Introduction

According to the latest data released by the International Agency for Research on Cancer (IARC), the prediction shows that by 2050, the number of newly diagnosed cancer cases worldwide will be as high as 35 million, increasing to 77% from 20221. In recent years, with the advancement of screening, diagnosis, and treatment technologies, cancer survival rates have increased. More and more patients with cancer survive for a longtime. Data from the National Cancer Center of China show that the 5-year cancer survival rate in China has rose from 30.9 to 40.5%, which means that the number of cancer survivors (CSs) who return to normal work and life is also growing rapidly2.

Since the 1970s, the Engel’s Bio-psycho-social (BPS) medical Model has been proposed which emphasize a holistic approach to patients’ health, focusing not only on physical and mental well-being but also on the restoration of social functioning3. For CSs, return to work (RTW) is an important sign of the recovery of their social functions which can improve their sense of self-identity, occupational belonging, and quality of life4,5 and, to some extent, reducing the economic burden on families and society6. In 2010, the model of Cancer and Work4 proposed that the work outcomes of CSs, such as RTW, work ability, and work adaptability are influenced by their personal, environmental, and social factors like physical and mental health, functional status, work environment, policies and financial circumstances. This theoretical framework is beneficial for optimizing, adjusting, and addressing RTW related issues in CSs. Based on the above two theoretical models, our research team proposed the CSs’ RTW adaptability (CSs-RTWA) theoretical model and identified the meaning of CSs-RTWA through grounded theory method. CSs-RTWA refers to the ability of CSs to utilize the internal and external resources to effectively cope with the various stresses during the period of RTW and to maintain a balance between work and physical and mental health7,8. The CSs-RTWA theoretical model holds that RTWA is the process of self-reconstruction utilizing the individual and external resources advantages, mainly including three aspects: focusing on rehabilitation, reconstruction efficacy, and adjustment planning7. Furthermore, the CSs-RTWA theoretical model also points out the individual’s coping resources during RTW, consisting of personal resources such as psychological resilience, belief, and faith, as well as external resources such as the support from family, peers, employer and professional. These resources can not only stimulate CSs’ internal driving force of RTW, but also provide a complete social support system for CSs to smoothly return to the workplace. Thus, the theoretical model of CSs-RTWA and coping resources provide a theoretical basis for the research hypothesis of this study.

Previous studies showed that CSs have an overall low level of RTWA9. It is affected by age, education level, occupation type, working status, income, cancer type, clinical stage, self-efficacy, coping style and other factors9. Therefore, the ability to adapt to challenging environments, seize opportunities, and engage in valuable vocational rehabilitation activities plays a crucial role in the CSs’ health status and quality of life10,11.

Latent profile analysis12 (LPA) is an individual-centered approach that includes individuals with similar characteristics in the same profile, distinguishes them into different categories with significant characteristics, and analyzes the characteristics of each potential category. In this study, LPA can be used to analyze the potential categories of CS-RTWA and help clinicians identify those with low CS-RTWA who may need precise interventions13. However, no researchers have yet explored the heterogeneity of CS-RTWA in CSs’ population or how demographic sociological factors influence these potential categories. Existing studies tend to use a variable-centered approach which may ignore individual differences in CS-RTWA, which may be the case that individuals who have large differences in their scores on individual terms, but are classified in the same group due to the same overall scores11.

It was found that economic status plays an important role in determining the work status of CSs14. Financial toxicity (FT) refers to the adverse economic side effects experienced by cancer patients due to high out-of-pocket costs during treatment, with potential consequences encompassing both objective financial burden (e.g., medical debt, reduced savings) and subjective financial distress (e.g., anxiety, coping strain)15. There is no doubt that FT has a profoundly negative impact on the psychosocial status and life experience of CSs, with psychological distress such as anxiety and depression, and adverse economic consequences such as altered employment status and increased household debt12,16. Previous studies found that FT can stimulate the behavior of CSs-RTW17. The study of chained mediation analysis revealed that FT can directly or indirectly influence CS-RTWA through stigma and/or social support18. There is ample evidence that good income is one of the protective factors influencing CSs-RTWA7. But, whether FT has an impact on the potential categories of CSs-RTWA has not been reported. Therefore, in order to strengthen the CSs-RTWA and help them better return to society, it is necessary to explore the association between the potential categories of CS-RTWA and FT.

