2.1. Protocol Registration

The protocol for this review has been registered with PROSPERO under the reference number CRD42023434406 (in Supporting Information (available here)).

2.2. Search Strategy

Eight databases (PubMed, Cochrane Library, EMBASE, CINAHL, Web of Science, China National Knowledge Infrastructure (CNKI), WanFang Data Knowledge Service Platform, and China Science and Technology Journal Database) were applied to search for relevant journal articles from the inception of each database to April 2024. The search words used were a combination of medical subject headings and keywords, such as “breast cancer”, “breast tumor,” “breast neoplasm,” “return-to-work,” “back to work,” “RTW,” “job reentry,” “reemployment,” “resume work,” and “employment status.” The detailed search strategies for each database will be presented in Supporting Information 1.

2.3. Inclusion and Exclusion Criteria

The inclusion criteria were as follows: (a) studies that analyzed a sample of patients with BC; (b) studies that measured RTW in that sample and specified the number or proportion of people who returned to work in the results; (c) language restrictions: Chinese and English; and (d) cross-sectional or cohort studies.

The exclusion criteria were (a) study samples that were mixed and lacked independent information on people with BC; (b) low-quality studies; and (c) lack of extractable effect sizes.

2.4. Study Selection

Duplicates were removed by importing documents using Endnote 20.0 software and using the software’s deduplication function, followed by a two-person (W.L. and J.W.) manual screening for duplicates. After titles and abstracts were screened by two reviewers (W.L. and J.W.) independently, a full-text evaluation based on the eligibility criteria was conducted by the same two reviewers. Any disagreement was resolved by a third review author (H.Y.).

2.5. Data Extraction

Herein, data were extracted by two reviewers based on the predefined criteria. The following variables were considered and recorded in a specific data definition record table: (a) authors; (b) year of publication; (c) country of publication; (d) study type; (e) sample size; (f) a percentage of BC patients who have returned to work in the sample. It should be noted that due to inconsistencies in RTW times between some studies and the fact that some studies did not report the time point chosen, for those studies that did report the time point, we extracted the time within 6–12 months of the patient’s RTW in order to standardize the criteria; and (g) factors influencing the RTW of BC patients, including their ORs and 95% confidence intervals. (All odd ratios mentioned in this study were extracted from the original study. We specifically selected those data that were derived using logistic regression analysis.)

2.6. Methodological Quality Assessment

Two investigators independently assessed the risk of bias in the included studies and cross-checked the results. In instances of disagreement, a third party was consulted to facilitate resolution. Cross-sectional studies were evaluated using the risk of bias criteria recommended by the Agency for Healthcare Research and Quality (AHRQ) [25]. There were 11 entries, each of which was evaluated using “Yes,” “No,” or “Unclear” as criteria, with “Yes” scoring one point and “No” or “Unclear” scoring 0. The total score for each entry was calculated by summing the individual scores, with 0–3 points as low quality, 4 to 7 points as medium quality, and 8 to 11 points as high quality [26]. The Newcastle - Ottawa Scale (NOS) [27] was used to evaluate the cohort study. The NOS is an eight-item scale ranging from 0 to 9. Scores between 0 and 3 indicate low quality, 4 to 6 suggest medium quality, and 7 to 9 signify high quality [28]. After evaluating the quality of the literature, only medium- and high-quality literature was included in the meta-analysis.

2.7. Statistical Analysis and Data Synthesis

The data were analyzed using Stata 12.0 software. The ORs and 95% CIs from the literature were chosen for data merging. I² was used to determine the extent of the heterogeneity of the data [29], and I² ≤ 50% indicated acceptable heterogeneity among studies, using a fixed-effect model for meta-analysis. I² > 50% suggested substantial heterogeneity, prompting the use of a random-effect model. Sensitivity analyses were performed by using a literature-by-literature exclusion method [30]. (One literature was excluded sequentially, and the remaining literature was subjected to meta-analysis. Any change in the significance of the results indicated instability). Funnel plots were drawn and used to analyze publication bias.

