ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1656

Physical activity and exerciseduring curative oncologicaltreatment

exploring the effects of exercise intensity andbehaviour change support, safety, and patients’ andexercise professionals’ experiences

ANNA HENRIKSSON

ISSN 1651-6206ISBN 978-91-513-0917-0urn:nbn:se:uu:diva-406981

Dissertation presented at Uppsala University to be publicly examined in Sal IV, Universitets huset, Biskopsgatan 3, Uppsala, Friday, 15 May 2020 at 09:15 for the degree of Doctor of Philosophy (Faculty of Medicine). The examination will be conducted in Swedish. Faculty examiner: Professor Yvonne Wengström (Karolinska Institutet).

AbstractHenriksson, A. 2020. Physical activity and exercise during curative oncological treatment. exploring the effects of exercise intensity and behaviour change support, safety, and patients’ and exercise professionals’ experiences. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1656. 98 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0917-0.

Aims: This thesis aimed to explore the effects of exercise intensity and behaviour change support (BCS), the safety of exercise, and experiences of exercise for both patients and exercise professionals during oncological treatment (e.g. neo/adjuvant chemotherapy, endocrine treatment, radiotherapy). This thesis is based on data from the Phys-Can (Physical training and Cancer) multicentre research program, consisting of a feasibility study, an observation study, and a randomised controlled trial (RCT). Methods: Paper I and II were quantitative studies. Paper I was a RCT with a 2x2 factorial design. Patients newly diagnosed with breast, prostate, or colorectal cancer about to start oncological treatment were randomised to six months of high intensity (HI) or low-moderate intensity (LMI) supervised group based resistance- and home-based endurance training, with or without additional BCS. The primary outcome, cancer related fatigue (CRF), was assessed by the Multidimensional Fatigue Inventory. Multiple linear regression and additional responder analysis for primary outcomes were performed. Paper II was a descriptive and comparative study based on secondary data from the observation study and RCT. Data were presented descriptively, and related factors to adverse events (AEs) were analysed with logistic regressions. Paper III and IV were qualitative studies. Participants were patients with breast, prostate, or colorectal cancer undergoing oncological treatment (Paper III) or coaches supervising exercise for participants in the RCT (Paper IV). Data were collected through semi-structured individual- (Paper III and IV) and focus group interviews (Paper III) and analysed with qualitative content analysis (Paper III) and thematic analysis (PaperIV). Main results and conclusions: The results from this thesis indicate that exercise at HImay not improve CRF in comparison with exercise at LMI in patients undergoing treatment,thus patients can be advised to exercise at either preferred intensity. Also, additional BCSdid not improve CRF in relatively motivated patients receiving supervised exercise (Paper I).Furthermore, exercise-related AEs in persons undergoing oncological treatment are minor, ofmusculoskeletal origin, and with a similar incidence as in healthy populations. However, ahigher risk of minor exercise-related AEs was reported in HI groups than in LMI groups. Moreserious AEs were rare, thus it seems safe to exercise even at HI for these patient groups (PaperII). The results also indicated that patients could experience side effects and concerns regardingthe safety of exercising during oncological treatment as barriers to engage in physical activity.Therefore, engaging in physical activity before the onset of side effects from treatment andproviding information regarding physical activity to patients could be beneficial (Paper III).Professionals supervising exercise for patients may find it highly rewarding, which is promisingfor implementation in cancer rehabilitation. However, patients may still receive contradictoryinformation regarding the safety of exercise from health care staff, which can be difficult forexercise professionals to counteract (Paper IV).

Keywords: Physical exercise, Physical activity, Oncology, Cancer rehabilitation

Anna Henriksson, Department of Public Health and Caring Sciences, Lifestyle and rehabilitation in long term illness, Box 564, Uppsala University, SE-75122 Uppsala, Sweden.

© Anna Henriksson 2020

ISSN 1651-6206ISBN 978-91-513-0917-0urn:nbn:se:uu:diva-406981 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-406981)

To my family, friends and relatives.

“Your speed doesn’t matter. Forward is for-ward.” (Unknown)

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Demmelmaier, I.1, Brooke, HL.1, Henriksson, A., Mazzoni, A-S., Helgesen Björke AC., Igelström, H., Ax, A-K., Sjövall, K., Hell-bom, M., Pingel, R., Lindman, H., Johansson, S., Velikova, G., Raastad, T., Buffart, LM., Åsenlöf, P., Aaronson, NK., Glime-lius, B., Nygren, P., Johansson, B., Börjeson, S., Berntsen, S., & Nordin, K. (2020) Does exercise intensity matter for fatigue dur-ing (neo)adjuvant cancer treatment?: The Phys-Can RCT. Sub-mitted manuscript.

II Henriksson, A., Johansson, B., Radu, C., Igelström, H., & Nor-din, K. (2020) Is it safe to exercise during oncological treatment? A study of adverse events during exercise – data from the Phys-Can study. Manuscript.

III Henriksson, A., Arving, C., Johansson, B., Igelström, H., & Nordin, K. (2016) Perceived barriers to and facilitators of being physically active during adjuvant cancer treatment. Patient Edu-cation and Counseling.

IV Henriksson, A., Igelström, H., Arving, C., Nordin, K., Johansson, B.,2 & Demmelmaier, I.2 (2020) The Phys-Can study: Meaning-ful and challenging – supervising physical exercise in a commu-nity-based setting for persons undergoing curative oncological treatment. Physiotherapy Theory and Practice.

1Shared first authorship. 2 Shared last authorship.

Reprints were made with permission from the respective publishers.

Abbreviations and dictionary

Adjuvant treatment Oncological treatment given after curative treatment (e.g. af-ter surgery). Given to reduce risk of relapse of cancer.

BCT Behaviour change techniques, e.g. goalsetting, setting graded tasks, self-monitoring.

Cancer survivor A person diagnosed with cancer until end of life.

CRF Cancer-related fatigue, side-effect/symptom to treatment and disease.

EORTC QLQ-C30 European Organisation of Research and Treatment of Cancer Quality of Life Questionnaire Core 30. PROM that measures health-related quality of life in persons with cancer.

FACIT-F Functional Assessment of Chronic Illness Therapy – Fatigue scale. PROM that measures CRF.

HADS Hospital Anxiety and Depression Scale. PROM that measures anxiety and depression.

HQoL Health-related quality of life

MFI Multidimensional fatigue inventory. PROM that measures CRF.

Neo-adjuvant treat-ment

Oncological treatment given before curative treatment (e.g. before surgery), often to reduce tumour burden.

Oncology treatment Non-surgical cancer treatment (i.e., chemotherapy, radiother-apy, endocrine treatment, and antibody treatment).

Phys-Can Physical training and Cancer. Multicentre research pro-gramme.

Physical activity Any bodily movement increasing energy expenditure above resting metabolism.

Physical exercise Structured and planned physical activity, aiming to maintain or improve physical function, strength, aerobic fitness, and health.

PICC Peripherally inserted central catheter. A type of central venous access, used for administration of chemotherapy and fluids and for blood sampling, etc.

PROM Patient reported outcome measurement. A measurement instru-ment that measures an aspect of a patient’s health that is di-rectly reported by the patient who experienced it.

RCT Randomised controlled trial.

WHODAS World Health Organization Disability Assessment Schedule. PROM that measures function in daily life.

Contents

Introduction ............................................................................................... 13Cancer and cancer treatments ............................................................... 13

Treatment-related side effects ......................................................... 14Changing incidence and mortality – an increased need for rehabilitating care ................................................................................. 16Cancer rehabilitation ............................................................................ 17

Physical activity and exercise in cancer rehabilitation .................... 17Exercise during oncological treatment ............................................ 18

Barriers to exercise during oncological treatment ................................ 19Behaviour change support ............................................................... 20

Safety of exercise during oncological treatment .................................. 21PICC-related complications and lymphoedema .............................. 22

Experiences of professionals supervising exercise .............................. 23

Rationale ................................................................................................... 24

Aims .......................................................................................................... 25Overall aim ........................................................................................... 25

Paper I .............................................................................................. 25Paper II ............................................................................................ 25Paper III ........................................................................................... 25Paper IV ........................................................................................... 25

Methods..................................................................................................... 26Setting ................................................................................................... 26

The Phys-Can research programme ................................................. 26Overview of methods and designs Paper I–IV ..................................... 27Paper I .................................................................................................. 28

Participants ...................................................................................... 28Procedure ......................................................................................... 28The exercise intervention ................................................................. 28The behaviour change support intervention .................................... 30Education of coaches ....................................................................... 31Power calculation ............................................................................ 33Data management and analysis ........................................................ 33

Paper II ................................................................................................. 34Participants ...................................................................................... 34

Outcomes ......................................................................................... 34Data analysis .................................................................................... 37

Paper III ................................................................................................ 37Participants ...................................................................................... 37Procedure ......................................................................................... 37Data collection ................................................................................. 38Data analysis .................................................................................... 39

Paper IV ................................................................................................ 39Participants ...................................................................................... 39Data collection ................................................................................. 40Data analysis .................................................................................... 40

Ethical considerations ............................................................................... 42

Results ....................................................................................................... 43Paper I .................................................................................................. 43

Adherence to the intervention .......................................................... 43Cancer-related fatigue ...................................................................... 43Secondary outcomes ........................................................................ 44

Paper II ................................................................................................. 47Coach- and participant-reported adverse events .............................. 47Lymphoedema, PICC-related complications, and new medical conditions ......................................................................................... 48

Paper III ................................................................................................ 49Paper IV ................................................................................................ 51

Results ............................................................................................. 51

Discussion ................................................................................................. 54The most important findings Paper I–IV .............................................. 54Effects of exercise intensity and behaviour change support on treatment-related side effects ............................................................... 55Safety of exercise during oncological treatment .................................. 57Experiences .......................................................................................... 58Methodological considerations ............................................................ 61

Generalisability and transferability ................................................. 62Study design and data collection ..................................................... 63Reflexivity ....................................................................................... 64

Conclusions ............................................................................................... 65

Implications and future research ............................................................... 66

Svensk sammanfattning ............................................................................ 67Metod ................................................................................................... 68Resultat ................................................................................................. 69Slutsatser .............................................................................................. 69

Tack ........................................................................................................... 71

References ................................................................................................. 73

Appendix ................................................................................................... 84

13

Introduction

Cancer is a major public health issue, with substantial impact on the individ-ual’s life and health. Approximately every third person in Sweden will be di-agnosed with cancer during her or his lifetime (1). Beyond the impact of hav-ing a life-threatening disease, cancer treatments also have side effects that can have negative impacts on the individual’s health. As many individuals live with and beyond cancer, some with negative, long-term consequences from treatment (2), it is necessary to find and provide effective interventions that can reduce and treat these consequences.

Historically, patients have been advised to rest during treatment, however research made during the 1990s and 2000s has questioned this advice and physical exercise have been suggested to attenuate, treat, and in some in-stances even prevent some of the physical and psychological issues that cancer survivors face (3,4). Also, a systematic review and meta-analysis of observa-tional research has suggest that physical activity after a cancer diagnosis may reduce cancer-related mortality (5). Physical activity and exercise are there-fore an important part of cancer rehabilitation and related research. In this the-sis, the focus is on exercise as a part of cancer rehabilitation during oncologi-cal treatment in a curative setting, including persons undergoing treatment against breast, prostate, or colorectal cancer.

Cancer and cancer treatments Cancer is not a single, specific disease, but a generic name for approximately 200 diseases. In Sweden, more than 63,000 individuals were diagnosed with cancer in 2018. The most common form among women was breast cancer, with more than 7,800 women diagnosed; among men, it was prostate cancer, with almost 10,900 men diagnosed. The same year, more than 6,800 persons were diagnosed with colorectal cancer, also a common form of cancer for women and men in Sweden (1).

For many years, the only treatment against cancer tumours was surgical removal. It was not until the beginning of the 20th century that new means of treatment were discovered and developed, enabling cure and improving treat-ment. Around mid-century, treatment was also developed for non-solid can-cers, like leukaemia (6). Cancer treatment can now entail surgical removal for

14

solid tumours, as well as oncological treatment, for example radiotherapy, chemotherapy, endocrine treatment, and antibody treatment.

Treatment can also have different aims: curing or preventing local spread of the cancer (e.g., surgically removing a tumour or radiotherapy), preventing relapse of cancer, i.e., adjuvant cancer treatment, or reducing tumour burden before curative treatment, i.e., neo-adjuvant cancer treatment. In some in-stances, patients receive both neo-adjuvant and adjuvant treatment. The choice of treatment depends on the type of tumour, its localisation and stage, if it is sensitive to the available treatments, and the patients age and present co-mor-bidity may also be taken into account (7). Thus, modern cancer treatments are individualised and efforts are made to reduce treatment intensity, and avoid under and over treating patients (8).

Treatment-related side effects All cancer treatments (both surgical and oncological) have side effects; their nature and severity may depend on the type of treatment, doses, and biological and psychological factors (9). Cancer-related fatigue (CRF) is the most com-mon side effect of treatment, as well as being a cancer symptom. It is defined as a subjective sense of persistent tiredness related to cancer, that interferes with usual functioning, and is not sufficiently relieved by rest or sleep. Fur-thermore, CRF seems to be multidimensional, with physical, mental, and psy-chological manifestations, such as feelings of general weakness, difficulties concentrating, decreased motivation, and emotional lability (10).

Moderate to severe CRF is present in 30 to 60 percent of patients during treatment and has a negative impact on health-related quality of life (HQoL) (9,11).

Although the aetiology is not completely determined, it is believed that fa-tigue is multifactorial. Research has suggested that one contributing to fatigue is pro-inflammatory cytokines (i.e., proteins and peptides which perform chemical signalling between cells). Therefore, CRF can be present as a symp-tom before the start of treatment and be a persistent side effect long after treat-ment has been completed. Typically, CRF increases during oncological treat-ment and may in some instances lead to reduction or early discontinuation of oncological treatment (9).

