Originals
The Impact of Breast Cancer on Employment among Japanese Women
2014 Volume 56 Issue 1 Pages 49-55
Details
2014 Volume 56 Issue 1 Pages 49-55
Objectives: Breast cancer (BC) is a cancer that affects working age women in Japan. The aim of this study was to examine the impact that BC has on the work-related life of Japanese women and identify factors that correlate with job resignation. Methods: A cross-sectional Internet survey of cancer survivors in Japan was conducted from December 2011 to February 2012. The questionnaire included questions regarding demographic characteristics, impact of cancer diagnosis and treatment on job resignation and consultation behavior of respondents regarding work-related issues. This study reports results obtained from 105 respondents with BC. Results: The mean age of respondents at diagnosis was 42.5 ± 6.4 years, and the median time since diagnosis was 40 months. Thirty-one respondents (29.5%) lost their jobs, and 12 could not find another job after BC diagnosis. Nearly half of the respondents (47.6%) reported a decrease in personal income after diagnosis. Multiple logistic regression analysis revealed that contract or part time workers were significantly more likely to lose their jobs compared with regular, full time workers (odds ratio, 2.52; 95% confidence interval, 1.39 to 4.55; p<0.001). Seventy-nine respondents (75.2%) consulted someone regarding work-related issues. The most frequently consulted person was the boss at the workplace. Conclusions: Our findings suggest that women with BC experience various job-related problems. In order to create a supportive work environment for BC survivors, focus should be placed on facilitating communication and coordination between BC survivors, healthcare providers and coworkers.
(J Occup Health 2014; 56: 49–55)
Employment serves an essential role in the quality of life of cancer survivors1–3). However, cancer survivors often face various work-related issues during the course of illness, including difficulties returning to work, decreased wages and job loss4, 5). Among these issues, work-related difficulties are widely regarded as an important theme of survivorship research and support practice6, 7). National cancer policies in Japan have also placed focus on effective countermeasures against work-related issues experienced by cancer survivors. In the Basic Plan to Promote Cancer Control Programs, which was approved in June 2012, “support for working survivors” was newly incorporated as an issue requiring particular attention8). Despite this, only a few studies have reported on the impact of cancer on Japanese survivors and factors that correlate with job resignation. Moreover, since Japan has signed but not ratified the Convention on the Rights of Persons with Disabilities, employers have no legal obligation to provide reasonable accommodations for disabled individuals.
In the present study, we decided to focus on breast cancer (BC) for two reasons. First, BC presents a major health problem for Japanese women, with the age-adjusted incidence rate of the disease being the highest among cancers9). An estimated 60,000 Japanese women are newly diagnosed with BC (including BC in situ) every year9). Second, BC is a cancer that affects working age women in Japan and has the highest age-specific incidence rate for women aged 40–69 years10).
This study aimed to examine the impact of a BC diagnosis on the work-related life of Japanese women and to identify factors that correlate with job resignation.
We conducted an anonymous, cross-sectional Internet survey with a convenience sample of cancer survivors in Japan from December 2011 to February 2012. A description of the survey was sent to newsletters and mailing lists of cancer self-help groups. The description was also published in newspapers and magazines to recruit cancer survivors who were not associated with self-help groups. We checked all IP addresses to ensure that there was no overlap among respondents. Eligibility criteria were (1) being diagnosed with cancer and (2) age ≥20 years.
A total of 431 cancer survivors responded to the survey, including 137 with BC. Of the 137 BC survivors, 32 were excluded for the following reasons: 11 had incomplete data, 12 were not employed at diagnosis and 9 were self-employed. Data from the remaining 105 BC survivors were analyzed.
