Use of inpatient rehabilitation for cancer patients in Switzerland: who undergoes cancer rehabilitation?

a Department of Cranio-Maxillo-Facial and Oral Surgery, University Hospital Zurich, University of Zurich, Switzerland b Institute of Complementary and Integrative Medicine, University Hospital Zurich, University of Zurich, Switzerland c RehaClinic Bad Zurzach, Switzerland d Department of Psychiatry and Psychotherapy, University Hospital Zurich, University of Zurich, Switzerland e Susenberg Klinik Zurich, Switzerland f RehaZentrum Davos, Switzerland g Oncology Centre Hirslanden Zurich, Switzerland h Department of Gynaecology, University Hospital Zurich, University of Zurich, Switzerland i Balgrist University Hospital, Switzerland j Department of Otorhinolaryngology, University Hospital Zurich, University of Zurich, Switzerland k Department of Psychology, University of Fribourg, Switzerland


Introduction
Cancer is one of the major issues in healthcare.It is the second most frequent cause of death overall and the most frequent cause of death in persons between 45 and 64 years of age in Switzerland [1].Over 38 000 new cancer cases are diagnosed and about 16 500 patients die in Switzerland each year [2].The 5-year survival rate after an initial diagnosis is 57% in women and 49% in men [3].Around 4% of the population in Switzerland live with a past cancer diagnosis [4].As a result of early diagnosis, prevention, and progress in medical therapies, the number of longterm cancer survivors has increased steadily over time.Nowadays, cancer may be considered a chronic disease rather than a lethal disease [1,5].Cancer patients frequently experience physical impairment and psychological distress, which is associated with poor quality of life (QoL) [6][7][8][9][10][11][12][13][14][15].After exhaustive, acute medical treatment, patients return to their former environments without the energy and abilities that they had before.They often experience difficulties in resuming their previous social and work roles.Physical and emotional burdens can impair social functioning, which includes maintenance of friendships and relationships with family members [16].Rehabilitation aims to reduce disabilities resulting from the disease and its treatment by improving physical, social, psychological and vocational functioning in order to return to a premorbid condition [17,18].Rehabilitation also helps patients to deal with possible residual disabilities so as to maintain and regain independence, social integration and participation in everyday life [19].Available studies suggest that cancer rehabilitation improves QoL, physical health and psychological well-being [20][21][22][23][24]. Cancer rehabilitation programmes vary across countries depending on the public health system and social security legislation [25].In the Anglo-Saxon countries and Nordic European countries, outpatient rehabilitation is more common.In Switzerland, inpatient rehabilitation is the primary pattern, as is the case in Germany [25][26][27].Many studies have focused on the benefits of specific rehabilitation intervention strategies, such as exercise and aerobic programmes designed in outpatient settings [28][29][30][31].Other studies have reported the benefits of inpatient rehabilitation in general [32][33][34][35][36].However, many of these studies used a pre-post design without control groups in evaluation studies [37], which limits their findings about the effectiveness of rehabilitation.Rehabilitation for cancer patients is often underused compared with that for patients with other diseases, such as cardiovascular diseases [38][39][40].Cheville speculated that the public perception of cancer as an unavoidable progressive and terminal disease might have slowed the development of cancer rehabilitation services in the past [39].Some studies have shown that many cancer patients were not referred to rehabilitation after hospital discharge [12,41,42].Patients were usually referred to rehabilitation at later stages of the disease, when QoL and physical health was already low.Hewitt suggested that more effective assessment and structured procedures are required [10].In Germany, the use of cancer rehabilitation in inpatient and outpatient settings has increased in recent years, because it is considered an inherent part of treatment [43].In Switzerland, cancer rehabilitation usually consists of physiotherapeutic treatment and other specific treatments (e. g. nutrition counselling, lymphatic drainage, psychotherapy).Inpatient rehabilitation usually lasts 2 to 3 weeks and can be prolonged on request.Cancer rehabilitation can also be performed in an outpatient setting.Bachmann et al. found that case management in outpatient rehabilitation can improve QoL of cancer patients after acute therapy in Switzerland [44].All Swiss residents must purchase basic health insurance, which covers a standardised basic benefit package including several health services.The basic health insurance can be supplemented by semiprivate or private insurance policies that give access to more extensive coverage than basic health insurance, including options such as better levels of accommodation or choice of physicians in hospitals [45].Referral to rehabilitation requires several decisions, which results in a rather complicated procedure.First, patients might have to express their wish to attend rehabilitation, and physicians have to estimate the potential for rehabilitation for a specific patient.However, this is rarely based on a standardised procedure.Second, even after such a positive decision the rehabilitation must be reimbursed by health insurance companies.Third, hospital social services have to find a rehabilitation clinic with sufficient capacity.The overall procedure presents several challenges.Good linkage of acute care and rehabilitation is required.Limited data are available for German-speaking countries on differences in sociodemographic or medical characteristics between cancer patients using (users) and not using (nonusers) rehabilitation.Studies from Germany report no differences in age, occupational position or marital status [46][47][48][49].Other studies have reported lower use of rehabilitation in breast cancer patients who are self-employed, are cohabiting with partners [27,[50][51][52][53] and have a shorter duration of chemotherapy [53].Geyer et al. found that breast cancer patients with lower education were more likely to undergo inpatient cancer rehabilitation [47].This is in contrast with more general findings that patients with a lower socioeconomic status were less likely to undergo rehabilitation [54].Studies from Germany have shown that breast cancer patients and younger patients with intestinal tumours are more likely to undergo rehabilitation than patients with other cancer diagnoses.In terms of gender, data also suggest that women are more likely to undergo cancer inpatient rehabilitation than men [43].In Switzerland, 18 rehabilitation clinics offer inpatient cancer rehabilitation.Most clinics are located in the Germanspeaking part of Switzerland.About 14 000 patients use inpatient rehabilitation, including internal/cancer rehabilitation, per year, with a mean duration of 24 days in the clinic [55][56][57].Only about 1% of cancer patients undergo inpatient rehabilitation after hospitalisation [58].About 90% of them are referred after acute inpatient treatment [58].To date, no empirical data are available about the sociodemographic and medical characteristics of patients undergoing inpatient cancer rehabilitation in Switzerland.More data would help to inform policy makers and practitioners about the use of rehabilitation among cancer patients.Therefore, the purposes of this study were (1) to describe differences in sociodemographic and medical characteristics between patients who undergo inpatient cancer rehabilitation and patients who do not; and (2) to analyse data stratified for women and men, since clinical characteristics differ between these populations.

