.01) in the experimental groups was better than that of the control groups. Furthermore, there were seven other kind of interventions which might contribute to improve resilience but with very few evidences available.


Introduction
Breast cancer is the leading cause of cancer death in women worldwide [1]. It has now overtaken lung cancer as the world's most commonly-diagnosed cancer, according to statistics released by the International Agency for Research on Cancer (IARC) in December 2020 [2]. In 2018, there were an estimated 2.1 million new cases of breast cancer and 627,000 deaths from breast cancer worldwide [3].
Breast cancer patients faced with high level of pressure in psychological and social respects. Also, both patients and their family undertook heavy economic and psychological burdens due to the painful and long process of disease diagnosis, treatment and rehabilitation. Breast cancer has adverse effects on patients' self-image, behavior, attitude towards life and quality of life [4]. Studies indicated that up to 50% of breast cancer patients suffered from signi cant psychosocial distress [5], and more than one-third of which developed anxiety (33.6%) and/or depression (58.6%) [6]. However, breast cancer still has a high chance of cure if diagnosed early and treated appropriately [2]. With the change of the social medicine model, the psychological problems of breast cancer patients have also received more and more attention from scholars.
Not everyone copes adversities in the negative way, many breast cancer patients may also develop resilience and produce positive psychological changes in the face of illness [7,8]. Resilience, which is also known as psychological resilience, was proposed by American psychologist Anthony in the 1970s, and the American Psychological Society de nes resilience as a process of 'bouncing back' from di cult experiences and 'adapting well in the face of adversity, trauma, tragedy, threats or signi cant sources of stress' [9]. In healthcare area, resilience were de ned as ''the capacity to adapt to challenges and changes at different system levels, to maintain high quality care'' [10]. Resilience is a dynamic developmental process that promotes a successful adaptation to cancer-related adversity [8,11]. Studies showed that psychological distress was the variable that most frequently linked to reductions in resilience, and there is a bidirectional relationship between them [8, 12,13]. A survey study showed that breast cancer patients with poor resilience may have poor body images and suffer more severe adverse effects of systemic therapy, such as arm symptoms and breast symptoms, and hair loss [13]. Resilience is related to many factors such as sociodemographic, clinical, psychosocial and physiological aspects in women with breast cancer [8]. Resilience it not solely a quality within individuals; it also can be enhanced by the external support from others [14]. The evidence showed that the resilience of breast cancer patients is at a moderate or lower level currently [15][16][17]. Thus, resilience-promoting interventions should be obtained to improve breast cancer patients' psychological well-being.
Various psychosocial interventions to enhance resilience were found recently, such as mindfulness meditation, yoga exercises, a healthy diet, peer support, group meetings and group counseling [12]. However, there is no clearly evidence to summarize the speci c effect of intervention measures for resilience of breast cancer patients, and overview of the quantitative studies on the interventions still lacks. In this study, we conducted a systematic review and meta-analysis using data from randomized controlled trials (RCTs). The aim of our study was to synthesis the available research evidence on the effectiveness of interventions designed to promote or enhance the resilience of breast cancer patients.

Methods
This review was adhered to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines [18] (see Additional le 1).

Inclusion and exclusion criteria
We included published studies meeting the following criteria: (1) Participants: Women with breast cancer; (2) Intervention: Interventions which aim to improve psychological resilience; (3) Control: Routine/Basic nursing care, wait-list or no intervention; (4) Outcomes: Psychological resilience score; and (5) Study design: RCTs.
Studies were excluded if they met the following criteria: (1) Studies in which full-text were not available; (3) Studies which reported unextractable or unrelated raw data; (4) Studies were not published in English or Chinese; and (5) Studies where types were reviews/ editorials/ books/ theses/ case reports/ news etc.

Search strategy
We performed a systematic and comprehensive search of 12 databases from inception to February 4,2021: Web of Science, PubMed, Embase, MEDLINE, Scopus, Cochrane Library, CINAHL, PsycINFO, Chinese National Knowledge Infrastructure (CNKI), China Biology Medicine disc (CBM), WanFang and VIP.
Keywords or medical subject headings used in the search were as following: ["resilience" OR " mental resilience" OR " psychological resilience" OR "Mental Elasticity"] AND ["breast cancer" OR " breast carcinoma" OR "mammary cancer" OR "Breast Neoplasm" OR "Breast Tumor"] AND ["randomized controlled trial*" OR " randomized clinical trial*" OR "RCT" OR "randomized"]. In addition, the references in the articles were manually tracked for relevant studies. All articles were managed by EndNote X9 software. Two reviewers searched the databases and screened of the literature searches independently based on the agreed-on criteria. Any disagreements were resolved by discussion or referred to a senior reviewer. Detailed search strategies were listed in Additional le 2.

