1 Introduction

Cancer diagnosis and treatment usually force survivors to limit their working activity. Following a cancer diagnosis, most Japanese either reduce their work hours, take sick leave, or stop working all together [1,2,3]. In 2010, it was estimated that about 30% (244,976 patients) of all diagnosed cancer patients were of working age (20–64 years of age) [4]. According to an investigation of data from the occupational health register of Japanese companies (which includes approximately 68,000 employees), of the 1,278 cancer survivors who experienced their first episode of sick leave due to cancer during the 12-year follow-up period, 47.1% of them returned to full-time work within 6 months of taking sick leave and 62.3% returned to full-time work within 12 months [1]. Quitting one’s job, reducing work hours, and taking sick leave due to cancer diagnosis and treatment are also associated with a decrease in HR-QOL [5, 6]. Additionally, for cancer survivors, returning to work after treatment can be a difficult and daunting task. Although many interventions to support cancer survivors’ return-to-work (RTW) status have been introduced (mostly in Europe), they have not yet advanced to the level by which statistically significant effects can be verified [7]. A literature review exploring barriers to employment in an investigation of the RTW status of cancer survivors revealed that it was more difficult for irregular employees, women, and those aged 50 and over to return to employment after treatment [8]. Recovery stages, as well as background factors, should be considered with regards to RTW. One study found that factors related to RTW changed in the process of cancer recovery [9] while another study suggested that changes in health-related quality of life (HR-QOL) play an important role in predicting RTW status [10].

For cancer survivors, returning to their work at an early stage may offer many advantages. For example, in a longitudinal study from the Netherlands, less time between cancer diagnosis and RTW was associated with persistently high work functioning [11]. However, for cancer survivors, returning to work does not necessarily mean their being in the best health condition. In a multi-country European study, self-employed survivors more often continued working even during treatment, and, in general, had worse financial outcomes than salaried cancer survivors [12]. Some cancer survivors are primarily obliged to work for their own livelihood. Why people seek to work depends on the person and their situation; however, considering their HR-QOL the ability or opportunity to return to work might be a very important step to recovery.

1.1 Conceptual framework

In this study, we focused on investigating the cross-sectional relationship between RTW status and HR-QOL of cancer survivors and identify the associated factors. In Japan, following hospital discharge, the initial follow-up appointment with the primary care physician (who made the cancer diagnosis) is an opportune time to assess recovery and RTW status. The study was based on the premise that the factors associated with RTW status constituted the following four domains: (1) HR-QOL, (2) social background, (3) medical condition(s), and (4) work context. HR-QOL is a concept that encompasses those aspects of overall QOL (an overall sense of well-being) that can be clearly shown to affect health and includes physical and mental health perceptions and their correlates [13]. Thus, to examine health status from multiple perspectives, five aspects of HR-QOL: (1) symptoms, (2) symptom-related interference, (3) physical function, (4) mental function, and (5) subjective well-being were measured. Health status of cancer survivors at the first follow-up appointment can be connected not only to their social background but also to their medical condition. Additionally, we involved a broad work context from regular (full-time) to irregular (part-time) employment to investigate the RTW status of cancer survivors. Alongside being a means to make a living, one’s occupation is a very important aspect of wellbeing and identity. Henderson [14] regards working as a fundamental human need; that is, even if one is not employed per se, committing oneself to a job, including unpaid work such as volunteering, provides a sense of accomplishment and positively affects wellbeing. We used cancer survivors’ own reporting of the work in which they have been engaged, regarding it as their primary work. To identify factors relating to RTW status, we surveyed whether and how the cancer survivors returned to their work and performed statistical analysis of the data. Deepening our understanding of the situation faced by Japanese cancer survivors, post-treatment, by examining health status from multiple perspectives and revealing the factors underlying RTW status, could help not only to facilitate their return to employment but also improve their HR-QOL.