To sum up, the objectives of this study were (a) to identify different potential categories of CSs-RTWA using potential profiling method; (b) to explore socio-demographic characteristics and disease-related variables that influence potential categories of CSs-RTWA; and (c) to analyze the effect of FT on potential categories of CSs-RTWA using multivariate logistic regression.

Methods

Participants

A total of 412 CSs from Affiliated Hospital of Nantong University and Nantong Cancer Rehabilitation Association were selected using the convenience sampling method for the study from April 2023 to August 2024. The inclusion criteria were as follows: (1) diagnosed with malignant tumors by clinical, imaging, and histopathological examinations, with pathological stages I to IV; (2) patients who had finished conventional cancer treatment and were in the follow-up stage, which were evaluated by specialist doctors; (3) working normally before diagnosis; (4) aged 18 to 60 years old; (5) having the ability of clear verbal expression and being able to comprehend and complete the assessment; and (6) being aware of the diagnosis of cancer and voluntarily participating in the study. Exclusion criteria: (1) other serious diseases or complications; (2) cognitive impairment; (3) history of mental disorder.

Measures

General socio-demographic information

The general socio-demographic questionnaire was designed by our own research team, which included patients’ socio-demographic information (age, gender, place of residence, education level, marital status, type of family, with or without minor children, occupation type, nature of work unit, nature of work, main economic sources, per capita monthly family income, types of medical insurance, current working status, and sick pay) and clinical information (cancer type, cancer stage, therapy method, time since diagnosis, any complications, and comorbidity with other chronic diseases).

The adaptability to return to work scale (ARTWS) for cancer patients

The scale was developed by our research group8 and consists of 3 dimensions, including focusing on rehabilitation, self-efficacy rebuilding, and plan adjustment, with a total of 24 items. Each entry was scored on a 5-point Likert scale ranging from 1 for “disagree” to 5 for “strongly agree”, with a total score of 24 to 120, with higher scores indicating greater adaptability to RTW for cancer patients. The Cronbach’s alpha coefficient of the total scale is 0.937, which has good reliability and validity.

Chinese version of the comprehensive scores for FT based on patient-reported outcome measures (COST-PROM)

This scale19 is used to measure the perceived degree of subjective and objective dilemmas in the economic and work status of cancer patients, which consists of 11 items consisting of 2 dimensions, positive wealth status (4 items) and negative psychosocial reactions (7 items), on a Likert 5-point scale from 0 to 4, with a total score of 0–44. The lower the scale score, the more serious the FT of cancer patients, and the higher the scale score, the lighter FT of cancer patients. In this study, the scale demonstrated good reliability and validity, with a Cronbach’s alpha coefficient of 0.935 and both I-CVI and S-CVI exceeding 0.80. COST-PORM score of ≥ 26 indicates no impact on the quality of life (level 0), score of 14–25 indicates a mild impact (level 1), score of 1–13 indicates a moderate impact (level 2), score of 0 indicates a high impact (level 3), and level 1 and above are defined as having FT.

Data collection and analysis

Before the survey, the researchers introduced the study purpose to CSs and explained the precautions for questionnaire completion. Data collection was completed when CSs visited the hospital for review. Face-to-face questioning and survey filling were adopted, and questionnaires were retrieved after filling. The data were entered independently by two members of the research group and repeatedly checked and verified to ensure accuracy.

