Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Work ability and return-to-work in cancer patients

Work ability and return-to-work in cancer patients Clinical Studies British Journal of Cancer (2008) 98, 1342 – 1347 & 2008 Cancer Research UK All rights reserved 0007 – 0920/08 $30.00 www.bjcancer.com ,1 1,2 1,3 4 5,9 6 AGEM de Boer , JHAM Verbeek , ER Spelten , ALJ Uitterhoeve , AC Ansink , TM de Reijke , 1,7 8 1 M Kammeijer , MAG Sprangers and FJH van Dijk 1 2 Coronel Institute for Occupational Heath, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Cochrane Collaboration 3 4 Occupational Health Field, Kuopio, Finland; NPVO, Amsterdam, The Netherlands; Department of Radiotherapy, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Department of Gynaecology and Obstetrics, Academic Medical Center, University of Amsterdam, 6 7 Amsterdam, The Netherlands; Department of Urology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Jan van Breemen Institute, Amsterdam, The Netherlands; Department of Medical Psychology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands The extent to which self-assessed work ability collected during treatment can predict return-to-work in cancer patients is unknown. In this prospective study, we consecutively included employed cancer patients who underwent treatment with curative intent at 6 months following the first day of sick leave. Work ability data (scores 0–10), clinical and sociodemographic data were collected at 6 months, while return-to-work was measured at 6, 12 and 18 months. Most of the 195 patients had been diagnosed with breast cancer (26%), cancer of the female genitals (22%) or genitourological cancer (22%). Mean current work ability scores improved significantly over time from 4.6 at 6 months to 6.3 and 6.7 at 12 and 18 months, respectively. Patients with haematological cancers and those who received chemotherapy showed the lowest work ability scores, while patients with cancer of urogenital tract or with gastrointestinal cancer had the highest scores. Work ability at 6 months strongly predicted return-to-work at 18 months, after correction for the influence of age and treatment (hazard ratio¼ 1.37, CI 1.27–1.48). We conclude that self-assessed work ability is an important factor in the return-to-work process of cancer patients independent of age and clinical factors. British Journal of Cancer (2008) 98, 1342–1347. doi:10.1038/sj.bjc.6604302 www.bjcancer.com Published online 18 March 2008 & 2008 Cancer Research UK Keywords: employment; work ability; return-to-work; longitudinal studies; prospective studies Cancer diagnoses in individuals who are still at the working age are work in cancer survivors seems, therefore, to be problematic in becoming more common, with almost half of the adult cancer some patients but certainly not in all. Hence, it is important to survivors being younger than 65 years (Short et al, 2005). With the identify those patients with a higher risk of lasting absence from sustained improvement in treatment and prognosis of many forms work and to provide them with the appropriate support and of cancer, an increasing number of survivors of cancer return-to- counselling in returning to work. work following treatment or continue to work during therapy To examine the factors that would influence this return-to-work (Hoffman, 2005). process, we previously studied a model based on the assumption Returning to work is important for both cancer patients that cancer-related symptoms would mediate return-to-work themselves and the society. Patients often regard returning to (Spelten et al, 2003). However, results showed that diagnosis and work as a symbol of complete recovery (Spelten et al, 2002) and treatment were much stronger predictors of return-to-work than regaining a normal life (Kennedy et al, 2007), while from the cancer-related symptoms such as fatigue, depressive symptoms or viewpoint of the society, it is an economic and social imperative to cognitive problems. In addition, recent empirical studies have encourage patients to return-to-work whenever possible. indicated the importance of patients’ expectations of recovery as Despite its importance, the impact of cancer and its treatment good predictors of return-to-work and rehabilitation independent on work (dis)continuation or resumption has not been studied of diagnosis and treatment (Ekbladh et al, 2004; Verbeek, 2006). frequently (Steiner et al, 2004). However, a number of studies have Studies in other disorders have also shown that a patient’s own documented the impact of cancer on employment and they assessment of work ability (Reiso et al, 2003), expectation of job reported that approximately 60% of the cancer patients return to success (Ekbladh et al, 2004) and work recovery expectations work within 1–2 years (Spelten et al, 2002; Maunsell et al, 2004; (Hogg-Johnson and Cole, 2003; Nieuwenhuijsen et al, 2006; Turner Bradley et al, 2005; Nieuwenhuijsen et al, 2006). The return-to- et al, 2006) do predict return-to-work. A theory that could explain these mechanisms is the well-known Leventhal’s ‘model of illness representations’, which states that *Correspondence: Dr AGEM de Boer; E-mail: A.G.deBoer@amc.uva.nl people’s cognitive representations of illness exert an important Current address: Erasmus MC, Daniel den Hoed Oncology Center, influence on their strategies for coping, which in turn influence Rotterdam, The Netherlands. illness outcomes (Leventhal et al, 1984). It has been shown in other Received 6 November 2007; revised 7 February 2008; accepted 11 diseases such as multiple sclerosis, rheumatoid arthritis and February 2008; published online 18 March 2008 kidney disease (Vaughan et al, 2003; Carlisle et al, 2005; Work ability and return-to-work in cancer patients AGEM de Boer et al Fowler and Baas, 2006) that, on the basis of this model, the Physical workload was measured with a seven-item scale and functional outcome might be worse or better, irrespective of work stress with an 11-item scale from the Dutch Questionnaire on the objective medical seriousness of the illness. This strongly Experience and Judgement of Work (VBBA) (van Veldhoven et al, suggests that the ideas a cancer patient has about the disabilities 2002). Patients were asked to assess their levels of workload and that might result from the diagnosis and treatment will encourage work stress for the work situation prior to diagnosis. The scores or hinder his or her return-to-work. range from 0 to 100, with higher scores indicating a higher level of With these new insights, our data were reanalysed with the focus physical work and more work stress, respectively. on the patients’ assessments of work ability as predictor of return- to-work. In our earlier publication on return-to-work of cancer Cancer-related and sociodemographic factors Information about survivors, we did not use information on the self-assessed ability diagnosis and treatment was reported by the patients. Twenty-two to work because at the time it was outside the focus of our study different diagnoses were then grouped according to cancer (Spelten et al, 2003). site into (1) breast cancer, (2) haematological oncology, (3) The aim of the current study is therefore (1) to examine any gastrointestinal cancer, (4) cancer of the female genitals, change in work ability scores in cancer patients over time and to (5) genitourological cancer and (6) other types of cancer. study differences among patient groups and (2) to assess the extent Treatments were classified into three categories: (1) surgery, (2) to which self-assessed work ability predicts return-to-work among radiotherapy or radiotherapy plus surgery and (3) chemotherapy cancer survivors independent of diagnosis, treatment and cancer- or chemotherapy plus radiotherapy and/or surgery. related symptoms. We measured cancer-related complaints with validated ques- tionnaires and converted all scores to a scale ranging from 0 to 100, with higher scores indicating more complaints (Spelten et al, MATERIALS AND METHODS 2003). The following complaints were measured: physical cancer- related complaints (de Haes et al, 1990), general fatigue (Smets Patients et al, 1995), sleep quality (Buysse et al, 1989), depressive Eligible patients had to be between 18 and 58 years to have a symptoms (Radloff, 1977), psychological distress (de Haes et al, primary diagnosis of cancer, to be in paid employment at the time 1990), cognitive dysfunction (Broadbent et al, 1982) and global of diagnosis, to be within 4–6 months following their first day of quality of life (de Haes et al, 1990). sick leave, and to have had treatment with curative intent. They Further information was enquired concerning marital status were consecutively recruited in three hospitals in The Netherlands (single, married, cohabitating or other), having children in the where the attending physician obtained the patients’ informed household, age, gender, education (lower education, high school, consent. The study has been carried out with the approval of the college/university) and work hours per week before the diagnosis hospitals’ medical ethical committees. of cancer. Questionnaires were distributed three times to the patients, at entry into the cohort and 6 and 12 months later, to obtain information on their return-to-work, diagnosis, treatment, work Statistical analysis ability and cancer-related symptoms. Details of the design and The work ability scores measured at 6, 12 and 18 months after the material of this prospective cohort study have been reported first day of sick leave were analysed with the mixed-model earlier (Spelten et al, 2003). The data were collected between 1998 procedure based on repeated measurements to examine any and 2002. For the current