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

Learn More →

Predictors of employment among cancer survivors after medical rehabilitation – a prospective study

Predictors of employment among cancer survivors after medical rehabilitation – a prospective study Downloaded from www.sjweh.fi on November 17, 2021 Original article Scand J Work Environ Health 2013;39(1):76-87 doi:10.5271/sjweh.3291 Predictors of employment among cancer survivors after medical rehabilitation  a prospective study by Mehnert A, Koch U Affiliation: Department and Outpatient Clinic of Medical Psychology, University Medical Center Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. mehnert@uke.de The following article refers to this text: 2020;46(4):382-391 Key terms: cancer; cancer survivor; cancer survivorship; employment; predictor; prospective study; psycho-oncology; rehabilitation; return to work; RTW This article in PubMed: www.ncbi.nlm.nih.gov/pubmed/22422040 This work is licensed under a Creative Commons Attribution 4.0 International License. Print ISSN: 0355-3140 Electronic ISSN: 1795-990X Original article Scand J Work Environ Health. 2013;39(1):76–87. doi:10.5271/sjweh.3291 Predictors of employment among cancer survivors after medical rehabilitation – a prospective study by Anja Mehnert, PhD,¹ Uwe Koch, PhD, MD ¹ ² Mehnert A, Koch U. Predictors of employment among cancer survivors after medical rehabilitation – a prospective study. Scand J Work Environ Health. 2013;39(1):76–87. doi:10.5271/sjweh.3291 Objectives This study aimed to (i) investigate cancer survivor’s employment status one year after the comple- tion of a medical rehabilitation program and (ii) identify demographic, cancer, and psychosocial, treatment-, and work-related predictors of return to work (RTW) and time until RTW. Methods A total of 1520 eligible patients were consecutively recruited on average 11 months post diagnosis and assessed at the beginning (t ) (N=1148) and end of rehabilitation (t ) (N=1060) and 12 months after rehabilitation 0 1 (t ) (N=750). Participants completed validated measures assessing functional impairments, pain, anxiety, depres- sion, quality of life, social support, and work-related characteristics including work ability, sick leave absence, job requirements, work satisfaction, self-perceived employer accommodation, and perceived job loss. Physicians estimated the degree of cancer-entity-specific functional impairment. Results In a mean time of six weeks after rehabilitation, 568 patients (76%) had returned to work. The multi- variate hierarchical logistic regression analysis indicated that baseline RTW intention [odds ratio (OR) 6.22, 95% confidence interval (95% CI) 1.98–19.51], perceived employer accommodation (OR 1.93, 95% CI 0.33–0.99), high job requirements (OR=1.84, 95% CI 1.02–3.30), cancer recurrence or progression (OR=0.27, 95% CI 0.12 – 0.63), baseline sick leave absence (OR=0.26, 95% CI 0.09–0.77), and problematic social interactions (OR=0.58, 95% CI 0.33–0.99) emerged as significant predictors for RTW. The explained variance of the total model was Nagelkerke’s R²=0.59 (P<0.001). Conclusion Our findings emphasize the high relevance of motivational factors. Occupational motivation and skepticism towards returning to work should be carefully assessed at the planning of the rehabilitation program. Key terms cancer survivorship; psycho-oncology; return to work; RTW. Since an increasing number of patients are likely to Previous research suggested that cancer increases return to work after diagnosis and treatment comple- the risk of unemployment among survivors compared to tion, there is an increasing recognition of the short healthy controls (20–22). On average, 63.5% of cancer and long-term impact of cancer and both its physical survivors (range 24–94%) return to work (17). Overall, and psychosocial consequences on employment during studies indicate a steady increase of return to work (RTW) the last years (1–13). Unfavorable cancer and treat - from on average 40% at six months post diagnosis to 62% ment consequences include a variety of physical and at 12 months, 73% at 18 months, and 89% at 24 months. functional disabilities, and psychological distress that Factors significantly associated with a greater likelihood may adversely affect a patient’s work ability, work sat- of being employed or RTW include perceived employer isfaction, as well as employment status (14–17). Since accommodation, flexible working arrangements, coun - work has the potential to help patients regain a sense of seling, training and rehabilitation services, younger age, meaning, normalcy and being valued, returning to work higher education, male gender, a lower physical symptom may comprise a range of positive consequences for the burden, cancer remission, shorter length of sick leave, and recovery and the psychological well-being (18, 19). continuity of care (7, 14, 17, 23–28). Department and Outpatient Clinic of Medical Psychology, Center for Psychosocial Medicine and University Cancer Center Hamburg (UCCH), University Medical Center Hamburg-Eppendorf, Hamburg, Germany. Deanery Medical Faculty and Department and Outpatient Clinic of Medical Psychology, Center for Psychosocial Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany. Correspondence to: Anja Mehnert PhD, Department and Outpatient Clinic of Medical Psychology, University Medical Center Hamburg- Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. [E-mail: mehnert@uke.de] 76 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch However, although important findings have emerged participation. Consecutive patients were recruited from in previous studies on work and employment among four cancer rehabilitation facilities and assessed at the cancer survivors (7, 29, 30), only limited knowledge beginning (t ) and end of rehabilitation (t ), and 12 0 1 exists about RTW after rehabilitation taking into account months after rehabilitation (t ). Inclusion criteria com- the impact of demographic characteristics, cancer and prised (i) age 18–60 years, (ii) the capability to complete treatment-related physical factors, psychosocial as well study measures, and (iii) the absence of permanent inva- as of work-related aspects of employment and RTW. lidity and early retirement. Since the old age pension in The Word Health Organization (WHO) (31) has defined Germany generally becomes effective at 65 years, the rehabilitation as “the use of all means aimed at reducing age limit of 60 years was chosen to enable patients to the impact of disabling and handicapping conditions and have a sufficient time period left to return to work. at enabling people with disabilities to achieve optimal A total of 1148 patients were enrolled at t (75.5% social integration”. Medical rehabilitation programs participation rate), whereas 372 patients declined to in Germany are provided to cancer patients accord - participate. Among those who participated at t , 1060 ing to the overall aim defined by the WHO and the completed the questionnaires at t . At t , questionnaires 1 2 International Classification of Functioning, Disability were mailed to an eligible 994 patients (36 patients and Health (ICF) (31). Thus, the (re-)integration of had moved to an unknown address and 30 had died), individuals with disabilities, chronic health conditions, 750 (75%, 65% of the total sample) of whom returned diseases, and handicaps into society and working life is questionnaires that could be evaluated (figure 1). Table one important aspect of rehabilitation by eliminating or 1 presents baseline sample characteristics. reducing the impact of chronic illness and disability. The aim is to maintain a patient’s optimal physical, sensory, Non-responder analyses psychological, and social functional levels. Rehabilita- tion also serves to prevent an impending disability or the At t , participants and non-participants differed in terms aggravation of existing physical damages. of age [mean 48.5, standard deviation (SD) 7.2, years Based on social laws in Germany, cancer patients have versus mean 50.4, SD 6.1, years] (P<0.001) (d=0.3) and a legal right to participate in at least one rehabilitation male gender (16% versus 22%) (P=0.007) (ϕ=0.07). No program after the completion of primary cancer treat- group differences in cancer entities were observed. At t , ments (32). Traditionally, cancer rehabilitation programs non-participating patients were more likely to be male are mainly carried out in in-patient settings in specialized (P=0.008) (ϕ=0.09), widowed (P=0.03) (ϕ=0.10), and rehabilitation clinics. Access to cancer rehabilitation pro- have head and neck or lung cancer (P=0.005) (ϕ=0.14). grams is usually facilitated by hospital doctors and social Participants at t were found to be significantly less workers immediately after completion of the primary depressed (P=0.01) (η²=0.007) and had a lower level of treatment (“follow-up rehabilitation”). However, cancer fear of cancer recurrence (P=0.009) (η²=0.007). rehabilitation at a later stage during the course of cancer treatment is also provided. Rehabilitation costs are covered Study variables and measures mainly by the pension and health insurances. A cancer rehabilitation program lasts three weeks following a mul- The main outcomes for this study were employment at tidimensional therapeutic approach that includes patient 12 months after rehabilitation (M=23.1 months after education, exercises, and physical therapy to regain physi- cancer diagnosis) and time until RTW. Employment was cal fitness and vitality, relaxation training, psychosocial defined according to a patient’s positive response to the counseling, and psychosocial support groups to enhance question “Are you currently working?” Time until RTW coping skills as well as individual psychotherapy. was measured in weeks after rehabilitation. This prospective longitudinal study aimed to (i) Demographic information was obtained at baseline identify the employment rate 12-months after cancer consisting of standardized questions concerning (age, rehabilitation, (ii) explore the work situation and expe- gender, marital status, and employment history). Educa- rienced work changes, and (iii) identify demographic, tion, monthly household net income, and occupational medical, functional, psychosocial, and work-related position were used to calculate a 3-factor social status predictors of the likelihood and time period of RTW. index (33). Medical information was collected at base - line [cancer entity, months since diagnosis, cancer stage as defined by the International Union against Cancer (UICC), clinical characteristics and disease phase]. In Methods addition to the baseline Karnofsky performance status (34), physicians estimated the degree of functional The study received research ethics committee approval. impairment using cancer-entity-specific physical func - All patients provided written informed consent prior to tioning scales. The Karnofsky status is a performance Scand J Work Environ Health 2013, vol 39, no 1 77 Employment predictors among cancer survivors Assessed for eligibility N=1653 Excluded: Non-participants t : Participation t : 0 0 Unclear employment status: N=23 N=372 N=1148 Housewife/houseman: N=110 Non-participants t : Participation t : 1 1 N=88 N=1060 Non-participants t : Participation t : 2 2 N=310 N=750 Figure 1. Enrollment of cancer survivors (t =beginning of cancer rehabilitation, t =end of a 3–4 week cancer rehabilita- Deceased: Moved to unknown Refused participation: tion, t =12 months after cancer N=30 N=244 2 address: N=36 rehabilitation) measure for rating the ability of a somatically ill person absence, (iii) 10 items about job requirements (eg, many to perform usual activities. A person is evaluated on work responsibilities, high pressure of competition, a score of 0–100, where 0=dead and 100=normal, no tight schedules) answered on a 4-point Likert scale complaints, no signs of disease. Using cancer-entity- ranging from 1=almost never to 4=quite often (Cron- specific physical functioning scales, the physician mea - bach’s α=0.87), and (iv) 12 items about work satisfac- sures the limitations specific for each tumor entity such tion answered on a 7-point Likert scale ranging from as shoulder mobility or lymphedema in breast cancer, 1=not satisfied at all to 7=totally satisfied (Cronbach’s incontinence in genital or colon cancers, swallowing α=0.92). Self-perceived work ability was evaluated on problems in head and neck cancers or dyspnea in lung a 5-point Likert item ranging from 1=totally limited to cancer. Pain intensity during the last week was evaluated 5=not limited at all; self-perceived