Other factors, such as physical inactivity and associated decreases in cardi-orespiratory fitness are also related to CRF (9). During treatment patient’s aer-obic capacity (i.e. cardiorespiratory fitness) typically decline (12), and it has been suggested that CRF, at least to some extent, mediates the relationship between physical fitness and HQoL (13).

Research have demonstrated that body composition may play a role when it comes to side effects from treatment. Individuals suffering from obesity or having unfavourable changes in body composition (i.e. decrease in muscles

15

mass and increase in body fat), are also at risk of experiencing more side ef-fects from oncological treatment. At the same time, an unfavourable change of body composition is a common side effect from several types of oncologi-cal treatments (14).

Psychological disorders, such as anxiety and depression, are also common complications of cancer. Prevalence can vary depending on the type of meas-urement instrument, the type of cancer, and where the patient is on the treat-ment continuum. A meta-analysis including adult patients with cancer found that prevalence of depression was highest during treatment; self-reported prevalence of depression was 27 percent and, when measured through diag-nostic interviews, prevalence was 14 percent (15). Furthermore, approxi-mately ten percent of cancer survivors has issues with anxiety and approxi-mately 30 to 40 percent had some combination of mood disorders when diag-nosed in psychiatric interviews (16,17).

Sleep disturbances (defined as any sleep-related outcome, including insom-nia) are a common problem in cancer survivors, with incidence rates of in-somnia ranging from twenty to over fifty percent (18,19). This can be com-pared with the general population, with less than ten percent prevalence of insomnia (17). The exact impact of sleep disturbances is difficult to estimate due to differing definitions, measurement methods, and study designs. Exer-cise may improve sleep dysfunctions in populations without cancer.

Side effects from radiotherapy, like skin erythema and cancer-related fa-tigue, are common in the treatment of breast cancer, and men with prostate cancer receiving radiotherapy experience urinary and bowel-related symp-toms (20).

Pain is a common side effect of cancer treatments. For instance, with a tax-ane, platinum and vinca alkaloid based chemotherapy about half of the pa-tients treated experienced peripheral neuropathy which besides pain includes symptoms of numbness, tingling, motor impairment and cold sensitivity (21) , in some cases these side-effects can be persistent (22). Aromatase inhibitors, a commonly used endocrine therapy used in the adjuvant setting against breast cancer, can cause arthralgia and muscular pain in women receiving treatment against breast cancer. Predictors for developing aromatase inhibitor-induced arthralgia is a Body Mass Index lower or higher than 25-30, and having taxane based chemotherapy was also associated with higher risk (23).

Other common side effects of chemotherapy are nausea and vomiting. Chemotherapy treatment with emetogenic effects is always combined with prophylactic antiemetic regimes. However, an observational study found that 31 percent of patients receiving moderately emetogenic chemotherapy did not have complete protection, and nausea was more challenging to control than vomiting (24).

16

Measuring side effects from cancer treatment Adverse events and side effects from treatment have in clinical studies usually been assessed by physicians, interpreting patients’ signs (i.e., discovered by physician) and symptoms (i.e., described by patient) with use of the Common Terminology Criteria for Adverse Events (25). Studies regarding the agree-ment between physician and patient assessment of adverse events indicate that intervals and severity of symptoms may be underestimated by physicians when the symptoms are more subjective, for example, in CRF (26,27). Instead, patient reported outcomes, defined as a measurement of an aspect of patient health that comes directly from the patient without interpretation from any-body else, have been introduced as effectiveness endpoints in clinical trials (28). A patient reported outcome measure (PROM) allows the patient to report the subjective severity of a symptom. The usefulness and importance of PROMs have gathered evidence, for instance, an epidemiologic study includ-ing almost 7000 cancer survivors found that a commonly used PROM that measured health related quality of life (HQoL) in cancer survivors, (EORTC QLQ C30) could predict all-cause mortality (29).

Changing incidence and mortality – an increased need for rehabilitating care Since 1958, the Swedish National Board of Health and Welfare has annually collected and registered data on national cancer incidence in Sweden. From about 1970 to 2016, the cancer incidence increased, while the overall mortality slowly decreased (1).

The change in cancer incidence is due to an ageing population and changes in lifestyle. The decrease in mortality, on its part, is due to new methods for detecting cancer at an early stage (i.e., screening methods) and new, more ef-fective cancer treatments (1). The demographic changes of cancer incidence and increased survival indicate that many people are going to live with and beyond cancer and have to handle the emotional and physical side effects of the disease and its treatment (26). For instance, cancer survivors (defined as persons who have been diagnosed with cancer until end of life) subjected to treatment may experience unfavourable changes in HQoL, CRF, body com-position, bone health, pain, sleep disturbances, and physical functioning. They also have an increased risk of cardiovascular disease compared with the non-cancer population (27).

These direct and long-term effects of cancer and its treatments may impair the ability to return to work (28). For instance, cancer survivors have a 1.4 times higher risk of unemployment compared with the healthy population (29). Reduced work ability in breast cancer survivors is associated with more

17

depression, financial problems, fatigue, more symptoms from the breast, and lower HQoL (28).

The impact of cancer and side effects from treatment not only cause human suffering, but also pose a challenge to public economy and health care sys-tems. In light of these challenges, interventions aimed to optimise functioning and reduce disability in cancer survivors, i.e., rehabilitation efforts (30), are needed.

Cancer rehabilitation The World Health Organisation has defined rehabilitation as interventions aimed to aid individuals who experience, or are expected to experience disa-bility to obtain and maintain optimal functioning in interaction with their en-vironment (30). A more specific definition for cancer rehabilitation has been provided by the Nordic Cancer Union stating that: “Cancer rehabilitation is a specific period of time during which the physical, psychological, social and existential consequences of cancer and the treatment are prevented and reduced”(31).

Cancer rehabilitation should be directed towards the need of the patient and can include psychological and physical interventions. Many professions can be included in the rehabilitating care; for instance, dieticians, physiotherapists, psychologists, physicians, and nurses may all be included, depending on the rehabilitation needs of the patient (30).

In the Swedish National cancer rehabilitation care program, it is recom-mended that all patients should be offered cancer rehabilitation throughout the entire care continuum and that physicians and contact nurses have the primary responsibility for cancer rehabilitation (32). In cancer rehabilitation, the con-tact nurse is responsible for assessing the patients’ needs, supporting the pa-tients in self-care and act as the main point of contact throughout the treatment and follow-up period (33). Furthermore, in the care program, physical activity and exercise have been highlighted as the most important interventions to achieve good results from cancer rehabilitation (32).

Physical activity and exercise in cancer rehabilitation Physical activity includes all physical motion that raises energy expenditure above that of the resting metabolism. Exercise is a subtype of physical activ-ity, including a planned and structured activity, and could be viewed as a be-haviour with the aim to improve or sustain physical function, strength, and/or aerobic fitness (34).

The general recommendation for cancer survivors is to engage in minimum 150 minutes per week of at least moderate intensity and resistance training including major muscle groups at least twice a week. Combination of endur-ance training and resistance training seems to have a greater effect than only

18

choosing one type of exercise (35). More specific guidelines regarding exer-cise for cancer survivors to prevent or reduce CRF are three weekly sessions of endurance training with at least moderate intensity and resistance training two to three times per week (4).

Exercise during oncological treatment As of yet, there is no consensus regarding when in the rehabilitation contin-uum exercise should be introduced for cancer survivors. However, early inter-vention may help prevent some of the side effects of surgery and oncological treatment. During the last decades, studies have investigated the effects of ex-ercise during oncological treatment (4). Exercise during treatment can help improve HQoL and reduce CRF, depression, anxiety, and sleep disturbance, as well as improving emotional wellbeing and physical function (36–39).

For instance, Courneya et al. (40) investigated the effects of different doses and modes of exercise on sleep quality in patients undergoing chemotherapy treatment against cancer. They found that higher volumes of endurance exer-cise, and endurance exercise combined with resistance training, improved sev-eral aspects of sleep quality compared with a standard dose of endurance ex-ercise.

Furthermore, exercise has been suggested to reduce symptoms of chemo-therapy-induced peripheral neuropathy. For instance, a six week home-based moderate-intensity walking and resistance training exercise intervention showed small to moderate effect in reducing these symptoms (41). However, it has not been studied if a longer duration exercise or higher intensity could improve outcomes further. Van Waart et al. (42) found that patients receiving chemotherapy against breast cancer that engaged in low intensity or moderate- to high-intensity resistance and endurance exercise programs had less nausea and vomiting when compared to patients receiving usual care.

Previous research has indicated that some of radiotherapy related side ef-fects may be alleviated by physical activity. A retrospective analysis (43) of radiotherapy-related side effects in men with prostate cancer during a moder-ate-intensity exercise intervention found statistically significantly less bowel-related symptoms in the intervention group.

Exercise intensity There is no consensus on if exercising at a high intensity improves outcomes for cancer survivors such as CRF or HQoL more than exercising at low-mod-erate intensities. For instance, one study including women with breast cancer receiving chemotherapy who were randomised to either low intensity or mod-erate to high intensity found a statistically significant difference in physical fatigue (42). However, exercise volume (i.e., intensity, duration, and fre-quency of the exercise) was not controlled for, and it is unknown if experi-enced intensity was important for the outcome of fatigue. Furthermore, two

19

intervention studies comparing exercise intensity after ended treatment did not find any differences regarding CRF between groups (44,45). This means that second generation studies comparing exercise intensities are needed (4,46).

Chemotherapy completion rates and exercise The efficacy of chemotherapy treatment is related not only to tumour charac-teristics but also to the timing of treatment initiation, the time intervals be-tween cycles (47), and the dose intensity. However, treatment-related side ef-fects and toxicity may lead to early discontinuation or delayed treatment. In-terrupted or premature discontinuation of treatment may reduce the overall and disease-free survival in breast cancer due to reduced dose intensity (48,49).

Relative dose intensity is a measure of the received dose intensity (in rela-tion to planned treatment). Maintaining relative dose intensity above 85% has been shown to improve survival rates for women receiving chemotherapy against breast cancer (49). Estimates of how many patients receive optimal dose intensity are difficult to make because of changing treatment regimens and differing use of prophylactic granulocyte colony-stimulating factor (a drug used to reduce duration of neutropenia and incidence of neutropenic fe-ver). There are few clinical studies on this; however, retrospective studies have reported that 16.2–26% of patients with breast cancer received less than 85% of relative dose intensity (50,51).

There are two studies indicating that supervised exercise at a high intensity, including resistance training and endurance exercise, can help improve chem-otherapy completion rates in women receiving adjuvant breast cancer treat-ment (42,52). However, there are several other studies where no effect on chemotherapy completion rates was found (53). These inconsistent results may be due to the studies investigating different doses, types, and delivery modes of exercise, as well as using different measurements of chemotherapy completion rates (53). For instance, van Waart et al. (42) used percentage of dose reduction as a dependent variable, while Courneya et al. measured chem-otherapy completion rates using percentage of relative dose intensity (40,52). Regardless, it seems that exercise does not worsen the tolerability of chemo-therapy treatments. Further research is needed to determine if exercise has any effect on chemotherapy completion rates (53).

Barriers to exercise during oncological treatment Though exercise has some benefits throughout the cancer continuum, exercis-ing during treatment may be a challenge. Epidemiological research has found that physical activity rates seem to decline during treatment and often remain at a lower level than before diagnosis. Also, patients with cancer are generally less physically active than the general population (54–57).

20

For instance, an epidemiological study including 3.000 breast cancer sur-vivors found that 42 percent decreased their time in moderate to vigorous physical activity and 73 percent maintained or increased time spent in seden-tary behaviour 24 months after diagnosis. Patients with higher education were more likely to have more time spent in moderate to vigorous physical activity. Furthermore, patients receiving chemotherapy, those diagnosed with a higher stage cancer, and those of older age were more likely to have unfavourable physical activity trajectories, indicating that these may be risk factors (58), why these individuals may need more support to maintain or increase physical activity. Other risk factors for inactivity that have been identified as important after completed treatment against breast cancer are high BMI, smoking, and worse physical functioning scores (59).

In contrast to the observed low adherence to physical activity recommen-dations, previous qualitative research has found that women who exercised during treatment had a positive attitude toward and experience of exercise (60,61). However, most qualitative research regarding experiences has been conducted on persons participating in exercise interventions. Thus, it is un-known if their experiences of physical activity and exercise can be transferred to patients not participating in exercise interventions.

Adhering to exercise recommendations can be challenging even for cancer survivors included in intervention studies. Shang et al. (62) reported low ad-herence to the intervention group exercise (32.5%) and the authors found a strong correlation between reduced adherence and symptoms such as CRF, anxiety, and depression. Courneya et al. (63) had more than seventy percent adherence in both their groups, when measuring adherence as being physically active at least 150 minutes/week at a moderate intensity or 60 minutes/week at a high intensity. In this study, over fifty percent of the non-adherence was due to disease or treatment-related side effects.

What appears to be the common denominator in studies that have high ad-herence to their exercise interventions is that all of them include some type of behavioural change techniques (BCT, i.e. strategies used to help change or govern the process of change), such as goal-setting and self-monitoring of ex-ercise (64,65).

Behaviour change support In health care, support provided to change behaviours, i.e., behaviour change support, can target a range of outcomes. Examples include preventing people from engaging in harmful behaviours, such as smoking or being sedentary, and promoting engagement with health-protective behaviours, such as physi-cal activity or exercise. As mentioned before, BCT may be helping patients change exercise behaviour, however, there are still uncertainties regarding this due to unclear descriptions of the BCT used and inconsistent reporting of ex-ercise adherence (66).