study, data on return-to-work and work change in work ability scores over time. We also used the mixed- ability were collected at study entry and 6 and 12 months later. model procedure to analyse any differences over time in work Data on work load, work stress, cancer-related factors and ability scores between several patient groups: age groups (18–27, sociodemographic factors had been collected at baseline. All 28–37, 38–47 and 48–58 years), education groups, men and questionnaires were mailed to the patients’ homes. women, diagnosis groups and treatment groups. Time, group and time group interaction effects were considered fixed effects and an Measures autoregressive covariance structure was selected because of Return-to-work Data on return-to-work were measured on the correlated work ability scores over time. In case of a statistically basis of two measures: time to return-to-work after sick leave and significant main effect, post hoc analyses were performed between rate of return-to-work at a specific point in time. All patients in time points and between groups with pairwise comparisons based The Netherlands typically have access to sick leave. Time to return- on the use of the mean difference of the estimated marginal means. to-work at 18 months after the first day of sick leave was calculated To examine whether self-assessed work ability can predict as the number of days between the first date of sick leave and the return-to-work in cancer patients a year later, taking the impact of first day the patient returned to work. Any kind of work clinical-, work- and subject-related factors into account, we used a resumption qualified as a return-to-work, irrespective of the two-step procedure. First, univariate analyses using Kaplan–Meier number of hours that the patients worked prior to their diagnosis. analyses were performed for the relationship between time taken In addition, patients were asked to indicate if they were still on sick to return-to-work (in days) at 18 months and each of the predictive leave (yes/no) at 6, 12 and 18 months following their first day of factors measured at baseline (on average 6 months after the first sick leave. day of sick leave): current work ability, mental work ability, physical work ability, physical work load, work stress, physical Work ability, work load and work stress Current work ability was complaints, fatigue, sleep impairments, depression, psychological measured with the first three items from the Work Ability Index distress, cognitive dysfunction, age, gender, education and the (WAI) (Ilmarinen and Tuomi, 1993, p 142; Tuomi et al, 1998), clinical factors (diagnosis and treatment type). Next, we analysed which is a reliable and valid measure of work ability (Ilmarinen the impact of work ability in addition to personal and clinical and Tuomi, 1993, p 142; de Zwart et al, 2002). First, current work factors in a multivariate Cox regression analysis. We entered all ability was assessed by asking the patients to estimate their current variables for which the log-rank test returned a P-valuep0.10 into work ability compared with their lifetime best (0¼ cannot work at a Cox regression analysis with forward selection of variables. With all to 10¼ best ever). In addition, we asked the cancer patients to this method, the best predictors of future return-to-work are rate both their current physical and mental work ability in relation selected (Altman, 1991). Because it was possible for patients to to job demands (0¼ very low to 5¼ very high). return to work before our first measurement at 6 months, we & 2008 Cancer Research UK British Journal of Cancer (2008) 98(8), 1342 – 1347 Clinical Studies Clinical Studies AGEM de Boer et al repeated both analyses with the exclusion of patients who had Table 1 Sociodemographic and cancer-related characteristics at base- returned before 6 months. Since this is a survival analysis, hazard line, on average 6 months after the first day of sick leave ratios (HRs) usually indicate the risk of dying, while in our case the N¼ 195 patients n (%) event is returning to work. Therefore, an HR higher than one indicates the higher ‘risk’ of return-to-work. Age (mean (s.d.)) 42.2 (9.3) Alpha was set at 0.05 unless stated otherwise and all tests were Sex (male) 78 (40%) two-sided. Analyses were conducted with SPSS 13. Returned-to-work 46 (24%) Education Lower 52 (27%) RESULTS High school 83 (42%) College/university 60 (31%) The first questionnaire was completed by 235 of the 264 eligible patients (a response of 89%), while a total of 29 patients declined Marital status participation in this study. The second questionnaire at 6 months Single 24 (12%) follow-up was completed by 221 of the 235 participating patients Married/cohabiting 160 (82%) (a follow-up response of 94%). At 12 months of follow-up, the Divorced 8 (4%) questionnaire was returned by 195 patients (an 83% follow-up Widower 3 (2%) response and 74% of the initially eligible patients), while 25 patients refused to return the questionnaire, 13 patients had died Diagnosis and 2 questionnaires got lost in the mail. Breast cancer 51 (26%) Haematological oncology 24 (12%) Table 1 shows sociodemographic and cancer-related characteris- Gastrointestinal cancer 23 (12%) tics at 6 months after the first day of sick leave. Half of the patients Cancer of the female genitals 43 (22%) had either breast cancer (26%) or cancer of the female genitals Genitourological cancer 43 (22%) (22%), while another 22% of the patients had been diagnosed with Other 11 (6%) genitourological cancer. Before diagnosis and treatment, patients worked an average of 34 h per week, and 6 months after the Treatment diagnosis, 46 patients (24%) had already returned to work or had Surgery 41 (21%) continued working. Data on work hours per week, children, Chemotherapy or chemotherapy plus radiotherapy and/or surgery 88 (45%) fatigue, depression, sleep problems, physical complaints, cognitive Radiotherapy or radiotherapy plus surgery 66 (34%) dysfunction, psychological distress, work load and work stress have been reported previously (Spelten et al, 2003). Table 2 depicts the mean values of current work ability at 6, 12 and 18 months after the first day of sick leave; the values improved Table 2 Mean value of current work ability according to socio- significantly over time (Po0.001) from 4.6 at 6 months to 6.3 at 12 demographic and disease-related factors at 6, 12 and 18 months after months and to 6.7 at 18 months. Post hoc analyses of work ability the first day of sick leave scores showed that all three time points were significantly different Workability score (mean (s.d.)) from each other (Po0.001 to P¼ 0.035). All age groups improved over time (Po0.001) with the 28- to 37-year-old patients N¼ 195 patients 6 months 12 months 18 months increasing most from 4.8 to 7.5. No differences in work ability scores were, however, found between age groups (P¼ 0.12). Work All patients 4.59 (3.2) 6.31 (2.7) 6.74 (2.7) ability scores of both men and women improved over time (Po0.001), but women improved more (P¼ 0.002). Male patients Age showed higher work ability scores at 6 months (5.8 vs 3.8, 18 – 27 years 5.33 (3.6) 5.71 (3.9) 7.27 (2.9) Po0.0001), but not at 12 months (6.8 vs 6.0, P¼ 0.053) or at 18 28 – 37 years 4.81 (3.4) 6.80 (2.5) 7.51 (2.4) months (6.9 vs 6.7, P¼ 0.52). Higher educated patients seemed to 38 – 47 years 4.53 (3.2) 6.65 (2.3) 6.80 (2.3) 48 – 58 years 4.31 (3.0) 5.73 (2.8) 5.97 (3.1) have higher work ability scores, but the differences were not statistically significant (P¼ 0.13). With regard to diagnosis, we Sex found significant differences between the different diagnosis Male 5.76 (3.0) 6.78 (2.6) 6.91 (2.8) groups (Po0.001). The haematological oncology patients showed Female 3.83 (3.1) 6.00 (2.7) 6.65 (2.6) the significantly (Po0.001) lowest scores of 3.3, 4.5 and 5.0 at 6, 12 and 18 months, respectively. The patients with genitourological Education cancer had the highest scores of 6.9 and 7.8 at 6 and 12 months Lower 3.90 (3.3) 6.14 (3.0) 6.26 (3.2) (Po0.001), and the patients with gastrointestinal cancer scored the High school 4.51 (3.0) 6.31 (2.6) 6.77 (2.5) highest work ability of 7.6 at 18 months (Po0.001). Patients with College/university 5.33 (3.4) 6.46 (2.6) 7.15 (2.6) cancer of the female genitals and breast cancer patients improved Diagnosis most over time (P¼ 0.01). Breast cancer 3.59 (3.1) 5.90 (2.3) 6.49 (2.5) Figure 1 shows the work ability scores for the three treatment Haematological oncology 3.29 (3.0) 4.46 (3.4) 4.95 (3.6) combinations: (1) surgery; (2) chemotherapy or chemotherapy Gastrointestinal cancer 5.52 (2.7) 6.95 (2.1) 7.57 (1.6) plus radiotherapy and/or surgery; and (3) radiotherapy or Cancer of the female genitals 3.91 (3.1) 6.29 (2.9) 7.00 (2.6) radiotherapy plus surgery, over time. Analyses revealed that scores Genitourological cancer 6.86 (2.5) 7.79 (1.7) 7.33 (2.7) improved over time for all three groups and that the group of Other 4.09 (3.4) 5.36 (2.9) 6.55 (2.5) patients that received chemotherapy or chemotherapy plus radio- a b Range 0 – 10; 10 indicating best work ability ever. Work ability score change over therapy and/or surgery consistently showed lower work ability time: Po0.01. Difference between groups: Po0.01. scores than the group that received surgery or radiotherapy (plus surgery) (Po0.001). Improvement was not statistically different in Results of univariate analyses using the Kaplan–Meier analyses the three groups (P¼ 0.45). showed that the time taken to return-to-work measured at 18 At 6 months after diagnosis, 24% of patients had returned to months was related to the following factors measured at 6 months: work, at 12 months 50%, and at 18 months 64% had returned. British Journal of Cancer (2008) 