Data extraction and synthesis
All relevant published data from the studies were extracted and recorded on a standardized data extraction form using Microsoft Excel. The extraction included author, year, origin, sample size, age, stage of cancer, study object, intervention description and outcome measurement. All the data were extracted by two authors independently and checked by a third reviewer.

Quality assessment
Two investigators evaluated the risk of bias of each article independently using the RoB 2.0, which is the latest tool, developed by the Cochrane Handbook for Systematic Reviews of Interventions (Version 6, 2019) [19]. RoB 2.0 includes ve domains and there are a series of "signalling questions" for each domain [19,20]. The authors should reach a judgement after these questions were answered and assign one of three levels ("low risk of bias", "some concerns", "high risk of bias") to each domain. The overall judgement for the articles were mapped from the results within domains. Any disagreements or uncertainty was resolved through discussion or consulted a senior reviewer.

Statistical analysis
The meta-analysis was performed by using the Cochrane Collaboration Review Manager Version 5.3 and STATA/SE Version 15.1 (StataCorp, College Station, TX, USA). ALL the outcome data were presented as the mean ± standard deviation (SD) and were expressed as mean difference (MD), standard mean difference (SMD), weighted mean difference (WMD) and 95% con dence interval (CI) because they were all continuous variables. Tests for heterogeneity were assessed statistically by using the standard chisquared test (χ 2 , or Cochran's Q test) and the I 2 statistic, with p ≤ 0.10 or I 2 > 50% indicating signi cant heterogeneity. A random-effects model was selected to pool the effect of the interventions on resilience of breast cancer patients because of the heterogeneity of interventions used and timing of outcome measurements [21]. Subgroup analysis was performed to identify the sources of heterogeneity and assess the effects of covariates on the pooled estimates [22,23]. Egger's linear regression test was performed to detect the publication bias [24]. For the data of which could not undertake meta-analyses, we synthesized the results and presented the evidence in a narrative form.

Search results
Initially, a total of 337 publications were identi ed from the 12 databases. Then, 101 duplicate records were excluded by automation tools and 24 duplicate records were excluded by a human. 212 records were screened based on title and abstract. 87 studies were sought for retrieval and 82 reports were assessed for eligibility. Following full texts screening, 59 studies were excluded. Finally, 23 studies met the eligibility criteria, of which 16 studies were included in the meta-analysis. We searched studies via other methods, but did not nd any additional studies that met the inclusion criteria. The owchart details the literature selection process using the PRISMA 2020 ow diagram [18] (Fig. 1).

Study characteristics
The characteristics of the included studies are shown in Table 1

Risk of bias assessment
Summary of quality assessments of all included RCT studies is shown in Fig. 2. With regard to the judgement of all the domains and signalling questions, twelve studies (78.3%) were judged to have low risk of bias, four studies (21.7%) were judged to raise some concerns, and none (0%) were judged to have high risk of bias for their outcomes. In general, the quality of the included studies was from moderate to high quality.

Resilience training program
Eight studies [25-28, 38, 43, 45, 46], with a total of 649 breast cancer women, reported data on the resilience training program. There was substantial heterogeneity among the studies (χ 2 = 65.83, p = 0.000, I 2 = 89.4%) according to the heterogeneity test. Therefore, we performed sensitivity analysis and found that Zhou K's study [26] substantially in uence the results. The heterogeneity has decreased signi cantly (χ 2 = 19.25, p = 0.004, I 2 = 69%) after removing the study. Therefore, the random effects model was utilized for the meta-analysis. The results showed that the resilience training program intervention was more effective than routine nursing care intervention in improving the resilience of breast cancer patients (SMD = 0.89; 95% CI 0.54, 1.24; p < 0.00001) (Fig. 3a).

Health education delivered via mobile devices
Five studies [34, 36, 37, 42, 44] provided information on the intervention of health education delivered via a mobile device, and these studies included 495 breast cancer women. There was substantial heterogeneity (χ 2 = 80.74, p < 0.00001, I 2 = 95%) according to the results of the heterogeneity test for these studies. Therefore, we used the random effects model to analyze these data, which con rmed that the intervention of health education delivered via a mobile device was more effective than routine nursing care in improving the resilience of breast cancer patients (SMD = 1.84; 95% CI 0.86, 2.82; p = 0.0002) (Fig.  3b).