3.1. Search Results

The phases of electronic searches, identification, and screening for eligible studies were depicted in the PRISMA flow diagram (Figure 1). Initially, 2762 studies were obtained through the outlined search strategy, which was conducted from the inception of each database to April 2024. After removing duplicates and screening titles and abstracts, studies were retained for full-text evaluation. Finally, a total of 23 studies were included in this systematic review and meta-analysis. These included 10 cross-sectional studies [31–40] and 13 cohort studies [41–53].

3.2. Basic Characteristics of the Included Literature

Table 1 outlines the basic characteristics of 23 studies conducted across 11 countries, spanning the years 2006–2023: USA (n = 5), Sweden (n = 3), Korea (n = 1), Germany (n = 2), Brazil (n = 1), Malaysia (n = 1), France (n = 5), China (n = 2), Norway (n = 1), the Netherlands (n = 1), and Portugal (n = 1); 10 cross-sectional studies and 13 cohort studies were included, with a sample size of 200–2000 cases and a total of 4384 participants.

3.3. Results of the Quality Assessment of the Included Literature

The 23 papers included in this study were of medium and high quality. Cross-sectional studies were evaluated according to AHRQ recommendations, with 4 papers of high quality and 6 papers of medium quality, as shown in Table 2. Cohort studies were evaluated based on NOS scores, with 2 papers of high quality and 11 papers of medium quality, as shown in Table 3.

3.4. Meta-Analysis Results

A total of 23 influencing factors were included in this meta-analysis (Table 4). The overall rate of RTW was estimated at 65.0% (95% CI: 55.0%–75.0%) (Figure 2). The results showed that the differences among the combined outcomes of various factors, such as age, chemotherapy and radiotherapy receipt, axillary clearance, surgical modality, anxiety symptoms, work pattern, cognitive function, and physical abilities, were minimal (p > 0.1, I² ≤ 50%), so a fixed-effect model was used. The meta-analysis revealed significant heterogeneity (p < 0.1, I² > 50%) across various factors such as education, marital status, cohabitation status, race, receipt of endocrine therapy, presence of comorbidities, experiencing arm symptoms, cancer stage, adjuvant therapy, having children, depression, type of work, social support, and breast-related symptoms. Therefore, a random-effect model was used. The meta-results showed that age, receipt of radiotherapy or not, presence of comorbidities or not, cancer stage, receipt of axillary lymph node dissection or not, surgical modality, having children or not, anxiety symptoms, work patterns, cognitive functioning, and physical functioning were the factors influencing the RTW of BC patients. The summary results of the meta-analysis are shown in Table 4 and Figures 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13.

3.5. Sensitivity Analysis

In this study, sensitivity analyses were performed on factors with heterogeneity including education level A, marital status B, cohabitation status C, race D, receipt of endocrine therapy E, presence of comorbidities F, presence of arm symptoms G, cancer stage H, adjuvant therapy I, whether children were born J, depressive symptoms K, type of work L, social support M, and breast-related symptoms N. (Capital letters represent the order of the corresponding pictures; see the Figures in Supporting Information 2) Sensitivity analyses of factors including receipt of endocrine therapy, presence of comorbidities, cancer stage, and presence of adjuvant therapy showed little difference between the results after the exclusion of each article and the original results, indicating the stability of the results. Sensitivity analyses involving the factor “whether having children” significantly impacted the RTW outcomes of BC patients, showing a large difference in the variation of results upon removing Agnes Dumas’s study [46]. This may be related to the large sample size of this study (n = 1874).

3.6. Publication Bias

In this study, we conducted a publication bias analysis focusing on comorbidities, and the results showed asymmetry in the funnel plots, suggesting potential publication bias (Figure 14).