98(8), 1342 – 1347 & 2008 Cancer Research UK Work ability and return-to-work in cancer patients AGEM de Boer et al Treatment Current work 1.0 ability at Surgery 6 months Radiotherapy (+ surgery 8 0.8 1 Chemotherapy (+ radiotherapy and /or surgery) 0.6 0.4 0.2 0.0 0.00 100.00 200.00 300.00 400.00 500.00 600.00 700.00 6 months 12 months 18 months Time to return to work (days) Time Figure 2 Plot of the work ability scores measured at 6 months in Figure 1 Mean value of current work ability at 6, 12 and 18 months after relation to time to return-to-work, after adjustment for age and treatment the first day of sick leave for the three treatment combinations: surgery; (n¼ 195). chemotherapy or chemotherapy plus radiotherapy and/or surgery; and radiotherapy or radiotherapy plus surgery (n¼ 195). age and work ability. For current work ability itself, every 1 point increase on the 11-point scale meant a 1.37 higher chance of returning to work earlier, after correction for the influence of age Table 3 Cox regression analysis on return to work and treatment. Figure 2 shows the plot of the work ability scores in relation to return-to-work after adjustment for age and treatment. Time to return to work The plot shows that of the patients with the lowest work ability N¼ 195 Hazard ratio 95% CI scores (0–5) at 6 months after the first day of sick leave, the majority (55–80%) did not return to work in the first year after Age, 10-year categories 0.78 0.65 – 0.94 diagnosis. Patients with very high work ability scores (8, 9, 10) did Current work ability 1.37 1.27 – 1.48 usually return to work within half a year, while virtually all of these high-scoring patients were back at work after the first year. Treatment Results of the analysis without the 46 patients who had already Surgery (reference) 1.00 returned to work at 6 months showed that in this model the Chemotherapy plus radiotherapy and/or surgery 0.41 0.25 – 0.69 Radiotherapy/radiotherapy plus surgery 0.63 0.39 – 1.0 strongest predictors of return-to-work were not only work ability (HR¼ 1.23; CI, 1.12–1.36), treatment (chemotherapy HR¼ 0.33; 95% CI¼ 95% confidence interval. CI, 0.18–0.60; radiotherapy HR¼ 0.52; CI, 0.29–0.95) and age (HR¼ 0.67; CI, 0.53–0.86) but also mental work ability (HR¼ 1.41; CI, 1.05–1.89) and cognitive dysfunction (HR¼ 1.03; CI, current work ability, mental work ability, physical work ability, 1.01–1.05). quality of life, fatigue, physical complaints, cognitive functioning, age, physical work load, work stress, gender, diagnosis and treatment (at the Pp0.10 level). Sleep impairments, depression, DISCUSSION psychological distress and education did not significantly predict return-to-work. Results of the analysis without the 46 patients who The aim of our study was to examine changes in work ability had already returned to work at 6 months showed the same factors scores in cancer patients over time and to study differences except for gender, which did not significantly predict return-to- between patient groups and, furthermore, to assess the extent to work. which self-assessed work ability predicts return-to-work among The factors that were predictive for return-to-work at 18 months cancer survivors independent of diagnosis, treatment and cancer- were entered in the Cox regression with a forward selection to related symptoms. identify the strongest predictors of return-to-work. Results in We found that the cancer patients’ work ability scores at 6, 12 Table 3 show that in the final model, age, current work ability and and 18 months after the first day of sick leave improved treatment are still significant. Current work ability, physical work significantly over time. Men scored higher on work ability than ability and mental work ability were highly correlated and, women but no differences were found between age or education therefore, only current work ability remained in the model. groups. Furthermore, the haematological oncology patients and Likewise, treatment and diagnosis were highly correlated, and only the patients who received chemotherapy or chemotherapy plus treatment was selected for the final model. Patients treated with radiotherapy and/or surgery consistently showed lower work surgery alone had the highest chance of returning to work quickly. ability scores. Finally, self-assessed work ability 6 months after Those who were treated with radiotherapy or radiotherapy plus the first day of sick leave proved to be a strong predictor of later surgery had an HR of 0.63 (95% CI: 0.39–1.0), corrected for age return-to-work in cancer survivors independent of age and and work ability, of returning to work and were thus 1.6 times therapy. more likely to stay off work than patients with surgery alone. This is the first longitudinal study in which the impact of work Patients treated with chemotherapy, either alone or in combination ability on return-to-work has been established in a systematic way. with other treatment modalities, had an HR of 0.41 (95% CI: The cohort has been followed for a considerable amount of time, 0.25–0.69) and their risk of staying off work was therefore 2.4 the number of patients lost to follow up was relatively small and all times higher than patients treated with surgery alone, corrected for factors have been measured with validated instruments. & 2008 Cancer Research UK British Journal of Cancer (2008) 98(8), 1342 – 1347 Mean work ability Proportion off work Clinical Studies Clinical Studies AGEM de Boer et al In our study, the mean current work ability scores at 6, 12 and the better predictor of return-to-work. Other studies have also 18 months after the first day of sick leave were 4.6, 6.3 and 6.7, found that fatigue influenced conditions of employment and respectively. Although we found a significant improvement of productiveness (Hofman et al, 2007). Results of the univariate current work ability, these scores are lower than the average analyses without the 46 patients who had already returned to work current work ability score of 7.9 found by Pohjonen (2001) in a at 6 months showed that gender was not a significant factor sample of female home care workers in the age group 40–44 years anymore. This might be caused by the fact that most of these old with an average of two diagnosed benign diseases. It might be returned patients were men (65%) and were diagnosed with testes possible that work ability scores in cancer patients will improve or prostate cancer (48%). In the model of best predictors of return- still further 2 years after the diagnosis or that their work ability to-work without those patients who had returned early, the factors scores might deteriorate because cancer has a larger impact on mental work ability and cognitive dysfunction were included. This work ability than other diseases. could imply that for patients who do not return early, the mental Research on the effect of cancer diagnosis and treatment on and psychological factors become more dominant in relation to work ability is scarce; however, studies have shown recently that return-to-work. most patients are employed but that both physical and mental Leventhal’s ‘model of illness representations’ states that people’s cognitive representations of illness play an important role in work ability can deteriorate owing to cancer (Gudbergsson et al, 2006; Steinbach et al, 2006; Kennedy et al, 2007; Taskila influencing their strategies for coping, which in turn influence et al, 2007). Patients in the recent study of Kennedy et al (2007), illness outcomes (Leventhal et al, 1984). On the basis of this who were 1–10 years after diagnosis, reported that they had model, the functional outcome might be worse or better, difficulties in coping and concentrating, and they worried irrespective of the objective medical seriousness of the illness. about their reduced capability. In the comparative study of Our results are congruent with this model. Irrespective of age, Gudbergsson et al (2006), it was found that cancer patients diagnosis, treatment, quality of life, fatigue, and physical or 2–6 years after diagnosis, who had returned to work after psychological complaints, self-assessed work ability strongly curative treatment, reported significantly poorer physical and predicted future return-to-work. This indicates that the ideas a mental work capacity compared to employed matched controls cancer patient has about his or her work disabilities that result from the general population. Most survivors of glioblastoma in the from the diagnosis and treatment of cancer are a reflection of the study of Steinbach et al (2006) also thought that their work true work capabilities. Therefore, the self-reported work ability ability was impaired. According to Taskila et al (2007), 26% of could be important in encouraging or hindering his or her return- cancer survivors reported that their physical work ability to-work. had deteriorated and 19% that their mental work ability had Employment outcomes can be improved with innovations deteriorated owing to cancer diagnosis and treatment. However, in treatment and with clinical and supportive services aimed at the work ability as measured with the WAI of these better management of symptoms, rehabilitation and accommoda- cancer survivors did not differ from that of a group of healthy tion of disabilities (Steiner et al, 2004). A recent study of referent persons. This is probably also caused by the fact that all Bouknight et al (2006) showed that a high percentage of employed their survivors with breast cancer, lymphoma and prostate cancer breast cancer patients returned to work after treatment and had already returned to work and that they were long-term that workplace accommodations played an important role in their survivors who had been diagnosed with cancer 