First, potential profile models with different numbers of categories were fitted through Mplus 8.0 software, using the scores of each item of CSs-RTWA as exogenous indicators. Starting from 1 category, the number of potential categories was gradually increased, the greatest likelihood method was applied to estimate each profile model, the test between the hypothetical model and the observed data was repeated, and the best model was determined by considering the model fit evaluation indexes and combining with the clinical judgment, and each potential category was named12. The fit metrics for potential profile models included the following 3 categories: (1) Akaike information criterion (AIC), Bayesian information criterion (BIC) and sample size-adjusted Bayesian Information Criteria (aBIC) are the information indicators. Criteria (aBIC) is the information criterion, the lower the value of these three indicators, the better the fitting effect of the model; Entropy is used to evaluate the accuracy of a model’s classification, and ranges from 0 to 1. The closer the value is to 1, the higher the classification accuracy. When the entropy value is ≥ 0.8, the classification accuracy exceeds 90%. (2) Calibration of Lo-Mendell-Rubin likelihood ratio test (LMRT) and Bootstrap-based likelihood ratio test (BLRT), LMR and BLRT have statistical significance (p < 0.05), indicating that the K-class model is superior to the k-1 class model. (3) The entropy value can be used to assess the accuracy of the model categorization, which takes the value of 0–1, and the closer the value is to 1, it indicates that the categorization accuracy is correspondingly higher. To sum up, an entropy value ≥ 0.8 indicates a classification accuracy of more than 90% for each potential category of the profile model20. Usually, lower values of AIC, BIC, and aBIC correspond to better fitting models, entropy values greater than 0.8 indicate high accuracy in classification, and BLRT and LMRT are statistically significant (p < 0.05) as the preferred choice for LPA models21. The posterior class probability refers to the probability that each individual belongs to different latent classes. Latent Profile Analysis (LPA) employs maximum likelihood estimation (MLE) as the core method for parameter estimation. In LPA, MLE is implemented using the Expectation-Maximization (EM) algorithm: in the E-step, the posterior probabilities of individuals belonging to each latent class are computed. In the M-step, model parameters are updated to optimize the likelihood function. This process iterates until convergence22. To ensure robustness, multiple random initializations were applied in practice. Moreover, the threshold for posterior probabilities is an empirically established standard rather than a fixed value in statistical theory. Therefore, we adopted an empirical criterion commonly used in latent profile analysis: If the model fits well and the entropy is sufficiently high, the probability of class membership should approach 1.0. Specifically, individuals were assigned to a class with high confidence if their maximum posterior probability exceeded 0.8. If the model’s entropy was below 0.8 and the maximum posterior probability fell between 0.5 and 0.8, the classification was considered “low certainty”. This “classify-analyze” approach minimizes information loss, thereby reducing the likelihood of misclassification23. Secondly, SPSS 27.0 software was used for statistical analysis. Naturally distributed measures were described by mean ± standard deviation, non-normally distributed measures were described by M (P25, P75), and count data were analyzed by frequency and percentage. After determining the optimal model, the general information of different categories of patients was compared by the chi-square test, Fisher’s exact probability method, and ANOVA. Pearson correlation analysis was utilized to explore the relationship between CS-RTWA and FT. Indicators with statistically significant differences in one-way analyses were analyzed by multivariate logistic regression, and socio-demographic, disease-related factors, and FT scores were used as independent variables, and LPA results were used as dependent variables to explore the effects of demographics, clinical characteristics, and FT on different categories of CSs-RTWA. The differences were statistically significant at p < 0.05.

Ethical considerations

This study was approved by the Ethics Committee of Affiliated Hospital of Nantong University (approval number: 2023-K097-01). The purpose and content of this study were explained to all participants, who signed a written informed consent form after agreeing to participate in the study. All participants were allowed to withdraw from the study at any time, and their medical rights were not affected. All participant information and privacy were kept strictly confidential. Informed consent was obtained from all subjects and/or their legal guardian(s). All experiments were performed in accordance with relevant guidelines and regulations.

Results

Demographic and clinical characteristics of the participants

A total of 426 questionnaires were distributed in this study, and 412 valid questionnaires were recovered with a validity rate of 96.71%. Of the 14 invalid questionnaires, 5 were discontinued due to participant reasons, and the remaining 9 questionnaires were answered regularly. There were 97 (23.5%) males and 315 (76.5%) females with an age range from 27 to 59 years (M = 44.95, SD = 6.71). Nearly one-third (27.4%) of CSs returned to work after treatment. 76.7% of CSs have no sick pay. Table 1 showed the demographic and clinical characteristics of the participants.