21

In order to understand physical activity and exercise behaviour and how to help patients exercise during treatment, using behaviour-oriented theories to inform interventions is important. Social Cognitive Theory developed by Ban-dura (67) includes determinants and describes the process of change. This the-ory describes outcomes expectations (i.e. the expectation an individual has on the effect a behaviour will produce) and outcomes expectancies (i.e. the value of an individual place on the outcome) as important determinants for behav-iour change. Furthermore, self-efficacy, is an important concept within the Social Cognitive Theory, which can be described as an individual’s degree of confidence that she or he can engage in a specific behaviour, for instance, exercise (68). Individuals with high self-efficacy are able to set higher goals and have higher commitment to the goals than individuals with low-self effi-cacy. Also, self-efficacy can determine how an individual handle barriers and impediments. According to Bandura, self-efficacy influences the behaviour directly, but also indirectly through impact on goals, outcome expectations and perceived barriers and facilitators (67). Besides self-efficacy, an individ-ual also needs behaviour capability, which included knowledge of the behav-iour and skill to perform the behaviour.

Furthermore, the theory behind using goal setting as a BCT, is that it may lead to an improved performance because the individual with a goal will exert herself to a higher degree, be more focused, be more persistent, and if needed develops strategies in order to perform the behaviour. The goal should be spe-cific and easy to measure, stated as a behaviour and preferably not too com-plex. Also the goal should be adapted to the individuals’ self-efficacy and goal setting can be applied to all behaviours in which feedback is possible (69).

Safety of exercise during oncological treatment One reason for not engaging in physical activity or exercise during treatment may be concern regarding the safety of exercise, both from the perspective of the patients and among health care providers (70–72).

Adverse events are any unfavourable or harmful complications or out-comes that occur during or after an intervention but are not necessarily caused by it. Studies that reported adverse events related to exercise during oncolog-ical treatment did not identify any specific risk from exercise and serious ad-verse events were rare. Therefore, it has repeatedly been stated that exercise is safe during treatment for most patients (36,73,74). However, it has also been recognised that definitions of adverse events and information on data collec-tion methods are not systematic or standardised across studies, and in some instances are completely missing (73–75). It is therefore uncertain whether there are no specific risks related to exercise during treatment or if there are such risks which have not been adequately reported or assessed.

22

Epidemiological research regarding adverse events in healthy individuals has indicated that the type of physical activity may influence the risk of injury. For instance, activities like walking are not related to increased risk of adverse events, while high-intensity exercise, like running and contact sports, in-creases the risk of musculoskeletal injuries (76–79). Furthermore, it seems that increasing time of physical activity at a higher intensity increases the risk of injury, especially in persons not previously physically active (76,79).

PICC-related complications and lymphoedema Most patients receiving chemotherapy in Sweden are fitted with venous access for drug administration. A common form of venous access is a peripherally inserted central catheter (PICC), a type of catheter that is usually inserted into a vein in the upper arm and guided into a large vein above the heart. Receiving a PICC is associated with increased risk of thrombosis, dislocations, and in-fections (80–82). Risk factors identified in observation studies for PICC-re-lated thromboembolism were receiving chemotherapy, older age, overweight, and diabetes (82–87). Previously, patients with central venous access (such as PICC) were advised to avoid resistance training of muscles in the area near the catheter in order to avoid dislodgment (88). The recommendation of avoid-ing upper limb resistance training is not in line with evidence of improved upper limb function from early exercise interventions in women after breast cancer surgery (89). Also, research has found that being less active with the inserted arm may increase the risk of thromboembolism (83,90). Today, there are no published studies on the risks of exercising with a PICC, except a Swe-dish study that investigated the effect of adding high intensity interval training to conventional exercise (including resistance training) in women receiving chemotherapy against breast cancer. This study reported that no adverse events regarding PICC occurred during the study period (91).

Lymphoedema is a side effect of breast cancer surgery and oncological treatment, affecting approximately one fifth of patients (92). Lymphoedema is the accumulation of interstitial fluid due to insufficient lymph drainage. Sur-gery or radiotherapy may lead to partial or complete destruction of the lym-phatic system of the treated area, and changes in the structure of the skin and subcutaneous tissues, like scarring and fibrosis, may also lead to insufficient drainage (93). Resistance training has historically been believed to increase lymph fluid production, which might in turn trigger or worsen lymphoedema (94). This is why patients with breast cancer at risk of or already suffering from lymphoedema have in the past been advised to avoid lifting heavy things, e.g., resistance training. Nonetheless, research has demonstrated that progres-sive resistance training is safe, and there have even been indications that ex-ercise can ameliorate lymphoedema (73,95). Still, there have been few studies comparing lymphoedema outcomes related to resistance training at different

23

exercise intensities (96). More knowledge regarding this could help guide rec-ommendations to patients at risk of developing lymphoedema.

Experiences of professionals supervising exercise Receiving professional support seems to be important for the effect of exercise on outcomes for cancer survivors. For instance, Sweegers et al. (97) found supervised exercise to be more efficient than non-supervised exercise in im-proving quality of life in patients undergoing oncological treatment, probably due to improved adherence to the prescribed exercise (72,97). Furthermore, exercising with other cancer survivors and having a skilled coach may also be facilitators (72,98). To date, studies describing the experiences of profession-als (e.g., physical therapists and personal trainers) who deliver exercise pro-grammes to cancer survivors are few. Knowledge regarding their experiences could help inform implementation.

Furthermore, due to a lack of financial and other resources, offering all pa-tients exercise facilities at hospitals may be impractical. Implementing com-munity-based exercise programmes might be an alternative. For instance, Covington et al. (99) found in their review that community-based programmes for cancer survivors could improve HQoL. However, they also stated that more implementation research regarding this was needed.

24

Rationale

Despite knowledge of the positive effects of exercise during oncological treat-ment on CRF and other outcomes, the role of exercise intensity and effect of behaviour change support to improve outcomes for patients are unclear.

Furthermore, it is important to systematically assess the possible risks of injury and other adverse events and to investigate if there are any risks that depend on exercise intensity in persons undergoing cancer treatment. Knowledge on this is imperative to address patient concerns regarding the risks of exercise during adjuvant treatment, and from a clinical point of view, to make adequate risk assessments and to provide evidence-based advice on exercise.

Previous research conducted with cancer survivors has identified a discrep-ancy between high levels of positivity towards physical activity and exercise and actual adherence to physical activity guidelines. This indicates that there are barriers preventing patients with cancer from being physically active and engaging in exercise. Identifying and understanding patients’ perceived barri-ers to and facilitators for being physically active is important. This knowledge is of significance for health care staff who will work with supporting cancer patients towards an active life.

Also, the experiences of professionals supervising exercise for patients un-dergoing treatment has not previously been addressed. Their experiences may inform implementation of programmes for supervised exercise in cancer reha-bilitation.

The fundamental purpose of increasing knowledge regarding these subjects is to improve cancer rehabilitation for persons undergoing oncological treat-ment.

25

Aims

Overall aim The overall aim of the thesis was to explore the effects of exercise intensity and behaviour change support, the safety of exercise, and experiences of ex-ercise for both patients and exercise professionals during oncological treat-ment.

Paper I The aim of Paper I was to determine the effects of high intensity versus low-to-moderate intensity exercise with or without additional behaviour change support on CRF in patients undergoing (neo-)adjuvant cancer treatment. Sec-ondary aims were to determine the effects on HQoL, anxiety/depression, func-tion in daily life, cardiorespiratory fitness, muscle strength, amount of physi-cal activity, sedentary time, time in sleep, and chemotherapy completion rates.

Paper II The aim of Paper II was to describe incidence and types of adverse events during exercise for persons undergoing oncological treatment, as well as re-lated factors. A second aim was to describe and compare incidence of lym-phoedema, PICC-related complications, and other medical conditions, be-tween different exercise intensities and a comparison group given usual care.

Paper III The aim of Paper III was to explore the barriers and facilitators to being phys-ically active during adjuvant cancer treatment, as perceived by cancer survi-vors with breast, prostate, or colorectal cancer.

Paper IV The aim of Paper IV was to explore how physical therapists and personal train-ers experienced supervision of exercise in a community-based setting for per-sons undergoing curative oncological treatment.

26

Methods

Setting The Phys-Can research programme Phys-Can (Physical training and Cancer) is a research programme consisting of three studies: a feasibility study, an observational study and an RCT (ran-domised controlled trial). Data from the three studies make up the foundation for this thesis.

The feasibility study In late 2013, the Phys-Can research programme began with a feasibility study conducted in Uppsala. The aims of this study were, in addition to evaluating the feasibility of an exercise intervention, to identify potential barriers to and facilitators for physical activity in patients receiving curative oncological treatment.

The observation study and RCT The Phys-Can observational study and RCT were multicentre studies con-ducted at three sites (Linköping, Lund, and Uppsala) in Sweden and have pre-viously been described in detail (100). The observational study aimed to de-scribe how the diseases and treatments affected the physical condition, mental well-being, and HQoL of persons during curative oncological treatment. Par-ticipants in the observational study did not receive an intervention, acting as a historical comparison group for those who were later included in the Phys-Can RCT (100). The recruitment to the Phys-Can observational study began in September 2014 and ended in 2015, before the start of the RCT.

The Phys-Can RCT was a study aiming to evaluate the effects of low to moderate intensity exercise (LMI) compared with high intensity exercise (HI), with or without behavioural change support, on CRF, in persons undergoing curative oncological treatment. In addition, the effects of exercise on HQoL, chemotherapy treatment completion rates, disease outcomes, adverse events, health-economic efficiency, etc., were evaluated. Recruitment of study partic-ipants to the Phys-Can RCT started immediately after the end of recruitment to the observational study in March 2015 and ended in April 2018. A more detailed description of the exercise intervention is presented in the methods section.

27

Overview of methods and designs Paper I–IV The four included papers were studies used both quantitative (Paper I and II) and qualitative (Paper III and IV) methods, see Table 1. Paper I is a RCT with a 2x2 factorial design. Paper II is a descriptive and comparing study including data from Phys-Can RCT and observation study. Paper III and IV are qualita-tive interview studies.

Table 1. Overview of methods and designs Paper I–IV

RCT = randomised controlled trial. BRC = Breast cancer. PRC = Prostate cancer. CRC = Colorectal cancer. OBS = observational study. VO2max= Maximal oxygen up take. 1RM = one repetition maximum PROMs= Patient reported outcomes measures.

Study method/ design

Data collection Participants Analyses

Paper I Quantitative. Multicentre RCT. 2x2 fac-torial design.

PROMs. VO2max. SenseWear®

1 RM test Medical record.

Participants with newly di-agnosed BRC, PRC, or CRC, N = 577

Descriptive statis-tics (including re-sponder analysis) and multivariate regression analysis.

Paper II Quantitative. Descriptive and comparative.

Coach reported and self-re-ported study specific ques-tionnaires. Medical record.

Participants with newly di-agnosed BRC, PRC, or CRC, N = 577 from RCT,N= 90 from OBS.

Descriptive statis-tics, logistic regres-sion analysis, and Chi2 tests.

Paper III Qualitative. In-terview study.

Focus group and individual interviews. Data from Phys-Can feasi-bility study.

Patients with BCR (n=9), PRC (n=8), or CRC (n=1) re-ceiving treat-ment and BRC (n=5) about to start cancer treatment. Total N=23

Qualitative content analysis.

Paper IV Qualitative. In-terview study.

Individual in-terviews. Data from Phys-Can intervention study.

Professional ex-ercise coaches (physiothera-pist, n=9 and personal train-ers, n =2), N = 11

Thematic analysis.

28

Paper I Participants Participants in Paper I were patients newly diagnosed with breast, prostate, or colorectal cancer about to start adjuvant (breast and colorectal cancer) or neo-adjuvant cancer treatment (breast cancer and prostate cancer). They were re-cruited at three Swedish university hospitals (Linköping, Lund, and Uppsala). In addition, men diagnosed with prostate cancer and scheduled to undergo cu-rative radiation therapy without endocrine treatment were included. Inclusion and exclusion criteria for Paper I are presented in Appendix 1. Between March 2015 and April 2018, 2600 patients were accessed for eli-gibility, and of those 549 were ineligible and 1451 individuals declined par-ticipation. Twenty-nine percent of the eligible patients participated in the RCT (Paper I). Six-hundred patients agreed to participate and, of these, 577 patients were randomised. Participants included were younger (mean age 58.7 vs. 63.6 years, p value < 0.001) and less likely to have prostate cancer or colorectal cancer than breast cancer (OR [95%CI]; CRC 0.58 [0.36–0.92], p-value = 0.02; PC 0.65 [0.51–0.83], p value = 0.001), than those who did not participate (Appendix 2).

Procedure Persons who were deemed eligible were given written and verbal information after a planned visit at a surgery or oncology office before the start of treat-ment. If a person consented to participate, they were asked to sign an informed consent form. Participants underwent a baseline measurement, including fill-ing out questionnaires, performing a maximal oxygen uptake test (VO2 max), and measuring physical activity (see outcomes for a detailed description of questionnaires and testing).

Participants were randomly assigned to one of four study conditions after baseline testing. The randomisation was computer-generated, stratified by di-agnosis (breast, prostate, or colorectal cancer) and study site (Linköping, Lund, and Uppsala), and randomised in blocks of eight. This was done in order to reduce the risk of uneven random allocation of participants regarding diag-nosis and to ensure approximately equal-sized intervention groups (101).

The exercise intervention The intervention consisted of HI or LMI supervised resistance training twice a week at a public gym and home-based endurance exercise during six months, with or without behaviour change support.

29

Resistance training Resistance training was group-based and supervised by professional coaches (physiotherapists and gym coaches) who received a study-specific education.