The questionnaire survey had three parts. First, respondents were asked questions regarding the following demographic characteristics: sex, age at diagnosis (years), time since diagnosis (months), religion (yes/no), presence of dependent family members (yes/no), education (junior high school, high school, junior college, university/graduate school and other) and cancer site and type (stomach, colorectal, esophagus, liver, gall bladder and bile duct, pancreatic, oral cavity and pharynx, larynx, lung, breast, uterine cervix, uterine corpus, ovarian, prostate, testicular, bladder, maxilla and mandible, brain, thyroid, skin, malignant lymphoma, leukemia, pediatric and other). Second, respondents were asked questions about the following factors related to the impact of the cancer diagnosis on work: change of workplace (“work for the same department at the same company”, “work for another department at the same company”, “resigned and re-employed”, “resigned and not re-employed” and “other”), the decision-maker when respondents resigned from their jobs at diagnosis or moved to another department of the same company (“according to own free will”, “ordered to do so by the company” and “other”), employment status at diagnosis and at the time of survey (“regular, full time worker”, “contract worker”, “part time worker”, “self-employed”, “unemployed including housewife”, “student” and “other”), type of work at diagnosis and at the time of survey (“specialist and technical worker”, “administrative and managerial worker”, “clerical worker”, “sales worker”, “service worker”, “security worker”, “agriculture, forestry and fisheries worker”, “transport and communication worker”, “production process and related worker” and “other”), the number of employees at diagnosis and at the time of survey (“less than nine people”, “10–49 people”, “50–99 people”, “100–299 people”, “300–999 people” and “1,000 people and over”), presence of an occupational health physician at diagnosis and at the time of survey (“yes”, “no” and “do not know”) and changes in personal and family income (“increased”, “no change” and “decreased”). For respondents who resigned from their jobs at diagnosis and were later re-employed, the questionnaire further asked about whether their new employer knew about the respondent's cancer history (yes/no). Third, respondents were asked about following: whether or not they consulted someone about work-related issues (yes/no), the person they consulted (“family”, “friend”, “social worker”, “member of cancer self-help group”, “physician in charge of cancer treatment”, “nurse”, “occupational health physician/nurse”, “boss at workplace”, “coworker”, “human resource department staff” and “other”) and the reason for not consulting someone regarding work-related issues (“the problem was not serious enough to consult with someone”, “did not know what to consult about”, “did not have a specific person to consult with”, “did not have the time to consult with someone”, “did not have energy to consult with someone”, “did not expect to get good advice”, “did not think to consult with someone”, “did not want anyone to know about the cancer diagnosis”, “feared being disadvantaged by consulting with someone” and “other”).
Statistical analysisQuestionnaire results were collated and analyzed using descriptive statistics. Multiple regression analysis was performed to identify factors that correlate with job resignation at diagnosis. Independent variables for the multiple logistic regression model included educational background (junior college or more vs. high school or less), religion (yes/ no), presence of dependent family members (yes/ no), type of employment at diagnosis (full time vs. contract or part time), type of work at diagnosis (professional, technical or managerial vs. other), number of employees at the company (<50 vs. >50), presence of an occupational health physician (yes vs. no or do not know) and consulting behavior regarding work-related issues (yes/ no). Spearman's correlation coefficients were <0.5 between each independent variable. All reported p values were two-tailed and all data analyses were conducted using SAS, version 9.3 (SAS Institute, Inc., Cary, USA). This study was approved by the ethics committee of Dokkyo Medical University.
Demographic characteristics of the respondents are shown in Table 1. All respondents were female (mean age at diagnosis, 42.5 ± 6.4 years; median time since diagnosis, 40 months; range, 1–185 months). At diagnosis, 24 respondents (22.9%) had dependent family members, 68 (64.8%) were full time workers, 34 (32.4%) had access to an occupational health physician at their workplace, 47 (44.8%) were specialists and technical workers and 43 (40.9%) were clerical workers.