2 Methods

2.1 Participants and procedures

This was a descriptive study using a questionnaire and a cross-sectional survey. Study participants were outpatients from four major hospitals in Japan that provide advanced cancer treatment. Although a wide range of cancer treatments are offered at those hospitals, the hematology and neurosurgery departments were excluded from the current study because we considered that the patients’ HR-QOL from these two departments would differ significantly from that of other patients. The inclusion criterion was age ≧20 and ≦75 years and the exclusion criteria were history of psychiatric disorder or having affective disorder as judged by a primary nurse or physician. From June, 2016 through November, 2017, nurse investigators recruited participants who were able to cooperate with our survey at the inpatient units of each hospital. Participants received a clear explanation of the study purpose and procedure and were asked to complete an anonymous questionnaire at their first follow-up appointment with their physician after hospital discharge. The participants were notified that the return of a completed questionnaire by post within one week was considered as consent to participate in the research. Before the start of the study, approval was obtained from the institutional review board (IRB) at the university to which the principal investigator belonged (Approval Number: 1040) and from all the IRBs of the hospitals in which the surveys were conducted. This study was strictly conducted to protecting the participants’ rights in terms of both privacy and confidentiality.

2.2 Questionnaire

Based on this study’s conceptual framework, the questionnaire included 14 original question items prepared for this survey that asked about socio-demographic information, disease and treatment-related information, and work-related information in addition to the Japanese versions of three HR-QOL scales: (1) the M.D. Anderson Symptom Inventory (MDASI) that measured symptoms and symptom-related interference; (2) the SF-8 Health Survey Short Form (SF-8) that measured physical function and mental function; and (3) the Ferrans and Powers Quality of Life Index Cancer Version III (QLI) that measured subjective well-being. The QLI measures subjective well-being/overall QOL over the four domains of (1) health and functioning, (2) psychological/spiritual, (3) social and economic, and (4) family [15]. Therefore, this study utilized these four subscales of the QLI as indicators of HR-QOL. As indicated by the subscales of these three HR-QOL instruments, they each take a different approach to measuring QOL or are focused on different aspects of QOL; thus, it was hypothesized that the granular evaluation of the patients’ HR-QOL rendered by the combined use of these scales would optimize the chance of identifying the pertinent factors associated with RTW status.

2.3 Original questionnaire items

The socio-demographic indicators used were: age (years), sex, marital status, length of education, annual household income, medical expense-related burden, and possession of employment insurance. Age was divided into two categories: (1) under 50 years old and (2) over 50 years old (according to the findings of the aforementioned review [8] of barriers related to RTW status of cancer survivors). Education was divided into three categories by length: (1) 9 years (period of compulsory education), (2) 12–15 years, and (3) over 16 years. Annual household income was divided into three categories: (1) less than 3.3 million yen, (2) 3.3–9 million yen, and (3) over 9 million yen (according to the statistical survey of National Tax Agency, an average annual salary in Japan in 2016 was 4.2 million yen). Regarding medical expense-related burden, participants were asked to indicate their view about the amount of burden imposed on their household budget by selecting from three choices: (1) high, (2) too high, and (3) no opinion. Employment insurance refers to a public insurance policy that provides coverage for an individual’s loss of employment. In Japan, not possessing employment insurance indicates that the individual is either unemployed, self-employed, a daily-rated worker, or a farmer.

The disease and treatment-related indicators used were: days after diagnosis, sites of cancer, and therapeutic status. Firstly, days after diagnosis was divided into two categories: (1) under 100 days and (2) over 100 days, based on the seminal work by Weisman and Worden [16], who first suggested that 100 days after a cancer diagnosis marked a point at which there would be a diminishing of the existential plight of cancer; a concept later described by Lee as “exacerbation of thoughts about one’s existence and potential for nonexistence following a diagnosis of cancer” [17]. Secondly, the sites of cancer were divided into three categories, labeled A, B and C, according to the five-year relative survival rate of Japanese, which was based on the 2016–18 data from the Center for Cancer Control and Information Services of the National Cancer Center (2018) [18]. Category A sites (less than 40% survival) were: esophagus, bladder, liver, pancreas, and lung cancers; B sites (50–80% survival) were: gastric, colon, renal, uterus, ovary, and bladder cancers; and C sites (over 90% survival) were: breast and prostate cancers.

Regarding the work-related items, participants were asked to write the primary work in which they were currently engaged, with example responses such as “company employee”, teacher”, and “homemaker” given. Those responses were subsequently allocated as either “regular” or “irregular work”. For RTW status, participants choose from one of four options: (1) “I am working as usual”, (2) “I am working less than usual”, (3) “I am on sick leave”, and (4) “I am not engaged in work.” Participants were asked to further categorize their work as either “physical work” or “mental work” and their work-related stress are either “high” or “low”.