2–6 years before return. Therefore, interventions should be developed to enable the time of the questionnaire. This could also explain the cancer survivors to return to work or to succeed in differences in mean work ability scores between their study and other appropriate employment, because no such interventions the patients in our study. For men in their study, the work ability aimed at work do exist at the present. These interventions scores were 8.0 (for prostate cancer) to 8.9 (for testicular cancer) should aim especially at patients who indicate that their work compared to 6.9 for the men in our study 18 months after the first ability is diminished, at older patients and at those treated with day of sick leave. Our female patients scored 6.7 at the end of chemotherapy, since they are at the greatest risk of prolonged follow-up compared to 8.2 (for breast cancer) and 8.5 (for work absence. Clinicians could play an important role in detecting lymphoma) in the Finnish study by Taskila et al (2007). Our those patients at risk because our study has shown that the study also showed that men initially showed higher scores of work indication of patients with possible return-to-work problems can ability, while women improved faster and no differences were be assessed very early in the treatment process when they have found after 1 year. It might be possible that women, who were diminished self-reported work ability. Physicians could help mainly diagnosed with breast cancer, received more chemother- patients in the return-to-work process and therefore help in apy, which would have prolonged the treatment period. Another improving their quality of life by asking patients if they have explanation might be that women could have more household returned to work or are experiencing problems in the return-to- activities than men and that they take these into account when work process. If so, referral to occupational specialists could be judging their work ability. considered. Our study indicated work ability as an independent predictor for In conclusion, the work ability of cancer patients who work at return-to-work, while quality of life was only found to be the time of their diagnosis is severely impaired in the first months predictive of time until return-to-work in the univariate analyses. after the first day of sick leave, but it does improve significantly in The same result was found earlier for Norwegian patients with the months afterwards. Self-assessed work ability 6 months after back disorders who had been certified as sick (Reiso et al, 2003). the first day of sick leave proved to be a strong predictor of later The authors of that study suggested that work ability questions return-to-work in cancer survivors independent of age and may be related more to function in a setting of sickness therapy. certification than a global quality of life question and therefore be more predictive. In an earlier analysis (Spelten et al, 2003), we found that fatigue at 6 months predicted a longer sick leave with an HR of 0.71, ACKNOWLEDGEMENTS adjusted for diagnosis, treatment, age and gender. Our present study indicated that fatigue was only a predictive factor of return- This study was supported by a grant from the Dutch Cancer to-work in the univariate analyses but not in the multivariate Society (AMC 97-1385). We thank Dr J van der Lelie for his analyses, which included work ability. Because work ability and contribution to this study. We are grateful to all of the patients for fatigue were correlated, only work ability remained in the model as their participation in this study. British Journal of Cancer (2008) 98(8), 1342 – 1347 & 2008 Cancer Research UK Work ability and return-to-work in cancer patients AGEM de Boer et al REFERENCES Altman DG (1991) Practical Statistics for Medical Research. Chapman and education regarding return to work in cancer survivors following Hall: London curative treatment: a pilot study. J Occup Rehabil 16: 647–657 Bouknight RR, Bradley CJ, Luo Z (2006) Correlates of return to work for Pohjonen T (2001) Perceived work ability of home care workers in relation breast cancer survivors. J Clin Oncol 24(3): 345–353 to individual and work-related factors in different age groups. Occup Bradley CJ, Neumark D, Luo Z, Bednarek H, Schenk M (2005) Employment Med (London) 51: 209 –217 outcomes of men treated for prostate cancer. JNatlCancerInst 97: 958–965 Radloff LS (1977) The CES-D Scale, a self-report depression scale for Broadbent DE, Cooper PF, Fitzgerald P, Parkes KR (1982) The Cognitive research in the general population. Appl Psych Meas 3: 385–401 Failures Questionnaire (CFQ) and its correlates. Br J Clin Psychol 21: 1–16 Reiso H, Nygard JF, Jorgensen GS, Holanger R, Soldal D, Bruusgaard D Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ (1989) The (2003) Back to work: predictors of return to work among patients with Pittsburgh Sleep Quality Index (PSQI): a new instrument for psychiatric back disorders certified as sick: a two-year follow-up study. Spine 28: practice and research. Psychiatric Res 28: 193–213 1468– 1473 Carlisle AC, John AM, Fife-Schaw C, Lloyd M (2005) The self-regulatory Short P, Vasey JJ, Tuncelli K (2005) Employment pathways in a large cohort model in women with rheumatoid arthritis: relationships between illness of adult cancer survivors. Cancer 103: 1292–1301 representations, coping strategies, and illness outcome. Br J Health Smets E, Garssen G, Bonke B, de Haes JCJM (1995) The Multidimensional Psychol 10: 571–587 Fatigue Inventory, psychometric qualities of an instrument to assess Ekbladh E, Haglund L, Thorell LH (2004) The worker role interview – fatigue. J Psychometric Res 39: 315–325 preliminary data on the predictive validity of return to work of clients Spelten ER, Sprangers MAG, Verbeek JH (2002) Factors reported to after an insurance medicine investigation. J Occup Rehabil 14: 131–141 influence the return to work of cancer survivors: a literature review. Fowler C, Baas LS (2006) Illness representations in patients with chronic Psychooncoloy 11: 124–131 kidney disease on maintenance hemodialysis. Nephrol Nurs J 33: Spelten ER, Verbeek JH, Uitterhoeve AL, Ansink AC, van der Lelie J, 173–174, 179– 186 de Reijke TM, Kammeijer M, de Haes JC, Sprangers MA (2003) Cancer, Gudbergsson SB, Fossa SD, Borgeraas E, Dahl AA (2006) A comparative fatigue and the return of patients to work-a prospective cohort study. Eur study of living conditions in cancer patients who have returned to work J Cancer 39: 1562–1567 after curative treatment. Support Care Cancer 14: 1020–1029 Steinbach JP, Blaicher HP, Herrlinger U, Wick W, Nagele T, Meyermann R, de Haes JCJM, van Knippenberg FCE, Neyt JP (1990) Measuring Tatagiba M, Bamberg M, Dichgans J, Karnath HO, Weller M (2006) psychological and physical distress in cancer patients: structure and Surviving glioblastoma for more than 5 years: the patient’s perspective. application of the RSCL. Br J Cancer 62: 1034–1038 Neurology 66: 239–242 Hoffman B (2005) Cancer survivors at work: a generation of progress. CA Steiner JF, Cavender TA, Main DS, Bradley CJ (2004) Assessing the impact Cancer J Clin 55: 271– 280 of cancer on work outcomes. Cancer 101: 1703–1711 Hofman M, Ryan JL, Figueroa-Moseley CD, Jean-Pierre P, Morrow GR Taskila T, Martikainen R, Hietanen P, Lindbohm ML (2007) Comparative (2007) Cancer-related fatigue: the scale of the problem. Oncologist study of work ability between cancer survivors and their referents. Eur J 12(Suppl 1): 4–10 Cancer 43: 914– 920 Hogg-Johnson S, Cole DC (2003) Early prognostic factors for duration on Tuomi K, Ilmarinen J, Jahkola A (1998) Work Ability Index, 2nd revised temporary total benefits in the first year among workers with edn. Finnish Institute of Occupational Health: Helsinki compensated occupational soft tissue injuries. Occup Environ Med 60: Turner JA, Franklin G, Fulton-Kehoe D, Sheppard L, Wickizer TM, Wu R, 244–253 Gluck JV, Egan K (2006) Worker recovery expectations and fear- Ilmarinen J, Tuomi K (1993) Work Ability Index for Aging Workers. Finnish avoidance predict work disability in a population-based workers’ Institute of Occupational Health: Helsinki compensation back pain sample. Spine 31: 682 –689 Kennedy F, Haslam C, Munir F, Pryce J (2007) Returning to work following Vaughan R, Morrison L, Miller E (2003) The illness representations of cancer: a qualitative exploratory study into the experience of returning to multiple sclerosis and their relations to outcome. Br J Health Psychol 8: work following cancer. Eur J Cancer Care 16: 17–25 287–301 Leventhal H, Nerenz DR, Steele DJ (1984) Illness representations and Verbeek JH (2006) How can doctors help their patients to return to work? coping with health threats. In Handbook of Psychology and Health, PLoS Med 3: e88 Baum A, Taylor SE, Singer JE (eds), Vol. 4, pp 219–252. Erlbaum: London van Veldhoven M, de Jonge J, Broersen S, Kompier M, Meijman T (2002) Maunsell E, Drolet M, Brisson J, Brisson C, Maˆsse B, Descheˆnes L (2004) Specific relationships between psychosocial job conditions and job- Work situation after breast cancer: results from a population-based related stress: a three-level analytic approach. Work Stress 16: 207– 228 study. J Natl Cancer Inst 96: 1813–1822 de Zwart BC, Frings-Dresen MH, van Duivenbooden JC (2002) Test-retest Nieuwenhuijsen K, Bos-Ransdorp B, Uitterhoeve LL, Sprangers MA, reliability of the Work Ability Index questionnaire. Occup Med (London) Verbeek JH (2006) Enhanced provider communication and patient 52: 177–181 & 2008 Cancer Research UK British Journal of Cancer (2008) 98(8), 1342 – 1347 Clinical Studies http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png British Journal of Cancer Springer Journals

Loading next page...