| Characteristics | Value (%) |
|---|---|
| Mean age at diagnosis, years (± SD) | 42.5 ± 6.4 |
| Median time since diagnosis, months | 40 |
| Education | |
| Junior high school | |
| High school or less | 19 (18.1) |
| Junior college | 27 (25.7) |
| College or more | 58 (55.2) |
| No answer | 1 (1.0) |
| Religion | |
| Yes | 9 (8.5) |
| No | 95 (90.5) |
| No answer | 1 (1.0) |
| Presence of dependent family members | |
| Yes | 24 (22.9) |
| No | 81 (77.2) |
| Employment status at diagnosis | |
| Full time | 68 (64.8) |
| Contracted/part time | 37 (35.2) |
| Type of work | |
| Professional/technical/managerial | 50 (47.6) |
| Other | 55 (52.4) |
| Number of employees in the workplace | |
| >9 people | 18 (17.1) |
| 10–49 people | 38 (36.2) |
| 50–100 people | 11 (10.5) |
| 100–299 people | 15 (14.3) |
| 300–999 people | 9 (8.6) |
| <1,000 people | 14 (13.3) |
| Presence of occupational health physician at workplace at diagnosis | |
| Yes | 34 (32.4) |
| No | 57 (54.3) |
| Do not know | 14 (13.3) |
| Work after diagnosis | |
| Work for the same department at the same company | 60 (57.2) |
| Work for another department at the same company | 14 (13.3) |
| Resigned from the job at diagnosis | 31 (29.5) |
| Re-employed | 19 (61.3)* |
| Not re-employed | 12 (38.7)* |
| Personal income | |
| Decreased | 50 (47.6) |
| No change | 39 (37.2) |
| Increased | 16 (15.2) |
| Type of work at diagnosis and at the time of survey | |
| Specialist and technical worker | 47 (44.8) |
| Administrative and managerial worker | 3 (2.9) |
| Clerical worker | 43 (40.9) |
| Sales worker | 4 (3.8) |
| Service worker | 4 (3.8) |
| Transport and communication worker | 4 (3.8) |
Of the 105 respondents, 60 (57.2%) continued to work for the same department at the same company after diagnosis, 14 (13.3%) were assigned to another department in the same company and 31 (29.5%) resigned from their jobs at diagnosis. Of the 31 who resigned, 19 (61.3%) were re-employed, while 12 (38.7%) had not been re-employed at the time of the survey. Of the 45 who resigned from their jobs at diagnosis or moved to another department at the same company, 22 (48.9%) answered they did so according to their own will, and 17 (37.8%) answered they were ordered to do so by the company. Of the 19 who were re-employed, five (26.3%) did not inform their new employers about their history of BC. Regarding employment status, 52 (76.5%) of 68 respondents who worked on a regular, full time basis at diagnosis were full time workers at the time of the survey, while the remaining 23.5% changed their employment status to either contract/part time, self-employed or unemployed. Similarly, 9 (53.0%) of 17 respondents who were contract workers at diagnosis and 14 (70.0%) of 20 respondents who worked part time remained in the same category at the time of the survey (Table 2).
Personal income decreased for 50 respondents (47.6%) and increased for 16 respondents (15.2%) after diagnosis. The proportion of respondents whose personal income decreased was significantly higher among those who resigned from their jobs at diagnosis compared with those who remained in the same company (80.6% vs. 33.8%, p<0.001).
Factors that correlate with job resignation at diagnosisMultiple logistic regression analysis revealed that contract/part time workers were significantly more likely to resign from their jobs at diagnosis compared with regular, full time workers (odds ratio, 2.52; 95% confidence interval [CI], 1.39 to 4.55; p<0.001) (Table 3).
Consulting behavior regarding work-related issuesIn response to the question, “Have you ever consulted with someone regarding your work-related issues?” 79 respondents (75.2%) answered yes. Table 4 lists the individuals with whom respondents consulted about their work-related issues (multiple answers). The most frequently consulted person was the boss at the workplace (41 respondents), followed by family (23 respondents) and friends (20 respondents).
Table 5 lists the reasons for not consulting with someone regarding work-related issues (multiple answers). “The problem was not serious enough to consult with someone” was the most frequently chosen reason (15 respondents), followed by “did not think to consult with someone” (7 respondents) and “did not have a specific person to consult with” (5 respondents).