During the first six weeks, the participants were familiarised with the exer-cise programme and the Omni scale of perceived exertion (102). The Omni scale has both a categorical rating format and a numerical response scale. The categorical rating format includes pictures of a man lifting a weightlifting bar, symbolising different resistance training intensities, and written descriptions. The numerical response scale ranges from 0 to 10.

The resistance training encompassed seated leg press, leg extension, seated leg curl, seated row, and chest press performed in machines, and seated or standing shoulder press performed with dumbbells. In addition, the partici-pants performed sit-ups, the plank, the bird-dog, and pelvic floor exercises (Appendix 3).

Intensity was determined after a regular strength test of 6 and 10 repetition maximum, i.e., the weight a person can maximally lift x number of times. The HI groups alternated between 3 sets x 6 repetition maximum with 2 min set rest and 3 sets x 10 repetition maximum with 1 min set rest, with the last set to failure and at 9–10 on the Omni scale. The LMI groups alternated between 3 sets x 12 repetitions at 50% of 6 repetition maximum with 2 min set rest and 3 sets x 20 repetitions at 50% of 10 repetition maximum with 1 min set rest. The last set was at 5–7 on the Omni scale.

Endurance training Before participants started home-based endurance exercise, they conducted four endurance training sessions with the coach, in order to learn how to use a heartrate monitor and to become familiar with the Borg rating of perceived exertion scale (103). Both heartrate monitor and Borg scaled were used to monitor exercise intensity. The LMI groups exercised 150 minutes per week at 40 to 50% of their measured heart rate reserve (calculated as: peak heartrate – resting heartrate) and rating of perceived exertion at 12–13.

The HI endurance exercise was conducted in intervals of two minutes with two minutes rest between work sessions, and exercise intensity of 80–90% of heart rate reserve and rating of perceived exertion 15–17 at the end of the last session. The interval exercise progressed with five sessions at the start, in-creasing to six after week 6 and adding one session every fourth week until a maximum of 10 sessions was reached.

The recommended frequency of endurance exercise was two times a week for all participants. The type of endurance training was based on each partici-pant’s individual preferences, e.g., walking, bicycling, running.

30

Safety measures during exercise and testing in Phys-Can Within the Phys-Can project, several precautionary measures were imple-mented in order to reduce the risk of adverse events related to exercise and testing. For instance, prior to inclusion, all participants were assessed by an oncologist or oncology surgeon regarding exclusion criteria and contraindica-tions to high intensity exercise or testing (Appendix 1).

Resistance training started at a low intensity, consisting of 1 set of 20 reps at 30 repetition maximum in week 1. From there, it slowly progressed for six weeks, with an increasing number of sets and power output, and a decreasing number of repetitions, for both low-moderate and high intensity groups.

The participants were instructed to warm up during 5–10 minutes at mod-erate intensity before every exercise session, both endurance and resistance training. Also, for resistance training, a specific warm-up for each resistance exercise, consisting of one set at 50% repetition maximum, was recom-mended. For participants receiving intravenously administered chemotherapy, no exercise or exercise testing (i.e., muscle strength test or VO2max test) was conducted for 24 hours after administration (104).

Exercises that were painful to a participant were substituted with other exer-cises activating the same muscle groups. If participants had severe side effects, weights were lowered temporarily, and then successively increased again.

In addition to these precautions, the participants were instructed to refrain from exercise if they had fever, a sore throat, or an ongoing infection.

The behaviour change support intervention Participants randomised to additional behaviour change support received face-to-face guidance from the coaches in using strategies to facilitate adherence to the exercise. These sessions were provided in connection with the supervised training or by phone. The primary focus of the behaviour change support was the home-based endurance training.

The coaches conducted initial exploration of the participants’ motivation by asking about a) previous experience of physical activity and exercise, b) expectations of outcomes if the exercise protocol was followed, c) anticipated barriers to and facilitators for following the exercise protocol, and d) self-ef-ficacy to take part in the planned weekly exercise. After the initial session, the participant and coach met on a regular basis, with the coach guiding and in-structing the participant in behaviour change techniques (BCTs).

The BCTs used were goal setting, short-term action planning, self-moni-toring, review of goals, basic functional behaviour analysis to identify indi-vidual determinants of exercise behaviour, and long-term coping planning to maintain exercise after the end of the exercise intervention. In this thesis, the focus is on the effects of the behaviour change support provided during the RCT (Paper I), not on long-term maintenance of exercise. The BCTs were individually tailored based on each participant’s needs and behaviours. In all,

31

approximately nine sessions were offered to the participants, with a gradually decreasing frequency during the course of the exercise intervention.

Education of coaches The coaches (13 physiotherapists and 2 personal trainers) supervising the ex-ercise in Phys-Can RCT received study-specific education. The education consisted of three course days with lectures and seminars on exercise physiol-ogy, cancer, and cancer treatment. It also encompassed practical training ses-sions in a gym, where the coaches practised providing supervised exercise and muscle strength testing, as described in the study protocol.

Coaches providing the behaviour change support intervention received ad-ditional education consisting of three course days and homework assignments.

To maintain the fidelity of the study protocol, research staff conducted re-peated on-site visits to monitor that the intervention was delivered in accord-ance with the protocol. Project meetings were held on five occasions. Further-more, issues regarding delivery were discussed in order to increase alignment on a regular basis. In order to make the adherence to the protocol easier, a checklist running week by week corresponding to the intervention protocol for each patient was used by the coaches.

Outcomes and data management All outcomes in Paper I (Table 2) were assessed at baseline (before randomi-sation) and immediately after the 6-month intervention period (100).

Primary outcome The primary outcome in Paper I was CRF, assessed with the Multidimensional Fatigue Inventory 20 (MFI). MFI is a scale validated and translated into Swe-dish (105,106). It consists of five subscales that measure different dimensions of fatigue: general fatigue, physical fatigue, reduced motivation, reduced ac-tivity, and mental fatigue. Patients answers 20 questions by indicating on a five-point Likert scale (ranging from yes, that is true to no, that is not true) the extent to which a particular statement applies. The statements refer to as-pects of fatigue experienced during the preceding few days and higher scores indicate a higher degree of fatigue (106). The subscales range from 4 to 20 and a change of 2 is considered minimally clinically relevant (107). i.e. the smallest amount an outcome must change to be meaningful to patients (108).

CRF was also measured with the Functional Assessment of Chronic Illness Therapy – Fatigue scale (FACIT-F), a single-scale questionnaire with scores ranging between 0 and 52. A higher score indicates a better outcome and a change of 3 is considered to minimally clinically relevant (109). FACIT-F is a reliable and validated instrument and was originally developed to assess can-cer-related fatigue. It is widely used in the cancer-affected population, as well as to assess fatigue in patients with other conditions (109,110).

32

Secondary outcomes HQoL was assessed using the European Organisation of Research and Treat-ment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30). This questionnaire consists of 30 questions and was developed for use in clin-ical trials to assess quality of life among cancer patients (111).

Anxiety and depression was assessed with the Hospital Anxiety and De-pression Scale (HADS), which is a reliable and validated instrument used in cancer populations (112). Function in daily life was assessed with the World Health Organization Disability Assessment Schedule (WHODAS) Swedish version (113).

All self-reported outcomes with composite scores were calculated in ac-cordance with published instructions (106,114,115).

The amount of physical activity, time spent sedentary, and time spent asleep were measured with the SenseWear® armband model MF-SW (Body Media®, Pittsburgh, Pennsylvania, United States of America). SenseWear® is an activity monitor that combines accelerometery with measurements of skin conductivity and heat production in order to measure total energy ex-penditure. Age, sex, weight, and height are entered into a computer program (proprietary algorithm) and used to estimate energy expenditure per minute. The participants wore the armbands for seven consecutive days. Only data for participants with 80% wear time for four out of seven days were included.

Cardiorespiratory fitness was measured with a test of maximal oxygen up-take (VO2max). Participants walked/ran until exhaustion on a treadmill. The test was performed using a modified Balke protocol which entailed that the treadmill inclination started two percent and the speed at four km/h. The in-clination increased by one percent each minute until reaching 12 percent in-clination, then the speed increased 0.5 km/h every minute. Two of the follow-ing criteria had to be fulfilled for data to be included: 1) the test was judged as maximal by testing staff, 2) the Borg rating of perceived exertion rating was ≥ 17, 3) the respiratory exchange ratio (i.e., carbon dioxide produced/ oxygen used) was ≥ 1.1.

Collection of data from each participant’s medical records was conducted based on a study-specific protocol. Data collected include type of cancer, on-cological treatment (including dose and dose adjustment), and treatment tox-icity. Information on tumour stage and surgical treatment (breast cancer and colorectal cancer) was gathered from the Swedish national quality registers for the respective cancer types.

Chemotherapy completion rates were calculated as mean relative dose in-tensity for the planned treatment based on the formula described by Longo et al. (116).

Background data relevant to the study such as education, tobacco and alco-hol use, and previous physical activity were self-reported and obtained through the study-specific questionnaires used at baseline.

33

Table 2. Overview and description of outcomes in Paper I.

Outcome Data source/measurement Analyses

CRF MFI FACIT-F

Descriptive, multiple linear regression, responder analy-sis (MFI physical fatigue subscale, FACIT-F)

HQoL EORTC QLQ-C30 summary score EORTC QLQ-C30 functional scales (physi-cal functioning, role functioning, emotional functioning, cognitive functioning, social functioning EORTC QLQ-C30 symptom scales (fatigue, nausea and vomiting, pain, dyspnoea, insom-nia, appetite loss, constipation, diarrhoea, fi-nancial difficulties

Descriptive, multiple linear regression, responder analy-sis (QLQ-C30 fatigue sub-scale)

Cardiovascu-lar fitness

VO2max Descriptive, multiple linear regression

Muscle strength

1 RM Descriptive, multiple linear regression

Sleep time, MVPA, sed-entary time

SW Descriptive, multiple linear regression

Chemother-apy comple-tion rates

RDI Descriptive, multiple linear regression

CRF = cancer-related fatigue. MFI = Multidimensional fatigue inventory, main outcome: subscale physical fatigue. FACIT-F = Functional Assessment of Chronic Illness Therapy – Fatigue. EORTC QLQ-C30= Eu-ropean Organisation of Research and Treatment of Cancer Quality of Life Questionnaire Core 30. VO2max = maximal oxygen uptake test. 1RM = one repetition maximum. RDI= relative dose intensity. MVPA = moderate to vigorous physical activity. SW = SenseWear.

Power calculation Based on previous research, the MFI physical fatigue subscale was used to conduct a power calculation of the main outcome (42,106). In order to detect a clinically and statistically significant difference of two points (SD 5) on the MFI physical fatigue subscale (107) with a 80% power at an alpha level of 0.05, a sample of 150 participants per randomisation group (total 600 partici-pants) was required. This calculation took account of interaction effects, as well as missing data due to drop-outs (100).

Data management and analysis In Paper I, the effects of exercise intensity and behaviour change support and their interactions (i.e. intensity x behaviour change support) were assessed us-ing multiple linear regression. Analyses were conducted according to inten-tion-to-treat and results were presented as adjusted mean differences with 95% confidence intervals (95% CI). Models included baseline measures of the out-come to increase precision and were adjusted for cancer diagnosis and site.

34

Multiple imputations with chained equations were used for missing data. The variables used for creating imputation values were age, education level, centre, diagnosis, chemotherapy, baseline values of outcome measure, and in-tervention group. If more than 10 percent of the data for a variable were miss-ing at baseline, it was not included as a supporting variable or as a variable in the main models, because the information would not add precision for the out-comes.

Additional responder analysis was used to present the proportion of patients whose change scores represented a minimum clinically important difference of 2 for the MFI subscale of physical fatigue (107), a minimum clinically im-portant difference of 3 for FACIT-F, or a minimum clinically important dif-ference of 6 for the QLQ-C30 fatigue subscale (117,118). All analyses were carried out in Stata version 15.0. No adjustments for multiple testing were made.

Paper II Participants Participants in Study II were included in the observational study (n = 90) or the RCT (n = 577). The observational study preceded the exercise trial and its recruitment ran from September 2014 to March 2015. The exercise trial started after enrolment in the observation study ended (March 2015) and recruitment stopped in May 2018.

The participants in the observational study (usual care) followed the same intervals in terms of performing physical activity testing, and answering ques-tionnaires as the participants in the intervention trial. Inclusion and exclusion criteria for Paper II were the same as in Paper I except that only participants with breast and colorectal cancer were included from the usual care group, as men with prostate cancer did not undergo chemotherapy and thus had no central venous access. Inclusion and exclusion criteria are presented in Appendix 1.

Breast cancer was the most common diagnosis in all groups (usual care: 84%, HI: 79%, LMI: 79%). There were fewer women with breast cancer who received chemotherapy in usual care (41%) than in HI (60%) or LMI (61%). Mean age was 58 years in usual care and 59 years in HI and LMI. (Appendix 4). In usual care group, there were more missing data (ranging from 0 to 69%, depending on the variable) than in HI or LMI (both ranging from 0 to 19%, depending on the variable) groups.

Outcomes See Table 3 for an overview and description of outcomes in Paper II.

35

Coach-reported adverse events Adverse events occurring during the intervention were registered, when they occurred by the coaches, in a study-specific checklist for each participant. Only events deemed by the coach to be caused by the exercise were reported. Each adverse event was graded based on whether the participant had to stop the specific exercise (grade 1) or the entire training session (grade 2). Descrip-tion of the symptom/injury was provided by coaches and coaches specified whether the adverse event was related to endurance or resistance training. All adverse events were summarised for each participant at the end of the exercise period. The adverse events were categorised as either musculoskeletal and connective tissue, accidents/injuries, cardiovascular symptoms/disorders, or other. Serious adverse events were defined as adverse events that required im-mediate hospital care and were reported to the principal investigator as soon as possible. Description of adverse events categories are presented in table 4.