 
/lp/springer-journals/work-ability-and-return-to-work-in-cancer-patients-ZxfRzw3gSG

References (73)

Publisher
Springer Journals
Copyright
Copyright © 2008 by The Author(s)
Subject
Biomedicine; Biomedicine, general; Cancer Research; Epidemiology; Molecular Medicine; Oncology; Drug Resistance
ISSN
0007-0920
eISSN
1532-1827
DOI
10.1038/sj.bjc.6604302
Publisher site
See Article on Publisher Site

Abstract

Clinical Studies British Journal of Cancer (2008) 98, 1342 – 1347 & 2008 Cancer Research UK All rights reserved 0007 – 0920/08 $30.00 www.bjcancer.com ,1 1,2 1,3 4 5,9 6 AGEM de Boer , JHAM Verbeek , ER Spelten , ALJ Uitterhoeve , AC Ansink , TM de Reijke , 1,7 8 1 M Kammeijer , MAG Sprangers and FJH van Dijk 1 2 Coronel Institute for Occupational Heath, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Cochrane Collaboration 3 4 Occupational Health Field, Kuopio, Finland; NPVO, Amsterdam, The Netherlands; Department of Radiotherapy, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Department of Gynaecology and Obstetrics, Academic Medical Center, University of Amsterdam, 6 7 Amsterdam, The Netherlands; Department of Urology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Jan van Breemen Institute, Amsterdam, The Netherlands; Department of Medical Psychology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands The extent to which self-assessed work ability collected during treatment can predict return-to-work in cancer patients is unknown. In this prospective study, we consecutively included employed cancer patients who underwent treatment with curative intent at 6 months following the first day of sick leave. Work ability data (scores 0–10), clinical and sociodemographic data were collected at 6 months, while return-to-work was measured at 6, 12 and 18 months. Most of the 195 patients had been diagnosed with breast cancer (26%), cancer of the female genitals (22%) or genitourological cancer (22%). Mean current work ability scores improved significantly over time from 4.6 at 6 months to 6.3 and 6.7 at 12 and 18 months, respectively. Patients with haematological cancers and those who received chemotherapy showed the lowest work ability scores, while patients with cancer of urogenital tract or with gastrointestinal cancer had the highest scores. Work ability at 6 months strongly predicted return-to-work at 18 months, after correction for the influence of age and treatment (hazard ratio¼ 1.37, CI 1.27–1.48). We conclude that self-assessed work ability is an important factor in the return-to-work process of cancer patients independent of age and clinical factors. British Journal of Cancer (2008) 98, 1342–1347. doi:10.1038/sj.bjc.6604302 www.bjcancer.com Published online 18 March 2008 & 2008 Cancer Research UK Keywords: employment; work ability; return-to-work; longitudinal studies; prospective studies Cancer diagnoses in individuals who are still at the working age are work in cancer survivors seems, therefore, to be problematic in becoming more common, with almost half of the adult cancer some patients but certainly not in all. Hence, it is important to survivors being younger than 65 years (Short et al, 2005). With the identify those patients with a higher risk of lasting absence from sustained improvement in treatment and prognosis of many forms work and to provide them with the appropriate support and of cancer, an increasing number of survivors of cancer return-to- counselling in returning to work. work following treatment or continue to work during therapy To examine the factors that would influence this return-to-work (Hoffman, 2005). process, we previously studied a model based on the assumption Returning to work is important for both cancer patients that cancer-related symptoms would mediate return-to-work themselves and the society. Patients often regard returning to (Spelten et al, 2003). However, results showed that diagnosis and work as a symbol of complete recovery (Spelten et al, 2002) and treatment were much stronger predictors of return-to-work than regaining a normal life (Kennedy et al, 2007), while from the cancer-related symptoms such as fatigue, depressive symptoms or viewpoint of the society, it is an economic and social imperative to cognitive problems. In addition, recent empirical studies have encourage patients to return-to-work whenever possible. indicated the importance of patients’ expectations of recovery as Despite its importance, the impact of cancer and its treatment good predictors of return-to-work and rehabilitation independent on work (dis)continuation or resumption has not been studied of diagnosis and treatment (Ekbladh et al, 2004; Verbeek, 2006). frequently (Steiner et al, 2004). However, a number of studies have Studies in other disorders have also shown that a patient’s own documented the impact of cancer on employment and they assessment of work ability (Reiso et al, 2003), expectation of job reported that approximately 60% of the cancer patients return to success (Ekbladh et al, 2004) and work recovery expectations work within 1–2 years (Spelten et al, 2002; Maunsell et al, 2004; (Hogg-Johnson and Cole, 2003; Nieuwenhuijsen et al, 2006; Turner Bradley et al, 2005; Nieuwenhuijsen et al, 2006). The return-to- et al, 2006) do predict return-to-work. A theory that could explain these mechanisms is the well-known Leventhal’s ‘model of illness representations’, which states that *Correspondence: Dr AGEM de Boer; E-mail: A.G.deBoer@amc.uva.nl people’s cognitive representations of illness exert an important Current address: Erasmus MC, Daniel den Hoed Oncology Center, influence on their strategies for coping, which in turn influence Rotterdam, The Netherlands. illness outcomes (Leventhal et al, 1984). It has been shown in other Received 6 November 2007; revised 7 February 2008; accepted 11 diseases such as multiple sclerosis, rheumatoid arthritis and February 2008; published online 18 March 2008 kidney disease (Vaughan et al, 2003; Carlisle et al, 2005; Work ability and return-to-work in cancer patients AGEM de Boer et al Fowler and Baas, 2006) that, on the basis of this model, the Physical workload was measured with a seven-item scale and functional outcome might be worse or better, irrespective of work stress with an 11-item scale from the Dutch Questionnaire on the objective medical seriousness of the illness. This strongly Experience and Judgement of Work (VBBA) (van Veldhoven et al, suggests that the ideas a cancer patient has about the disabilities 2002). Patients were asked to assess their levels of workload and that might result from the diagnosis and treatment will encourage work stress for the work situation prior to diagnosis. The scores or hinder his or her return-to-work. range from 0 to 100, with higher scores indicating a higher level of With these new insights, our data were reanalysed with the focus physical work and more work stress, respectively. on the patients’ assessments of work ability as predictor of return- to-work. In our earlier publication on return-to-work of cancer Cancer-related and sociodemographic factors Information about survivors, we did not use information on the self-assessed ability diagnosis and treatment was reported by the patients. Twenty-two to work because at the time it was outside the focus of our study different diagnoses were then grouped according to cancer (Spelten et al, 2003). site into (1) breast cancer, (2) haematological oncology, (3) The aim of the current study is therefore (1) to examine any gastrointestinal cancer, (4) cancer of the female genitals, change in work ability scores in cancer patients over time and to (5) genitourological cancer and (6) other types of cancer. study differences among patient groups and (2) to assess the extent Treatments were classified into three categories: (1) surgery, (2) to which self-assessed work ability predicts return-to-work among radiotherapy or radiotherapy plus surgery and (3) chemotherapy cancer survivors independent of diagnosis, treatment and cancer- or chemotherapy plus radiotherapy and/or surgery. related symptoms. We measured cancer-related complaints with validated ques- tionnaires and converted all scores to a scale ranging from 0 to 100, with higher scores indicating more complaints (Spelten et al, MATERIALS AND METHODS 2003). The following complaints were measured: physical cancer- related complaints (de Haes et al, 1990), general fatigue (Smets Patients et al, 1995), sleep quality (Buysse et al, 1989), depressive Eligible patients had to be between 18 and 58 years to have a symptoms (Radloff, 1977), psychological distress (de Haes et al, primary diagnosis of cancer, to be in paid employment at the time 1990), cognitive dysfunction (Broadbent et al, 1982) and global of diagnosis, to be within 4–6 months following their first day of quality of life (de Haes et al, 1990). sick leave, and to have had treatment with curative intent. They Further information was enquired concerning marital status were consecutively recruited in three hospitals in The Netherlands (single, married, cohabitating or other), having children in the where the attending physician obtained the patients’ informed household, age, gender, education (lower education, high school, consent. The study has been carried out with the approval of the college/university) and work hours per week before the diagnosis hospitals’ medical ethical committees. of cancer. Questionnaires were distributed three times to the patients, at entry into the cohort and 6 and 12 months later, to obtain information on their return-to-work, diagnosis, treatment, work Statistical analysis ability and cancer-related symptoms. Details of the design and The work ability scores measured at 6, 12 and 18 months after the material of this prospective cohort study have been reported first day of sick leave