Peer support
Three studies [31-33], with a total of 240 breast cancer women, reported data on the peer support. The results showed that the heterogeneity between the two groups was very signi cant (χ 2 = 127.50, p < 0.00001, I 2 = 98%), and we utilized a random effects model for the meta-analysis. The results showed that peer support was more effective than routine nursing care in improving the resilience of breast cancer patients (SMD = 2.62; 95% CI 0.61, 4.63; p = 0.01) (Fig. 3c).

Other kind of interventions
Seven studies [29, 30, 35, 39-41, 47], with a total of 617 breast cancer women, reported seven other kind of interventions which may contribute to improve resilience: 1) trinity nursing intervention, 2) nursing of family affection and psychological counseling, 3) synchronous stage cognitive therapy of family members combined with psychological intervention, 4) supportive psychological nursing, 5) psychological phased change nursing intervention, 6) admission and commitment therapy, and 7) collaboration care model. However, meta-analysis was abandoned and the evidence was presented in a narrative form because there was signi cant heterogeneity within the implemented interventions and the data cannot be classi ed and integrated. Further details about the summary of all the included studies' interventions of all the included studies are shown in Table 2.

Subgroup analysis
Two subgroup analyses were undertaken based on the intervention time of resilience training program. The p-value for interaction was 0.85, which suggests that there was no interaction, in other words, the intervention time was not a source of heterogeneity. The results for the intervention times showed that the difference was not signi cant among the three subgroups: the intervention time was less than 12 weeks (SMD = 1.00; 95%CI 0.26, 1.74), the intervention time was 12 weeks (SMD = 1.19; 95%CI 0.10, 2.28), and the intervention time was more than 12 weeks (SMD = 0.88; 95%CI 0.57, 1.19). Overall, the results of the subgroup analysis showed that the resilience training program was more effective than the routine nursing care intervention, indicating the results of resilience training program had no change (Fig. 4).

Publication bias
We used Egger's linear regression test to detect the publication bias caused by the in uencing factors quantitatively. The results for the resilience training program (95% CI -11.37, 9.41; p = 0.825) and the intervention of health education delivered via a mobile device (95% CI -10.38, 33.46; p = 0.192) showed that there was no signi cant publication bias among the included studies. However, the p-value for peer support (95% CI 3.81, 29.78; p = 0.039) was less than 0.05 and the results reversed signi cantly after the data were analysed by the trim and ll method, which indicated that there was signi cant publication bias.