Participant-reported adverse events Participant reports on exercise-related adverse events and PICC complications were gathered through a study-specific questionnaire at the end of the exercise period. The questions concerned injuries or discomfort experienced (no/yes) during the six-month exercise intervention. If any such had occurred, the par-ticipant was asked to specify the injury/discomfort. Participant-reported ad-verse events were categorised as either musculoskeletal and connective tissue, accidents/injuries, cardiovascular symptoms/disorders, or other. PICC-related complications were categorised as thrombosis, infection, dislocation, throm-bophlebitis, or other. Description of adverse events categories are presented in table 4.

Lymphoedema, PICC-related complications and new medical conditions Collection of data from participants’ medical records was conducted based on a study-specific protocol. Data collected included oncological treatment (background information), lymphoedema, PICC data (insertion and removal date, and complications), and new medical conditions that occurred during the intervention period. Description of adverse events categories are presented in table 4.

Background data relevant to the study such as education and tobacco and alcohol use were self-reported and obtained from the study-specific question-naires filled out at baseline. Baseline physical activity was measured with SenseWear® armband model MF-SW.

36

Table 3. Overview and description of outcomes in Paper II. Outcome Participants Data source Analysis

Exercise-related adverse events

HI, LMI Coach-reported Participant-re-ported

Descriptive, logistic regres-sion Descriptive, logistic regres-sion

PICC-related complica-tions

HI, LMI, UC HI, LMI

Medical record Participant-re-ported

Descriptive, Chi-2 test Descriptive

Lymphoedema HI, LMI, UC

Medical record Descriptive, Chi-2 test

New medical conditions HI, LMI Medical record Descriptive, Chi-2 test HI = high intensity groups, LMI = low-moderate intensity groups, UC = usual care. Low-moderate intensity vs. high-intensity group, exercise adherence, no chemotherapy vs. chemotherapy.

Table 4. Description of adverse events categories Paper II.

Description of coach- and participant-reported exercise-related adverse events Musculoskeletal and connective tis-sue

muscle strains, joint pain and muscular pain

Accidents/inju-ries

falling/tripping resulting in fractures, bruising, and swelling

Cardiovascular symptoms/disor-ders

feelings of lightheadedness, dizziness, fainting, and heart palpitations

Other fatigue, hernia, and eczema from heartrate monitor Description of PICC-related complications Medical record arm swelling, occlusion/no back flow, leakage, pain at the insertion

site, not being able to remove PICC, and too narrow blood vessel Participant-re-ported

feelings of discomfort in the arm with PICC inserted, PICC chafed dur-ing exercise, and discomfort in the arm during chest press

Description of new medical conditions Musculoskeletal and connective tis-sue

arthritis, swollen ankle, back pain, knee pain, impingement of shoulder, and herniated disc

Accidents/inju-ries

tripping/falling resulting in fractures, and spraining of ankle

Cardiovascular symptoms condi-tions

nose bleeding, atrial fibrillation, fainting, and chest pain during interval training. Atrial fibrillation was treated with anticoagulant medication and in one instance cardioversion therapy. Chest pain was further in-vestigated by physician without finding an underlying cause and the pa-tient continued exercising without experiencing pain.

Pulmonary symptoms/disor-ders

asthma, coughing, and shortness of breath

Other migraine, myomas, inguinal hernia, skin cancer, infections, psychiatric conditions and colon polyps, hyperlipidemia, endometriosis, and gastri-tis

37

Data analysis In Paper II, comparisons were made between participants randomised to HI and LMI, regardless of whether they received behaviour change support or not, because differences in adverse events due to behaviour change support were not expected.

In Paper II, logistic regression analyses were used to explore if any coach-reported adverse events (no = 0, yes = 1) or participant-reported adverse events (no = 0, yes = 1) were associated with exercise intensity (LMI = 0, HI = 1), chemotherapy (no = 0, yes = 1), or adherence to exercise (continuous variable). Odds ratios with 95% confidence intervals (CI) were calculated for variables included in regression analyses. Differences between usual care, HI, and LMI groups regarding the occurrence (no/yes) of lymphoedema and PICC complications were analysed with the chi-square test. Differences between HI and LMI groups regarding the inclusion (no/yes) of any new medical condi-tion in medical records were analysed with the chi-square test. The analyses in Paper II were performed in IBM SPSS Statistics version 25.

Paper III Participants The participants in Paper III were patients with breast, prostate, or colorectal cancer undergoing or about to start adjuvant cancer treatment. In order to in-clude patients with different experiences, patients on different treatments and at different stages of treatment were included, i.e., the inclusion was purpose-ful and not consecutive. The inclusion of participants ended when no new in-formation seemed to occur from the interviews.

During the period of recruitment to the Phys-Can feasibility study, 102 pa-tients were identified as suitable to invite. Of the 102 patients approached, 32 agreed to participate in the interview study. The most common reason to de-cline participation was having a long distance to travel to the university. Nine participants initially included could not participate in the interviews because none of the dates were suitable or because of illness.

In total, five focus group interviews were conducted. The focus groups con-sisted of ten women and eight men. Individual interviews were conducted with five women with breast cancer.

Procedure The patients were approached at a planned visit at the oncology clinic or on-cology day care unit at a university hospital in Sweden during autumn 2013

38

and spring 2014. The patients received verbal and written information regard-ing the study, and if they consented to participate, they were later contacted to book an appointment for the interview.

The recruitment of participants was continued until no new information was believed to be gained from further interviews. Due to the short time frame between recruitment and start of treatment, it was not feasible to gather a focus group for the participants about to start treatment; instead, the five participants about to start cancer treatment were interviewed individually.

The focus groups were conducted separately for women and men, and this was done in order to make the participants less reluctant to address sensitive topics. An experienced researcher conducted four of the focus group inter-views and all of the individual interviews. One of the focus groups was inter-viewed by two nursing students, as part of their student thesis. In order to prepare the nursing students for the task, they received tuition from the expe-rienced researcher conducting the other interviews. The interviews were con-ducted in a quiet and secluded room at the university. During the focus group interviews, AH was present and took notes. The interviews lasted for approx-imately one hour and were recorded with a digital audio recorder. The indi-vidual interviews lasted between 22 to 43 minutes.

Data collection The participants answered a questionnaire including questions regarding de-mographic information (gender, age, and education), their physical activity level in the previous six months, their level of motivation for being physically active during treatment, and their perceived ability to be physically active dur-ing treatment. Motivation and perceived ability to be physically active during treatment were measured using a 100 mm visual analogue scale. Information regarding diagnosis and oncological treatment was obtained from each partic-ipant’s medical journal.

The interviews were conducted with the aid of an interview guide. The in-terview guide was constructed with open-ended questions regarding how the participants experienced daily life, the disease, how treatment affected daily activities, and their perceived ability to be physically active and exercise dur-ing treatment. The main focus of the questions was physical activity and ex-ercise. The interview guide was not tested before the start of interviews; in-stead, it was pilot-tested on the first focus group. After the first interview, the guide was evaluated. The participants’ answers were deemed to cover the re-search aim and the participants seemed to understand the questions asked; therefore, no further adjustments of the interview guide were made.

39

Data analysis The audio recordings of the interviews were listened to several times and tran-scribed verbatim. The text from the transcriptions was analysed with qualita-tive content analysis. Qualitative content analysis is derived from quantitative content analysis, which was originally a method for analysing large quantities of text, for instance from newspapers, and therefore widely used in communi-cation and media research (119).

When conducting the analysis in Paper III an inductive approach was cho-sen. This means that the codes and categories (and later themes) created were determined during the analysis process, not in advance. This approach is used when participant responses are unknown. The opposite of this approach is de-ductive analysis, which means that the researcher uses a predetermined theo-retical framework or structure (120). The analysis was conducted with a low level of interpretation and the main focus was on the manifest content.

A qualitative content analysis involves several steps. The analysis has been explained and described by several researchers and, in Paper III, the descrip-tion made by Granheim and Lundman was used (121). The analysis is not a linear process; therefore, the steps are not always consecutive. Instead, the researcher may move forward and backward in the process several times be-fore finalising the analysis.

The first step of the analysis is to become familiar with the text. This is done by reading the transcripts several times. This is also beneficial in order to get at sense of the whole. After this first step of the analysis, content areas are identified and extracted from the text. A content area is a part of the text which covers a specific subject. In this case, the content area is the part of the transcribed text that correspond to the aim of the study, i.e., what the partici-pants say about physical activity and exercise during cancer treatment. After the content areas are identified, the next step in the analysis is to extract mean-ing units and condense the text to make it manageable, but without it losing its meaning. The condensed meaning units are then coded, and this is done to simplify identification of meaning units during development of categories. Categories and sub-categories were developed through several group discus-sions with all the article authors until a consensus was achieved.

Paper IV Participants Coaches from all three study sites who had provided exercise instruction to the participants of Phys-Can RCT were considered eligible to participate. However, to include participants with adequate experience and to reduce the risk of memory bias, coaches who had less than six months’ experience of working in the project or who had ended their employment more than six

40

months ago (n = 3) were excluded. Also, coaches who were researchers in the Phys-Can project were excluded (n = 2). During the spring of 2017, 13 coaches were asked to participate and received written information regarding the study. Of these, 11 chose to participate.

The included coaches were physiotherapists (n = 9) and personal trainers (n = 2) with different educational and professional backgrounds. Ten coaches were women and one was a man. Median age was 46 years (min–max 27–69). Four physiotherapists had previously supervised exercise for patients with cancer, with a median experience of 10 years (min–max 1–15). Six physio-therapists had experience of supervising exercise for patients with other diag-noses, with a median experience of four years (min–max 3 months–21 years). The personal trainers and three of the physiotherapists had a median experi-ence of 13 years (min–max 6–47 years) of supervising exercise in a public setting (e.g., public gym.) Also they had high outcome expectations regarding exercise for patients during oncological treatment as well as high self-efficacy regarding supervising exercise for persons undergoing oncological treatment (Table 5).

Table 5. Outcome expectations and supervision self-efficacy of participants Paper IV. Outcome expectations (scale 1–5) • How certain are you that exercise during oncological treatment is healthy in the long term?

Median (min–max) 5 (4–5)

• How certain are you that exercise during treatment can reduce side effects from treatment?

Median (min–max) 5 (4–5)

Supervision self-efficacy (scale 1–5) • How certain are you about your ability to supervise en-durance training and resistance training at a minimum of moderate intensity for persons undergoing oncological treatment?

Median (min–max) 5 (4–5)

Data collection Before the interviews, all participants signed a consent form, received verbal information, and were asked if they had any questions regarding the study. They also answered a questionnaire containing questions about age, educa-tion, profession, and previous experience of instructing exercise in various groups. Furthermore, participants also rated how confident they felt regarding the benefits of exercise during adjuvant cancer treatment and how confident they felt regarding their own ability to instruct exercise to persons with cancer undergoing cancer treatment. The rating was made on a Likert scale from 1, not confident, to 5, completely confident (Table 5).

Data analysis The analysis was conducted thematically based on the description of the anal-ysis made by Brown and Clarke (122). A thematic analysis is an appropriate

41

method to use in order to identify and analyse patterns across and within a dataset (122,123).

The analysis approach of this study was inductive, i.e., the analysis was derived from data and not from theory, which is an appropriate approach to use when the area of research is novel or sparsely explored (120).

The interviews were transcribed verbatim and before the start of coding, the first three interviews were test-coded by BJ, ID, and AH separately. Then, a discussion followed regarding the coding and chosen code areas, until the authors reached consensus. AH then coded the rest of the interviews using Nvivo Pro 11 software to organise the data. During the coding procedure, ideas for themes were written down and the codes with similar content were sorted into the initial themes. When the initial coding was completed, BJ, ID, and AH selected and developed the themes at three separate occasions, and AH made revisions between the meetings. When consensus on preliminary themes and sub-themes was achieved, the authors reviewed these together and decided on names for the themes and sub-themes.

42

Ethical considerations Eligible persons in Paper I–IV received written and verbal information regard-ing the respective study. If they decided to participate, they signed an informed consent form containing information regarding the study, how data would be presented, and their right to withdraw from the study at any time. Participation was voluntary and consent to participate could be withdrawn at any time. The Regional Ethical Review Board in Uppsala approved the Phys-Can feasibility study, i.e. Paper III (registration number 2013/248), the Phys-Can observation study, i.e., Paper II, and the Phys-Can RCT, i.e. Paper I (registration number 2014/249).

An application to the Regional Ethical Review Board was also made for Paper IV, but the Review Board responded that this was not required for the study. The audio recordings and transcripts were stored so that only authorised persons had access to them. Because the informants in this study were also co-workers of the authors in Phys-Can RCT, it was decided that the person inter-viewing should not have worked with the participating coach, and preferably not have had any or only very little previous contact with the person. Further-more, due to the nature of this study, with coaches being asked about the ex-perience of their work, they might feel obliged to participate, because of this work relationship. However, emphasis was placed on participation being vol-untary, in both written and verbal information to the participants.

All studies were conducted in accordance with Swedish legislation (124) and the ethical guidelines of the Declaration of Helsinki (125).

Physical activity and exercise are known to be beneficial for patients even during oncological treatment. All participants included in the RCT received an exercise intervention which could be beneficial for them. However, exer-cising may also include risks, thus all potential participants were assessed by a physician before they were deemed eligible. Furthermore, several precau-tions in order to minimise risk for the participants were instituted. The partic-ipants included in the observation study did not receive an exercise interven-tion, however they received usual care, which entails information regarding physical activity.