2.4 HR-QOL scales

The MDASI consists of a 13-item symptom scale (MDASI-S) and a 6-item interference scale (MDASI-I) (National Cancer Center, Japan, n.d.). The former assesses the frequently occurring and distressing symptoms, and the latter assesses how much those symptoms interfere with various aspects of the patient’s life. The possible range for both scales is from 0 to 10, with a higher score indicating more severe symptoms and interference during the last 24 h. The reliability of the Japanese version of the MDASI-S and MDASI-I was previously verified and the criterion validities for MDASI-S and MDASI-I correlate with the EORTC QLQ-C 30 [19].

The SF-8 is a shorter alternative form of the SF-36, which yields an eight-scale profile of functional health [20]. The eight items composing the SF-8 are in ordinal response format, and each item can be scored by assigning weights that are estimated from standardized scores from a Japanese general population [21]. Physical functional health summary scores (PCS) and mental functional health summary scores (MCS) are calculated using norm-based scoring. The license agreement of the Japanese version of the SF-8 in this study was approved by iHope International [20].

The QLI consists of two parts that respectively measure satisfaction and importance of QOL (using a 1–6 scale) across 33 paired items covering various aspects well-being. The QLI yields five scores (overall QOL, health and functioning, psychological/spiritual, social and economic, and family) that represent how satisfied someone is with the things in life that they value most [22]. The possible range for the final scores is from 0 to 30, with a lower score indicating that the subjective wellbeing is affected more negatively. The cultural adaptation and the psychometric properties of the Japanese version of QLI were respectively evaluated through cognitive interviewing with cancer patients [23] and a questionnaire survey of cancer patients [24].

2.5 Statistical analysis

Participants who chose the option of “I am not engaged in work” for the question about RTW status or simply listed “homemaker” as their current work were excluded from the study analyses because, given their background characteristics, the former was not expected to return to work in the future and for the latter, criteria for selecting RTW status was unclear. Differences between the remaining three categories of RTW status (working as usual, working less than usual, on sick leave) regarding socio-demographic variables, disease and treatment-related variables, work-related variables, and HR-QOL scores were analyzed using Chi-square test or Kruskal Wallis test because of the sample size. To guide the regression analysis, residual analysis and Bonferroni’s correction (to avoid Type I error) were used for multiple post-hoc comparisons after Chi-square test and Kruskal Wallis test, respectively. In the last phase, multinomial logistic regression analysis was carried out to verify the effects of relevant factors on RTW status. SPSS Statistics 28 (IBM, New York, NY, USA) was used for the statistical analysis procedures. For all analyses, two-sided probability values (P) below 0.05 were considered significant.

3 Results

3.1 Characteristics of study sample

A total of 293 questionnaires were distributed during the survey period, of which 176 (60.1%) were returned. Thirty questionnaires with significant defects were excluded from the analyses. Of those 30 questionnaires, 15 had many missing values, 17 had no response to the question about RTW status, and two were duplicates. A further 52 questionnaires were excluded from the analyses because the participants chose the option of “I am not engaged in work” (29 cases) or simply listed “homemaker” as their current work (23 cases). Finally, 94 (32.1%) questionnaires were used for the analysis.

Participant characteristics, based on data from the self-administered questionnaire, are shown in Table 1. The average age of the participants was 60.1 years (SD, 8.7) (data not shown), and 52 (55.3%) were male. Regarding annual household income, 43 (45.7%) of the participants selected the category of “3.3–9 million, yen” and 30 (31.9%) selected the category of “over 9 million yen”. With regards medical expense-related burden, 7 (7.4%) of participants answered that medical expenses were “too high”, and 65 (69.1%) of the participants chose “no opinion”. The average number of days after diagnosis was 236.0 (SD, 635.7) days (data not shown), and 42 (44.7%) of the participants were still under treatment. Eighty (85.1%) of the participants had regular employment. The number of participants who were working as usual was 25 (26.6%), working less than usual was 27 (28.7%), and on sick leave was 42 (44.7%). Of two kinds of work category options, 47 (50.0%) of the participants selected “physical work” of and 45 (47.9%) selected “mental work”.