Table 1 Sample demographic and clinical characteristics (N = 412).
Full size table

The score of CSs-RTWA and FT

The total CSs-RTWA score was 87.01 (SD = 12.45), with the score of focusing on rehabilitation dimension being 22.73 (SD = 3.13), the score of self-efficacy rebuilding dimension being 31.02 (SD = 6.25), and the score of plan adjustment dimension being 33.27 (SD = 5.45). The total FT score for CSs was 22.23 (SD = 9.29), with the positive wealth status and negative psychosocial reaction dimension scores being 4.67 (SD = 3.14) and 17.57 (SD = 6.89), respectively. Pearson correlation analysis showed that the FT scores were positively correlated with the CSs-RTWA scores (r = 0.304, p < 0.001). According to the scoring method of the scale, a higher FT score indicated lighter FT.

Characterization and name of potential categories of the CSs-RTWA

In this study, Mplus software was used to fit latent class models with 1 to 5 classes, and the detailed fit indices were presented in Table 2. The results showed that as the number of classes increased, the AIC, BIC, and aBIC values gradually decreased, while the entropy values were all above 0.9. When the number of classes reached 4 or 5, the LMRT test was no longer statistically significant (p > 0.05), leading to the exclusion of the 4-class and 5-class models. In conclusion, based on a comprehensive consideration of model parsimony, statistical robustness, and practical interpretability, the 3-class model was deemed to best align with the research objectives. As shown in Table 3, the probabilities of CSs correctly belonging to each category were 97.3%, 94.2%, and 96.3%, indicating that the model had good discriminative properties and reliable categorization.

Table 2 Latent profile analysis models and fit indices.
Full size table
Table 3 The probability of latent profile analysis.
Full size table

Three potential categories were named based on their scores on each dimension of ARTWS for cancer patients. Category 1 had low scores on all dimensions of CSs-RTWA and was named the “poor CSs-RTWA group” (N = 85, 20.5%). Category 2 was named the “moderate CSs-RTWA-adjustment group” (N = 134, 32.9%) because it had generally moderate scores on all dimensions of CSs-RTWA and generally higher mean scores on the focusing on rehabilitation and plan adjustment dimensions than on the self-efficacy rebuilding dimension. Category 3 had higher scores on all three dimensions of CSs-RTWA, so it was named the “high CSs-RTWA-harmonization group” (N = 193, 46.6%). The distribution of potential category characteristics of CSs-RTWA was shown in Fig. 1.

Fig. 1
figure 1

Latent profiles of cancer survivors’ adaptability to return to work.

Full size image

Univariate analysis of potential categories of CSs-RTWA

The intergroup differences between the three potential categories of CSs-RTWA in terms of age group, place of residence, education level, type of family, occupation type, nature of work unit, nature of work, main economic sources, per capita monthly family income, types of medical insurance, current working status, sick pay, and cancer type were statistically significant (p < 0.05), and the results were shown in Table 4. The differences between the score groups for each potential category of CSs-RTWA in the total FT score, the positive wealth status dimension, and the negative psychosocial reaction dimension were statistically significant (p < 0.05). Subsequently, post hoc multiple comparisons with Bonferroni correction were performed. The results showed statistically significant differences (p < 0.05) among the three RTWA categories in terms of the total FT scores and its two dimensions. This indicated that the latent classification of RTWA in this study was valid and can effectively differentiate individuals’ RTWA levels. The results were shown in Table 5.

Table 4 The general information of the participants and single factor analysis of 3 latent categories among cancer survivors.
Full size table
Table 5 Total financial toxicity scores and scores on dimensions for different latent profiles of return to work adaptability.
Full size table

Multiple logistic regression analysis of potential categories of CSs-RTWA

The results of the multiple logistic regression analyses, using the variables with statistically significant differences in the univariate analyses of variance as the independent variables and the three potential categories of CSs-RTWA as the dependent variables, were shown in Table 6. Using the “poor CSs-RTWA group” as the reference group, (1) the “moderate CSs-RTWA-adjustment group” was affected by the factors of age ≤ 40 years old, per capita monthly family income more than 3000 yuan, occupation other than being a housewife or househusband, and total score of FT and score of positive wealth status (p < 0.05). (2) The “high CSs-RTWA -harmonization group” was affected by the factors of age ≤ 40 years, family types other than single-parent families, occupation other than being a housewife, depending on one’s own or one’s spouse’s income predominantly, on-the-job status or paid sick leave, and having sick pay (p < 0.05). (3) The “poor CSs-RTWA group” was affected by the factors of urban residence, working in public unit and national administrative authority, manual labor, no sick pay, educational level of primary school or below, per capita monthly family income less than 1000 yuan (p < 0.05).