43

Results Paper I

Adherence to the intervention The adherence to the resistance training was similar across all intervention groups (mean 47.5–52.7%, p = 0.438). However, adherence to endurance training was higher in the LMI group with behaviour change support com-pared with the HI group with behaviour change support (13.7, 95%CI 3.0–24.5). There was a similar number of minutes of exercise performed at too high intensity (> 60% of heart rate reserve) in both LMI groups (p = 0.197). Cancer-related fatigue Participants exercising at HI had lower physical fatigue (MFI) than partici-pants exercising at LMI (adjusted mean difference -1.11 [95% CI -1.91 to -0.30, p value = 0.007]) No statistically significant differences were found for the other MFI dimensions or for FACIT-F. There were no statistically signif-icant differences between groups with or without additional behaviour change support and no interaction of exercise intensity and behaviour change support for any of the CRF outcomes (Table 6).

The results from the responder analysis showed that over half of the partic-ipants in the HI groups, with and without behaviour change support (56% and 55%, respectively), had a clinically relevant improvement on the MFI physical fatigue scale (Appendix 9) in comparison with the LMI groups (45% and 47%, respectively). However, the responder analysis of FACIT-F (Appendix 10) showed smaller differences of clinically relevant improvements ranging be-tween 34 and 33 percent in HI with and without behaviour change support, versus 42 and 35 percent in LMI with and without behaviour change support, respectively. EORTC QLQ-C30 fatigue subscale (Appendix 11) showed a similar result with 34 and 39 percent of participants in HI with and without behaviour change support, respectively, having a clinically significant im-provement of CRF, and 38 and 37 percent in LMI with and without behaviour change support, respectively.

44

Table 6. Main effects of exercise intensity, additional behaviour change support and interaction post-intervention CRF

Exercise inten-sity AMD (95%CI)

p- value

BCS AMD (95%CI)

p- value

Interaction AMD (95%CI)

p- value

MFI General Fatigue

-0.34 (-1.03 to 0.34)

0.328 -0.20 (-0.91 to 0.50)

0.570 0.19 (-0.53 to 0.91)

0.609

MFI Physical Fatigue

-1.11 (-1.91 to -0.30)

0.007 -0.50 (-1.29 to 0.28)

0.207 0.30 (-0.51 to 1.12)

0.469

MFI Reduced Activity

0.22 (-0.49 to 0.92)

0.543 -0.35 (-1.05 to 0.34)

0.316 -0.06 (-0.77 to 0.65)

0.874

MFI Reduced Motivation

0.06 (-0.52 to 0.64)

0.847 -0.26 (-0.84 to 0.32)

0.378 0.30 (-0.26 to 0.87)

0.291

MFI Mental Fatigue

-0.26 (-0.95 to 0.43)

0.459 -0.19 (-0.88 to 0.50)

0.585 0.36 (-0.32 to 1.05)

0.297

FACIT-F -0.44 (-1.87 to 0.99)

0.546 -0.12 (-1.62 to 1.38)

0.871 -0.57 (-2.00 to 0.85)

0.428

Included multiple imputation by chained equations to account for missing data, presented as adjusted mean difference and 95% confidence intervals (CI) (n=577). CRF: Cancer-related fatigue; MFI: Multidimen-sional Fatigue Inventory; FACIT: Functional Assessment of Chronic Illness Therapy. Bold indicates p-value<0.05.

Secondary outcomes Based on EORTC QLQ-C30 total score, no main effects of exercise intensity or behaviour change support on HQoL were found (Appendix 7). There was an interaction effect of behaviour change support indicating that the HI group receiving behaviour change support had a lower HQoL compared with the LMI group with behaviour change support. In contrast, among those not re-ceiving behaviour change support, the HI group had a higher HQoL than the LMI group (Appendix 7). There was an interaction indicating that HI exercise was associated with a lower quality of life, compared with LMI exercise, for participants receiving behaviour change support, and with a higher quality of life compared with LMI exercise among those who didn’t received behaviour change support (adjusted mean difference -3.04 [95% Confidence Interval -4.77, -1.31]). This was also the case in five of the EORTC subscales: i.e. Phys-ical functioning, Cognitive functioning, Nausea and vomiting, Pain, and In-somnia (Appendix 8).

Cardiorespiratory fitness was improved in HI in comparison with LMI (ad-justed mean difference 1.60 ml/kg/min [95% CI 0.12–3.07], p value 0.034). Furthermore, participants exercising at HI had greater leg strength in compar-ison with participants exercising at LMI (adjusted mean difference 3.98 kg [95% CI 1.65–7.30, p value = 0.019). There were no main interaction effects of additional behaviour change support on these outcomes (Appendix 5).

45

No effects of the intervention (main or interaction) were found on anxiety, depression (HADS), functioning in daily life (WHODAS), time in sleep, sed-entary behaviour, amount of moderate to vigorous physical activity, or chem-otherapy completion rates (Appendix 7).

Out of 298 participants who received chemotherapy, forty percent needed a dose reduction or discontinuation of treatment (Table 7), while no partici-pants required dose adjustment of radiotherapy. The most common reason for chemotherapy adjustments or discontinuation were myelosuppression and neuropathy (Table 8).

Table 7. Chemotherapy completion rates for participants with breast cancer and col-orectal cancer. HI BCS HI LMI BCS LMI

Breast cancer Mean relative dose intensity (mean [SD])**

92.4 (13.7) 93.3 (12.9) 95.4 (7.0) 94.7 (9.8)

Relative dose in-tensity <85% (n [%])***

13 (18.6) 8 (12.1) 5 (7.2) 8 (11.3)

Chemotherapy dose reduction/dis-continuation (n [%])

28 (40.0) 23 (34.8) 28 (40.6) 22 (31.0)

Colorectal cancer Mean relative dose intensity (mean [SD])**

59.8 (39.3) 84.9 (12.8) 73.2 (16.7) 68.0 (16.7)

Relative dose in-tensity <85% (n [%])***

3 (60.0) 2 (33.3) 3 (50.0) 4 (80.0)

Chemotherapy dose reduction/dis-continuation (n [%])

4 (80.0) 3 (50.0) 5 (83.3) 5 (100.0)

HI BCS= high-intensity with behaviour change support, HI= high-intensity without behaviour change sup-port, LMI BCS=Low-moderate intensity with behaviour change support, LMI=Low-moderate intensity.

Tabl

e 8.

Rea

son

for c

hem

othe

rapy

dos

e re

duct

ion/

disc

ontin

uatio

n

n (%

) B

C

CR

C

BC

C

RC

B

C

CR

C

BC

C

RC

M

yelo

supr

essi

on1

14 (5

0.0)

0 8

(34.

8)

1 (3

3.3)

12

(42.

9)

3 (6

0.0)

10

(45.

5)

0 In

fect

ion

5 (1

7.9)

1

(25.

0)

3 (1

3.0)

0

5 (1

7.9)

0

4 (1

8.2)

0

Nau

sea/

vom

iting

2(7

.1)

0 2

(7.1

) 2

(9.1

)Pa

in5

(17.

9)

1 (4

.3)

4 (1

4.3)

1

(4.5

)N

euro

path

y 4

(14.

3)

2 (5

0.0)

5

(21.

7)

3 (1

00.0

) 1

(3.6

) 2

(40.

0)

2 (9

.1)

0 C

onst

ipat

ion/

diar

rhoe

a3

(10.

7)

1 (2

5.0)

1

(4.3

) 0

2 (7

.1)

0 1

(4.5

) 1

(20.

0)

Car

diov

ascu

lar s

igns

/sym

ptom

s 3

(10.

7)

0 2

(7.1

) 2

(9.1

)H

and-

foot

synd

rom

e 4

(14.

3)

1 (2

5.0)

2

(8.7

) 1

(33.

3)

1 (3

.6)

0

1 (4

.5)

4 (8

0.0)

El

evat

ed li

ver e

nzym

es

2 (7

.1)

2 (8

.7)

4 (1

4.3)

2

(9.1

)O

ther

2 4

(14.

3)

2 (5

0.0)

3

(13.

0)

0

5 (1

7.9)

1

(20.

0)

6 (2

7.3)

1

(20.

0)

Unk

now

n

1 (3

.6)

5 (2

1.7)

3 (1

0.7)

1

(4.5

)H

I BC

S= h

igh-

inte

nsity

with

beh

avio

ur c

hang

e su

ppor

t, H

I= h

igh-

inte

nsity

with

out b

ehav

iour

cha

nge

supp

ort,

LMI B

CS=

Low

-mod

erat

e in

tens

ity w

ith b

ehav

iour

cha

nge

supp

ort,

LMI=

Low

-mod

erat

e in

tens

ity. B

C=

brea

st c

ance

r. C

RC

= c

olor

ecta

l can

cer.

Parti

cipa

nts c

ould

hav

e m

ore

than

one

reas

on. 1 =

Incl

udin

g fe

brile

neu

trope

nia,

low

blo

od c

ell c

ount

. 2 =

Incl

udin

g fa

tigue

, alle

rgic

reac

tions

such

as U

rtica

ria, a

nd re

duce

d ge

nera

l con

ditio

n.

HI B

CS

H

I

LMI B

CS

LM

I

47

Paper II

Coach- and participant-reported adverse events One hundred and eight (20%) of 526 participants (26% in HI and 15% in LMI) had at least one coach-reported adverse events, while 93 (22%) of 420 partic-ipants (29% in HI and 15% in LMI) reported at least one ADVERSE EVENT. The most common exercise-related adverse events was musculoskeletal for both resistance and endurance training (coach-reported 54 and 70%, and par-ticipant-reported 82 and 85%, respectively). adverse events were more com-mon during resistance training (coach-reported n = 115, participant-reported n = 105) than endurance training (coach-reported n = 23, participant-reported n = 50). Of the 108 participants with a coach-reported adverse event, 74 (68%) had a grade 1 adverse event. Thirty-one (29%) had a grade 2 adverse event. Three participants (3%) had grade 2 adverse events that were considered seri-ous.

The three serious adverse events occurred in the HI group during resistance training (n = 2) and endurance training (n = 1). Two women with breast cancer had an episode of syncope and were taken to hospital for observation. Both recovered swiftly. A possible reason for one of these syncope episodes was not eating before exercise. The reason for the episode in the other participant was unknown. After the event, the participant with a known cause for syncope continued the exercise intervention in accordance with the protocol. The par-ticipant with an unknown reason for syncope was in the exercise familiarisa-tion period and decided to withdraw from the study. The third participant with a serious adverse event was a man with prostate cancer who tripped over a rowing machine when attempting to sit down. The accident resulted in a frac-tured finger that had to be repositioned and sutured. The participant missed three resistance training sessions before he could continue exercising.

Coach-reported adverse events were two times more likely in HI than in LMI (OR: 2.01 [95% CI 1.26–3.20], p = 0.003), after adjustments for chemo-therapy and exercise adherence. Participant-reported adverse events were more than three times more common in HI than in LMI (OR: 3.30 [95% CI 2.06–5.28], p = 0.000) after these adjustments. Exercise adherence and chem-otherapy were not associated with exercise-related adverse events for either coach-reported (OR for exercise adherence: 0.996 [95% CI 0.987–1.005], p = 0.424, OR for chemotherapy: 0.810 [95% CI 0.517–1.271], p = 0.359) or par-ticipant-reported adverse events (OR for exercise adherence: 1.005 [95% CI 0.995–1.015], p = 0.367, OR for chemotherapy: 0.932 [95% CI 0.595–1.459], p = 0.758).

PICC complications reported by participants included two dislocations in the HI group, one venous thrombosis in the HI group, and seven participants (HI n = 2, LMI n = 5) who experienced discomfort in the inserted arm.

48

Lymphoedema, PICC-related complications, and new medical conditions Information regarding lymphoedema was available for 443 participants (usual care n=50, HI n = 198, LMI n = 198) and missing for 52 participants (usual care = 13, HI n = 15, LMI n = 18). The numbers of participants who had developed lymphoedema from the time of inclusion until the 6-month follow-up were two (4%) in usual care, 17 (9%) in HI, and 11 (6 %) in LMI. There was no statistically significant difference between HI, LMI, or usual care re-garding the incidence of lymphoedema (X2 (2, N = 443) = 2.03, p = 0.361).

In total 303 participant had a PICC (Appendix 4) and 51 participants had at least one PICC-related complication. Information regarding PICC-compli-cations were missing for 34 participants (usual care n = 5, HI n = 19, LMI n=10). Seven participants (23%) in usual care, 23 (17%) in HI, and 21(15 %) in LMI had a PICC-related complication. The most common PICC-related complication was other (34%), including arm swelling, occlusion/no back flow, leakage, pain at the insertion site, not being able to remove PICC, and too narrow blood vessels. There was no statistically significant difference be-tween HI, LMI, or usual care regarding the incidence of PICC complications (X2 (2, N = 269) = 2.20, p = 0.332).

Thirty-three (12%) in HI and 26 (10%) in LMI had a new other medical condition record. The most common category for new conditions that arose from the time of inclusion in the RCT until the 6-month follow-up was other (including migraine, myomas, inguinal hernia, skin cancer, infections, psychi-atric conditions, colon polyps, hyperlipidaemia, endometriosis, and gastritis). There was no statistically significant difference between groups (X2 (1, N = 475) = 1.27, p = 0.26).

49

Paper III The analysis of patient experiences of being physically active during cancer treatment resulted in three categories (Table 8).

Physical and emotional barriers The first category included descriptions of treatment-related side effects, co-morbid conditions, and emotional distress that prevented physical activity. For instance side effects like fatigue, bowel and bladder related symptoms, pain and infections could prevent physical activity. Also co-morbid conditions like osteoarthritis and unstable diabetes or atrial fibrillation could also be a barrier. Furthermore, feelings of listlessness were also described as a barrier that could hinder patients from staying active.

Perspectives and attitudes This category described how self-efficacy, self-image, exercise preferences, concerns, and expectations regarding physical activity, previous experiences of exercise, and new perspectives regarding one’s health (teachable moments) influenced motivation and ability to engage in physical activity. For instance, participants described that they did not believe that they could exercise or not viewing themselves as persons who exercise. Also, feeling awkward about being in a changing room because of the loss of hair or scarring could prevent them from going to a gym.