The focus of this study was to assess the impact of a BC diagnosis on the work-related life of Japanese women, identify factors that correlate with job resignation and reveal consulting behavior regarding workrelated
| Employment status at diagnosis | Employment status at the time of survey (%) | ||||||
|---|---|---|---|---|---|---|---|
| Regular, full-time worker | Contract worker | Part-time worker | Self-employed | Unemployed* | NA | ||
| Regular, full-time worker | 68 | 52 (76.5) | 1 (1.5) | 5 (7.3) | 8 (11.8) | 2 (2.9) | 0 (0.0) |
| Contract worker | 17 | 1 (5.9) | 9 (53.0) | 3 (17.6) | 1 (5.9) | 2 (11.7) | 1 (5.9) |
| Part-time worker | 20 | 1 (5.0) | 1 (5.0) | 14 (70.0) | 0 (0.0) | 4 (20.0) | 0 (0.0) |
| Total | 105 | 54 (51.4) | 11 (10.5) | 22 (21.0) | 9 (8.6) | 8 (7.6) | 1 (0.9) |
NA: No answer.
| Variable | OR | 95% CI |
|---|---|---|
| Age at diagnosis | 0.94 | 0.87–1.02 |
| Education (0=junior college or more, 1=high school or less) | 2.35 | 0.59–9.35 |
| Religion (0=yes, 1=no) | 3.22 | 0.31–33.79 |
| Presence of dependent family members (0=yes, 1=no) | 0.59 | 0.18–1.97 |
| Employment status at diagnosis (0=full time, 1=contract/part-time) | 2.52* | 1.39–4.55 |
| Type of work (0=professional/technical/managerial, 1=other) | 0.69 | 0.20–2.31 |
| Number of employees in the workplace (0=50 and over, 1=less than 50) | 1.07 | 0.33–3.16 |
| Presence of occupational health physician at workplace at diagnosis (0=yes, 1=no) | 1.90 | 0.52–6.92 |
| Consulting regarding work-related issues (0=yes, 1=no) | 1.13 | 0.36–3.55 |
OR, odds ratio; CI, confidence interval. Each variable was adjusted for all other factors listed.
| N (%) | |
|---|---|
| Boss at workplace | 41 (51.9) |
| Family | 23 (29.1) |
| Friend | 20 (25.3) |
| Physician in charge of cancer treatment | 20 (25.3) |
| Member of cancer self-help group | 12 (15.2) |
| Other | 10 (12.7) |
| Colleague | 6 (7.6) |
| Nurse | 5 (6.3) |
| Social worker | 5 (6.3) |
| Occupational health physician/nurse | 4 (5.1) |
| Human resource department staff | 4 (5.1) |
| N (%) | |
|---|---|
| The problem was not serious enough to consult with someone | 15 (60.0) |
| Did not think to consult with someone | 7 (28.0) |
| Did not have a specific person to consult with | 5 (20.0) |
| Did not have energy to consult with someone | 4 (16.0) |
| Did not expect to get good advice | 3 (12.0) |
| Feared being disadvantaged by consulting with someone | 3 (12.0) |
| Did not know what to consult about | 2 (8.0) |
| Did not have the time to consult with someone | 2 (8.0) |
| Other | 2 (8.0) |
| Did not want anyone to know about the cancer diagnosis | 1 (4.0) |
We found that a BC diagnosis and BC treatments significantly impacted the work-related lives of BC survivors. Nearly 30% of respondents resigned from the company they worked for at diagnosis and a considerable number of respondents changed their employment status. This was accompanied by a reduction of personal income in nearly half of the respondents. These results are consistent with previous studies on the influence of cancer on the work- related lives of survivors. In a survey of Korean BC survivors, 14.4% resigned from the company they worked for at diagnosis11). In a questionnaire survey of 2,625 Japanese cancer patients who worked at diagnosis, 910 (34.7%) voluntarily resigned or were laid off from the company they worked for at diagnosis12). Our present findings are in agreement with such studies, in that a BC diagnosis impacts work-related life.
Our present survey revealed that 48.9% of respondents resigned or moved to another department according to their own will. However, this result is difficult to interpret because we cannot rule out the possibility that some survivors may have been forced into voluntary retirement by their companies. Future studies should address this by assessing the respondents’ level of acceptance of the decision.