Discussion
This systematic review and meta-analysis estimated the effect of the interventions for resilience of breast cancer patients. These interventions included resilience training program, health education delivered via mobile devices, peer support, and seven other kind of interventions. The effectiveness of the interventions included in the meta-analysis was examined through the assessment of resilience score. The measurement of the outcome for most studies were CD-RISC-25 and RSA.
Resilience training program is helpful to improving the psychological resilience of breast cancer patients. In recent years, more and more researches suggest that resilience training may play a pivotal role in the realm of public health and prevention [48], and it also has a great number of wider bene ts such as enhanced psychosocial functioning and improved performance [49]. There was considerable variation in the type of resilience training program provided in this study, although most involved Acceptance and Commitment Therapy (ACT), Cognitive Behavioral Therapy (CBT), Mindfulness-Based techniques [9,48].
Our meta-analysis showed that the resilience training program was more effective than the routine nursing care intervention (SMD = 0.89; 95% CI 0.54, 1.24; p < 0.00001). This result is consistent with the study of resilience training programmes and interventions in adults, re ecting a moderate positive effect (SMD = 0.44; 95% CI 0.23, 0.64) in favor of the resilience training program group [9]. It may be because psychological resilience training program helps patients control their emotions, cope with setbacks and adapt to adversity. The in uence of psychological training on future functions depends heavily on the type of measurement results and the setting of the training [50]. Therefore, further researches about the effect of the resilience training should be conducted for a more de nitive conclusion due to the difference in content, delivery and length.
Health education delivered via mobile devices is a new and effective healthcare method [51], and the included studies about it were all conducted in China. WeChat is a free communication application, which is the most widely and extensively used mobile social networking App in China. Mobile health (mHealth) or WeChat-based intervention programs have become a popular health service model in China [26]. Our meta-analysis showed that the intervention of health education delivered via mobile devices was more effective than the routine nursing care intervention (SMD = 1.84; 95% CI 0.86, 2.82; p = 0.0002). This result is consistent with a study of people living with schizophrenia (PLS), which reported that WeChat-based intervention was related to a series of positive health outcomes, including decreased symptoms and depression, as well as improved functioning, recovery, quality of life and mental well-being [52]. It may be because that the online delivery of resilience training has the potential of delivering effective training in a more exible, personalized, and cost-effective way [53]. This new method can hold great promise for mental health care by extending the reach of services and supplementing existing models of care [54]. Therefore, health education delivered via mobile devices is worthy of clinical promotion for breast cancer patients.
Peer support was more effective than routine nursing care in improving the resilience of breast cancer patients. Peer support refers to the approach whereby individuals with the same disease or condition meet in order to exchange information, share experiences, and provide emotional/appraisal/informational assistance [55,56]. The results of this study showed that peer support was more effective than routine nursing care in improving the resilience of breast cancer patients (SMD = 2.62; 95% CI 0.61, 4.63; p = 0.01), which was similar to the results of a systematic review [56]. However, this conclusion should be treated with caution because only three trials [31-33], were included. In addition, the small sample size and publication bias may also lead to false negative results. Peer support was underutilised amongst women with breast cancer in recent years [57,58]. Future studies with better design/execution and larger sample size are needed to investigate the effectiveness of peer support on breast cancer patients.
There were seven other kind of interventions included which may be favorably associated with the breast cancer women's resilience [29,30,35,[39][40][41]47]. However, at present, these studies on resilience interventions for breast cancer patients had many limitations such as scattered research content and low quality of some researches, all of which bring di culties to data integration. These studies were not included in the meta-analysis due to the inability to classify and integrate the data, soour study cannot draw de nitive conclusions regarding the effectiveness of them. Thus, there is no clearly evidence to summarize the effect of these intervention measures for resilience of breast cancer patients. Further studies are needed to clarify the effectiveness of these interventions, respectively.
Although the existing interventions seem to be effective for promoting resilience of breast cancer patients, the current evidences are still in the preliminary research stage and the quality of them were not high. Recently, more and more studies have shown that in addition to social and psychological factors, biological mechanisms such as the amygdala, hippocampus, ventral striatum and frontal-associated regions of the brain are related to resilience strength [11,59]. An improved understanding of the biological mechanisms-based intervention could reveal novel therapeutic targets and help clarifying which psychosocial interventions can improve the psychological elasticity of breast cancer patients [60]. Therefore, we should formulate a staged, personalized and reasonable intervention plan from physiological, psychological social dimensions and in uencing factors in the future.
One of the advantages of this study is that we performed a systematic and comprehensive search of 12 databases. Another major strength of our review and meta-analysis is that all the included studies are RCTs. We used the latest tool RoB 2.0 to assess the risk of bias, and the quality of the included studies was moderate to high quality. This review was adhered to the PRISMA 2020 checklist and PRISMA 2020 ow diagram, which are the latest versions of PRISMA.
There were several limitations in our study. First, we considered RCTs published in English and Chinese only. Second, the included studies were exclusively conducted in the USA, Spain or China, which limit generalizability across other countries. Third, the small sample size (ranges from 20 to 175) might have decreased the reliability and could cause bias. Furthermore, this study performed meta-analysis on two interventions only due to data limitations and more RCTs are needed for data integration and analysis of other interventions.

Conclusions
It has been found in this study that the resilience training program, health education delivered via mobile devices and peer support are more effective than that of the routine nursing care. However, the heterogeneities between the studies were obvious and the sample sizes were not large. In addition, this study cannot draw de nitive conclusions regarding the effectiveness of other seven kind of interventions included in the systematic review because the sample size of these studies is small and cannot be conducted in the meta-analysis as there is only one study included regarding each kind of intervention. More studies are needed in the future to clarify the effectiveness of these interventions.

Declarations
Funding information This research was supported by Health research project from Health Commission of Sichuan Province (19PJ121) and Special Scienti c Research Fund of Sichuan Academy of Medical Sciences & Sichuan Provincial People's Hospital (2021ZX08). The funding bodies had no role in study design, data collection and analysis, decision to publish or writing the manuscript.

Con icts of interest/Competing interests
All authors certify that they have no a liations with or involvement in any organization or entity with any nancial interest or non-nancial interest in the subject matter or materials discussed in this manuscript.
Availability of data and material The data sets analyzed during the current study will be available upon reasonable request of the corresponding author.    Forest plots of the effectiveness for the intervention group versus the routine nursing care group. a Forest plot for the effectiveness the resilience training program. b Forest plot for the effectiveness of the intervention of health education delivered via a mobile device. c Forest plot for the effectiveness of peer support.