employer accommo- using the Brief Pain Inventory (BPI) (35). dation was evaluated on a 5-point Likert item ranging The psychosocial and work-related variables were from 1=not at all to 5=extremely. Perceived threat of job further assessed at baseline. Anxiety and depression loss was measured using single-item questions (No/Yes). were assessed using the Hospital Anxiety and Depres - sion Scale (HADS) (36). Fear of cancer recurrence was Statistical analysis measured using the 12-item short version of the Fear of Recurrence Questionnaire (FoP-Q-SF) scored on a In order to identify significant predictors of RTW, demo - 5-point Likert scale ranging from 1=never to 5=very graphic, medical, functional, psychosocial, and work- often (37). The Short-Form Health Survey assesses related factors were entered separately (block-wise) into dimensions of quality of life (QoL): Here, the two a multivariate hierarchical logistic regression analysis summary scores for physical (PCS) and mental health against the outcome variable “RTW”. Step-wise back- (MCS) were calculated. Higher scores indicate better wards elimination was used, testing each candidate vari- QoL (38). The Illness-Specific Social Support Scale able for removal using Wald statistic. Before testing the (ISSS) measures the degree to which partners/friends regression model, we performed correlations (Pearson provide positive support (eg, “listened to you”) or act and Spearmans correlation coefficients) for all predictor in a non-supportive way (detrimental interactions) (eg, candidates between variables in one block and between “tried to change the way you’re coping with your illness candidate variables and the outcome criteria. In order in a way you didn’t like”). Items are scored on a 5-point to identify significant predictors of the time period of Likert scale ranging from 0=ever to 4=always (39). RTW, candidate predictor variables were entered into Occupational and work-related characteristics were a Cox’s proportional hazards model to calculate hazard assessed using questions and brief questionnaires devel- ratios. To provide an estimate of the magnitude of the oped and psychometrically evaluated by Bürger et al group differences, Cohen’s standardized effect size ( ϕ, (40). Occupational information included (i) profes- d, η²) was calculated. Two-tailed significance tests were sional status, (ii) work ability and periods of sick leave conducted using a significance level of P<0.05. 78 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch Table 1. Baseline demographic and medical sample characteristics. [SD=standard deviation; UICC=International Union against Cancer] Study sample P- d / ϕ value Total sample (N=750) Employed (N=702) Unemployed (N=48) Mean SD N % Mean SD N % Mean SD N % Demographic characteristics Age 48.7 6.8 48.7 6.7 48.8 8.0 0.93 Female gender 643 85.7 597 85.0 46 95.8 0.04 0.08 Married 498 66.4 473 67.4 25 52.0 0.03 0.11 Partnership 566 75.5 537 76.5 29 60.4 0.01 0.09 High school/university degree 218 29.1 207 29.5 11 22.9 0.55 Monthly household net income (€) <0.001 0.44 <1000 39 9.3 18 4.9 21 43.8 1000–<2000 136 32.5 122 33.0 14 29.2 2000–<3000 124 29.7 115 31.1 9 18.8 ≥3000 119 28.5 115 31.1 4 8.3 Social class Lower 203 27.1 176 25.1 27 56.3 <0.001 0.18 Middle 441 58.8 421 60.0 20 41.7 Upper 106 14.1 105 15.0 1 2.1 Medical characteristics Cancer entity Breast cancer 446 59.5 419 59.7 27 56.3 0.53 Gynecological cancers 109 14.5 100 14.2 9 18.8 Head and neck cancers 67 8.9 62 8.8 5 10.4 Skin cancer 46 6.1 45 6.4 1 2.1 Colon/rectum cancer 42 5.6 39 5.6 3 6.3 Lung cancer 23 3.1 20 2.8 3 6.3 Hematological neoplasias 17 2.3 17 2.4 0 Months since diagnosis 11.1 8.5 11.1 8.5 10.2 7.3 0.47 Cancer stage (UICC) 0.20 In situ 8 1.1 6 0.9 2 4.2 I 305 41.6 288 42.0 17 35.4 II 268 36.6 248 36.2 20 41.7 III 118 16.1 112 16.4 6 12.5 IV 34 4.6 31 4.5 3 6.3 Clinical characteristics First primary tumor 719 95.9 673 95.9 46 95.8 0.99 Second primary tumor 31 4.1 29 4.1 2 4.2 Disease phase 0.61 Remission 653 87.1 612 87.2 41 85.4 Recurrence/progress/metastasis 72 9.6 68 9.7 1 2.1 Unclear 25 3.3 22 3.1 3 6.3 Treatment phase 0.57 Curative treatment intention 693 92.4 648 92.3 45 93.8 Palliative treatment intention 36 4.8 35 5.0 1 2.1 Unclear 21 2.8 19 2.7 2 4.2 Months since active cancer 7.7 12.5 7.7 12.6 6.5 10.4 0.52 treatment Cancer treatments Surgery 716 95.5 669 95.3 47 97.9 0.40 Radiation therapy 494 65.9 464 66.1 30 62.5 0.61 Chemotherapy 454 60.5 427 60.8 27 56.3 0.53 Hormonal treatment 328 43.7 308 43.9 20 41.7 0.77 Total number of therapies 2.7 1.1 2.7 1.1 2.6 1.1 0.58 Karnofsky performance status 91.7 7.5 91.7 7.5 91.5 6.2 0.81 Without hematological neoplasias. Scand J Work Environ Health 2013, vol 39, no 1 79 Employment predictors among cancer survivors Results ees (97.2%), and the lowest percentage among work- ers (66.4%) (P=0.02) (ϕ=0.12). No group differences were found in clinical characteristics. However, among Baseline occupational characteristics patients with a higher UICC cancer stage (P<0.001) At the beginning of the rehabilitation program, 702 (ϕ=0.18) and palliative treatment, a lower percent- patients (93.6%) were employed and 48 (6.4%) were age (38.9%) returned to work or were re-employed unemployed. Among the employed, 54.9% were on compared to patients with curative treatment (78.1%) physician-classified sick leave (usually the primary care (P<0.001) (ϕ=0.20). Also, the highest percentage of physician). The majority of the working participants patients who did not return to work was observed among worked as employees (75.1%), 18.4% were workers; patients with lung cancer (43%) and head and neck 5.6% were self-employed, and 1.0% worked as civil cancers (58%). Among patients with cancer progress servants. The mean duration of sick leave within the or metastatic cancer, a significantly lower percentage 12-months period prior to the rehabilitation program (38.9%) returned to work or got re-employed compared was 150.6 (SD 107.4, range 1–365) days. The majority to patients in remission (78.6%) (P<0.001) (ϕ=0.17). of patients (84.5%) were motivated either to return to work or be re-employed after rehabilitation. RTW time period and re-employment after rehabilitation Forty-nine percent of patients returned to work imme - RTW and re-employment 12 months after rehabilitation diately after rehabilitation. The mean time until RTW At t , 568 patients (75.7%) had returned to work or were or re-employment was 5.7 (SD 8.6, range 1–45) weeks. re-employed. The percentage of patients who returned to The highest percentage of patients who returned to work work was highest among participants not on sick leave immediately were among those who were not on sick at baseline (92.4%). Twenty-one percent of the patients leave at baseline, belonged to the upper social class, and who were unemployed at baseline managed to get had skin cancer (table 2). employed at a follow-up time point (table 4). Patients who returned to work or were re-employed were slightly Work situation at follow-up younger [mean 48.2, (SD 7.0) years versus mean 50.5 (SD 6.1) years] (P<0.001) (η²=0.021). No gender differ- Among the 568 patients who returned to work, the ences were observed. The highest percentage of patients majority (81.2%) returned to their former position and who returned to work was observed among employ- workplace; 46% worked full-time. More women (77.8%) Table 2. Time period of return to work (RTW) and reemployment after rehabilitation (N=568). Employed patients N (t ) Time period to RTW / re-employment P-value 0 ϕ / η² Patients who returned to work Weeks until RTW / immediately after rehabilitation re-employment N % Mean SD a b a b Baseline employment status <0.001 / <0.001 0.68 / 0.24 Patients working 268 228 85.1 1.64 2.46 Patients on sick leave 290 48 16.6 8.90 9.83 Unemployed patients 10 2 20.0 21.60 16.67 Total 568 278 48.9 5.68 8.60 a b a Social class 0.008 / 0.63 0.13 Lower social class 133 53 39.8 6.23 8.23 Middle social class 343 169 49.3 5.41 8.44 Upper social class 92 56 60.9 5.89 9.64 a b a b Cancer entity 0.001 / <0.001 0.20 / 0.05 Breast cancer 348 171 49.1 5.19 7.98 Gynecological cancers 81 38 46.9 5.23 6.77 Head and neck cancers 39 14 35.9 8.03 10.31 Skin cancer 38 30 78.9 2.76 6.00 Colon/rectum cancer 36 16 44.4 10.21 13.41 Lung cancer 10 1 10.0 13.70 14.21 Hematological neoplasias 16 8 50.0 5.13 7.53 Significance and effect size refer to group differences in patients who returned to work immediately after rehabilitation. Significance and effect size refer to group differences in weeks until RTW/reemployment. 80 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch worked part-time compared to 22.2% of men (P<0.001) sus palliative treatment intention), months since active (ϕ=0.25). Fifty-two percent of the patients were on sick cancer treatment, total numbers of therapy, anxiety, dis- leave absence at least once after rehabilitation for an tress, and the expected period of RTW. Table 3 presents average duration of 62 days. the correlations between the final candidate predictor variables and the outcome variables. The first block entered into the multivariate hierarchi - Predictors of RTW – logistic regression model cal logistic regression model consisted of demographic Before testing the logistic regression model, we per- factors (table 4). The variables education and social sta- formed correlations between variables in one block for tus were excluded from the regression model, whereas all predictor candidates. In candidate predictor vari- age and monthly household net income remained in the ables with an intercorrelation of r≥0.60, one variable model (Nagelkerke’s R²=0.07) (P<0.001). was removed from the regression analysis in order to The second block entered consisted of medical and avoid multicollinearity. The following variables were functional factors. The variables cancer entity, months removed: marital status, treatment phase (curative ver- since current diagnosis, UICC cancer stage, and number Table 3. Correlations between candidate predictor variables (t ) and outcome variables. [UICC=International Union against Cancer.] Candidate predictor variables (t ) 12-months return to work Time period of return to work (weeks) r P-value r P-value Demographic factors Age -0.14 >0.001 -0.13 <0.001 Education 0.12 0.001 -0.12 0.001 Monthly household net income 0.18 >0.001 -0.20 <0.001 Social status 0.16 >0.001 -0.16 <0.001 Medical and functional factors Cancer entity Breast cancer 0.07 0.08 -0.08 0.06 Gynecological cancers -0.01 0.71 0.03 0.55 Head and neck cancers -0.13 >0.001 0.05 0.26 Skin cancer 0.04 0.26 -0.09 0.05 Colon/rectum cancer 0.06 0.12 0.11 0.01 Lung cancer -0.13 >0.001 0.13 0.003 Hematological neoplasias 0.07 0.07 -0.03 0.52 Months since current diagnosis 0.09 0.03 -0.30 >0.001 Cancer stage (UICC) -0.15 >0.001 -0.15 >0.001 Disease phase Remission 0.17 >0.001 0.007 0.87 Recurrence/progress/metastasis -0.12 0.004 0.07 0.12 Number of functional impairments -0.19 >0.001 0.26 >0.001 Number of physical problems -0.19 >0.001 0.26 >0.001 Karnofsky status 0.30 >0.001 -0.34 >0.001 Pain -0.25 >0.001 0.32 >0.001 Psychosocial factors Fear of cancer recurrence -0.21 >0.001 0.08 0.05 Depression -0.31 >0.001 0.09 0.03 Mental quality of life 0.23 >0.001 -0.06 0.17 Physical quality of life 0.32 >0.001 -0.24 >0.001 Social support 0.10 0.007 0.02 0.71 Detrimental interactions -0.21 >0.001 0.04 0.32 Work-related factors Employment and work ability Working 0.31 >0.001 -0.63 >0.001 Sick leave -0.14 >0.001 0.46 >0.001 Unemployed -0.34 >0.001 0.30 >0.001 Duration of sick leave (days) prior to rehabilitation -0.30 >0.001 0.47 >0.001 Self-perceived work ability -0.48 >0.001 0.37 >0.001 Perceived threat of job loss -0.17 >0.001 0.02 0.64 Intention to return to work 0.50 >0.001 -0.05 0.28 Expected period of return to work after rehabilitation -0.33 >0.001 0.53 >0.001 Perceived employer accommodation 0.34 >0.001 -0.07 0.13 Job requirements -0.11 0.004 0.08 0.06 Overall job satisfaction 0.13 >0.001 -0.02 0.65 Scand J Work Environ Health 2013, vol 39, no 1 81 Employment predictors among cancer survivors Table 4. Multivariate logistic regression model for the identification of significant predictors of 12-months (re-) employment after rehabilitation. [OR=odds ratio; SD=standard deviation; SE=standard error; 95% CI=95% confidence interval] (Re-)employment ß SE OR 95% CI P-value Yes (N=568) No (N=182) Mean SD N % Mean SD N % Block 1 Demographic <0.001 factors (Nagelkerke’s R²=0.07) Age 48.2 7.0 50.5 6.1 -0.05 0.03 0.94 0.90–1.00 0.054 Monthly household net income (€) <1000 29 46.8 33 53.2 0.21 1000–<2000 183 73.8 65 26.2 1.20 0.72 3.33 0.87–13.73 0.10 