Participants and material
Patients aged at least 18 years with acute inpatient treatment for cancer or for benign or malignant brain tumours were recruited at the University Hospital Zurich and the Balgrist University Hospital (Switzerland) between April 2013 and November 2014.Both the University Hospital Zurich and the Balgrist University Hospital exclusively provide acute health care and no rehabilitation for cancer patients.Some 18.3% of all cancer cases in the area of Zurich are treated at these hospitals [59].We only included patients from these two hospitals to ensure a similar referral organisation procedure throughout the entire sample.As some tumour sites are associated with cognitive impair-ment, which might be more severe in older subjects, we screened patients aged ≥50 years for cognitive impairment (Mini Mental State Examination [MMSE], cut-off of at least 25 points) to ensure their cognitive ability to complete the questionnaires [60].The cut-off indicates a high chance for mild to severe cognitive impairment.Furthermore, sufficient German knowledge was required.These patients were referred to six rehabilitation clinics in several regions of Switzerland.Rehabilitation programmes are adapted to the abilities and needs of the patient, in compliance with the International Classification of Functioning, Disability and Health (ICF) [61].The study was approved by the Cantonal Ethics Committee of Zurich (KEK-ZH Nr. 2012-0563).Sociodemographic and medical data were obtained from the patients' clinical records.Additional sociodemographic information regarding employment status, highest education attained, marital/cohabitating status, children in household, income and treatments after hospitalisation was obtained during hospitalisation with a standardised questionnaire, which was developed especially for this study.Completing the questionnaire took about 5 minutes.Referral allocation to rehabilitation was decided by the physician responsible for acute management in agreement with the patient.Patients' health insurance companies decided whether inpatient rehabilitation was reimbursed or not.

Measures
Tumour stage was classified according to the Tumour-Node-Metastasis (TNM) Classification of Malignant Tumours [62].Thus, brain tumours and haematological malignancies were not included in the analysis of tumour stage.The cancer sites were categorised according to the World Health Organization (WHO) / International Agency for Research on Cancer (IARC) Classification of Tumours, fourth edition [63].