were analysed with the mixed-model earlier (Spelten et al, 2003). The data were collected between 1998 procedure based on repeated measurements to examine any and 2002. For the current study, data on return-to-work and work change in work ability scores over time. We also used the mixed- ability were collected at study entry and 6 and 12 months later. model procedure to analyse any differences over time in work Data on work load, work stress, cancer-related factors and ability scores between several patient groups: age groups (18–27, sociodemographic factors had been collected at baseline. All 28–37, 38–47 and 48–58 years), education groups, men and questionnaires were mailed to the patients’ homes. women, diagnosis groups and treatment groups. Time, group and time group interaction effects were considered fixed effects and an Measures autoregressive covariance structure was selected because of Return-to-work Data on return-to-work were measured on the correlated work ability scores over time. In case of a statistically basis of two measures: time to return-to-work after sick leave and significant main effect, post hoc analyses were performed between rate of return-to-work at a specific point in time. All patients in time points and between groups with pairwise comparisons based The Netherlands typically have access to sick leave. Time to return- on the use of the mean difference of the estimated marginal means. to-work at 18 months after the first day of sick leave was calculated To examine whether self-assessed work ability can predict as the number of days between the first date of sick leave and the return-to-work in cancer patients a year later, taking the impact of first day the patient returned to work. Any kind of work clinical-, work- and subject-related factors into account, we used a resumption qualified as a return-to-work, irrespective of the two-step procedure. First, univariate analyses using Kaplan–Meier number of hours that the patients worked prior to their diagnosis. analyses were performed for the relationship between time taken In addition, patients were asked to indicate if they were still on sick to return-to-work (in days) at 18 months and each of the predictive leave (yes/no) at 6, 12 and 18 months following their first day of factors measured at baseline (on average 6 months after the first sick leave. day of sick leave): current work ability, mental work ability, physical work ability, physical work load, work stress, physical Work ability, work load and work stress Current work ability was complaints, fatigue, sleep impairments, depression, psychological measured with the first three items from the Work Ability Index distress, cognitive dysfunction, age, gender, education and the (WAI) (Ilmarinen and Tuomi, 1993, p 142; Tuomi et al, 1998), clinical factors (diagnosis and treatment type). Next, we analysed which is a reliable and valid measure of work ability (Ilmarinen the impact of work ability in addition to personal and clinical and Tuomi, 1993, p 142; de Zwart et al, 2002). First, current work factors in a multivariate Cox regression analysis. We entered all ability was assessed by asking the patients to estimate their current variables for which the log-rank test returned a P-valuep0.10 into work ability compared with their lifetime best (0¼ cannot work at a Cox regression analysis with forward selection of variables. With all to 10¼ best ever). In addition, we asked the cancer patients to this method, the best predictors of future return-to-work are rate both their current physical and mental work ability in relation selected (Altman, 1991). Because it was possible for patients to to job demands (0¼ very low to 5¼ very high). return to work before our first measurement at 6 months, we & 2008 Cancer Research UK British Journal of Cancer (2008) 98(8), 1342 – 1347 Clinical Studies Clinical Studies AGEM de Boer et al repeated both analyses with the exclusion of patients who had Table 1 Sociodemographic and cancer-related characteristics at base- returned before 6 months. Since this is a survival analysis, hazard line, on average 6 months after the first day of sick leave ratios (HRs) usually indicate the risk of dying, while in our case the N¼ 195 patients n (%) event is returning to work. Therefore, an HR higher than one indicates the higher ‘risk’ of return-to-work. Age (mean (s.d.)) 42.2 (9.3) Alpha was set at 0.05 unless stated otherwise and all tests were Sex (male) 78 (40%) two-sided. Analyses were conducted with SPSS 13. Returned-to-work 46 (24%) Education Lower 52 (27%) RESULTS High school 83 (42%) College/university 60 (31%) The first questionnaire was completed by 235 of the 264 eligible patients (a response of 89%), while a total of 29 patients declined Marital status participation in this study. The second questionnaire at 6 months Single 24 (12%) follow-up was completed by 221 of the 235 participating patients Married/cohabiting 160 (82%) (a follow-up response of 94%). At 12 months of follow-up, the Divorced 8 (4%) questionnaire was returned by 195 patients (an 83% follow-up Widower 3 (2%) response and 74% of the initially eligible patients), while 25 patients refused to return the questionnaire, 13 patients had died Diagnosis and 2 questionnaires got lost in the mail. Breast cancer 51 (26%) Haematological oncology 24 (12%) Table 1 shows sociodemographic and cancer-related characteris- Gastrointestinal cancer 23 (12%) tics at 6 months after the first day of sick leave. Half of the patients Cancer of the female genitals 43 (22%) had either breast cancer (26%) or cancer of the female genitals Genitourological cancer 43 (22%) (22%), while another 22% of the patients had been diagnosed with Other 11 (6%) genitourological cancer. Before diagnosis and treatment, patients worked an average of 34 h per week, and 6 months after the Treatment diagnosis, 46 patients (24%) had already returned to work or had Surgery 41 (21%) continued working. Data on work hours per week, children, Chemotherapy or chemotherapy plus radiotherapy and/or surgery 88 (45%) fatigue, depression, sleep problems, physical complaints, cognitive Radiotherapy or radiotherapy plus surgery 66 (34%) dysfunction, psychological distress, work load and work stress have been reported previously (Spelten et al, 2003). Table 2 depicts the mean values of current work ability at 6, 12 and 18 months after the first day of sick leave; the values improved Table 2 Mean value of current work ability according to socio- significantly over time (Po0.001) from 4.6 at 6 months to 6.3 at 12 demographic and disease-related factors at 6, 12 and 18 months after months and to 6.7 at 18 months. Post hoc analyses of work ability the first day of sick leave scores showed that all three time points were significantly different Workability score (mean (s.d.)) from each other (Po0.001 to P¼ 0.035). All age groups improved over time (Po0.001) with the 28- to 37-year-old patients N¼ 195 patients 6 months 12 months 18 months increasing most from 4.8 to 7.5. No differences in work ability scores were, however, found between age groups (P¼ 0.12). Work All patients 4.59 (3.2) 6.31 (2.7) 6.74 (2.7) ability scores of both men and women improved over time (Po0.001), but women improved more (P¼ 0.002). Male patients Age showed higher work ability scores at 6 months (5.8 vs 3.8, 18 – 27 years 5.33 (3.6) 5.71 (3.9) 7.27 (2.9) Po0.0001), but not at 12 months (6.8 vs 6.0, P¼ 0.053) or at 18 28 – 37 years 4.81 (3.4) 6.80 (2.5) 7.51 (2.4) months (6.9 vs 6.7, P¼ 0.52). Higher educated patients seemed to 38 – 47 years 4.53 (3.2) 6.65 (2.3) 6.80 (2.3) 48 – 58 years 4.31 (3.0) 5.73 (2.8) 5.97 (3.1) have higher work ability scores, but the differences were not statistically significant (P¼ 0.13). With regard to diagnosis, we Sex found significant differences between the different diagnosis Male 5.76 (3.0) 6.78 (2.6) 6.91 (2.8) groups (Po0.001). The haematological oncology patients showed Female 3.83 (3.1) 6.00 (2.7) 6.65 (2.6) the significantly (Po0.001) lowest scores of 3.3, 4.5 and 5.0 at 6, 12 and 18 months, respectively. The patients with genitourological Education cancer had the highest scores of 6.9 and 7.8 at 6 and 12 months Lower 3.90 (3.3) 6.14 (3.0) 6.26 (3.2) (Po0.001), and the patients with gastrointestinal cancer scored the High school 4.51 (3.0) 6.31 (2.6) 6.77 (2.5) highest work ability of 7.6 at 18 months (Po0.001). Patients with College/university 5.33 (3.4) 6.46 (2.6) 7.15 (2.6) cancer of the female genitals and breast cancer patients improved Diagnosis most over time (P¼ 0.01). Breast cancer 3.59 (3.1) 5.90 (2.3) 6.49 (2.5) Figure 1 shows the work ability scores for the three treatment Haematological oncology 3.29 (3.0) 4.46 (3.4) 4.95 (3.6) combinations: (1) surgery; (2) chemotherapy or chemotherapy Gastrointestinal cancer 5.52 (2.7) 6.95 (2.1) 7.57 (1.6) plus radiotherapy and/or surgery; and (3) radiotherapy or Cancer of the female genitals 3.91 (3.1) 6.29 (2.9) 7.00 (2.6) radiotherapy plus surgery, over time. Analyses revealed that scores Genitourological cancer 6.86 (2.5) 7.79 (1.7) 7.33 (2.7) improved over time for all three groups and that the group of Other 4.09 (3.4) 5.36 (2.9) 6.55 (2.5) patients that received chemotherapy or chemotherapy plus radio- a b Range 0 – 10; 10 indicating best work ability ever. Work ability score change over therapy and/or surgery consistently showed lower work ability time: Po0.01. Difference between groups: Po0.01. scores than the group that received surgery or radiotherapy (plus surgery) (Po0.001). Improvement was not statistically different in Results of univariate analyses using the Kaplan–Meier analyses the three groups (P¼ 0.45). showed that the time taken to return-to-work measured at 18 At 6 months after diagnosis, 24% of patients had returned to months was related to the