2000–<3000 190 78.5 52 21.5 0.74 0.70 2.09 0.53–8.31 0.29 ≥3000 166 83.8 32 16.2 0.41 0.72 1.50 0.37–6.18 0.57 Block 2 Medical and <0.001 functional factors (Nagelkerke’s R²=0.32) Disease phase Remissiona 513 78.6 140 21.4 0.002 Recurrence/progress/ 55 56.7 42 43.3 -1.31 0.43 0.27 0.12–0.63 metastasis Number of functional 2.3 3.0 4.2 4.6 -0.01 0.04 1.00 0.91–1.08 0.78 impairments Karnofsky status 93.0 7.5 87.8 7.1 0.05 0.02 1.05 1.01–1.09 0.01 Pain 2.5 2.1 4.1 2.8 -0.09 0.09 0.91 0.77–1.08 0.28 Physical quality of life 50.6 9.0 43.7 7.8 0.02 0.03 1.02 0.99–1.08 0.51 Block 3 Psychosocial <0.001 factors (Nagelkerke’s R²=0.39) Mental quality of life 53.2 10.1 47.6 11.1 0.03 0.02 1.03 1.00–1.07 0.08 Detrimental interactions 1.0 0.6 1.3 0.6 -0.55 0.28 0.58 0.33–0.99 0.05 Block 4 Work-related <0.001 factors (Nagelkerke’s R²=0.56) Employment Working 268 92.4 22 7.6 0.05 Sick leave 290 70.4 122 29.6 -1.36 0.56 0.26 0.09–0.77 0.02 Unemployed 10 20.8 38 79.2 -1.66 1.12 0.19 0.02–1.68 0.14 Self-perceived work 3.4 1.0 2.2 1.0 0.41 0.22 1.50 0.99–2.30 0.06 ability Intention to return to 538 84.9 96 15.1 1.83 0.58 6.22 1.98–19.51 0.002 work Perceived employer 3.8 0.9 2.8 1.2 0.66 0.16 1.93 1.41–2.65 <0.001 accommodation Job requirements 3.8 0.9 2.8 1.2 0.61 0.30 1.84 1.02–3.30 0.04 Reference category. of physical problems were excluded from the regression to rehabilitation, perceived threat of job loss, expected model (Nagelkerke’s R²=0.32) (P<0.001). Patients with period of RTW after rehabilitation, and overall job cancer recurrence or cancer progress had a significant satisfaction. Patients who intended to return to work lower chance to return to work compared to patients in at baseline were more than six times as likely to do so cancer remission. (P<0.001). Also, patients who perceived their employer The third block entered consisted of psychosocial as being cooperative and accommodating of their can- factors. The variables fear of recurrence, depression, cer were more likely to return to work. Although the and social support were excluded from the regression variable job requirements showed a negative correla- model (Nagelkerke’s R²=0.39) (P<0.001). Individuals tion with RTW, it also emerged as a positive predictor with a higher amount of problematic social interactions for RTW. This is likely an effect of suppression since were less likely to return to work. high job requirements are positively associated with The forth block entered consisted of work-related RTW among individuals in the higher social class but factors. The following variables were excluded from negatively associated with RTW among patients in the regression model: duration of sick leave (days) prior lower social class. When the variable is entered as a 82 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch single factor into a regression analysis, it remains as symptom burden at the beginning of the rehabilitation a negative predictor of RTW [ß=-0.41, OR=0.67, 95% program. In our study, the employment rate was 76% confidence interval (95% CI) 0.48–0.92, P=0.01]. The on average 23 months after diagnosis and consider - explained variance of the total model (Nagelkerke’s R²) ably higher compared to findings showing that overall, was 0.59 (P<0.001). about 63.5% of cancer patients (range 24–94%) manage to return to work (17). However, comparable results (73%, range 64–82%) have been found at 18 months Predictors of time period to RTW after the cancer diagnosis among breast cancer patients The following variables remained in the model as signif- (14) and among samples with mixed tumor entities (1, icant predictors: UICC cancer stage, Karnofsky index, 8, 25). Concordant with previous studies showing work physical and mental quality of life, employment charac- changes in 8–17% of cancer survivors, we found similar teristics, intention to return to work, perceived employer results (25, 30). accommodation, and job requirements (overall model Both the absence of sick leave and belonging to fit - 2 log likelihood = 3858.2; χ²[total model] = 175.6, a higher social class at baseline were associated with df = 13, P<0.001) (table 5). a significant higher rate of employment at follow-up. Belonging to a higher social class is related to work environments that may provide more favorable work- ing conditions in terms of flexibility, a lower degree of Discussion manual and physically exhausting work, better earnings, and better living conditions. Accordingly, it has been The majority of patients (85%) were motivated either stated that employees with cancer or other persistent to return to work or be re-employed after rehabilitation. health problems generally need some flexibility at vari - These findings are consistent with results indicating ous aspects and times at work (42). Atkinson et al (43) the significance of work and the strong motivation to and Gudbergson et al (44) pointed out that living condi - continue work during treatment or return to work after tions include the social indicators that stimulate social treatment completion among survivors (41). The explo- inclusion and reduce social exclusion. These indicators ration of work characteristics revealed that slightly more are economy, education, employment, health, housing, than half of participants (55%) were on sick leave at and social participation. baseline indicating a high physical and/or psychological The highest percentage of patients who did not return Table 5. Cox’s proportional hazards regression model for the identification of significant predictors of the time to return-to-wor k. [HR=hazard ratio; SD=standard deviation; SE=standard error; UICC=International Union against Cancer; 95% CI=95% confidence interval. ] Time period (weeks) to SE HR 95% CI P-value return to work Mean SD Months since current diagnosis 0.01 0.01 1.01 0.99–1.03 0.35 Cancer stage (UICC) I 13.90 19.65 0.02 II 18.07 21.44 -0.27 0.12 0.76 0.60–0.97 0.03 III 20.05 22.66 -0.37 0.16 0.69 0.51–0.94 0.02 IV 29.18 24.74 -0.59 0.26 0.56 0.34–0.92 0.02 Karnofsky status 0.02 0.01 1.02 1.00–1.04 0.04 Number of functional impairments -0.01 0.02 0.99 0.96–1.02 0.49 Physical quality of life 0.02 0.01 1.02 1.01–1.04 0.001 Mental quality of life 0.01 0.01 1.01 1.00–1.03 0.009 Employment Working 5.48 13.56 <0.001 Sick leave 21.61 21.37 -0.99 0.16 0.37 0.27–0.51 <0.001 Unemployed 45.67 14.45 -1.41 0.41 0.25 0.11–0.55 0.001 Intention to return to work No 40.53 20.18 -0.49 0.20 0.61 0.41–0.91 0.02 Yes 12.59 18.38 Perceived employer accommodation 0.17 0.06 1.18 1.06–1.32 0.002 Job requirements 0.17 0.09 1.19 1.01–1.41 0.04 Without hematological neoplasias Reference category Scand J Work Environ Health 2013, vol 39, no 1 83 Employment predictors among cancer survivors to work was observed among patients with lung cancer a positive predictor for RTW likely due to suppression (43%) and head and neck cancers (58%), among patients effects. Our findings emphasize that more research is with advanced cancer stage, progress or metastatic can- needed particularly to investigate the kind of job require- cer, and palliative treatment intention. The adverse effect ments (such as having plenty of work or high time pres- of cancer progress and poor physical functioning on sure) that might lead to early retirement, unemployment, RTW has been shown in several previous studies (7, 23, or a higher probability to RTW. Our results point toward 27). Corresponding to previous research, we found re- the fact that high job requirements are positively associ- employment significantly associated with younger age ated with RTW among individuals in the higher social and a professional status as an employee (24, 27, 45). class but are negatively associated with RTW among Our regression analyses findings emphasize the patients in the lower social class. Individuals belonging importance of volitional factors for RTW. Patients who to a higher social class are more likely to have better expressed their intention to return to work at the begin- working conditions such as flexible work, responsibility, ning of the cancer rehabilitation program were six times and a considerable amount of decision-making freedom more likely to do so compared to patients who did not that might compensate the high workload. intend to return to work. Our findings show that the Although this prospective study includes a large patient can best predict RTW at an early stage of the sample size compared to previous research, this study rehabilitation process. However, so far only limited has several methodological limitations. Both the initial knowledge exists about motivational and volitional fac- and follow-up non-response lead to a bias in several tors and its association with demographic, family and sociodemographic and psychosocial outcome variables work-related aspects (46, 47). Occupational motivation of interest in this research. With regard to the generality or skepticism towards RTW should be carefully assessed and interpretation of the findings, a sample bias must be at the beginning of rehabilitation programs and during considered toward: (i) female gender, (ii) younger age, the joint establishment of rehabilitation aims between (iii) being in a cancer rehabilitation program, (iv) cancer the patients and the professional team (48). Furthermore, entities associated with a better physical health status and factors influencing occupational motivation among prognosis, and (v) better psychological well-being. Given cancer survivors need to be understood in more detail. this bias, our findings with regard to employment might The strong influence of work-related aspects next overestimate the degree to which cancer patients return to medical factors such as disease phase confirm the to work or stay employed. Nevertheless, systematic dif- current state of the literature with regard to factors ferences between participants and non-participants were positively and adversely influencing RTW among can - small in view of effect sizes and at least partially a conse- cer survivors (7, 14, 23, 25, 26, 45). Consistent with quence of the relatively large sample size. Although 86% the literature, the perceived employer accommodation of the patients were women, the overall large sample size indicated a fairly positive attitude towards employment, including 107 men justifies gender analyses; however, work-related support, and necessary job changes (14, gender did not correlate with employment status and 27, 49) and emerged as a significant predictor for RTW. therefore was excluded from further regression analyses. In our study, analyses also demonstrated that the Another limitation is that due to the allocation pro- patients who were working at baseline were more likely cess mainly regulated by the German pension insur - to return to work after cancer rehabilitation and were ance, it was not feasible to randomize the study sample. more likely to do so at an earlier stage than patients on Despite the fact that cancer rehabilitation programs are sick leave or – unsurprisingly – unemployed patients. provided to every cancer patient in Germany, our sample Self-perceived work ability, in contrast, did not emerge consists only of patients who use the services provided. as a significant predictor for RTW. The relevance of This might lead to a bias towards a sample with high perceived work ability has been emphasized in a num - physical and psychosocial impairments. Furthermore, ber of studies (3, 15, 44, 50–53). However, only limited the inclusion of a matched control group could not be knowledge exists about the association between per- realized since most patients with rehabilitation needs ceived work ability and sick leave absence. are referred to a rehabilitation program. Thus, this study Furthermore, we found that detrimental social inter- could not determine possible effects of the rehabilitation actions were inversely associated with RTW. Patients program on RTW. with problematic social interactions in their personal The (re-)integration of cancer survivors into working environment might also lack social skills at the work- life is one important aspect of participation according place or might have fewer personal resources to effec- to the ICF (31). Rehabilitation programs are important tively adapt to the challenges of being a cancer survivor not only for the physical and psychosocial recovery, but in the work environment. for the labor market reintegration of patients. Profound The variable “high job requirements” showed a neg- understanding of cancer and treatment-induced impair- ative correlation with RTW. However, it also emerged as ments and their impact on daily activities and work is 84 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch an essential basis for the development of better educa- 9. Tiedtke C, Rijk A de, Dierckx Casterlé B de, Christiaens M, Donceel P. Experiences and concerns about ‘returning to work’ tional, rehabilitative, and occupational interventions in for women breast cancer survivors: a literature review. Psycho- cancer care (54). A better understanding of cancer and Oncology. 