Table 6 Multiple logistic regression analysis of potential categories of return to work adaptability for cancer survivors.
Full size table

Discussion

This study identified for the first time three potential categories of CSs-RTWA based on latent profile analysis, namely, the poor CSs-RTWA group, the moderate CSs-RTWA-adjustment group, and the high CSs-RTWA-harmonization group. The results suggest group heterogeneity exits in CSs-RTWA. In addition, the results of this study showed that age, place of residence, education level, type of family, per capita monthly family income, main economic sources, occupation type, nature of work, nature of work unit, current working status, sick pay, and FT affect the potential categories of CSs-RTWA.

It is noteworthy that 20.5% of CSs were in the poor CSs-RTWA group and had low probabilities on each of the ARTWS dimensions, which implies that patients in this category were less able to focus on physical and psychological recovery, rebuilding self-efficacy, and career plans adjustment. According to this study, CSs living in urban areas were more likely to have poor CSs-RTWA. This was likely due to the higher cost of living in urban areas than in rural areas. Meanwhile, urban residents may have higher job expectations and therefore may develop greater dissatisfaction when faced with the challenge of RTW. CSs were also physically and psychologically challenged for the fast pace of work and high work pressure in cities24. The findings also indicated that CSs who work in hard physical labor, have a per capita monthly household income of less than 1,000 RMB, have no sick pay and have education level of elementary school or lower were less able to adjust to RTW. The possible reason may be that this group of CSs was of low socioeconomic status and poor social support. Compared with mental workers and mixed workers, the group engage in physical labor mostly has jobs of a temporary employment nature and has less autonomy in choosing the type and the duration of their work, which resulted in poorer psychosocial adjustment, as well as greater stress and difficulties when they RTW, which was consistent with the results of previous studies25. Similar to the results of previous studies7, compared with CSs with high educational attainment, those with low educational attainment often had a lower economic and social status. They may not have sufficient savings and stable sources of income. Under more severe FT, their RTWA was mostly at a lower level. Compared with the CSs with sick leave pay and with a relatively higher per capita monthly income of the family, the CSs without sick leave pay and with a lower per capita monthly income of the family less than 1,000 RMB had less support from units and employers, and they had fewer opportunities to obtain health insurance and social security. The FT of this group of CSs was more severe. They had to “force” themselves to RTW earlier to maintain normal treatment and living needs and their RTWA was poor, which was similar to the previous research results26,27,28. In addition, this study discovered that CSs working in public units and national administrative authority were not well adjusted in RTW, which contradicted earlier research findings26. This might be due to the fact that only 7.3% of the sample worked in public unit and national administrative authority in this study, resulting in sampling bias; nevertheless, we will employ multicenter sampling and enlarge sample size to improve the sample’s representativeness.

The moderate CSs-RTWA-adjustment group, with 32.9% of the total number of CSs, had a more even probability on the three dimensions of the ARTWS. In contrast to the poor CSs-RTWA group, CSs in this group maintained a relatively balanced profile in terms of facilitating recovery and career planning adjustment, but poor performance in self-efficacy reconstruction. Analyzing the items with the lowest probability, we found that this group of CSs faces insufficient support from leaders and units and poor professional guidance from medical staff and less encouragement from patients who had already returned to work. Some studies have emphasized the positive role of workplace support plays a positive role in cancer patients’ efforts to adapt to changes in their working conditions and external environment29. At the same time, Hallgren et al. identified the failure of employers to provide work assistance, as well as the lack of work-related information by healthcare workers and peers, are significant factors preventing cancer patients from RTW30. Therefore, suitable job positions, flexible working hours and job security provided by the employers, comprehensive medical support and disease monitoring provided by professional doctors, a good working atmosphere created by colleagues, and life care and positive feedback supported by family play an important role in the CSs’ reintegration into society and realization their personal value31. As a result of this study, it was found that age under 40 years had a significant influence on moderate and high CSs-RTWA groups. STONE et al. also indicated the younger cancer group aged 15 to 39 years work harder to cope with physical and psychosocial changes caused by disease and actively improve their abilities to adapt to work changes32. The study also found that per capita monthly family income greater than 3000 RMB had a significant effect on CSs-RTWA, which is similar with the previous findings7, that is, better family income can protect patients from FT and thus promote readjust to work. It is worth emphasizing that, for the first time, we found that CSs with better positive wealth status are more likely to be categorized in the moderate CSs-RTWA group, probably because they have more generous salary and benefits, better medical insurance, and better financial stability to support the cost of their cancer treatment. As a result, their effects in disease treatment and physical and mental rehabilitation are better. The economic advantages bring greater flexibility in career choices to this group of CSs and they have a higher willingness to return to work. Therefore, under the combined effect of these favorable conditions, patients can not only maintain a better health state, but also rebuild their professional identity more smoothly, and ultimately demonstrate a higher RTWA33. This fully demonstrates the crucial role of favorable economic conditions in the vocational rehabilitation process of CSs34. Compared with housewives/househusbands, people with specialized occupations have better adaptability to return to work and may be related to the fact that they have a clear social and professional identity35,36.