following factors measured at 6 months: work, at 12 months 50%, and at 18 months 64% had returned. British Journal of Cancer (2008) 98(8), 1342 – 1347 & 2008 Cancer Research UK Work ability and return-to-work in cancer patients AGEM de Boer et al Treatment Current work 1.0 ability at Surgery 6 months Radiotherapy (+ surgery 8 0.8 1 Chemotherapy (+ radiotherapy and /or surgery) 0.6 0.4 0.2 0.0 0.00 100.00 200.00 300.00 400.00 500.00 600.00 700.00 6 months 12 months 18 months Time to return to work (days) Time Figure 2 Plot of the work ability scores measured at 6 months in Figure 1 Mean value of current work ability at 6, 12 and 18 months after relation to time to return-to-work, after adjustment for age and treatment the first day of sick leave for the three treatment combinations: surgery; (n¼ 195). chemotherapy or chemotherapy plus radiotherapy and/or surgery; and radiotherapy or radiotherapy plus surgery (n¼ 195). age and work ability. For current work ability itself, every 1 point increase on the 11-point scale meant a 1.37 higher chance of returning to work earlier, after correction for the influence of age Table 3 Cox regression analysis on return to work and treatment. Figure 2 shows the plot of the work ability scores in relation to return-to-work after adjustment for age and treatment. Time to return to work The plot shows that of the patients with the lowest work ability N¼ 195 Hazard ratio 95% CI scores (0–5) at 6 months after the first day of sick leave, the majority (55–80%) did not return to work in the first year after Age, 10-year categories 0.78 0.65 – 0.94 diagnosis. Patients with very high work ability scores (8, 9, 10) did Current work ability 1.37 1.27 – 1.48 usually return to work within half a year, while virtually all of these high-scoring patients were back at work after the first year. Treatment Results of the analysis without the 46 patients who had already Surgery (reference) 1.00 returned to work at 6 months showed that in this model the Chemotherapy plus radiotherapy and/or surgery 0.41 0.25 – 0.69 Radiotherapy/radiotherapy plus surgery 0.63 0.39 – 1.0 strongest predictors of return-to-work were not only work ability (HR¼ 1.23; CI, 1.12–1.36), treatment (chemotherapy HR¼ 0.33; 95% CI¼ 95% confidence interval. CI, 0.18–0.60; radiotherapy HR¼ 0.52; CI, 0.29–0.95) and age (HR¼ 0.67; CI, 0.53–0.86) but also mental work ability (HR¼ 1.41; CI, 1.05–1.89) and cognitive dysfunction (HR¼ 1.03; CI, current work ability, mental work ability, physical work ability, 1.01–1.05). quality of life, fatigue, physical complaints, cognitive functioning, age, physical work load, work stress, gender, diagnosis and treatment (at the Pp0.10 level). Sleep impairments, depression, DISCUSSION psychological distress and education did not significantly predict return-to-work. Results of the analysis without the 46 patients who The aim of our study was to examine changes in work ability had already returned to work at 6 months showed the same factors scores in cancer patients over time and to study differences except for gender, which did not significantly predict return-to- between patient groups and, furthermore, to assess the extent to work. which self-assessed work ability predicts return-to-work among The factors that were predictive for return-to-work at 18 months cancer survivors independent of diagnosis, treatment and cancer- were entered in the Cox regression with a forward selection to related symptoms. identify the strongest predictors of return-to-work. Results in We found that the cancer patients’ work ability scores at 6, 12 Table 3 show that in the final model, age, current work ability and and 18 months after the first day of sick leave improved treatment are still significant. Current work ability, physical work significantly over time. Men scored higher on work ability than ability and mental work ability were highly correlated and, women but no differences were found between age or education therefore, only current work ability remained in the model. groups. Furthermore, the haematological oncology patients and Likewise, treatment and diagnosis were highly correlated, and only the patients who received chemotherapy or chemotherapy plus treatment was selected for the final model. Patients treated with radiotherapy and/or surgery consistently showed lower work surgery alone had the highest chance of returning to work quickly. ability scores. Finally, self-assessed work ability 6 months after Those who were treated with radiotherapy or radiotherapy plus the first day of sick leave proved to be a strong predictor of later surgery had an HR of 0.63 (95% CI: 0.39–1.0), corrected for age return-to-work in cancer survivors independent of age and and work ability, of returning to work and were thus 1.6 times therapy. more likely to stay off work than patients with surgery alone. This is the first longitudinal study in which the impact of work Patients treated with chemotherapy, either alone or in combination ability on return-to-work has been established in a systematic way. with other treatment modalities, had an HR of 0.41 (95% CI: The cohort has been followed for a considerable amount of time, 0.25–0.69) and their risk of staying off work was therefore 2.4 the number of patients lost to follow up was relatively small and all times higher than patients treated with surgery alone, corrected for factors have been measured with validated instruments. & 2008 Cancer Research UK British Journal of Cancer (2008) 98(8), 1342 – 1347 Mean work ability Proportion off work Clinical Studies Clinical Studies AGEM de Boer et al In our study, the mean current work ability scores at 6, 12 and the better predictor of return-to-work. Other studies have also 18 months after the first day of sick leave were 4.6, 6.3 and 6.7, found that fatigue influenced conditions of employment and respectively. Although we found a significant improvement of productiveness (Hofman et al, 2007). Results of the univariate current work ability, these scores are lower than the average analyses without the 46 patients who had already returned to work current work ability score of 7.9 found by Pohjonen (2001) in a at 6 months showed that gender was not a significant factor sample of female home care workers in the age group 40–44 years anymore. This might be caused by the fact that most of these old with an average of two diagnosed benign diseases. It might be returned patients were men (65%) and were diagnosed with testes possible that work ability scores in cancer patients will improve or prostate cancer (48%). In the model of best predictors of return- still further 2 years after the diagnosis or that their work ability to-work without those patients who had returned early, the factors scores might deteriorate because cancer has a larger impact on mental work ability and cognitive dysfunction were included. This work ability than other diseases. could imply that for patients who do not return early, the mental Research on the effect of cancer diagnosis and treatment on and psychological factors become more dominant in relation to work ability is scarce; however, studies have shown recently that return-to-work. most patients are employed but that both physical and mental Leventhal’s ‘model of illness representations’ states that people’s cognitive representations of illness play an important role in work ability can deteriorate owing to cancer (Gudbergsson et al, 2006; Steinbach et al, 2006; Kennedy et al, 2007; Taskila influencing their strategies for coping, which in turn influence et al, 2007). Patients in the recent study of Kennedy et al (2007), illness outcomes (Leventhal et al, 1984). On the basis of this who were 1–10 years after diagnosis, reported that they had model, the functional outcome might be worse or better, difficulties in coping and concentrating, and they worried irrespective of the objective medical seriousness of the illness. about their reduced capability. In the comparative study of Our results are congruent with this model. Irrespective of age, Gudbergsson et al (2006), it was found that cancer patients diagnosis, treatment, quality of life, fatigue, and physical or 2–6 years after diagnosis, who had returned to work after psychological complaints, self-assessed work ability strongly curative treatment, reported significantly poorer physical and predicted future return-to-work. This indicates that the ideas a mental work capacity compared to employed matched controls cancer patient has about his or her work disabilities that result from the general population. Most survivors of glioblastoma in the from the diagnosis and treatment of cancer are a reflection of the study of Steinbach et al (2006) also thought that their work true work capabilities. Therefore, the self-reported work ability ability was impaired. According to Taskila et al (2007), 26% of could be important in encouraging or hindering his or her return- cancer survivors reported that their physical work ability to-work. had deteriorated and 19% that their mental work ability had Employment outcomes can be improved with innovations deteriorated owing to cancer diagnosis and treatment. However, in treatment and with clinical and supportive services aimed at the work ability as measured with the WAI of these better management of symptoms, rehabilitation and accommoda- cancer survivors did not differ from that of a group of healthy tion of disabilities (Steiner et al, 2004). A recent study of referent persons. This is probably also caused by the fact that all Bouknight et al (2006) showed that a high percentage of employed their survivors with breast cancer, lymphoma and prostate cancer breast cancer patients returned to work after treatment and had already returned to work and that they were long-term that workplace accommodations played