2010;19:677–83. http://dx.doi.org/10.1002/ treatment-induced physical, cognitive and psychosocial pon.1633. treatment consequences related to work-related prob- lems will help to develop interventions and educational 10. Feuerstein M, Todd BL, Moskowitz MC, Bruns GL, Stoler MR, Nassif T et al. Work in cancer survivors: a model for programs for patients, healthcare professionals, and practice and research. J Cancer Surviv. 2010;4:415–37. http:// employers to better address the professional needs of dx.doi.org/10.1007/s11764-010-0154-6. individuals with cancer. 11. Thijs KM, Boer AGEM, Vreugdenhil G, Wouw AJ, Houterman S, Schep G. Rehabilitation Using High-Intensity Physical Training and Long-Term Return-to-Work in Cancer Survivors. J Occup Rehabil. 2011 Nov 12. [Epub ahead of print]. http:// Acknowledgements dx.doi.org/10.1007/s10926-011-9341-1. We thank Birgit Leibbrand, MD, Jürgen Barth, MD, 12. McGrath PD, Hartigan B, Holewa H, Skarparis M. Returning to work after treatment for haematological Manfred Gaspar, MA, Gerhard Friedrich, MD, Wilhelm cancer: findings from Australia. Support Care Can cer. 2011 Bootsveld, MD, Ulrich Gärtner, MD, and Christine- Oct 28. [Epub ahead of print] . http://dx.doi.org/10.1007/ Daniela Kegel, MD, for their data collection and man - s00520-011-1298-2. agement support. This research was funded by the Nordrhein-Westfalen Association for the Fight Against 13. Roelen CAM, Koopmans PC, Groothoff JW, Klink JJL, Bültmann U. Return to Work After Cancer Diagnosed in 2002, Cancer, Germany. 2005 and 2008. J Occup Rehabil. 2011;21:335–41. http:// dx.doi.org/10.1007/s10926-011-9319-z. 14. Bouknight RR, Bradley CJ, Zhehui L. Correlates of Return to References Work for Breast Cancer Survivors. J Clin Oncol. 2006;24:345– 53. http://dx.doi.org/10.1200/JCO.2004.00.4929. 1. Spelten E. Cancer, fatigue and the return of patients to work—a 15. Taskila T, Lindbohm ML. Factors affecting cancer survivors‘ prospective cohort study. Eur J Cancer. 2003;39:1562–7. employment and work ability. Acta Oncol. 2007;46:446–51. http://dx.doi.org/10.1016/S0959-8049(03)00364-2. http://dx.doi.org/10.1080/02841860701355048. 2. Verbeek J. Return to work of cancer survivors: a prospective 16. Park J, Park E, Park J, Kim S, Lee S. Job loss and re-employment cohort study into the quality of rehabilitation by occupational of cancer patients in Korean employees: a nationwide physicians. Occup Environ Med. 2003;60:352–7. http:// retrospective cohort study. J Clin Oncol. 2008;26:1302–9. dx.doi.org/10.1136/oem.60.5.352. http://dx.doi.org/10.1200/JCO.2007.14.2984. 3. Bradley CJ, Neumark D, Luo Z, Bednarek H, Schenk M. 17. Mehnert A. Employment and work-related issues in cancer Employment Outcomes of Men Treated for Prostate Cancer. survivors. Crit Rev Oncol Hemat. 2011;77:109–30. http:// J Natl Cancer I. 2005;97:958–65. http://dx.doi.org/10.1093/ dx.doi.org/10.1016/j.critrevonc.2010.01.004. jnci/dji171. 18. Peteet JR. Cancer and the meaning of work (eng). Gen Hosp Psychiatry. 2000;22:200–5. http://dx.doi.org/10.1016/S0163- 4. Hoffman B. Cancer Survivors at Work: A Generation of 8343(00)00076-1. Progress. CA-Cancer J Clin. 2005;55:271–80. http://dx.doi. org/10.3322/canjclin.55.5.271. 19. Rasmussen DM, Elverdam B. The meaning of work and working life after cancer: an interview study. Psycho- 5. Feuerstein M. Cancer Survivorship and Work. J Occup Oncology. 2008;17:1232–8. http://dx.doi.org/10.1002/ Rehabil. 2005;15:1–2. http://dx.doi.org/10.1007/s10926- pon.1354. 005-0868-x. 6. Feuerstein M, Harrington CB. Recommendations for the 20. Choi KS, Kim E, Lim J, Kim S, Lim MK, Park J et al. Job U.S. National Occupational Research Agenda: Research on loss and reemployment after a cancer diagnosis in Koreans—a Cancer Survivorship and Musculoskeletal Disorders and prospective cohort study. Psycho-Oncology. 2007;16:205–13. Work Disability. J Occup Rehabil. 2006;16:1–5. http://dx.doi. http://dx.doi.org/10.1002/pon.1054. org/10.1007/s10926-005-9004-1. 21. Carlsen K, Dalton SO, Diderichsen F, Johansen C. Risk for unemployment of cancer survivors: A Danish cohort study. 7. Pryce J, Munir F, Haslam C. Cancer survivorship and work: Eur J Cancer. 2008;44:1866–74. http://dx.doi.org/10.1016/j. Symptoms, supervisor response, co-worker disclosure and ejca.2008.05.020. work adjustment. J Occup Rehabil. 2007;17:83–92. http:// dx.doi.org/10.1007/s10926-006-9040-5. 22. Boer AGEM de, Taskila T, Ojajärvi A, van Dijk FJH, Verbeek JHAM. Cancer survivors and unemployment: a meta-analysis 8. Boer AGEM de, Verbeek JHAM, Spelten ER, Uitterhoeve and meta-regression. J Am Med Assoc. 2009;301:753–62. ALJ, Ansink AC, Reijke TM de et al. Work ability and return- http://dx.doi.org/10.1001/jama.2009.187. to-work in cancer patients. Br J Cancer. 2008;98:1342–7. http://dx.doi.org/10.1038/sj.bjc.6604302. Scand J Work Environ Health 2013, vol 39, no 1 85 Employment predictors among cancer survivors 23. Spelten ER, Sprangers MA, Verbeek JH. Factors reported to 37. Mehnert A, Herschbach P, Berg P, Henrich G, Koch U. Fear of influence the return to work of cancer survivors: a literature progression in breast cancer patients--validation of the short review. Psycho-Oncology. 2002;11:124–31. http://dx.doi. form of the Fear of Progression Questionnaire (FoP-Q-SF). Z org/10.1002/pon.585. Psychosom Med Psychother. 2006;52:274–88. 24. Taskila-Abrandt T, Martikainen R, Virtanen S, Pukkala E, 38. Ware JE, Kosinski M, Dewey JE, Gandek B. How to Score Hietanen P, Lindbohm M. The impact of education and and Interpret Single-Item Health Status Measures: A Manual occupation on the employment status of cancer survivors. for Users of the SF-8TM Health Survey. Lincoln, RI: Eur J Cancer. 2004;40:2488–93. http://dx.doi.org/10.1016/j. QualityMetric Incorporated; 1999. ejca.2004.06.031. 39. Ramm GC, Hasenbring M. Die deutsche Adaptation des Illness-specific Social Support Scale und ihre teststatistische 25. Amir Z, Moran T, Walsh L, Iddenden R, Luker K. Return to Überprüfung beim Einsatz an Patienten vor und nach paid work after cancer: a British experience. J Cancer Surviv. Knochenmarktransplantation [The German adaptation of the 2007;1:129–36. http://dx.doi.org/10.1007/s11764-007- Illness-specific Social Support Scale and the test statistical 0021-2. evaluation on the basis of patients undergoing bone marrow 26. Chan F, Strauser D, da Silva Cardoso E, Xi Zheng L, Chan transplantation]. Z Med Psychol 2003;12, 29-38. JYC, Feuerstein M. State vocational services and employment in cancer survivors. J Cancer Surviv. 2008;2:169–78. http:// 40. Bürger W, Dietsche S, Morfeld M, Koch U. Multiperspective dx.doi.org/10.1007/s11764-008-0057-y. estimates on the probability of patient return to work following orthopaedic rehabilitation: findings and predictive relevance. 27. Molina Villaverde R, Feliu Batlle J, Villalba Yllan A, Jimenez Rehabilitation. 2001;40:217–25. Gordo AM, Redondo Sanchez A, San Jose Valiente B et al. Employment in a cohort of breast cancer patients. Occup Med. 41. Mahar KK, BrintzenhofeSzoc K, Shields JJ. The impact of 2008;58:509–11. http://dx.doi.org/10.1093/occmed/kqn092. changes in employment status on psychosocial well-being: a study of breast cancer survivors. J Psychosoc Oncol. 2008;26:1– 28. Ross L, Petersen MA, Johnsen AT, Lundstroem LH, Carlsen K, 17. http://dx.doi.org/10.1080/07347330802115400. Groenvold M. Factors associated with Danish cancer patients’ return to work. A report from the population-based study ‘The 42. Feuerstein M, Luff GM, Harrington CB, Olsen CH. Pattern of Cancer Patient‘s World’. Cancer Epidemiol. 2011 Aug 12. workplace disputes in cancer survivors: a population study of [Epub ahead of print]. ADA claims. J Cancer Surviv. 2007;1:185–92. http://dx.doi. org/10.1007/s11764-007-0027-9. 29. Short PF, Vasey JJ, Tunceli K. Employment pathways in a large cohort of adult cancer survivors. Cancer. 2005;103:1292–301. 43. Atkinson T, Cantillon B, Marlier E, Nolan B. Social indicators. http://dx.doi.org/10.1002/cncr.20912. The EU and social inclusion. Oxford: Oxford University Press; 30. Gudbergsson SB, Fosså SD, Dahl AA. A study of work changes due to cancer in tumor-free primary-treated cancer patients. A 44. Gudbergsson SB, Fosså SD, Borgeraas E, Dahl AA. A NOCWO study. Support Care Cancer. 2008;16:1163–71. comparative study of living conditions in cancer patients who http://dx.doi.org/10.1007/s00520-008-0407-3. have returned to work after curative treatment. Support Care Cancer. 2006;14:1020–9. http://dx.doi.org/10.1007/s00520- 31. World Health Organization (WHO). ICF—International 006-0042-9. Classification of Functioning, Disability and Health. Genf; 45. Hensel M. Quality of life and rehabilitation in social and professional life after autologous stem cell transplantation. 32. Hellbom M, Bergelt C, Bergenmar M, Gijsen B, Loge JH, Ann Oncol. 2002;13:209–17. http://dx.doi.org/10.1093/ Rautalathi M et al. Cancer rehabilitation: A Nordic and annonc/mdf031. European perspective. Acta Oncol. 2011;50:179–86. http:// dx.doi.org/10.3109/0284186X.2010.533194. 46. Nachreiner N, Dagher R, McGovern P, Baker B, Alexander B, Gerberich S. Successful return to work for cancer survivors. 33. Winkler J, Stolzenberg H. Der Sozialschichtindex AAOHN J. 2007;55:290–5. im Bundesgesundheitssurvey [The social class index according to the national health survey]. Gesundheitswesen. 47. Johnsson A, Fornander T, Olsson M, Nystedt M, Johansson H, 1999;61:S178-S183. Rutqvist L. Factors associated with return to work after breast cancer treatment. Acta Oncol. 2007;46:90–6. http://dx.doi. 34. Karnofsky DA, Burchenal JH. The Clinical Evaluation of org/10.1080/02841860600857318. Chemotherapeutic Agents in Cancer. In: MacLeod CM, editor. Evaluation of Chemotherapeutic Agents. New York: Columbia 48. Thies S, Lehmann C, Kriz D, Nubling R, Mehnert A. University Press; 1949, p. 191–205. Patientenfragebogen zur Erfassung der Reha-Motivation (PAREMO-20) - Testtheoretische Überprüfung und Validierung 35. Cleeland CS, Ryan KM. Pain assessment: global use of an einer Stichprobe von Krebspatienten unterschiedlicher the Brief Pain Inventory (eng). Ann Acad Med Singap. Diagnosegruppen . Rehabilitation. 2008;47:308–18. http:// 1994;23:129–38. dx.doi.org/10.1055/s-2008-1076709. 36. Herrmann C, Buss U, Snaith RP. HADS-D – Hospital Anxiety and Depression Scale – Deutsche Version: Ein Fragebogen 49. Steiner JF, Cavender TA, Nowels CT, Beaty BL, Bradley CJ, zur Erfassung von Angst und Depressivität in der somatischen Fairclough DL et al. The impact of physical and psychosocial Medizin. Bern: Huber; 1995. factors on work characteristics after cancer. Psycho-Oncology. 2008;17:138–47. http://dx.doi.org/10.1002/pon.1204. 86 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch 50. Short PF, Vasey JJ, BeLue R. Work disability associated 53. Torp S, Nielsen RA, Gudbergsson SB, Dahl AA. Worksite with cancer survivorship and other chronic conditions. adjustments and work ability among employed cancer Psycho-Oncology. 2008;17:91–7. http://dx.doi.org/10.1002/ survivors. Support Care Cancer. 2011 Nov 16. [Epub ahead pon.1194. of print]. http://dx.doi.org/10.1007/s00520-011-1325-3. 51. Lee MK, Lee KM, Bae J, Kim S, Kim Y, Ryu KW et al. 54. Tamminga S, de B, Verbeek J, Taskila T, Frings-Dresen M. Employment status and work-related difficulties in stomach Enhancing return-to-work in cancer patients, development of cancer survivors compared with the general population. an intervention and design of a randomised controlled trial. Br J Cancer. 2008;98:708–15. http://dx.doi.org/10.1038/ BMC Cancer. 2010;10:345. http://dx.doi.org/10.1186/1471- sj.bjc.6604236. 2407-10-345. 52. Ahn E, Cho J, Shin DW, Park BW, Ahn SH, Noh D et al. Impact of breast cancer diagnosis and treatment on work- related life and factors affecting them. Breast Cancer Res Treat. 2009;116:609–16. http://dx.doi.org/10.1007/s10549- Received for publication: 12 December 2012 008-0209-9. Scand J Work Environ Health 2013, vol 39, no 1 87 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Scandinavian Journal of Work, Environment & Health Unpaywall

Predictors of employment among cancer survivors after medical rehabilitation – a prospective study

Scandinavian Journal of Work, Environment & HealthMar 15, 2012

Loading next page...