4.1. Socio-Demographic Factors

The present study showed that among the socio-demographic factors, the age of BC survivors and whether they have children or not are important factors associated with RTW. Younger age and having children increase the likelihood of RTW than older BC patients, consistent with previous studies [23, 37]. This tendency can be attributed to their superior physical capabilities, increased energy levels, existing employment history, and accumulated knowledge prior to their cancer diagnosis [55]. In contrast, older patients are more likely to have complications due to the gradual deterioration of their physical functioning, impacting their social engagement and work capacity. This can be attributed to their retired status or lack of inclination to work [56], alongside greater economic stability at this stage of life [57]. Therefore, it is necessary to focus on patients of working age, analyze their difficulties and obstacles to returning to work, and develop individualized interventions to help them RTW. Having children often motivates patients to RTW, possibly due to a sense of familial duty and the desire to contribute to the family’s financial well-being [58]. This aligns with the study by Porro et al. [59] but contrary to the findings of Caumette et al. [60], Johnsson et al. [32], and Lee et al. [61]. The differences could be due to variations in cultural contexts, sample characteristics, and methodologies used across studies. Additionally, economic stability and healthcare systems in different countries could also account for these discrepancies. This combined result stands out as the only diverse finding among the significant factors in this study. For future analyses, it is crucial to broaden the scope by incorporating studies from various countries and cultural backgrounds. Factors such as the number and age of children, as well as economic situations, should be considered. Lee’s findings, for example, showed that many patients would give up returning to work because of the need to care for their children at a young age [61]. In the future, consideration can be given to supporting patients who have recently given birth by offering flexible work options such as telecommuting and adjustable work hours. This will help them effectively manage both their family responsibilities and work commitments. The support of their spouses is also an aspect that needs to be taken into account.

4.2. Factors Related to Disease Status

In this study, the stage of cancer and comorbidities emerged as significant disease-related factors influencing RTW among BC survivors. Patients diagnosed with early-stage BC are more likely to RTW than those with advanced BC. Cancer treatment duration and intensity vary based on the stage. Early-stage cancers often require shorter treatments, primarily surgical, aimed at tumor removal. In contrast, advanced stages necessitate longer, comprehensive approaches due to the need for combined therapies such as chemotherapy and radiotherapy, which bring about more adverse effects, demanding extended recovery periods [62, 63]. The side effects of these treatments can affect recovery, potentially impacting a patient’s ability to work during this time [64]. Patients with more advanced cancers often face greater physical burdens and recovery challenges. Treatment may take a greater toll on the body, leading to a prolonged and more challenging recovery process [65]. Also, people with more advanced cancer may face greater psychological burdens and emotional distress [66]. This may include uncertainty about the future and concern about the effectiveness of treatment. Psychological factors can lead patients to approach their RTW cautiously, often requiring additional time to address emotional challenges. The results found that the presence of comorbidities can prevent BC patients from returning to work. BC-related comorbidities such as fatigue, pain, and lymphedema are important factors affecting patients’ RTW [22]. The present studies showed that patients with one or more comorbidities are more difficult to RTW than those without comorbidities [67, 68]. The presence of comorbidities may lead to instability or deterioration of the patient’s physical condition, which may affect the patient’s ability to work. Patients may need more time to recover and adjust to life, thus delaying or adjusting plans to RTW [69, 70]. Besides, comorbidities may increase the psychological burden on patients, including emotional problems such as anxiety and depression [71, 72]. These mental health issues may affect the patient’s interest and confidence in work and reduce their motivation to RTW [73, 74]. When evaluating a BC patient’s RTW, a thorough assessment of how comorbidities affect their physical and mental health, along with the challenges they present in adapting to work, helps to better grasp the role of these additional health conditions in the decision-making process. The impact of these factors may vary with individual patients, and each patient’s decision is influenced by a combination of factors.

4.3. Factors Related to Treatment

This study revealed significant correlations between a patient’s RTW and specific treatment factors, such as receiving radiotherapy, undergoing axillary lymph node dissection, and the type of surgery (breast-conserving or mastectomy). Patients who have radiotherapy treatment or axillary lymph node dissection or mastectomy are more likely to RTW, a result that is consistent with previous studies [23, 56, 75]. The reason for this is that these treatments help to control the progression of BC and improve the health of the patient [76]. When treatment succeeds and patients experience improved health, their likelihood of returning to work increases [77]. It also needs to be taken into account that after successfully undergoing these treatments, patients may experience increased confidence and a more positive mindset [78]. This positive mindset may contribute to their earlier return to normal life, including the RTW [79]. It is important to note that BC treatment encompasses a range of approaches, and individual variations along with specific circumstances can significantly influence treatment and recovery outcomes. It is crucial to highlight that our study results only represent a trend. Indeed, each patient’s recovery and RTW may vary depending on a variety of factors. For example, some studies [80, 81] reported that undergoing lymph node dissection surgery results in an increased likelihood of upper limb disability affecting the patient’s RTW. More studies should be included in future analyses.