Table 1 Characteristics of study sample
Full size table

3.2 Factors related to return-to-work

Among the socio-demographic variables (age, sex, marital status, length of education, annual household income, medical expense-related burden, and possession of employment insurance), disease and treatment-related variables (days after diagnosis, sites of cancer, and therapeutic status), and work-related variables (job style, work category, and work-related stress), only the work category (physical work and mental work) revealed a significant interaction among three RTW status categories (χ2 = 13.70, df = 2; P = 0.001). Compared to cancer survivors who were doing physical work, those who were doing mental work were more likely to be working as usual and were less likely to be on sick leave (Table 2).

Table 2 Cross table of work category and return-to-work
Full size table

Summary of the statistical analyses of HR-QOL scores on RTW status is shown in Table 3. HR-QOL scores corresponded to RTW status with scores being from highest to lowest respectively for cancer survivors working as usual, those working less than usual, and those on sick-leave. However, only the four factors of symptom-related interference, physical function, mental function, and health and functioning, identified through the MDASI-I, SF-8 PCS, SF-8 MCS, QLI-HF scales, respectively, had significant interaction among three RTW status categories according to Kruskal–Wallis’s test (H = 17.25, P < 0.001; H = 10.62, P = 0.005; H = 13.98, P < 0.001; and H = 8.10, P = 0.017, respectively). In the post-hoc multiple comparisons, (a) interference and mental function revealed significant differences between cancer survivors working as usual and those working less than usual (P = 0.030 and P = 0.001, respectively) and between cancer survivors working as usual and those on sick-leave (P < 0.001, and P = 0.013, respectively); and (b) physical function and health and functioning revealed significant differences between cancer survivors working as usual and those on sick-leave (P = 0.007 and P = 0.013, respectively).

Table 3 Summary of means, standard deviations, and differences of scores of HR-QOL scales on return-to-work
Full size table

3.3 Factors predictive of return-to-work

In the multinomial logistic regression model, only work category and the HR-QOL scales, which were significantly related to RTW status, were set as independent variables. The model revealed that only the factor of “work category” was a variable with significant effects on RTW status (Table 4). The model fit statistic was supported by Chi-square testing (χ2 = 34.60, df. = 10, p < 0.001) and rates of contribution in the model were determined using Cox-Snell, Nagelkerke, and McFadden (R2 = 0.31; R2 = 0.36; and R2 = 0.18, respectively). In the model where the RTW status category “working as usual” was a reference category, those engaging in a physical work were 4.98 times (95% CI 1.29, 19.15) more likely than those engaging in a mental work to be working less than usual and were 7.07 times (95% CI 1.87, 26.79) more likely to be on sick leave.

Table 4 Variables having significant effects on return-to-work at multinomial logistic regression analysis
Full size table

4 Discussion

As mentioned earlier, in a cohort study of Japanese cancer survivors, Endo and colleagues found that 47.1% and 62.3% of survivors returned to work within 6 and 12 months after initial day of sick leave, respectively [1]. Similarly, for the participants of the current study, cancer survivors with both regular and irregular employment had difficulty returning to work as usual at an early stage following their hospital discharge; only 26.6% reported that they had returned to work by the first follow-up appointment with their physician and 44.7% of them were on sick leave.

While the previously-mentioned literature study indicated that barriers to returning to work after cancer treatment included being irregular employees, women, and aged 50 and over [7], the current study did not show these factors to be related to RTW status. The finding of the multi-country European study showed that self-employed survivors more often continued working even during treatment than salaried cancer survivors [12]; however, in the current study, possession of employment insurance (for which sole proprietors are not eligible) and differences in annual household income showed no association with RTW status.

The HR-QOL of cancer survivors is closely related to their RTW status. Some studies have evidenced that returning to work had a beneficial effect on HR-QOL [25], and that being on sick leave leads to the vitiation of various aspects of HR-QOL [6]. In the current study, although symptoms and subjective well-being did not show statistical differences with respect to RTW status, symptom-related interference and mental function revealed significant interactions not only between cancer survivors working as usual and those on sick-leave but also between cancer survivors working as usual and those working less than usual. Physical function and health and functioning showed significant interactions between cancer survivors working as usual and those on sick-leave. However, our multinomial logistic regression analysis revealed only the factor of “work category” to have a significant effect on RTW status. Thus, HR-QOL, including physical and mental functioning was associated with, but not predictive of, RTW status. Our analyses showed that RTW status was impacted only by the factor of being engaged in physical work.