 
/lp/unpaywall/predictors-of-employment-among-cancer-survivors-after-medical-SMO8vvSXsp

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0355-3140
DOI
10.5271/sjweh.3291
Publisher site
See Article on Publisher Site

Abstract

Downloaded from www.sjweh.fi on November 17, 2021 Original article Scand J Work Environ Health 2013;39(1):76-87 doi:10.5271/sjweh.3291 Predictors of employment among cancer survivors after medical rehabilitation  a prospective study by Mehnert A, Koch U Affiliation: Department and Outpatient Clinic of Medical Psychology, University Medical Center Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. mehnert@uke.de The following article refers to this text: 2020;46(4):382-391 Key terms: cancer; cancer survivor; cancer survivorship; employment; predictor; prospective study; psycho-oncology; rehabilitation; return to work; RTW This article in PubMed: www.ncbi.nlm.nih.gov/pubmed/22422040 This work is licensed under a Creative Commons Attribution 4.0 International License. Print ISSN: 0355-3140 Electronic ISSN: 1795-990X Original article Scand J Work Environ Health. 2013;39(1):76–87. doi:10.5271/sjweh.3291 Predictors of employment among cancer survivors after medical rehabilitation – a prospective study by Anja Mehnert, PhD,¹ Uwe Koch, PhD, MD ¹ ² Mehnert A, Koch U. Predictors of employment among cancer survivors after medical rehabilitation – a prospective study. Scand J Work Environ Health. 2013;39(1):76–87. doi:10.5271/sjweh.3291 Objectives This study aimed to (i) investigate cancer survivor’s employment status one year after the comple- tion of a medical rehabilitation program and (ii) identify demographic, cancer, and psychosocial, treatment-, and work-related predictors of return to work (RTW) and time until RTW. Methods A total of 1520 eligible patients were consecutively recruited on average 11 months post diagnosis and assessed at the beginning (t ) (N=1148) and end of rehabilitation (t ) (N=1060) and 12 months after rehabilitation 0 1 (t ) (N=750). Participants completed validated measures assessing functional impairments, pain, anxiety, depres- sion, quality of life, social support, and work-related characteristics including work ability, sick leave absence, job requirements, work satisfaction, self-perceived employer accommodation, and perceived job loss. Physicians estimated the degree of cancer-entity-specific functional impairment. Results In a mean time of six weeks after rehabilitation, 568 patients (76%) had returned to work. The multi- variate hierarchical logistic regression analysis indicated that baseline RTW intention [odds ratio (OR) 6.22, 95% confidence interval (95% CI) 1.98–19.51], perceived employer accommodation (OR 1.93, 95% CI 0.33–0.99), high job requirements (OR=1.84, 95% CI 1.02–3.30), cancer recurrence or progression (OR=0.27, 95% CI 0.12 – 0.63), baseline sick leave absence (OR=0.26, 95% CI 0.09–0.77), and problematic social interactions (OR=0.58, 95% CI 0.33–0.99) emerged as significant predictors for RTW. The explained variance of the total model was Nagelkerke’s R²=0.59 (P<0.001). Conclusion Our findings emphasize the high relevance of motivational factors. Occupational motivation and skepticism towards returning to work should be carefully assessed at the planning of the rehabilitation program. Key terms cancer survivorship; psycho-oncology; return to work; RTW. Since an increasing number of patients are likely to Previous research suggested that cancer increases return to work after diagnosis and treatment comple- the risk of unemployment among survivors compared to tion, there is an increasing recognition of the short healthy controls (20–22). On average, 63.5% of cancer and long-term impact of cancer and both its physical survivors (range 24–94%) return to work (17). Overall, and psychosocial consequences on employment during studies indicate a steady increase of return to work (RTW) the last years (1–13). Unfavorable cancer and treat - from on average 40% at six months post diagnosis to 62% ment consequences include a variety of physical and at 12 months, 73% at 18 months, and 89% at 24 months. functional disabilities, and psychological distress that Factors significantly associated with a greater likelihood may adversely affect a patient’s work ability, work sat- of being employed or RTW include perceived employer isfaction, as well as employment status (14–17). Since accommodation, flexible working arrangements, coun - work has the potential to help patients regain a sense of seling, training and rehabilitation services, younger age, meaning, normalcy and being valued, returning to work higher education, male gender, a lower physical symptom may comprise a range of positive consequences for the burden, cancer remission, shorter length of sick leave, and recovery and the psychological well-being (18, 19). continuity of care (7, 14, 17, 23–28). Department and Outpatient Clinic of Medical Psychology, Center for Psychosocial Medicine and University Cancer Center Hamburg (UCCH), University Medical Center Hamburg-Eppendorf, Hamburg, Germany. Deanery Medical Faculty and Department and Outpatient Clinic of Medical Psychology, Center for Psychosocial Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany. Correspondence to: Anja Mehnert PhD, Department and Outpatient Clinic of Medical Psychology, University Medical Center Hamburg- Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. [E-mail: mehnert@uke.de] 76 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch However, although important findings have emerged participation. Consecutive patients were recruited from in previous studies on work and employment among four cancer rehabilitation facilities and assessed at the cancer survivors (7, 29, 30), only limited knowledge beginning (t ) and end of rehabilitation (t ), and 12 0 1 exists about RTW after rehabilitation taking into account months after rehabilitation (t ). Inclusion criteria com- the impact of demographic characteristics, cancer and prised (i) age 18–60 years, (ii) the capability to complete treatment-related physical factors, psychosocial as well study measures, and (iii) the absence of permanent inva- as of work-related aspects of employment and RTW. lidity and early retirement. Since the old age pension in The Word Health Organization (WHO) (31) has defined Germany generally becomes effective at 65 years, the rehabilitation as “the use of all means aimed at reducing age limit of 60 years was chosen to enable patients to the impact of disabling and handicapping conditions and have a sufficient time period left to return to work. at enabling people with disabilities to achieve optimal A total of 1148 patients were enrolled at t (75.5% social integration”. Medical rehabilitation programs participation rate), whereas 372 patients declined to in Germany are provided to cancer patients accord - participate. Among those who participated at t , 1060 ing to the overall aim defined by the WHO and the completed the questionnaires at t . At t , questionnaires 1 2 International Classification of Functioning, Disability were mailed to an eligible 994 patients (36 patients and Health (ICF) (31). Thus, the (re-)integration of had moved to an unknown address and 30 had died), individuals with disabilities, chronic health conditions, 750 (75%, 65% of the total sample) of whom returned diseases, and handicaps into society and working life is questionnaires that could be evaluated (figure 1). Table one important aspect of rehabilitation by eliminating or 1 presents baseline sample characteristics. reducing the impact of chronic illness and disability. The aim is to maintain a patient’s optimal physical, sensory, Non-responder analyses psychological, and social functional levels. Rehabilita- tion also serves to prevent an impending disability or the At t , participants and non-participants differed in terms aggravation of existing physical damages. of age [mean 48.5, standard deviation (SD) 7.2, years Based on social laws in Germany, cancer patients have versus mean 50.4, SD 6.1, years] (P<0.001) (d=0.3) and a legal right to participate in at least one rehabilitation male gender (16% versus 22%) (P=0.007) (ϕ=0.07). No program after the completion of primary cancer treat- group differences in cancer entities were observed. At t , ments (32). Traditionally, cancer rehabilitation programs non-participating patients were more likely to be male are mainly carried out in in-patient settings in specialized (P=0.008) (ϕ=0.09), widowed (P=0.03) (ϕ=0.10), and rehabilitation clinics. Access to cancer rehabilitation pro- have head and neck or lung cancer (P=0.005) (ϕ=0.14). grams is usually facilitated by hospital doctors and social Participants at t were found to be significantly less workers immediately after completion of the primary depressed (P=0.01) (η²=0.007) and had a lower level of treatment (“follow-up rehabilitation”). However, cancer fear of cancer recurrence (P=0.009) (η²=0.007). rehabilitation at a later stage during the course of cancer treatment is also provided. Rehabilitation costs are covered Study variables and measures mainly by the pension and health insurances. A cancer rehabilitation program lasts three weeks following a mul- The main outcomes for this study were employment at tidimensional therapeutic approach that includes patient 12 months after rehabilitation (M=23.1 months after education, exercises, and physical therapy to regain physi- cancer diagnosis) and time until RTW. Employment was cal fitness and vitality, relaxation training, psychosocial defined according to a patient’s positive response to the counseling, and psychosocial support groups to enhance question “Are you currently working?” Time until RTW coping skills as well as individual psychotherapy. was measured in weeks after rehabilitation. This prospective longitudinal study aimed to (i) Demographic information was obtained at baseline identify the employment rate 12-months after cancer consisting of standardized questions concerning (age, rehabilitation, (ii) explore the work situation and expe- gender, marital status, and employment history). Educa- rienced work changes, and (iii) identify demographic, tion, monthly household net income, and occupational medical, functional, psychosocial, and work-related position were used to calculate a 3-factor social status predictors of the likelihood and time period of RTW. index (33). Medical information was collected at base - line [cancer entity, months since diagnosis, cancer stage as defined by the International Union against Cancer (UICC), clinical characteristics and disease phase]. In Methods addition to the baseline Karnofsky performance status (34), physicians estimated the degree of functional The study received research ethics committee approval. impairment using cancer-entity-specific physical func - All patients provided written informed consent prior to tioning scales. The Karnofsky status is a performance Scand J Work Environ Health 2013, vol 39, no 1 77 Employment predictors among cancer survivors Assessed for eligibility N=1653 Excluded: Non-participants t : Participation t : 0 0 Unclear employment status: N=23 N=372 N=1148 Housewife/houseman: N=110 Non-participants t : Participation t : 1 1 N=88 N=1060 Non-participants t : Participation t : 2 2 N=310 N=750 Figure 1. Enrollment of cancer survivors (t =beginning of cancer rehabilitation, t =end of a 3–4 week cancer rehabilita- Deceased: Moved to unknown Refused participation: tion, t =12 months after cancer N=30 N=244 2 address: N=36 rehabilitation) measure for rating the ability of a somatically ill person absence, (iii) 10 items about job requirements (eg, many to perform usual activities. A person is evaluated on work responsibilities, high pressure of competition, a score of 0–100, where 0=dead and 100=normal, no tight schedules) answered on a 4-point Likert scale complaints, no signs of disease. Using cancer-entity- ranging from 1=almost never to 4=quite often (Cron- specific physical functioning scales, the physician mea - bach’s α=0.87), and (iv) 12 items about work satisfac- sures the limitations specific for each tumor entity such tion answered on a 7-point Likert scale ranging from as shoulder mobility or lymphedema in breast cancer, 1=not satisfied at all to 7=totally satisfied (Cronbach’s incontinence in genital or colon cancers, swallowing α=0.92). Self-perceived work ability was evaluated on problems in head and neck cancers or dyspnea in lung a 5-point Likert item ranging from 1=totally limited to cancer. Pain intensity during the last week was evaluated 5=not limited at