Analysis
First, participants and nonparticipants were compared.Second, the participating patients were divided into users and nonusers.We compared sociodemographic and medical characteristics.Since cancer sites differed substantially between male and female patients, gender was included in further analyses.Descriptive and medical data for women and men with regard to use of inpatient rehabilitation are provided in table 3 in numbers and percentages.All statistical analyses were performed using the Statistical Package for the Social Sciences version 21.0.As a result of non-normal distribution, a Mann-Whitney-U test and a Pearson chi-square test or a Fisher's exact test as univariate analysis were conducted to compare nonparticipants and participants in order to determine biases in age, sex, nationality, employment status, insurance type, cancer stage (excluding brain tumours and haematological malignancies), cancer type, cancer site, treatment and duration of hospitalisation.The sociodemographic and medical data of the nonparticipants are noted in appendix 1.The same statistical analysis procedures were conducted to compare differences between inpatient rehabilitation users and nonusers and for gender.All tests were two-tailed, and the level of statistical significance was set at 95% confidence level (p <0.05).We included age, sex, marital/cohabitating status, children in household, insurance type, employment status, cancer stage (excluding brain tumours and haematological malignancies), cancer type, cancer site, treatment and duration of hospitalisation in the evaluation.

Participants vs nonparticipants
A total of 341 patients were assessed for eligibility using a consecutive sampling approach between April 2013 and November 2014 (fig. 1 shows the patient flow).Fortytwo patients were excluded from completing the questionnaires because of cognitive or physical impairment (n = 19, 45.2% of the excluded patients) or limited knowledge of German (n = 23, 55.8% of the excluded patients).Of the 299 eligible patients, 61 (20.4%) refused to participate.Finally, 238 (79.6% of the total assessed) patients were included in the study.No significant differences were found between participants and nonparticipants in age, nationality, health insurance type, tumour stage, cancer type or cancer site.Participants did not differ in treatment and use of inpatient rehabilitation from nonparticipants.However, study participants stayed in hospital longer than nonparticipants (p = 0.014) (appendix 1).

Patients with and without use of inpatient rehabilitation
Of the 238 cancer patients, 101 (42.4%) patients used inpatient rehabilitation and 137 (57.6%) did not.Descriptive information on users and nonusers is presented in table 1.For the total sample of all patients,the median age was 61 years, with 52.1% (n = 124) aged below 61 years.The sample consisted of 51.7% (n = 123) male patients.Three quarters of the patients were married or cohabiting with a partner (73.5%, n = 175), 15.5% (n = 37) had minors in their household, and 48.7% (n = 116) were employed.Half of the patients (n = 117) had completed apprenticeships, 10.1% (n = 24) basic school level, 14.3% (n = 34) high school, and 22.2% (n = 62) university.No significant differences in sociodemographic characteristics were found between rehabilitation users and rehabilitation nonusers excepting employment status (p = 0.029): rehabilit- Flowchart of study participants vs nonparticipants, rehabilitation users vs rehabilitation nonusers.

Original article
Swiss Med Wkly.2015;145:w14214 ation users were less likely to be employed than rehabilitation nonusers.As table 2 shows, rehabilitation users stayed longer in hospital (M = 20.7;SD = 10.5)than rehabilitation non-users (M = 13.9;SD = 10.4) (p <0.001).The majority of patients had basic insurance (71.8%, n = 171).Patients differed in insurance type, since the percentage of patients with semiprivate or private supplementary health insurance was higher for rehabilitation users (p = 0.030).All tumour stages were represented: 17.0% (n = 33) in stage I, 24.7% (n = 48) in II, 27.3% (n = 53) in III and 30.9% (n = 60) IV.The majority of patients underwent surgery (85.7%, n = 204) and suffered from carcinoma (63.9%, n = 152).There were no differences between rehabilitation users and rehabilitation nonusers regarding tumour stage, cancer type or type of treatment.The most frequent diagnoses were tumours of the head and neck (23.9%, n = 57), of the digestive organs (14.7%, n = 35) and haematological malignancies (12.6%, n = 30).Rehabilitation users and rehabilitation nonusers differed in their cancer site (p = 0.001).Patients with tumours of the digestive organs (p = 0.003) or of the thoracic organs (p = 0.013) used rehabilitation more often, whereas breast cancer patients were less likely to use rehabilitation (p = 0.013).

Use of inpatient rehabilitation and gender
Women who underwent rehabilitation were longer in acute treatment (M = 20.4;SD = 12) than those who did not use rehabilitation (M = 12.3; SD = 10.3;p <0.001) (table 3).Similar results were found for men (rehabilitation users M = 21.0;SD = 9.1; rehabilitation nonusers M = 15.5;SD = 10.2;p <0.001).Neither female nor male patients differed by age in the use of rehabilitation.Men who underwent rehabilitation differed in the distribution of level of education (p = 0.039) from those who did not.Employed men were less likely to undergo rehabilitation (p = 0.051).Male patients with semiprivate or private supplementary health insurance used rehabilitation more often than men with basic health insurance (p = 0.037).Women with an advanced tumour stage used inpatient rehabilitation more often (p = 0.012).Women with tumours of the female genital organs or with breast cancer were less likely to use rehabilitation (p = 0.023).Differences in cancer site were found in both sexes (women p = 0.023; men p = 0.046).Men with tumours of the digestive or of the thoracic organs were more likely to undergo inpatient rehabilitation, while men with haematological malignancies were less likely to do so (trend).