The participants expressed different preferences regarding what type of ex-ercise they wanted to engage in, and also some wanted to exercise in groups, and some preferred to exercise alone. Furthermore, having concerns regarding the side effects of treatment could be a barrier because they were uncertain regarding what side effects they would experience. In contrast, other partici-pants had positive expectations of physical activity; for instance, that exercise could improve strength and promote one’s health.

Other facilitating factors were having positive previous experience of phys-ical activity and changing priorities in their life with a focus on themselves and their health.

Support and practicalities The third category included descriptions of how overprotective family and friends could prevent physical activity by being too considerate (inhibiting care). This category also included descriptions of the importance of surround-ing factors like access to training facilities, not having time to exercise due to work, the need to receive information and support from health care staff (mo-tivational support) and how the participants used different strategies to be able to engage in physical activity.

For example, the participants expressed high confidence in the information that health care staff provided, and receiving information regarding physical

50

activity helped increase their motivation to be physically active. Participants that did not receive information about physical activity expressed a wish for this and that they would like to have information about what is safe to do dur-ing treatment. Furthermore, participants expressed a desire to receive profes-sional support to be physically active, which would make them feel safe and more motivated.

Participants also described different strategies to be physically active, for instance, setting goals, doing things they enjoy, and making the most of the periods during treatment they felt at their best, as well as using strategies to maintain energy levels throughout the day was regarded as helpful.

Table 8. Overview of categories and sub-categories Paper III Physical and emotional barriers Perspectives and attitudes Support and practicalities

The side effects of treatment Self-efficacy and self-image Inhibiting care Co-morbidity Exercise preferences Environmental support and

practical issues Emotional distress

Concerns and expectations Motivational support

Experience of physical activity

Strategies

Teachable moments

51

Paper IV

Results The analysis of the coaches’ experience of supervising exercise resulted in two overarching themes: Supervising exercise to persons undergoing onco-logical treatment is a meaningful task and a challenging task, with four and five subthemes, respectively.

Supervising exercise to persons undergoing oncology treatment is a meaningful task The sub-theme, making an important difference, is a description regarding how the coaches felt that they made a difference and contributed to the partic-ipants' cancer care. They felt appreciated, and the work felt meaningful on a personal and professional level. Supervising exercise in a public gym contrib-uted further to the positive experience because the coaches felt like they be-came a link between health care and health promotion.

Linking health care and health promotion is a sub-theme including experi-ences of that the setting (i.e., being in a public gym) and exercise helped both the participants and themselves to shift their focus from illness to health. Coaches described the exercise as something the participants could do for themselves to feel better. Supervising exercise at a public gym felt like being a link between oncology care and health promotion, but the coaches expressed that experience from oncology care was necessary. This experience could be either from their own practice of treating patients as a physical therapist at the hospital or from having a colleague in the coaching team with this experience.

More skills and confidence is a sub-theme regarding that the coaches ini-tially could feel hesitant because they thought that it might be difficult to ex-ercise during oncological treatment, and this made them reluctant to push the participants. As the coaches gained more experience, it became more comfort-able for them to push the participants, because they had seen the positive ef-fects of exercise and become skilled in identifying who they could push and how. Furthermore, coaches experienced that participants who were new to ex-ercise could often underestimate their ability to exercise. In such cases, they tried to encourage and support the participants to try even if the participants felt hesitant. However, this was done while respecting their autonomy and not being too insistent, because if participants were pushed too much, it might make them lose interest. Having previous experience of supervising exercise for cancer survivors facilitated for the coaches, while not having such experi-ence could make them feel unprepared for the task. Also, having previous ex-perience of supervising exercise in a gym was perceived as an advantage be-cause this helped make the coaches who were physical therapists feel com-fortable with the environment. Furthermore, supervising group exercise was

52

considered as educational, because the coaches could learn how to organize exercise sessions and speak in front of a group. They also learned about group leadership and became more skilled in guiding participants.

Long-term interaction enables individualized supervision is a sub-theme about that supervising exercise regularly over a long period made it possible for the coaches to get to know the participants. This made it possible to learn more about their physical and psychological capacities and also to gain their trust, which facilitated individualized supervision. However, getting to know a participant required that he or she participated in the exercise regularly. The coaches who were physical therapists reflected over the fact that they, in their clinical work at the hospital, usually did not have this opportunity to meet patients this regularly and over such a long period.

Supervising exercise to persons undergoing oncology treatment is a challenging task The sub-theme facing severe illness was a description of the emotional chal-lenges supervising exercise for persons with cancer entailed. For instance, if a participant had a relapse of disease. These challenges that they experienced were handled by talking to their colleagues, and on one occasion, coaches re-ceived support from an external counsellor to handle a difficult situation.

Further challenges that the coaches experienced was handling the barriers to exercise that the participants had. For instance, side effects from oncologi-cal treatment could cause participants to give up. They also experienced that participants sometimes held back because of fear or discomfort during exer-cise, and it could be challenging to know if the participants were experiencing actual pain or just the feeling of physical exertion. The coaches believed that making the participant feel safe were necessary for their ability to exercise, and the coaches provided safety by giving accurate information and being at-tentive to their concerns. Also, having knowledge about cancer and side effect related to treatment were considered important to create security.

If questions regarding medical issues arose that the coaches did not know how to answer, a nurse or physician was contacted. However, getting in touch with these health professionals was sometimes complicated, and the coaches felt that it would have been valuable to have someone (e.g., a designated nurse or physician) they could contact directly with their questions. Furthermore, the coaches experienced that the home-based endurance training was more challenging for the participants to succeed with, especially for those in the HI groups. Besides, addressing barriers such as not having enough time or not prioritizing exercise, and lack of motivation to exercise was considered diffi-cult to solve. However, providing feedback from follow-up tests was per-ceived as encouraging to the participants, but they could also be disappointed if their results did not improve.

The sub-theme contradictory information addressed the concern that health care staff had on occasions given outdated or incorrect information regarding

53

the safety of exercise during cancer treatment to the participants. For instance, when the participants received information from nurses, physicians or physi-cal therapists, that exercise was not advised or that they should avoid public areas during chemotherapy because of the risk of infections the participants became worried and reluctant to engage in exercise. It was difficult for the coaches to address this, as they wanted to avoid giving the participants con-tradictory information.

The sub-theme group dynamics described the challenges of relating to sev-eral people at once and making sure that new participants felt welcomed and included in the group. It could be difficult if participants did not get on with others in the group or if they had different needs concerning talking about their illness and situation. Although peer support between participants was per-ceived as valuable by coaches, if participants had too much to talk about, it could also be challenging to get them to do the exercise. Also, it could be difficult to handle competitiveness between participants, which could be prob-lematic if it became extreme.

The last sub-theme adhering to the study protocol described how it was sometimes difficult when participants needed adjustments to their exercises. The coaches could experience that there was a conflict between their respon-sibility to follow the protocol and a participant's need for adjustments. Another challenge was not to provide extra BCTs in the groups that were not random-ized for this, especially to participants who the coached believed needed them. Furthermore, handling time pressure and knowing what to prioritize in the study protocol could, at times, be challenging. In such instances, being more than one coach in a group was an advantage because it created a feeling of security and provided opportunities to discuss problems and share the respon-sibility.

54

Discussion

The most important findings Paper I–IV The results from Paper I showed that high intensity exercise yielded statisti-cally significantly lower physical fatigue compared with LMI exercise in pa-tients undergoing (neo-)adjuvant treatment. However, the effect did not reach the pre-set levels of a minimal clinically important difference (100). Also, no differences between groups were found regarding other CRF dimensions. Fur-thermore, there were no effects of additional behaviour change support on any outcome. There was a statistically significant difference in muscle strength and cardiorespiratory fitness, favouring HI exercise. However, the difference in cardiorespiratory fitness was small and of unknown clinical relevance.

The results from Paper II indicates that most adverse events were minor and of musculoskeletal origin. There was a higher risk to have a minor adverse event when exercising at HI in comparison with exercising at LMI. No in-creased risk of PICC-related complications or lymphoedema when exercising on HI or LMI was seen in comparison with usual care. No association of new medical conditions when exercising at HI was seen in comparison with at LMI.

The results from Paper III indicates that patients experienced physical and emotional barriers, such as side effects from treatment and feelings of listless-ness, preventing physical activity. Receiving information regarding physical activity from health care staff helped increase their motivation to be physically active. Participants that did not receive information about physical activity expressed a wish that they would like to have information about what is safe to do during treatment and that receiving professional support would make them feel safe and more motivated.

The results from Paper IV indicates that coaches experienced supervising exercise for persons undergoing oncological treatment as a “meaningful task” and a “challenging task.” They felt they made an important difference, became a link between health care, but were challenged in managing participants’ bar-riers to exercise, and that participants received contradictory information about exercise from health care staff.

55

Effects of exercise intensity and behaviour change support on treatment-related side effects Exercise intensity did not seem to have a clinical impact in reducing CRF. Previous research conducted by van Waarts et al. found a statistically signifi-cant improvement on physical fatigue, that did not reach clinical relevance at the group level, which is in line with the findings of Paper I (107). Further-more, two RCTs investigating effects of exercise intensity on CRF after end of treatment did not find any additional benefits from exercise at HI in com-parison with exercise at LMI (44,45).

Interestingly, the additional descriptive responder analysis regarding phys-ical fatigue indicated higher clinically relevant improvements in both HI groups in comparison with LMI groups, but this was not the case for FACIT-F or the EORTC QLQ-C30 fatigue subscale. These results were consistent with the regression analysis, with a statistically significant difference for MFI physical fatigue, and no statistically significant differences between groups for FACIT-F or the EORTC QLQ-C30 fatigue subscale. The reason for these different results is unknown; however it could be due to that MFI physical fatigue subscale is more sensitive to changes in physical fatigue than the other fatigue instruments.

The lack of effects from behaviour change support was contrary to what was hypothesised. One suggested reason for this is that all participants re-ceived support from coaches which included some BCTs, such as self-moni-toring of endurance training, follow-up, and feedback (from heartrate moni-tors and muscle strength test). All these are BCTs suggested to improve ad-herence to exercise interventions in cancer survivors (65). Furthermore, group-based exercise, which may entail unstructured peer support, as provided during resistance training, could be considered a form of specific support (126). In order to maintain fidelity to the exercise protocol, this extensive sup-port was necessary in the groups not receiving behaviour change support and could be considered a trade-off between investigating the effects of exercise intensity and investigating the effects of behaviour change support. Further, participants included in a six-month exercise trial are probably motivated to exercise; future research should include a broader and more heterogenous pop-ulation to determine if there are any benefits from behaviour change support in this population. Furthermore, since behaviour change support entails learn-ing behaviour change techniques, i.e., the individual learns how to set goals and how to realise these, investigating if there are any long-term effects of this support is essential.

Furthermore, the association between high intensity exercise and lower HQoL in the group that had behaviour change support was also surprising. However, the group differences were small and not of clinical relevance (118). Furthermore, these findings were not supported by the results from qualitative interviews and questionnaire data conducted by Mazzoni et al. (127), where

56

the participants in Phys-Can describe that they perceived the behaviour change support as helpful.

Muscle strength increased more in the HI groups in comparison with the LMI groups. This result is in line with previous research (128), indicating that higher loads (i.e., amount of weight lifted) increases maximal strength in ex-ercise, even when controlled for exercise volume, also during oncological treatment. However, previous research has also found that untrained individ-uals may benefit from lower loads of 45 to 50 percent of one repetition maxi-mum, which should be considered if patients are not able to exercise at higher loads (128).

There were statistically significant differences regarding cardiorespiratory fitness between groups, but these were small, and could be due to similar ex-ercise volume in all groups. This is in line with a meta-analysis concluding that exercise volume may be more important in improving respiratory fitness than exercise intensity alone (129).

No other secondary outcomes differed between groups. Based on these re-sults, with small improvements in the main outcome CRF, HI may not be more beneficial than LMI for reducing or preventing fatigue. Therefore, because it is well-known that exercise during treatment is beneficial for numerous health outcomes (130), patients can be advised to exercise at either intensity.

None of the participants (breast and prostate cancer) had a dose adjustment regarding the radiation therapy. Chemotherapy completion rates were high for participants with breast cancer and were moderate for participants with colo-rectal cancer in all randomisation groups, which is in line with previous re-search (131,132).

Adherence to the intervention In Paper I, the exercise adherence was lower than described in other exercise interventions for cancer survivors, however different methods for calculating and defining adherence makes it difficult to make comparisons. (133,134). These adherence numbers may well reflect what is feasible when undergoing (neo-)adjuvant oncology treatment, if measured as percentage of prescribed exercise.

Adherence is important for understanding the feasibility and effect of an intervention. One suggested way to improve attendance is to implement peri-odisation of the exercise, i.e., it being decided beforehand that intensity will be lowered when side effects from treatment are expected (135).

Another aspect that is interesting to consider is that the intervention was six months long (with four exercise sessions/week) and a meta-analysis found that exercise interventions longer than 12 weeks do not have the same effects on cancer-related fatigue as shorter interventions (136). Why this is the case is not known; it might be a matter of adherence.

57

Safety of exercise during oncological treatment Approximately twenty percent of the participants in Paper II had an exercise-related adverse event reported either by coaches or themselves. The most com-mon adverse event was musculoskeletal, and adverse events were more com-mon during resistance training than during endurance training.

The results indicate that exercising in HI groups was associated with higher risk of receiving an exercise-related adverse events than exercising in LMI groups, while receiving chemotherapy or exercise adherence were not.