all; self-perceived employer accommo- using the Brief Pain Inventory (BPI) (35). dation was evaluated on a 5-point Likert item ranging The psychosocial and work-related variables were from 1=not at all to 5=extremely. Perceived threat of job further assessed at baseline. Anxiety and depression loss was measured using single-item questions (No/Yes). were assessed using the Hospital Anxiety and Depres - sion Scale (HADS) (36). Fear of cancer recurrence was Statistical analysis measured using the 12-item short version of the Fear of Recurrence Questionnaire (FoP-Q-SF) scored on a In order to identify significant predictors of RTW, demo - 5-point Likert scale ranging from 1=never to 5=very graphic, medical, functional, psychosocial, and work- often (37). The Short-Form Health Survey assesses related factors were entered separately (block-wise) into dimensions of quality of life (QoL): Here, the two a multivariate hierarchical logistic regression analysis summary scores for physical (PCS) and mental health against the outcome variable “RTW”. Step-wise back- (MCS) were calculated. Higher scores indicate better wards elimination was used, testing each candidate vari- QoL (38). The Illness-Specific Social Support Scale able for removal using Wald statistic. Before testing the (ISSS) measures the degree to which partners/friends regression model, we performed correlations (Pearson provide positive support (eg, “listened to you”) or act and Spearmans correlation coefficients) for all predictor in a non-supportive way (detrimental interactions) (eg, candidates between variables in one block and between “tried to change the way you’re coping with your illness candidate variables and the outcome criteria. In order in a way you didn’t like”). Items are scored on a 5-point to identify significant predictors of the time period of Likert scale ranging from 0=ever to 4=always (39). RTW, candidate predictor variables were entered into Occupational and work-related characteristics were a Cox’s proportional hazards model to calculate hazard assessed using questions and brief questionnaires devel- ratios. To provide an estimate of the magnitude of the oped and psychometrically evaluated by Bürger et al group differences, Cohen’s standardized effect size ( ϕ, (40). Occupational information included (i) profes- d, η²) was calculated. Two-tailed significance tests were sional status, (ii) work ability and periods of sick leave conducted using a significance level of P<0.05. 78 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch Table 1. Baseline demographic and medical sample characteristics. [SD=standard deviation; UICC=International Union against Cancer] Study sample P- d / ϕ value Total sample (N=750) Employed (N=702) Unemployed (N=48) Mean SD N % Mean SD N % Mean SD N % Demographic characteristics Age 48.7 6.8 48.7 6.7 48.8 8.0 0.93 Female gender 643 85.7 597 85.0 46 95.8 0.04 0.08 Married 498 66.4 473 67.4 25 52.0 0.03 0.11 Partnership 566 75.5 537 76.5 29 60.4 0.01 0.09 High school/university degree 218 29.1 207 29.5 11 22.9 0.55 Monthly household net income (€) <0.001 0.44 <1000 39 9.3 18 4.9 21 43.8 1000–<2000 136 32.5 122 33.0 14 29.2 2000–<3000 124 29.7 115 31.1 9 18.8 ≥3000 119 28.5 115 31.1 4 8.3 Social class Lower 203 27.1 176 25.1 27 56.3 <0.001 0.18 Middle 441 58.8 421 60.0 20 41.7 Upper 106 14.1 105 15.0 1 2.1 Medical characteristics Cancer entity Breast cancer 446 59.5 419 59.7 27 56.3 0.53 Gynecological cancers 109 14.5 100 14.2 9 18.8 Head and neck cancers 67 8.9 62 8.8 5 10.4 Skin cancer 46 6.1 45 6.4 1 2.1 Colon/rectum cancer 42 5.6 39 5.6 3 6.3 Lung cancer 23 3.1 20 2.8 3 6.3 Hematological neoplasias 17 2.3 17 2.4 0 Months since diagnosis 11.1 8.5 11.1 8.5 10.2 7.3 0.47 Cancer stage (UICC) 0.20 In situ 8 1.1 6 0.9 2 4.2 I 305 41.6 288 42.0 17 35.4 II 268 36.6 248 36.2 20 41.7 III 118 16.1 112 16.4 6 12.5 IV 34 4.6 31 4.5 3 6.3 Clinical characteristics First primary tumor 719 95.9 673 95.9 46 95.8 0.99 Second primary tumor 31 4.1 29 4.1 2 4.2 Disease phase 0.61 Remission 653 87.1 612 87.2 41 85.4 Recurrence/progress/metastasis 72 9.6 68 9.7 1 2.1 Unclear 25 3.3 22 3.1 3 6.3 Treatment phase 0.57 Curative treatment intention 693 92.4 648 92.3 45 93.8 Palliative treatment intention 36 4.8 35 5.0 1 2.1 Unclear 21 2.8 19 2.7 2 4.2 Months since active cancer 7.7 12.5 7.7 12.6 6.5 10.4 0.52 treatment Cancer treatments Surgery 716 95.5 669 95.3 47 97.9 0.40 Radiation therapy 494 65.9 464 66.1 30 62.5 0.61 Chemotherapy 454 60.5 427 60.8 27 56.3 0.53 Hormonal treatment 328 43.7 308 43.9 20 41.7 0.77 Total number of therapies 2.7 1.1 2.7 1.1 2.6 1.1 0.58 Karnofsky performance status 91.7 7.5 91.7 7.5 91.5 6.2 0.81 Without hematological neoplasias. Scand J Work Environ Health 2013, vol 39, no 1 79 Employment predictors among cancer survivors Results ees (97.2%), and the lowest percentage among work- ers (66.4%) (P=0.02) (ϕ=0.12). No group differences were found in clinical characteristics. However, among Baseline occupational characteristics patients with a higher UICC cancer stage (P<0.001) At the beginning of the rehabilitation program, 702 (ϕ=0.18) and palliative treatment, a lower percent- patients (93.6%) were employed and 48 (6.4%) were age (38.9%) returned to work or were re-employed unemployed. Among the employed, 54.9% were on compared to patients with curative treatment (78.1%) physician-classified sick leave (usually the primary care (P<0.001) (ϕ=0.20). Also, the highest percentage of physician). The majority of the working participants patients who did not return to work was observed among worked as employees (75.1%), 18.4% were workers; patients with lung cancer (43%) and head and neck 5.6% were self-employed, and 1.0% worked as civil cancers (58%). Among patients with cancer progress servants. The mean duration of sick leave within the or metastatic cancer, a significantly lower percentage 12-months period prior to the rehabilitation program (38.9%) returned to work or got re-employed compared was 150.6 (SD 107.4, range 1–365) days. The majority to patients in remission (78.6%) (P<0.001) (ϕ=0.17). of patients (84.5%) were motivated either to return to work or be re-employed after rehabilitation. RTW time period and re-employment after rehabilitation Forty-nine percent of patients returned to work imme - RTW and re-employment 12 months after rehabilitation diately after rehabilitation. The mean time until RTW At t , 568 patients (75.7%) had returned to work or were or re-employment was 5.7 (SD 8.6, range 1–45) weeks. re-employed. The percentage of patients who returned to The highest percentage of patients who returned to work work was highest among participants not on sick leave immediately were among those who were not on sick at baseline (92.4%). Twenty-one percent of the patients leave at baseline, belonged to the upper social class, and who were unemployed at baseline managed to get had skin cancer (table 2). employed at a follow-up time point (table 4). Patients who returned to work or were re-employed were slightly Work situation at follow-up younger [mean 48.2, (SD 7.0) years versus mean 50.5 (SD 6.1) years] (P<0.001) (η²=0.021). No gender differ- Among the 568 patients who returned to work, the ences were observed. The highest percentage of patients majority (81.2%) returned to their former position and who returned to work was observed among employ- workplace; 46% worked full-time. More women (77.8%) Table 2. Time period of return to work (RTW) and reemployment after rehabilitation (N=568). Employed patients N (t ) Time period to RTW / re-employment P-value 0 ϕ / η² Patients who returned to work Weeks until RTW / immediately after rehabilitation re-employment N % Mean SD a b a b Baseline employment status <0.001 / <0.001 0.68 / 0.24 Patients working 268 228 85.1 1.64 2.46 Patients on sick leave 290 48 16.6 8.90 9.83 Unemployed patients 10 2 20.0 21.60 16.67 Total 568 278 48.9 5.68 8.60 a b a Social class 0.008 / 0.63 0.13 Lower social class 133 53 39.8 6.23 8.23 Middle social class 343 169 49.3 5.41 8.44 Upper social class 92 56 60.9 5.89 9.64 a b a b Cancer entity 0.001 / <0.001 0.20 / 0.05 Breast cancer 348 171 49.1 5.19 7.98 Gynecological cancers 81 38 46.9 5.23 6.77 Head and neck cancers 39 14 35.9 8.03 10.31 Skin cancer 38 30 78.9 2.76 6.00 Colon/rectum cancer 36 16 44.4 10.21 13.41 Lung cancer 10 1 10.0 13.70 14.21 Hematological neoplasias 16 8 50.0 5.13 7.53 Significance and effect size refer to group differences in patients who returned to work immediately after rehabilitation. Significance and effect size refer to group differences in weeks until RTW/reemployment. 80 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch worked part-time compared to 22.2% of men (P<0.001) sus palliative treatment intention), months since active (ϕ=0.25). Fifty-two percent of the patients were on sick cancer treatment, total numbers of therapy, anxiety, dis- leave absence at least once after rehabilitation for an tress, and the expected period of RTW. Table 3 presents average duration of 62 days. the correlations between the final candidate predictor variables and the outcome variables. The first block entered into the multivariate hierarchi - Predictors of RTW – logistic regression model cal logistic regression model consisted of demographic Before testing the logistic regression model, we per- factors (table 4). The variables education and social sta- formed correlations between variables in one block for tus were excluded from the regression model, whereas all predictor candidates. In candidate predictor vari- age and monthly household net income remained in the ables with an intercorrelation of r≥0.60, one variable model (Nagelkerke’s R²=0.07) (P<0.001). was removed from the regression analysis in order to The second block entered consisted of medical and avoid multicollinearity. The following variables were functional factors. The variables cancer entity, months removed: marital status, treatment phase (curative ver- since current diagnosis, UICC cancer stage, and number Table 3. Correlations between candidate predictor variables (t ) and outcome variables. [UICC=International Union against Cancer.] Candidate predictor variables (t ) 12-months return to work Time period of return to work (weeks) r P-value r P-value Demographic factors Age -0.14 >0.001 -0.13 <0.001 Education 0.12 0.001 -0.12 0.001 Monthly household net income 0.18 >0.001 -0.20 <0.001 Social status 0.16 >0.001 -0.16 <0.001 Medical and functional factors Cancer entity Breast cancer 0.07 0.08 -0.08 0.06 Gynecological cancers -0.01 0.71 0.03 0.55 Head and neck cancers -0.13 >0.001 0.05 0.26 Skin cancer 0.04 0.26 -0.09 0.05 Colon/rectum cancer 0.06 0.12 0.11 0.01 Lung cancer -0.13 >0.001 0.13 0.003 Hematological neoplasias 0.07 0.07 -0.03 0.52 Months since current diagnosis 0.09 0.03 -0.30 >0.001 Cancer stage (UICC) -0.15 >0.001 -0.15 >0.001 Disease phase Remission 0.17 >0.001 0.007 0.87 Recurrence/progress/metastasis -0.12 0.004 0.07 0.12 Number of functional impairments -0.19 >0.001 0.26 >0.001 Number of physical problems -0.19 >0.001 0.26 >0.001 Karnofsky status 0.30 >0.001 -0.34 >0.001 Pain -0.25 >0.001 0.32 >0.001 Psychosocial factors Fear of cancer recurrence -0.21 >0.001 0.08 0.05 Depression -0.31 >0.001 0.09 0.03 Mental quality of life 0.23 >0.001 -0.06 0.17 Physical quality of life 0.32 >0.001 -0.24 >0.001 Social support 0.10 0.007 0.02 0.71 Detrimental interactions -0.21 >0.001 0.04 0.32 Work-related factors Employment and work ability Working 0.31 >0.001 -0.63 >0.001 Sick leave -0.14 >0.001 0.46 >0.001 Unemployed -0.34 >0.001 0.30 >0.001 Duration of sick leave (days) prior to rehabilitation -0.30 >0.001 0.47 >0.001 Self-perceived work ability -0.48 >0.001 0.37 >0.001 Perceived threat of job loss -0.17 >0.001 0.02 0.64 Intention to return to work 0.50 >0.001 -0.05 0.28 Expected period of return to work after rehabilitation -0.33 >0.001 0.53 >0.001 Perceived employer accommodation 0.34 >0.001 -0.07 0.13 Job requirements -0.11 0.004 0.08 0.06 Overall job satisfaction 0.13 >0.001 -0.02 0.65 Scand J Work Environ Health 2013, vol 39, no 1 81 Employment predictors among cancer survivors Table 4. Multivariate logistic regression model for the identification of significant predictors of 12-months (re-) employment after rehabilitation. [OR=odds ratio; SD=standard deviation; SE=standard error; 95% CI=95% confidence interval] (Re-)employment ß SE OR 95% CI P-value Yes (N=568) No (N=182) Mean SD N % Mean SD N % Block 1 Demographic <0.001 factors (Nagelkerke’s R²=0.07) Age 48.2 7.0 50.5 6.1 -0.05 0.03 0.94 0.90–1.00 0.054 Monthly household net income (€) <1000 29 46.8 33 53.2 0.21 