Particularly noteworthy is the finding that 26.3% of respondents did not inform their new employers about their BC history. These respondents may have been afraid of being disadvantaged by disclosing their health status and history. This is understandable given the social image in Japan that cancer is a disease with a poor prognosis13).
Our present survey also revealed that 47.6% of respondents experienced a decrease in personal income after diagnosis. Such findings have been reported in studies from abroad14, 15) and in Japan16). The importance of providing information to survivors and families regarding assistance with medical expenses cannot be overemphasized. Interestingly, 15.2% of respondents reported an increase in personal income after diagnosis. Further studies are needed to clarify the diverse nature of income changes after a BC diagnosis.
Regarding employment status, about one fourth (23.5%) of respondents who worked on a regular, full time basis at diagnosis changed their employment status to either contract/part time, self-employed or unemployed. Further studies examining the rationale for changing employment status are warranted.
Multiple logistic regression analysis revealed that contract/part time workers were significantly more likely to resign from their jobs compared with regular, full time workers. This result is consistent with previous studies of BC survivors revealing the disadvantaged position of contract/part time workers. For example, a Canadian retrospective cohort study reported that having a part time job was a predictor of not working three years after a BC diagnosis17). Also, working full time was a significant predictor for returning to work after a BC diagnosis18). This finding highlights the insecure employment conditions of non-regular workers. Since more than half of Japanese women worked on a non-regular basis (contract/part time work) in 201019), supportive measures for BC survivors who work on a non-regular basis are an urgent issue.
With respect to consulting behavior, 75.2% of respondents consulted with someone regarding work- related issues. The most frequently chosen reason for not consulting with someone was “the problem was not serious enough to consult with someone.” Many of the respondents may have devised ways to deal with work-related issues and somehow managed to overcome the situation on their own. Sharing such measures may help other BC survivors cope with their work-related issues. Some respondents answered “Did not think to consult with someone” and “Did not have a specific person to consult with”. In order to facilitate consultation behavior, easy access to a supportive and reliable adviser regarding work-related issues would be necessary.
The boss at the workplace was the most frequently consulted person regarding work-related issues. This result underscores the importance of empowering the boss at the workplace to serve as an adviser regarding work-related issues. Occupational health staff may play an important role as a support resource such that bosses can function as both good advisers and providers of work-related information.
Our study has several limitations worth noting. First, the respondents were not a representative sample of working BC survivors in Japan. Our respondents may have been biased to survivors who had experienced work-related difficulties and thus were more interested in the survey. Moreover, the respondents of the present study had a higher education level than average female workers20), and the proportion of respondents who were contract/part time workers was lower than that reported in the Labour Force Survey conducted by the Ministry of Internal Affairs and Communications21). Therefore, caution should be exercised when generalizing the findings of this study. Second, there was substantial variation (1–185 months) regarding the time since diagnosis. Respondents who had a longer time since diagnosis may have resigned from their jobs for non-cancer-related reasons. In addition, recall bias may exist given that some respondents were diagnosed over a decade ago. In the future, we will need to conduct longitudinal studies to examine the timing of and reasons for job resignation. Third, because respondents were recruited from outside clinical settings, information regarding cancer stage could not be used in the analysis. Previous studies have reported that advanced cancer stage is an important factor that correlates with unemploy- ment11, 22). However, as patients with early-stage BC often receive combined modality therapy, cancer stage may have a smaller impact on their work-related lives.
Despite these limitations, we believe that our findings provide important information regarding the impact that a BC diagnosis has on the work-related life of Japanese women. In order to provide effective support, more research and efforts to facilitate collaboration among health-care providers, coworkers, occupational health staff and survivors themselves are needed.
Acknowledgments: This work was supported by a Japan Society for the Promotion of Science (JSPS) Grant-in-Aid for Scientific Research (B) (KAKENHI No. 20390158 to M.T.) and a Grant-in-Aid for Cancer Research from the Japanese Ministry of Health, Labour and Welfare (H22-ganrinsho-ippan-008 to M.T., H25-ganrinsho-ippan-004 to M.T.).