In this study, nearly half (46.6%) of the CSs was in the high CSs-RTWA-harmonization group and showed high levels of RTW acceptance and adaptability, with relatively balanced scores on the three dimensions. However, we still found that patients in this group received weak role model support from their friends or peers, which suggests that clinicians should encourage and guide patients to learn from examples of successful RTW after cancer treatment and share their experiences and positive energy, which can help CSs better coping with the changes in work and life after treatment, and actively explore new ways to enhance self-management and adapt to the new challenging environment37. Similar to the moderate CSs-RTWA-adjustment, the results of this category also showed that CSs with age ≤ 40, having sick pay have a greater probability of being categorized in the high CSs-RTWA-harmonization group, for possible reasons has described above. The results of this study also revealed that people who have returned to work or have paid sick leave have multiple advantages over those who are in unemployment or retirement status. For CSs who returned to work showed better physical and mental status and social functioning. They efficiently mobilized and utilized internal and external resources to cope with stresses in their lives, thus keeping a balance between work maintaining and physical and mental health, meeting their personal needs and adapting to the external environment7,38. CSs on paid sick leave have better economic security and social support, and they are more adaptable to return to work. At the same time, we identified an important factor which had not been explored previously: the effect of family type on CSs-RTWA. Specifically, compared with single-parent families, CSs with the family types of core, trunk, and extended families tend to receive stronger financial, emotional, and practical support from family members. Relying on the comprehensive system of family support, the family can play its full role as a strong backbone, which will increase CSs-RTWA and healing efficiency39. Second, this study also found that, compared with CSs who mainly relied on external economic sources such as social assistance or children’s support, those who relied on their own labor income or spouse’s labor income as their core economic pillars demonstrated stronger adaptability. Such survivors, with both financial and emotional support from their spouses, hope to RTW as early as possible through positive physical and psychological adjustment. Hence, the financial support from the spouse is not only crucial to their personal recovery process but also has far-reaching implications for maintaining family economic stability and emotional harmony40.

In addition, the present study also found that the FT score is positively correlated with the CSs-RTWA score (r = 0.304, P < 0.001), which indicated that the lighter the FT, the better the CSs-RTWA, and the lighter the FT, the greater the likelihood of being categorized as a “moderate or high CSs-RTWA-adjustment group”. So, to some extent, light FT has a substantial impact on improving CSs-RTWA. When FT is light, the financial burden on CSs and their families is less. This allows CSs to focus more attention on treatment and recovery without too much worry on their medical costs. The financial relief reduces CSs’ psychological pressure, enabling them face the disease and life with a more positive and optimistic mindset. A healthy physical and mental state is crucial to enable CSs to quickly adapt to a new environment after RTW41,42. At the same time, they have more time and energy to adjust their family and personal lives and improve their quality of life. This helps CSs better adapt to the work environment and alleviate negative emotions triggered by work stress43. For those who suffer severe FT, the side effects of long-term economic hardship will erode their ability to work continuously and may trigger discrimination and social isolation in the workplace44. This situation is more likely to exacerbate CSs’ vicious cycle of poor economic footing-difficulty in seeking medical care-deteriorating health outcomes-reduced adaptability to RTW-hindered RTW process-sharp decline in income and savings-exacerbated FT45,46. Therefore, in response to this vicious cycle, the application of a nurse-led multidisciplinary teamwork model of financial navigation enables patients to allocate their limited financial resources more rationally and reduce FT47. At the same time, at the macro level, the Chinese government had attached great importance to people’s health and introduce a series of strategic guidelines and policies such as “Healthy China” in recent years, improve medical and health policies, social security, and other systems to provide welfare policy support for cancer patients. Provide better welfare policy support for cancer patients, especially those surviving patients with severe FT. For example, apply for economic assistance or a larger proportion of expense reimbursement for cancer survivors who have fallen back into poverty due to cancer, so as to alleviate their financial burden and promote the full recovery of physical and mental health and social functions. In addition, this study is supported by the project of “The research on the social supportive system for CSs’ RTW”, which had been funded by the National Social Science Foundation of China in 2021. The results of this study will be used as the recommendation reports to help Chinese government better provide social support for CSs’ RTW. It is worth emphasizing that China’s social insurance system is generally consistent, suggesting that our findings may be applicable to most CSs in China. However, economically backward regions may face greater economic pressure and the gap in social support for CSs48, indicating that further research is needed to explore differentiation strategy to support CSs to have lighter FT and higher RTWA.