1000–<2000 183 73.8 65 26.2 1.20 0.72 3.33 0.87–13.73 0.10 2000–<3000 190 78.5 52 21.5 0.74 0.70 2.09 0.53–8.31 0.29 ≥3000 166 83.8 32 16.2 0.41 0.72 1.50 0.37–6.18 0.57 Block 2 Medical and <0.001 functional factors (Nagelkerke’s R²=0.32) Disease phase Remissiona 513 78.6 140 21.4 0.002 Recurrence/progress/ 55 56.7 42 43.3 -1.31 0.43 0.27 0.12–0.63 metastasis Number of functional 2.3 3.0 4.2 4.6 -0.01 0.04 1.00 0.91–1.08 0.78 impairments Karnofsky status 93.0 7.5 87.8 7.1 0.05 0.02 1.05 1.01–1.09 0.01 Pain 2.5 2.1 4.1 2.8 -0.09 0.09 0.91 0.77–1.08 0.28 Physical quality of life 50.6 9.0 43.7 7.8 0.02 0.03 1.02 0.99–1.08 0.51 Block 3 Psychosocial <0.001 factors (Nagelkerke’s R²=0.39) Mental quality of life 53.2 10.1 47.6 11.1 0.03 0.02 1.03 1.00–1.07 0.08 Detrimental interactions 1.0 0.6 1.3 0.6 -0.55 0.28 0.58 0.33–0.99 0.05 Block 4 Work-related <0.001 factors (Nagelkerke’s R²=0.56) Employment Working 268 92.4 22 7.6 0.05 Sick leave 290 70.4 122 29.6 -1.36 0.56 0.26 0.09–0.77 0.02 Unemployed 10 20.8 38 79.2 -1.66 1.12 0.19 0.02–1.68 0.14 Self-perceived work 3.4 1.0 2.2 1.0 0.41 0.22 1.50 0.99–2.30 0.06 ability Intention to return to 538 84.9 96 15.1 1.83 0.58 6.22 1.98–19.51 0.002 work Perceived employer 3.8 0.9 2.8 1.2 0.66 0.16 1.93 1.41–2.65 <0.001 accommodation Job requirements 3.8 0.9 2.8 1.2 0.61 0.30 1.84 1.02–3.30 0.04 Reference category. of physical problems were excluded from the regression to rehabilitation, perceived threat of job loss, expected model (Nagelkerke’s R²=0.32) (P<0.001). Patients with period of RTW after rehabilitation, and overall job cancer recurrence or cancer progress had a significant satisfaction. Patients who intended to return to work lower chance to return to work compared to patients in at baseline were more than six times as likely to do so cancer remission. (P<0.001). Also, patients who perceived their employer The third block entered consisted of psychosocial as being cooperative and accommodating of their can- factors. The variables fear of recurrence, depression, cer were more likely to return to work. Although the and social support were excluded from the regression variable job requirements showed a negative correla- model (Nagelkerke’s R²=0.39) (P<0.001). Individuals tion with RTW, it also emerged as a positive predictor with a higher amount of problematic social interactions for RTW. This is likely an effect of suppression since were less likely to return to work. high job requirements are positively associated with The forth block entered consisted of work-related RTW among individuals in the higher social class but factors. The following variables were excluded from negatively associated with RTW among patients in the regression model: duration of sick leave (days) prior lower social class. When the variable is entered as a 82 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch single factor into a regression analysis, it remains as symptom burden at the beginning of the rehabilitation a negative predictor of RTW [ß=-0.41, OR=0.67, 95% program. In our study, the employment rate was 76% confidence interval (95% CI) 0.48–0.92, P=0.01]. The on average 23 months after diagnosis and consider - explained variance of the total model (Nagelkerke’s R²) ably higher compared to findings showing that overall, was 0.59 (P<0.001). about 63.5% of cancer patients (range 24–94%) manage to return to work (17). However, comparable results (73%, range 64–82%) have been found at 18 months Predictors of time period to RTW after the cancer diagnosis among breast cancer patients The following variables remained in the model as signif- (14) and among samples with mixed tumor entities (1, icant predictors: UICC cancer stage, Karnofsky index, 8, 25). Concordant with previous studies showing work physical and mental quality of life, employment charac- changes in 8–17% of cancer survivors, we found similar teristics, intention to return to work, perceived employer results (25, 30). accommodation, and job requirements (overall model Both the absence of sick leave and belonging to fit - 2 log likelihood = 3858.2; χ²[total model] = 175.6, a higher social class at baseline were associated with df = 13, P<0.001) (table 5). a significant higher rate of employment at follow-up. Belonging to a higher social class is related to work environments that may provide more favorable work- ing conditions in terms of flexibility, a lower degree of Discussion manual and physically exhausting work, better earnings, and better living conditions. Accordingly, it has been The majority of patients (85%) were motivated either stated that employees with cancer or other persistent to return to work or be re-employed after rehabilitation. health problems generally need some flexibility at vari - These findings are consistent with results indicating ous aspects and times at work (42). Atkinson et al (43) the significance of work and the strong motivation to and Gudbergson et al (44) pointed out that living condi - continue work during treatment or return to work after tions include the social indicators that stimulate social treatment completion among survivors (41). The explo- inclusion and reduce social exclusion. These indicators ration of work characteristics revealed that slightly more are economy, education, employment, health, housing, than half of participants (55%) were on sick leave at and social participation. baseline indicating a high physical and/or psychological The highest percentage of patients who did not return Table 5. Cox’s proportional hazards regression model for the identification of significant predictors of the time to return-to-wor k. [HR=hazard ratio; SD=standard deviation; SE=standard error; UICC=International Union against Cancer; 95% CI=95% confidence interval. ] Time period (weeks) to SE HR 95% CI P-value return to work Mean SD Months since current diagnosis 0.01 0.01 1.01 0.99–1.03 0.35 Cancer stage (UICC) I 13.90 19.65 0.02 II 18.07 21.44 -0.27 0.12 0.76 0.60–0.97 0.03 III 20.05 22.66 -0.37 0.16 0.69 0.51–0.94 0.02 IV 29.18 24.74 -0.59 0.26 0.56 0.34–0.92 0.02 Karnofsky status 0.02 0.01 1.02 1.00–1.04 0.04 Number of functional impairments -0.01 0.02 0.99 0.96–1.02 0.49 Physical quality of life 0.02 0.01 1.02 1.01–1.04 0.001 Mental quality of life 0.01 0.01 1.01 1.00–1.03 0.009 Employment Working 5.48 13.56 <0.001 Sick leave 21.61 21.37 -0.99 0.16 0.37 0.27–0.51 <0.001 Unemployed 45.67 14.45 -1.41 0.41 0.25 0.11–0.55 0.001 Intention to return to work No 40.53 20.18 -0.49 0.20 0.61 0.41–0.91 0.02 Yes 12.59 18.38 Perceived employer accommodation 0.17 0.06 1.18 1.06–1.32 0.002 Job requirements 0.17 0.09 1.19 1.01–1.41 0.04 Without hematological neoplasias Reference category Scand J Work Environ Health 2013, vol 39, no 1 83 Employment predictors among cancer survivors to work was observed among patients with lung cancer a positive predictor for RTW likely due to suppression (43%) and head and neck cancers (58%), among patients effects. Our findings emphasize that more research is with advanced cancer stage, progress or metastatic can- needed particularly to investigate the kind of job require- cer, and palliative treatment intention. The adverse effect ments (such as having plenty of work or high time pres- of cancer progress and poor physical functioning on sure) that might lead to early retirement, unemployment, RTW has been shown in several previous studies (7, 23, or a higher probability to RTW. Our results point toward 27). Corresponding to previous research, we found re- the fact that high job requirements are positively associ- employment significantly associated with younger age ated with RTW among individuals in the higher social and a professional status as an employee (24, 27, 45). class but are negatively associated with RTW among Our regression analyses findings emphasize the patients in the lower social class. Individuals belonging importance of volitional factors for RTW. Patients who to a higher social class are more likely to have better expressed their intention to return to work at the begin- working conditions such as flexible work, responsibility, ning of the cancer rehabilitation program were six times and a considerable amount of decision-making freedom more likely to do so compared to patients who did not that might compensate the high workload. intend to return to work. Our findings show that the Although this prospective study includes a large patient can best predict RTW at an early stage of the sample size compared to previous research, this study rehabilitation process. However, so far only limited has several methodological limitations. Both the initial knowledge exists about motivational and volitional fac- and follow-up non-response lead to a bias in several tors and its association with demographic, family and sociodemographic and psychosocial outcome variables work-related aspects (46, 47). Occupational motivation of interest in this research. With regard to the generality or skepticism towards RTW should be carefully assessed and interpretation of the findings, a sample bias must be at the beginning of rehabilitation programs and during considered toward: (i) female gender, (ii) younger age, the joint establishment of rehabilitation aims between (iii) being in a cancer rehabilitation program, (iv) cancer the patients and the professional team (48). Furthermore, entities associated with a better physical health status and factors influencing occupational motivation among prognosis, and (v) better psychological well-being. Given cancer survivors need to be understood in more detail. this bias, our findings with regard to employment might The strong influence of work-related aspects next overestimate the degree to which cancer patients return to medical factors such as disease phase confirm the to work or stay employed. Nevertheless, systematic dif- current state of the literature with regard to factors ferences between participants and non-participants were positively and adversely influencing RTW among can - small in view of effect sizes and at least partially a conse- cer survivors (7, 14, 23, 25, 26, 45). Consistent with quence of the relatively large sample size. Although 86% the literature, the perceived employer accommodation of the patients were women, the overall large sample size indicated a fairly positive attitude towards employment, including 107 men justifies gender analyses; however, work-related support, and necessary job changes (14, gender did not correlate with employment status and 27, 49) and emerged as a significant predictor for RTW. therefore was excluded from further regression analyses. In our study, analyses also demonstrated that the Another limitation is that due to the allocation pro- patients who were working at baseline were more likely cess mainly regulated by the German pension insur - to return to work after cancer rehabilitation and were ance, it was not feasible to randomize the study sample. more likely to do so at an earlier stage than patients on Despite the fact that cancer rehabilitation programs are sick leave or – unsurprisingly – unemployed patients. provided to every cancer patient in Germany, our sample Self-perceived work ability, in contrast, did not emerge consists only of patients who use the services provided. as a significant predictor for RTW. The relevance of This might lead to a bias towards a sample with high perceived work ability has been emphasized in a num - physical and psychosocial impairments. Furthermore, ber of studies (3, 15, 44, 50–53). However, only limited the inclusion of a matched control group could not be knowledge exists about the association between per- realized since most patients with rehabilitation needs ceived work ability and sick leave absence. are referred to a rehabilitation program. Thus, this study Furthermore, we found that detrimental social inter- could not determine possible effects of the rehabilitation actions were inversely associated with RTW. Patients program on RTW. with problematic social interactions in their personal The (re-)integration of cancer survivors into working environment might also lack social skills at the work- life is one important aspect of participation according place or might have fewer personal resources to effec- to the ICF (31). Rehabilitation programs are important tively adapt to the challenges of being a cancer survivor not only for the physical and psychosocial recovery, but in the work environment. for the labor market reintegration of patients. Profound The variable “high job requirements” showed a neg- understanding of cancer and treatment-induced impair- ative correlation with RTW. However, it also emerged as ments and their impact on daily activities and work is 84 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch an essential basis for the development of better educa- 9. Tiedtke C, Rijk A de, Dierckx Casterlé B de, Christiaens M, Donceel P. Experiences and concerns about ‘returning to work’ tional, rehabilitative, and occupational interventions in for women breast cancer survivors: a literature review. Psycho- cancer care (54). A better understanding of cancer and Oncology. 