Cancer survivors engaging in a physical work were almost five times more likely than those engaging in mental work to be working less than usual or to be on sick leave. For years, studies have shown that work category is associated with the RTW status of cancer survivors. For example, Handschel and colleagues showed that blue-collar workers managed to return less frequently and later to work than white-collar workers [26]. Our finding is in accord with those of Takahashi and colleagues, who, in their investigation of job resignation after cancer diagnosis, found that the most frequently given reason for a cancer survivor leaving their job after returning to their employment was being physically unable to perform their work [3]. It could be hard for cancer survivors to change the work category in which they have been employed; thus, nurses could help facilitate a patient’s RTW status by advising them about the adjustment they could make to ease the physical burden of work.

For cancer survivors to return to work, an important initial step might be to set out to work less than usual—through a reduced schedule or lighter workload—before striving to work as usual. For example, one study showed that accommodations that allow cancer survivors to work fewer hour or ease the burden of work could generate health benefits [27]. In the current study, RTW status was more closely associated with symptom-related interference, physical function, mental function, and health and functioning rather than the socio-demographic and disease- and treatment-related factors. Of these, the factors of symptom-related interference and mental function were significantly related to the RTW status category of working less than usual. Thus, when planning support for cancer survivors to return to work, it may be important to assess the attainable level of RTW status attainable for each individual patient considering their different socio-demographic backgrounds and treatment status regarding life- and health-related functions, specifically, symptom-related interference and mental function.

In Japan, most cancer treatment is provided at general hospitals or specialized cancer hospitals. In such hospitals, cancer survivors increasingly have the opportunity to receive advice on employment, such as how to return to work, and use various social insurance programs from nationally certified labor and social security attorneys. If adequately assessment of cancer survivors’ HR-QOL, including symptom-related interference and mental function can be made, it can ease collaboration between nurses and other professionals such as labor and social security attorneys as well as pharmacists and managerial dieticians and, ultimately, facilitate the provision of effective support for the survivors. In addition, specialist nursing care, such as a wound, ostomy, continence, and lymphedema care can reduce the effects of symptom-related interference. In Japan, however, such specialist nursing support is usually only available at large hospitals.

The current study adds to our understanding of the challenges faced by cancer survivors in continuing their careers by highlighting that “work category” is the key factor in this journey; the employment of those who do physical work was far more likely be negatively impacted by the cancer diagnosis and treatment. Given that “work category” is such an important factor, a multidisciplinary approach involving physicians, nurses, and physical and occupational therapists, could be effective for preparing the cancer survivor to return to work. Communications between the cancer survivor, the care team, and the survivor’s employer to discuss strategies for reintegrating into the workplace could be very effective. Guidance for employers, such as those issued by the Japanese Ministry of Health, Labour and Welfare [28], on ways to effectively reintegrate the cancer survivor back into the workplace, for example, by allocating less physically demanding work, could be an effective strategy. As Endo and colleagues suggest, it is important for companies to improve their RTW support systems for cancers survivors [1]. Finally, to enable more cancer survivors to return to work in a favorable physical and mental health condition, it is necessary for the care team to assess HR-QOL from multiple perspectives, particularly focusing on symptom-related interference and mental function. The granularity of the findings regarding the various aspects of cancer survivors’ HR-QOL as provided by the three HR-QOL instruments used in this study evidence this assertion.

5 Research limitations

The participants of this study were recruited via convenience sampling at major Japanese cancer hospitals; therefore, the applicability of the findings to different populations, such as patients who have undergone treatment at a local hospital in rural areas, should be examined in future research. The fact that the analysis was conducted with a limited sample size is a barrier to generalizing the findings of this study. Additionally, considering the heterogeneity of the participants’ social backgrounds, cancer sites, stages, and treatments, further research is necessary using a larger and more homogeneous sample to generalize any results.

6 Conclusion

In our study, we examined the RTW status and HR-QOL of cancer survivors at their first follow-up appointment with their physician after discharge from hospital. We found that RTW status was significantly associated with work category, symptom-related interference, physical function, mental function, and health and functioning. RTW status was chiefly impacted by the factor of “work category”, and the HR-QOL aspects of symptom-related interference and mental function were significantly related to the RTW status category of working less than usual. Our findings suggest that to help cancer survivors return to work, it is important to consider the work category they are engaged in. Interventions that are focused on the individual’s health functioning including physical and mental functions rather than social background factors are likely to be valuable for promoting cancer survivors’ return to the workforce.