To summarize, medical staff need to focus on the poor CSs-RTWA group, especially those who live in towns and cities with heavy manual labor, lower family income, lower educational level, no sick leave pay and more serious FT. CSs should be actively advocate and encourage to proactively seek various internal and external resources, including their own and social support, to better adapt to the new changes in life and work after being diagnosed with cancer. CSs should be helped to rebuild their self-efficacy through multidisciplinary supportive rehabilitation intervention such as medical support, workplace support, family support and government support to successfully RTW and achieve comprehensive recovery. At the same time, CSs should be encouraged to RTW or adjust their employment strategies flexibly to alleviate the negative economic impact brought by cancer treatment. If necessary, we should actively call on social charitable organizations, government relief and security systems, and other relevant departments to provide financial assistance to CSs with severe FT. CSs should be prevented from falling into the predicament of becoming poor due to illness and abandoning treatment due to poverty. Overall, CSs’ RTW is not only the CSs’ deep desire to resume a normal role in society, but also has a profound impact on the stability of their families and even the whole society. Therefore, CSs with serious FT and low CS-RTWA should be accurately identified and a series of effective interventions should be implemented to effectively resolve the problems.

Limitations

There are several limitations in this study. It was a cross-sectional study which did not assess the dynamic trajectory of CSs-RTWA. In the future, longitudinal studies should be conducted to track changes in the CSs-RTWA, further explore the causal relationship between variables. Second, this study used convenience sampling with an uneven distribution of sample areas; in future study, a rigorous stratified sampling technique should be created to improve the representativeness of the study sample. Third, data was collected from only one tertiary-level hospital and the Cancer Rehabilitation Association in China, which may have resulted in sample bias. In light of this, a multicenter large-sample study should be conducted to collect data from a diverse range of samples in various geographic regions and contexts, ensuring the diversity and comprehensiveness of the information obtained and improving the generalizability of the findings. Finally, specific socio-economic indicators such as income and sick leave benefits were not included in this study. There may be collinearity issues among these variables and the collinearity issues should be systematically tested or adjusted in case of affecting the stability of the model results. In addition, although interaction effects between variables (e.g., the interaction between FT and occupational type) are potentially significant, due to limitations in research conditions and time, they are not explored in depth in this study. Future research should address collinearity through methods such as variance inflation factor (VIF) analysis or principal component analysis (PCA) to further examine interaction effects to refine the conclusions.

Conclusion

In this study, the characteristics of CSs-RTWA were categorized into three groups by latent profile analysis, i.e., “poor CSs-RTWA group”, “moderate CSs-RTWA-adjustment group”, and “high CSs-RTWA-harmonization group”, and it was found that the majority of CSs are in poor and moderate levels of CSs-RTWA. FT has a significant effect on the population heterogeneity classification of CSs-RTWA. We must focus on the poor CSs-RTWA group and actively call on social relief and security departments and other livelihood departments to provide targeted financial assistance to those with high FT and implement precise poverty alleviation so as to effectively reduce FT and enhance their CSs-RTWA.