2010;19:677–83. http://dx.doi.org/10.1002/ treatment-induced physical, cognitive and psychosocial pon.1633. treatment consequences related to work-related prob- lems will help to develop interventions and educational 10. Feuerstein M, Todd BL, Moskowitz MC, Bruns GL, Stoler MR, Nassif T et al. Work in cancer survivors: a model for programs for patients, healthcare professionals, and practice and research. J Cancer Surviv. 2010;4:415–37. http:// employers to better address the professional needs of dx.doi.org/10.1007/s11764-010-0154-6. individuals with cancer. 11. Thijs KM, Boer AGEM, Vreugdenhil G, Wouw AJ, Houterman S, Schep G. Rehabilitation Using High-Intensity Physical Training and Long-Term Return-to-Work in Cancer Survivors. J Occup Rehabil. 2011 Nov 12. [Epub ahead of print]. http:// Acknowledgements dx.doi.org/10.1007/s10926-011-9341-1. We thank Birgit Leibbrand, MD, Jürgen Barth, MD, 12. McGrath PD, Hartigan B, Holewa H, Skarparis M. Returning to work after treatment for haematological Manfred Gaspar, MA, Gerhard Friedrich, MD, Wilhelm cancer: findings from Australia. Support Care Can cer. 2011 Bootsveld, MD, Ulrich Gärtner, MD, and Christine- Oct 28. [Epub ahead of print] . http://dx.doi.org/10.1007/ Daniela Kegel, MD, for their data collection and man - s00520-011-1298-2. agement support. This research was funded by the Nordrhein-Westfalen Association for the Fight Against 13. Roelen CAM, Koopmans PC, Groothoff JW, Klink JJL, Bültmann U. Return to Work After Cancer Diagnosed in 2002, Cancer, Germany. 2005 and 2008. J Occup Rehabil. 2011;21:335–41. http:// dx.doi.org/10.1007/s10926-011-9319-z. 14. Bouknight RR, Bradley CJ, Zhehui L. Correlates of Return to References Work for Breast Cancer Survivors. J Clin Oncol. 2006;24:345– 53. http://dx.doi.org/10.1200/JCO.2004.00.4929. 1. Spelten E. Cancer, fatigue and the return of patients to work—a 15. Taskila T, Lindbohm ML. Factors affecting cancer survivors‘ prospective cohort study. Eur J Cancer. 2003;39:1562–7. employment and work ability. Acta Oncol. 2007;46:446–51. http://dx.doi.org/10.1016/S0959-8049(03)00364-2. http://dx.doi.org/10.1080/02841860701355048. 2. Verbeek J. Return to work of cancer survivors: a prospective 16. Park J, Park E, Park J, Kim S, Lee S. Job loss and re-employment cohort study into the quality of rehabilitation by occupational of cancer patients in Korean employees: a nationwide physicians. Occup Environ Med. 2003;60:352–7. http:// retrospective cohort study. J Clin Oncol. 2008;26:1302–9. dx.doi.org/10.1136/oem.60.5.352. http://dx.doi.org/10.1200/JCO.2007.14.2984. 3. Bradley CJ, Neumark D, Luo Z, Bednarek H, Schenk M. 17. Mehnert A. Employment and work-related issues in cancer Employment Outcomes of Men Treated for Prostate Cancer. survivors. Crit Rev Oncol Hemat. 2011;77:109–30. http:// J Natl Cancer I. 2005;97:958–65. http://dx.doi.org/10.1093/ dx.doi.org/10.1016/j.critrevonc.2010.01.004. jnci/dji171. 18. Peteet JR. Cancer and the meaning of work (eng). Gen Hosp Psychiatry. 2000;22:200–5. http://dx.doi.org/10.1016/S0163- 4. Hoffman B. Cancer Survivors at Work: A Generation of 8343(00)00076-1. Progress. CA-Cancer J Clin. 2005;55:271–80. http://dx.doi. org/10.3322/canjclin.55.5.271. 19. Rasmussen DM, Elverdam B. The meaning of work and working life after cancer: an interview study. Psycho- 5. Feuerstein M. Cancer Survivorship and Work. J Occup Oncology. 2008;17:1232–8. http://dx.doi.org/10.1002/ Rehabil. 2005;15:1–2. http://dx.doi.org/10.1007/s10926- pon.1354. 005-0868-x. 6. Feuerstein M, Harrington CB. Recommendations for the 20. Choi KS, Kim E, Lim J, Kim S, Lim MK, Park J et al. Job U.S. National Occupational Research Agenda: Research on loss and reemployment after a cancer diagnosis in Koreans—a Cancer Survivorship and Musculoskeletal Disorders and prospective cohort study. Psycho-Oncology. 2007;16:205–13. Work Disability. J Occup Rehabil. 2006;16:1–5. http://dx.doi. http://dx.doi.org/10.1002/pon.1054. org/10.1007/s10926-005-9004-1. 21. Carlsen K, Dalton SO, Diderichsen F, Johansen C. Risk for unemployment of cancer survivors: A Danish cohort study. 7. Pryce J, Munir F, Haslam C. Cancer survivorship and work: Eur J Cancer. 2008;44:1866–74. http://dx.doi.org/10.1016/j. Symptoms, supervisor response, co-worker disclosure and ejca.2008.05.020. work adjustment. J Occup Rehabil. 2007;17:83–92. http:// dx.doi.org/10.1007/s10926-006-9040-5. 22. Boer AGEM de, Taskila T, Ojajärvi A, van Dijk FJH, Verbeek JHAM. Cancer survivors and unemployment: a meta-analysis 8. Boer AGEM de, Verbeek JHAM, Spelten ER, Uitterhoeve and meta-regression. J Am Med Assoc. 2009;301:753–62. ALJ, Ansink AC, Reijke TM de et al. Work ability and return- http://dx.doi.org/10.1001/jama.2009.187. to-work in cancer patients. Br J Cancer. 2008;98:1342–7. http://dx.doi.org/10.1038/sj.bjc.6604302. Scand J Work Environ Health 2013, vol 39, no 1 85 Employment predictors among cancer survivors 23. Spelten ER, Sprangers MA, Verbeek JH. Factors reported to 37. Mehnert A, Herschbach P, Berg P, Henrich G, Koch U. Fear of influence the return to work of cancer survivors: a literature progression in breast cancer patients--validation of the short review. Psycho-Oncology. 2002;11:124–31. http://dx.doi. form of the Fear of Progression Questionnaire (FoP-Q-SF). Z org/10.1002/pon.585. Psychosom Med Psychother. 2006;52:274–88. 24. Taskila-Abrandt T, Martikainen R, Virtanen S, Pukkala E, 38. Ware JE, Kosinski M, Dewey JE, Gandek B. How to Score Hietanen P, Lindbohm M. The impact of education and and Interpret Single-Item Health Status Measures: A Manual occupation on the employment status of cancer survivors. for Users of the SF-8TM Health Survey. Lincoln, RI: Eur J Cancer. 2004;40:2488–93. http://dx.doi.org/10.1016/j. QualityMetric Incorporated; 1999. ejca.2004.06.031. 39. Ramm GC, Hasenbring M. Die deutsche Adaptation des Illness-specific Social Support Scale und ihre teststatistische 25. Amir Z, Moran T, Walsh L, Iddenden R, Luker K. Return to Überprüfung beim Einsatz an Patienten vor und nach paid work after cancer: a British experience. J Cancer Surviv. Knochenmarktransplantation [The German adaptation of the 2007;1:129–36. http://dx.doi.org/10.1007/s11764-007- Illness-specific Social Support Scale and the test statistical 0021-2. evaluation on the basis of patients undergoing bone marrow 26. Chan F, Strauser D, da Silva Cardoso E, Xi Zheng L, Chan transplantation]. Z Med Psychol 2003;12, 29-38. JYC, Feuerstein M. State vocational services and employment in cancer survivors. J Cancer Surviv. 2008;2:169–78. http:// 40. Bürger W, Dietsche S, Morfeld M, Koch U. Multiperspective dx.doi.org/10.1007/s11764-008-0057-y. estimates on the probability of patient return to work following orthopaedic rehabilitation: findings and predictive relevance. 27. Molina Villaverde R, Feliu Batlle J, Villalba Yllan A, Jimenez Rehabilitation. 2001;40:217–25. Gordo AM, Redondo Sanchez A, San Jose Valiente B et al. Employment in a cohort of breast cancer patients. Occup Med. 41. Mahar KK, BrintzenhofeSzoc K, Shields JJ. The impact of 2008;58:509–11. http://dx.doi.org/10.1093/occmed/kqn092. changes in employment status on psychosocial well-being: a study of breast cancer survivors. J Psychosoc Oncol. 2008;26:1– 28. Ross L, Petersen MA, Johnsen AT, Lundstroem LH, Carlsen K, 17. http://dx.doi.org/10.1080/07347330802115400. Groenvold M. Factors associated with Danish cancer patients’ return to work. A report from the population-based study ‘The 42. Feuerstein M, Luff GM, Harrington CB, Olsen CH. Pattern of Cancer Patient‘s World’. Cancer Epidemiol. 2011 Aug 12. workplace disputes in cancer survivors: a population study of [Epub ahead of print]. ADA claims. J Cancer Surviv. 2007;1:185–92. http://dx.doi. org/10.1007/s11764-007-0027-9. 29. Short PF, Vasey JJ, Tunceli K. Employment pathways in a large cohort of adult cancer survivors. Cancer. 2005;103:1292–301. 43. Atkinson T, Cantillon B, Marlier E, Nolan B. Social indicators. http://dx.doi.org/10.1002/cncr.20912. The EU and social inclusion. Oxford: Oxford University Press; 30. Gudbergsson SB, Fosså SD, Dahl AA. A study of work changes due to cancer in tumor-free primary-treated cancer patients. A 44. Gudbergsson SB, Fosså SD, Borgeraas E, Dahl AA. A NOCWO study. Support Care Cancer. 2008;16:1163–71. comparative study of living conditions in cancer patients who http://dx.doi.org/10.1007/s00520-008-0407-3. have returned to work after curative treatment. Support Care Cancer. 2006;14:1020–9. http://dx.doi.org/10.1007/s00520- 31. World Health Organization (WHO). ICF—International 006-0042-9. Classification of Functioning, Disability and Health. Genf; 45. Hensel M. Quality of life and rehabilitation in social and professional life after autologous stem cell transplantation. 32. Hellbom M, Bergelt C, Bergenmar M, Gijsen B, Loge JH, Ann Oncol. 2002;13:209–17. http://dx.doi.org/10.1093/ Rautalathi M et al. Cancer rehabilitation: A Nordic and annonc/mdf031. European perspective. Acta Oncol. 2011;50:179–86. http:// dx.doi.org/10.3109/0284186X.2010.533194. 46. Nachreiner N, Dagher R, McGovern P, Baker B, Alexander B, Gerberich S. Successful return to work for cancer survivors. 33. Winkler J, Stolzenberg H. Der Sozialschichtindex AAOHN J. 2007;55:290–5. im Bundesgesundheitssurvey [The social class index according to the national health survey]. Gesundheitswesen. 47. Johnsson A, Fornander T, Olsson M, Nystedt M, Johansson H, 1999;61:S178-S183. Rutqvist L. Factors associated with return to work after breast cancer treatment. Acta Oncol. 2007;46:90–6. http://dx.doi. 34. Karnofsky DA, Burchenal JH. The Clinical Evaluation of org/10.1080/02841860600857318. Chemotherapeutic Agents in Cancer. In: MacLeod CM, editor. Evaluation of Chemotherapeutic Agents. New York: Columbia 48. Thies S, Lehmann C, Kriz D, Nubling R, Mehnert A. University Press; 1949, p. 191–205. Patientenfragebogen zur Erfassung der Reha-Motivation (PAREMO-20) - Testtheoretische Überprüfung und Validierung 35. Cleeland CS, Ryan KM. Pain assessment: global use of an einer Stichprobe von Krebspatienten unterschiedlicher the Brief Pain Inventory (eng). Ann Acad Med Singap. Diagnosegruppen . Rehabilitation. 2008;47:308–18. http:// 1994;23:129–38. dx.doi.org/10.1055/s-2008-1076709. 36. Herrmann C, Buss U, Snaith RP. HADS-D – Hospital Anxiety and Depression Scale – Deutsche Version: Ein Fragebogen 49. Steiner JF, Cavender TA, Nowels CT, Beaty BL, Bradley CJ, zur Erfassung von Angst und Depressivität in der somatischen Fairclough DL et al. The impact of physical and psychosocial Medizin. Bern: Huber; 1995. factors on work characteristics after cancer. Psycho-Oncology. 2008;17:138–47. http://dx.doi.org/10.1002/pon.1204. 86 Scand J Work Environ Health 2013, vol 39, no 1 Mehnert & Koch 50. Short PF, Vasey JJ, BeLue R. Work disability associated 53. Torp S, Nielsen RA, Gudbergsson SB, Dahl AA. Worksite with cancer survivorship and other chronic conditions. adjustments and work ability among employed cancer Psycho-Oncology. 2008;17:91–7. http://dx.doi.org/10.1002/ survivors. Support Care Cancer. 2011 Nov 16. [Epub ahead pon.1194. of print]. http://dx.doi.org/10.1007/s00520-011-1325-3. 51. Lee MK, Lee KM, Bae J, Kim S, Kim Y, Ryu KW et al. 54. Tamminga S, de B, Verbeek J, Taskila T, Frings-Dresen M. Employment status and work-related difficulties in stomach Enhancing return-to-work in cancer patients, development of cancer survivors compared with the general population. an intervention and design of a randomised controlled trial. Br J Cancer. 2008;98:708–15. http://dx.doi.org/10.1038/ BMC Cancer. 2010;10:345. http://dx.doi.org/10.1186/1471- sj.bjc.6604236. 2407-10-345. 52. Ahn E, Cho J, Shin DW, Park BW, Ahn SH, Noh D et al. Impact of breast cancer diagnosis and treatment on work- related life and factors affecting them. Breast Cancer Res Treat. 2009;116:609–16. http://dx.doi.org/10.1007/s10549- Received for publication: 12 December 2012 008-0209-9. Scand J Work Environ Health 2013, vol 39, no 1 87

Journal

Scandinavian Journal of Work, Environment & HealthUnpaywall

Published: Mar 15, 2012

There are no references for this article.