4.4. Factors Related to Physical and Mental Status

In this study, anxiety levels, cognitive function, and physical well-being emerged as crucial factors influencing the RTW among BC survivors, highlighting their significance in both physical and mental aspects. A diagnosis of anxiety symptoms can make it harder for a patient to RTW. Anxiety adds to the psychological load, making it tough to handle the stresses and challenges of the job [82, 83]. This may lead to resistance to work and affect their willingness and ability to RTW. Attention also needs to be paid to the provision of psychological support services to help patients to overcome anxiety and improve their self-confidence, so that they can RTW with greater confidence. The results of this study show a positive correlation between cognitive function and RTW. Previous studies [84, 85] have shown that approximately 34% of BC patients who undergo cancer treatment experience cognitive dysfunction. This is strongly associated with reduced ability to work, including memory issues impacting timely task completion, significantly influencing their ability to RTW [34]. Improved cognitive function leads to enhanced learning and memory capabilities. In the work environment, it is necessary to constantly learn new information and adapt to new work requirements, and enhanced cognitive function facilitates a smoother adjustment to new job demands, expediting the RTW [86]. In the rehabilitation treatment of BC patients, attention should be paid to the provision of enhanced cognitive function training, including professional cognitive rehabilitation programs, to help patients improve their cognitive abilities in learning, memory, decision-making, and other aspects. Currently, there are relevant exercise studies [87, 88] showing that it can promote cognitive recovery to a certain extent, mainly through internet-based cognitive forging games, including speed training, memory training, reasoning training, imagery training, and storytelling. The better the physical functioning, the greater the likelihood of RTW in this study. Improving physical functioning enhances a patient’s capacity to perform tasks at work and adapt to the work environment more effectively [89]. Enhanced physical functioning may increase productivity and reduce the physical burden of being at work, thus facilitating the likelihood of RTW [90]. Good physical functioning may also be seen by employers as a sign of the patient’s recovery, indicating that the patient can perform the job [91]. This may increase the employer’s confidence in the patient’s ability and make it easier for him/her to integrate into the work team [92].

4.5. Factors Related to Job Characteristics

This study revealed that the employment status of BC survivors, whether they worked part-time or full-time prior to their illness, is a significant factor influencing their RTW outcomes. Besides, patients who work part-time before diagnosis have more difficulty returning to work. Firstly, different work patterns often correspond to different occupational requirements and skills, and patients may have chosen to work part-time based on their current life situation and ability level [93]. Cancer patients may need more time to adapt to the skills and occupational requirements needed for their original job [92]. Secondly, patients opting for part-time work may rely more on social support networks, be it from employers, coworkers, or other interested parties, potentially impacting the stability of their recovery process [94]. Establishing a social support network in part-time versus full-time work settings differs, requiring time for patients to build new support systems. It is important to note that this study demonstrates only general trends, yet the specifics may be affected by several factors such as individual patient differences, the degree of rehabilitation, and the specific work environment.

4.6. Other Factors

Although the results of this meta-analysis showed that the p values in the combined results of possible influences such as education, marital status, and type of work were greater than 0.05, this does not mean that these factors are meaningless in the process of returning to work for BC patients. Firstly, the p value is not the only criterion for assessing the strength of the variable’s impact on the outcome [95]. Factors that do not reach statistical significance can still influence patients’ RTW, especially considering individual differences and cultural backgrounds. For example, one study [54] showed that patients with higher levels of education chose to RTW, and another study [96] showed that those in high-paying occupations, such as professional technicians, tended to RTW more quickly. Secondly, factors such as the education level, marital status, and type of work may play different roles in different contexts and groups. For example, in Emerson’s study [97], blacks and whites, irrespective of their life statuses, demonstrated varied RTW results. There is also a need to take into account factors such as the heterogeneity of the study sample and the potential lack of control variables, the effects of which may not have been fully included in the analyses, leading to statistically insignificant results. Future studies can further explore the potential impact of these factors on the RTW of BC patients through a more refined design and more comprehensive data collection. This will help to provide a more comprehensive understanding of the multidimensional factors involved in the work reintegration of BC patients and provide a more nuanced reference for the development of individualized rehabilitation plans.