an important role in their survivors who had been diagnosed with cancer 2–6 years before return. Therefore, interventions should be developed to enable the time of the questionnaire. This could also explain the cancer survivors to return to work or to succeed in differences in mean work ability scores between their study and other appropriate employment, because no such interventions the patients in our study. For men in their study, the work ability aimed at work do exist at the present. These interventions scores were 8.0 (for prostate cancer) to 8.9 (for testicular cancer) should aim especially at patients who indicate that their work compared to 6.9 for the men in our study 18 months after the first ability is diminished, at older patients and at those treated with day of sick leave. Our female patients scored 6.7 at the end of chemotherapy, since they are at the greatest risk of prolonged follow-up compared to 8.2 (for breast cancer) and 8.5 (for work absence. Clinicians could play an important role in detecting lymphoma) in the Finnish study by Taskila et al (2007). Our those patients at risk because our study has shown that the study also showed that men initially showed higher scores of work indication of patients with possible return-to-work problems can ability, while women improved faster and no differences were be assessed very early in the treatment process when they have found after 1 year. It might be possible that women, who were diminished self-reported work ability. Physicians could help mainly diagnosed with breast cancer, received more chemother- patients in the return-to-work process and therefore help in apy, which would have prolonged the treatment period. Another improving their quality of life by asking patients if they have explanation might be that women could have more household returned to work or are experiencing problems in the return-to- activities than men and that they take these into account when work process. If so, referral to occupational specialists could be judging their work ability. considered. Our study indicated work ability as an independent predictor for In conclusion, the work ability of cancer patients who work at return-to-work, while quality of life was only found to be the time of their diagnosis is severely impaired in the first months predictive of time until return-to-work in the univariate analyses. after the first day of sick leave, but it does improve significantly in The same result was found earlier for Norwegian patients with the months afterwards. Self-assessed work ability 6 months after back disorders who had been certified as sick (Reiso et al, 2003). the first day of sick leave proved to be a strong predictor of later The authors of that study suggested that work ability questions return-to-work in cancer survivors independent of age and may be related more to function in a setting of sickness therapy. certification than a global quality of life question and therefore be more predictive. In an earlier analysis (Spelten et al, 2003), we found that fatigue at 6 months predicted a longer sick leave with an HR of 0.71, ACKNOWLEDGEMENTS adjusted for diagnosis, treatment, age and gender. Our present study indicated that fatigue was only a predictive factor of return- This study was supported by a grant from the Dutch Cancer to-work in the univariate analyses but not in the multivariate Society (AMC 97-1385). We thank Dr J van der Lelie for his analyses, which included work ability. Because work ability and contribution to this study. We are grateful to all of the patients for fatigue were correlated, only work ability remained in the model as their participation in this study. British Journal of Cancer (2008) 98(8), 1342 – 1347 & 2008 Cancer Research UK Work ability and return-to-work in cancer patients AGEM de Boer et al REFERENCES Altman DG (1991) Practical Statistics for Medical Research. Chapman and education regarding return to work in cancer survivors following Hall: London curative treatment: a pilot study. J Occup Rehabil 16: 647–657 Bouknight RR, Bradley CJ, Luo Z (2006) Correlates of return to work for Pohjonen T (2001) Perceived work ability of home care workers in relation breast cancer survivors. J Clin Oncol 24(3): 345–353 to individual and work-related factors in different age groups. Occup Bradley CJ, Neumark D, Luo Z, Bednarek H, Schenk M (2005) Employment Med (London) 51: 209 –217 outcomes of men treated for prostate cancer. JNatlCancerInst 97: 958–965 Radloff LS (1977) The CES-D Scale, a self-report depression scale for Broadbent DE, Cooper PF, Fitzgerald P, Parkes KR (1982) The Cognitive research in the general population. Appl Psych Meas 3: 385–401 Failures Questionnaire (CFQ) and its correlates. Br J Clin Psychol 21: 1–16 Reiso H, Nygard JF, Jorgensen GS, Holanger R, Soldal D, Bruusgaard D Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ (1989) The (2003) Back to work: predictors of return to work among patients with Pittsburgh Sleep Quality Index (PSQI): a new instrument for psychiatric back disorders certified as sick: a two-year follow-up study. Spine 28: practice and research. Psychiatric Res 28: 193–213 1468– 1473 Carlisle AC, John AM, Fife-Schaw C, Lloyd M (2005) The self-regulatory Short P, Vasey JJ, Tuncelli K (2005) Employment pathways in a large cohort model in women with rheumatoid arthritis: relationships between illness of adult cancer survivors. Cancer 103: 1292–1301 representations, coping strategies, and illness outcome. Br J Health Smets E, Garssen G, Bonke B, de Haes JCJM (1995) The Multidimensional Psychol 10: 571–587 Fatigue Inventory, psychometric qualities of an instrument to assess Ekbladh E, Haglund L, Thorell LH (2004) The worker role interview – fatigue. J Psychometric Res 39: 315–325 preliminary data on the predictive validity of return to work of clients Spelten ER, Sprangers MAG, Verbeek JH (2002) Factors reported to after an insurance medicine investigation. J Occup Rehabil 14: 131–141 influence the return to work of cancer survivors: a literature review. Fowler C, Baas LS (2006) Illness representations in patients with chronic Psychooncoloy 11: 124–131 kidney disease on maintenance hemodialysis. Nephrol Nurs J 33: Spelten ER, Verbeek JH, Uitterhoeve AL, Ansink AC, van der Lelie J, 173–174, 179– 186 de Reijke TM, Kammeijer M, de Haes JC, Sprangers MA (2003) Cancer, Gudbergsson SB, Fossa SD, Borgeraas E, Dahl AA (2006) A comparative fatigue and the return of patients to work-a prospective cohort study. Eur study of living conditions in cancer patients who have returned to work J Cancer 39: 1562–1567 after curative treatment. Support Care Cancer 14: 1020–1029 Steinbach JP, Blaicher HP, Herrlinger U, Wick W, Nagele T, Meyermann R, de Haes JCJM, van Knippenberg FCE, Neyt JP (1990) Measuring Tatagiba M, Bamberg M, Dichgans J, Karnath HO, Weller M (2006) psychological and physical distress in cancer patients: structure and Surviving glioblastoma for more than 5 years: the patient’s perspective. application of the RSCL. Br J Cancer 62: 1034–1038 Neurology 66: 239–242 Hoffman B (2005) Cancer survivors at work: a generation of progress. CA Steiner JF, Cavender TA, Main DS, Bradley CJ (2004) Assessing the impact Cancer J Clin 55: 271– 280 of cancer on work outcomes. Cancer 101: 1703–1711 Hofman M, Ryan JL, Figueroa-Moseley CD, Jean-Pierre P, Morrow GR Taskila T, Martikainen R, Hietanen P, Lindbohm ML (2007) Comparative (2007) Cancer-related fatigue: the scale of the problem. Oncologist study of work ability between cancer survivors and their referents. Eur J 12(Suppl 1): 4–10 Cancer 43: 914– 920 Hogg-Johnson S, Cole DC (2003) Early prognostic factors for duration on Tuomi K, Ilmarinen J, Jahkola A (1998) Work Ability Index, 2nd revised temporary total benefits in the first year among workers with edn. Finnish Institute of Occupational Health: Helsinki compensated occupational soft tissue injuries. Occup Environ Med 60: Turner JA, Franklin G, Fulton-Kehoe D, Sheppard L, Wickizer TM, Wu R, 244–253 Gluck JV, Egan K (2006) Worker recovery expectations and fear- Ilmarinen J, Tuomi K (1993) Work Ability Index for Aging Workers. Finnish avoidance predict work disability in a population-based workers’ Institute of Occupational Health: Helsinki compensation back pain sample. Spine 31: 682 –689 Kennedy F, Haslam C, Munir F, Pryce J (2007) Returning to work following Vaughan R, Morrison L, Miller E (2003) The illness representations of cancer: a qualitative exploratory study into the experience of returning to multiple sclerosis and their relations to outcome. Br J Health Psychol 8: work following cancer. Eur J Cancer Care 16: 17–25 287–301 Leventhal H, Nerenz DR, Steele DJ (1984) Illness representations and Verbeek JH (2006) How can doctors help their patients to return to work? coping with health threats. In Handbook of Psychology and Health, PLoS Med 3: e88 Baum A, Taylor SE, Singer JE (eds), Vol. 4, pp 219–252. Erlbaum: London van Veldhoven M, de Jonge J, Broersen S, Kompier M, Meijman T (2002) Maunsell E, Drolet M, Brisson J, Brisson C, Maˆsse B, Descheˆnes L (2004) Specific relationships between psychosocial job conditions and job- Work situation after breast cancer: results from a population-based related stress: a three-level analytic approach. Work Stress 16: 207– 228 study. J Natl Cancer Inst 96: 1813–1822 de Zwart BC, Frings-Dresen MH, van Duivenbooden JC (2002) Test-retest Nieuwenhuijsen K, Bos-Ransdorp B, Uitterhoeve LL, Sprangers MA, reliability of the Work Ability Index questionnaire. Occup Med (London) Verbeek JH (2006) Enhanced provider communication and patient 52: 177–181 & 2008 Cancer Research UK British Journal of Cancer (2008) 98(8), 1342 – 1347 Clinical Studies

Journal

British Journal of CancerSpringer Journals

Published: Mar 18, 2008

There are no references for this article.