Discussion
To our best knowledge, this is the first study that has analysed differences in sociodemographic and medical characteristics between users and nonusers of inpatient cancer rehabilitation in Switzerland.We found that rehabilitation users and nonusers differed in medical and in sociodemographic characteristics.Users stayed longer in hospital, and inpatient rehabilitation was more likely to be used by pa- tients with tumours of the digestive organs or of the thoracic organs and less likely by breast cancer patients.Interestingly, inpatient rehabilitation users were more likely to have semiprivate or private supplementary health insurance and were less likely to be employed than nonusers.This is an important finding that needs further empirical investigation, since it indicates that social factors might facilitate rehabilitation use or be obstacles for rehabilitation use.
Our study is in line with reports in Switzerland that patients with semiprivate or private supplementary health insurance are more likely to use inpatient rehabilitation [57,64,65].This finding might be explained by the higher chance of such patients obtaining the rehabilitation more promptly.This finding does not coincide with better education of users, since education did not predict rehabilitation use in the current study.Indeed, the opposite was the case: Similarly to Geyer et al., we found that rehabilitation users had a lower education than nonusers, especially among men [47].This finding might be explained by the higher motivation of more educated patients to return to work [66][67][68].According to Lehmann et al., our study reveals that employed patients, especially men, were less likely to undergo cancer inpatient rehabilitation [48].Returning to work, and therefore to normal, as soon as possible might be a strategy for coping with cancer among employed cancer patients [69,70].Mehnert has shown that men with cancer in particular often return to work and show a shorter period of occupational disability [71].Women might tend to make their decision in relation to family or children [72], a possibility which unfortunately was not assessed in our study.Another possible explanation is that nonusers were less impaired than users before rehabilitation and consequently more often in employment.
Our results suggest that women with more advanced cancer more often use rehabilitation.One explanation might be that women with less impairment prefer to stay at home with their relatives to reassume their social role as soon as possible [72].However, they might recognise the need for rehabilitation later.In our study, women with breast or female genital organs cancers did not often use inpatient rehabilitation, which is in contrast to findings from Germany [26,47,53].In Germany, cancer patients might use rehabilitation even after acute outpatient treatment (e. g. chemotherapy, radiotherapy) with less severe impairment [58].In our study, women with breast or female genital organ can- Our study has some limitations that should be considered in the interpretation of our findings.The broad inclusion criteria increase the external validity of our results, but such an approach limits the interpretation for specific diagnoses and patient groups.We were only able to run stratified analyses for gender, but other variables such as education or age could provide meaningful results in a larger sample.In addition, the value of our dataset could have been increased by inclusion of measures of psychopathology and quality of life, as well as stress tests such as the 6-minute walk test or the sit-to-stand test.Unfortunately, limited funding did not allow for a more comprehensive data collection in this study.
A strength of our study is the design, which facilitated the collection of data on nonusers of rehabilitation.The balanced number of users and nonusers made the comparison of both groups feasible.Another strength of our study was the data assessment within different healthcare settings.Acute care and rehabilitation should be well connected, but such studies are rarely done, since the transfer of patients to another system increases problems in data collection.Studies often focus either on acute care or rehabilitation, so this study fills an important gap.
This study provides a basis for further research on this issue.Studies focused on specific diagnoses or with a larger sample are required.

Conclusion
This is the first study that examined the use of inpatient rehabilitation among cancer patients in Switzerland.Our study revealed that referral to inpatient rehabilitation is highly associated with more severe illness, but also with sociodemographic factors such as health insurance and employment, particularly in men.However, the latter indicates that decisions for referrals to inpatient rehabilitation may not be regularly based on medical factors, so cancer patients with regular work and no semiprivate or private supplementary health insurance might underuse rehabilitation.Additionally, our findings provide evidence that gender might influence the use of inpatient cancer rehabilitation.Consequently, more specific and standardised tools assessing medical and psychosocial factors are needed to better and more reliably identify patients who need inpatient cancer rehabilitation.

Figures (large format)
Flowchart of study participants vs nonparticipants, rehabilitation users vs rehabilitation nonusers.

Table 3 :
Characteristics for women and men with regard to use of inpatient rehabilitation.