Not finding an association between adherence and risk was surprising, be-cause one might expect to find an increased risk of adverse events due to in-creased exposure to exercise. However, previous epidemiological research re-garding healthy individuals has indicated that sedentary individuals occasion-ally engaging in high intensity exercise are at higher risk of sustaining an in-jury than individuals habitually engaging in exercise. The risk is also increased for individuals habitually engaging in high intensity exercise who increases their exercise time (79). Thus, adherence alone may not capture the complex-ity of risk for adverse events, and additional factors, such as previous exercise experience, previous injuries, and exercise intensity, may be more important.

The incidence of exercise-related (coach- and participant-reported) adverse events was higher in this study than previously reported in exercise trials in-cluding patients with cancer. This may be because of the structured gathering of data and inclusion of minor adverse events (75). When comparing these results to epidemiological research conducted in healthy individuals, a similar incidence of exercise-related musculoskeletal injuries is described. Minor ad-verse events may occur in individuals with and without cancer, making it im-portant to report these in a structured way in exercise trials, whether they in-clude cancer survivors or not.

In Paper II, data were gathered from three different sources (coaches, par-ticipants, and medical records) and the results indicated that serious adverse events were uncommon, which previous research has also reported (73–75). Most of the adverse events were minor, transient, and did not prevent further exercise participation.

The incidence rates of lymphoedema were similar in the exercise groups and the usual care group. However, it should be recognised that lymphoedema can develop later than six months after surgery, meaning that a longer follow-up period is necessary in order to draw more certain conclusions. As an exam-ple, one study found that in a pre-post measurement after 14 months of heavy load exercise intervention, more than 27 percent of the women with breast cancer was diagnosed with lymphoedema (137).

The incidence of PICC-related complications was comparable or lower than that previously described in research (138–141). Less than ten percent had a PICC-related thromboembolism; this could be compared with a Swedish RCT in which patients with breast or colorectal cancer were randomised to

58

either subcutaneous infusion port or PICC. Eight percent of the participants randomised to PICC were diagnosed with a deep vein thrombosis (78). While it does not seem like exercising with a PICC increased risk for complications further, research is needed to strengthen this conclusion.

Exercising in HI groups was not associated with higher incidence of new medical conditions in the medical records. The incidence of adverse events reported in the medical record was lower than those of adverse events reported by coaches or by the participants themselves. One reason for this is that infor-mation included in the medical record is only of clinical relevance and minor, transient adverse events may not be recorded.

The results from Paper II indicate that exercise during chemotherapy does not increase the risk of having an adverse event in this sample. However, on-cological treatments change, and it is unknown how the results from Paper II translate to other patient groups receiving different types of chemotherapy. Therefore, it is important to continually evaluate the safety and risks of exer-cise when introducing new oncological treatments (73).

To enable comparison between studies, it is important that intervention characteristics, such as preventive strategies to reduce injuries (e.g., warm-up, progression, etc.), are described in detail. Also, when exercise interventions are conducted during oncological treatment, it is important to include infor-mation on treatment completion (75).

Experiences

Patient experiences The results from Paper III suggest that several factors influence patients’ abil-ities to be physically active and exercise during treatment.

Examples of barriers mentioned were side effects from treatment and hav-ing co-morbid conditions. Several of the side-effects described as barriers by participants, for instance CRF, could be prevented or reduced by physical ac-tivity and exercise (37). This indicates that introducing physical activity and exercise to reduce side effects from treatment should be started as soon as possible after diagnosis. Furthermore, the participants experienced fluctua-tions regarding the severity of side effects during treatment, which should be considered when helping patients plan their physical activity, i.e., take ad-vantage of the better periods during treatment (142).

Another way to facilitate physical activity for patients during treatment is by considering practical issues. For instance, when choosing what type of in-travenous access to use, a patient’s preferred physical activity (e.g., swim-ming) should be considered.

Furthermore, the results from Paper III indicate that some patients may feel reluctant to go to a training facility (e.g. a public gym) because they feel self-

59

conscious about scarring or side effects of treatment (e.g. loss of hair). Offer-ing these patients somewhere to exercise in a relaxed atmosphere with other patients in the same situation might be helpful in overcoming this barrier. Other practical aspects, like patients wishing to work during treatment or not having access to training facilities, should be considered when helping pa-tients plan exercise.

Though undergoing oncological treatment is challenging, findings from this study suggest that patients may be susceptible to lifestyle interventions during this period, and that taking advantage of ‘teachable moments’ is im-portant. Therefore, early exercise interventions, even before starting treat-ment, may be feasible for most patients with cancer.

Self-care activities (i.e., what individuals do to maintain or improve their own health and well-being) may depend on the individual understanding why and how an activity is to be performed. Low health literacy (i.e., low ability to obtain and act on information regarding health) (143) may co-exist with other barriers and improving health literacy will not necessarily affect other important factors, like self-efficacy and motivation to exercise. However, providing information that is easy for the patient to understand and asking about physical activity and exercise habits should be the first step in compre-hensive support, including follow-up, prescribing physical activity, and refer-ring patients to professionals who can provide the support needed (144).

The results from Paper III suggest that information about the benefits and safety of physical activity for patients with cancer is crucial for motivating and facilitating physical activity during treatment. Not receiving information regarding physical activity during treatment may cause patients to feel unsure about what they are allowed and able to do. Inhibition of care from friends and family may be avoided by providing adequate information. Patients should be informed of current recommendations regarding physical activity and that ex-ercise is an important component in the management of their health-related quality of life and side effects from treatment (145). Moreover, one study re-garding information about physical activity provided by oncologists suggests that information about physical activity during cancer treatment is optimised if a motivational aspect is included (146).

Participants expressed a need for professional support to help them be ac-tive during this period, saying that this would help their motivation and make physical activity feel safer, which is in line with a review of qualitative re-search including women with breast cancer (147). However, what this support should include, especially for long-term maintenance of physical activity is not yet determined (46). According to previous research, the most salient mod-ifiable predictors are readiness and intention to change, self-efficacy for phys-ical activity, and perceived behavioural control (64,148–151). These are im-portant aspects in behaviour change support targeting physical activity behav-iour.

60

Using BCTs comprising both motivational and self-regulatory components and functional behaviour analysis could help reduce barriers such as low self-efficacy and help patients manage pre-existing uncertainties regarding physi-cal activity. Kampshoff et al. (151) identified exercise history as a determinant of exercise adherence and maintenance in cancer patients. While exercise his-tory is a non-modifiable factor, it is an important aspect to ask about, since previous experience of physical activity and exercise could affect a person’s self-efficacy. This highlights the need to explore a patient’s exercise history in order to identify negative experiences and low self-efficacy. For instance, Husebø et al. (150) suggest that supporting patients’ self-efficacy for changing health behaviour and considering the impact of treatment-related side effects may reduce perceived barriers to physical activity in women with breast can-cer.

Coach experiences The results from Paper IV indicate that professionals supervising exercise for persons receiving oncological treatment is meaningful but also challenging. The coaches became a link between oncology care and health promotion and felt that they made a difference in the participants’ lives. The results also in-dicated that the coaches grew more confident in their professional role and more convinced that exercising during treatment was feasible. The challenges that the coaches encountered included managing participants’ barriers to ex-ercise, such as treatment side effects, and that participants received contradic-tory information from oncology care staff, advising them not to exercise.

Lack of consistent support could cause reluctance in patients to engage in exercise, and handling this issue was not easy for the coaches. Therefore, ef-forts should be made to increase the knowledge and support so that oncology care staff can communicate a consistent message to patients regarding the ben-efits and safety of exercise. Support could consist of information updates to oncology care staff and an individualised care plans with information, goal-setting, planning, and follow-up for the patients (152).

The results also indicate that practical experience is essential for the coaches to feel confident in that it is feasible and safe to exercise during treat-ment. This finding is in line with previous research, including physiotherapists using BCTs within a physical activity intervention for patients with rheuma-toid arthritis. It was reported that education and support were necessary for the physiotherapist to transition into the role of a coach (157). Furthermore, it is important to help coaches develop strategies to cope with the emotional challenges that may arise when supervising exercise for patients with cancer. One coping strategy that may be useful is talking to colleagues about difficult situations. Exercising with peers may be a way for cancer survivors to gain a sense of normality and a place where they can meet others in the same situa-tion (72,147). Paper IV found that the coaches could benefit as well, because they gained skills in supervising exercise, organising exercise session, and

61

talking to groups. Supervised exercise groups for cancer survivors can be meaningful for both patients and professionals and are possibly a more cost-effective way to organise exercise in cancer rehabilitation than individual ses-sions.

The results from Paper IV suggest that having a regular contact, over a long period of time, was helpful for the coaches to get to know the participants, and thus tailor the support after each participant’s needs. This finding may shed additional light on why supervised exercise is more effective in improving outcomes for cancer survivors than unsupervised exercise (97). On the other hand, the coaches sometimes felt limited in adapting the support to the needs of each participant because they needed to adhere to the study protocol. It may be that cancer survivors that experience severe side effects would benefit from a non-study setting, where the support can be completely tailored to each in-dividual’s preferences and needs.

To provide this extensive support within the financial boundaries of a health care system may not be feasible (153). Therefore, finding a model where different actors within and outside the health care system cooperate to create physical activity programmes for cancer survivors may be a more suit-able solution. For instance, oncology nurses and physicians could refer per-sons undergoing oncological treatment to various exercise programmes or fur-ther assessment by a physical therapist based on each patient’s specific needs (145).

In line with previous research (72), barriers like lack of time and not prior-itising exercise were difficult for the coaches to address. Simply providing advice may not be sufficient. One way to address this could be to use commu-nication strategies inspired by motivational interviewing (154). This approach takes its starting point in exploring the patient’s previous experiences and preferences, guiding and supporting in the choice of suitable activities, and helping the patient set motivating and feasible exercise goals. This approach also takes the patient’s readiness to make a change into account. For instance, if a patient is not ready to start exercising, it may be appropriate to ask for permission to offer some information regarding the benefits of physical activ-ity and allow the patient time to consider this. If a patient is uncertain, it may be helpful to perform a decisional balance exercise, where the pros and cons of a change are listed and talked about (155). Training in using these tools is already incorporated in several educational programmes for health care pro-fessionals, such as physiotherapists, but may require further implementation.

Methodological considerations This thesis includes both quantitative and qualitative methods and several methodological considerations need to be addressed.

62

Generalisability and transferability Generalisability and transferability are two key concepts when discussing for whom the results are valid. Generalisability is commonly used in quantitative studies, while transferability is common in qualitative studies. However, they both address issues with external validity, i.e., was the sampling and setting representative of the population that the research programme intended to study?

The Phys-Can RCT is one of the largest exercise interventions studies con-ducted to date. It encompassed 577 randomised participants and included par-ticipants with the three most common types of cancer. However, the inclusion rate was low in comparison with other exercise trials including patients re-ceiving oncological treatment. This may be due to participants in the Phys-Can RCT being willing, or at least having agreed to be randomised to either exercise at low-moderate or high intensity as well as exercising four times per week (two sessions of resistance training and two sessions of endurance train-ing) for a period of six months.

A common procedure in research is to use exclusion criteria to reduce risks to participants (38). Therefore, fewer adverse events are expected in RCTs than in the clinical setting, and the generalisability of results (Paper II) to pa-tients with more severe co-morbidity is unknown. Also, the participants re-ceived supervised exercise, which may have contributed to a low incidence of serious adverse events. Furthermore, it is unknown how results from this the-sis translate to other patient groups with other types of cancers, receiving other types of treatments, or with more advanced disease. Furthermore, the sample included in the RCT (Paper I and II) consisted of relatively healthy individuals and the majority were well-educated women with breast cancer.

We aimed to include patients with colorectal cancer in Phys-Can. However, few such participants were included in the RCT (Paper I and II), the observa-tional (Paper II), and the feasibility study (Paper III). Thus, the generalisability and transferability of the results from this patient group may be questioned. The reason for not including more participants with colorectal cancer was that fewer individuals with colorectal cancer who received adjuvant treatment were deemed eligible by physicians than patients with breast or prostate can-cer. Secondly, fewer among the eligible patients were willing to participate. Apart from these considerations regarding inclusion of patients with colorectal cancer, the study sample in Paper III consisted of patients of various ages, socioeconomic statuses and both those meeting and those not meeting the guidelines for recommended physical activity, improving the transferability of these results.

In Paper IV, most of the coaches were physiotherapists; only two coaches were personal trainers. It is therefore unknown how the results may be trans-ferred to this important group of professionals. Furthermore, the number of

63

potential informants in Paper IV was limited and it is difficult to assess if sat-uration was reached. It is unknown if the coaches who did not participate could have contributed with additional information. However, the information pro-vided by the interviewed coaches were rich and repetitive. As Paper IV was conducted within an intervention study, it should also be acknowledged that some of the results may not be transferred to exercise professionals working in non-study settings, furthermore the coaches had high outcome expectations and high self-efficacy in supervising exercise for persons undergoing onco-logical treatment (Table 5).

Study design and data collection A major strength with Paper I was that it was a large RCT with the same ex-ercise volume in all groups, enabling a comparison of exercise intensity.

Also, objective gold standard methods were used to assess cardiorespira-tory fitness and muscle strength. Furthermore, intervention delivery was closely monitored, to ensure fidelity to exercise and behaviour change support protocol.

A strength of Paper II was the systematic and structured collection of coach-reported adverse events and that data were collected from multiple sources, i.e., also from participants and medical records. This increased the probability of capturing both minor and more serious adverse events and dif-ferent aspects of adverse events. However, the study was based on secondary data and may have been underpowered to detect differences between groups. Furthermore, the usual care group was not randomised and had a much smaller sample size than the groups in the RCT. Also, there was a lot of missing data, especially from the medical records for participants in usual care. This should be considered when interpreting the results. For instance, the usual care group had very low incidence of lymphoedema, and one possible explanation for this may be missing data.

The methods of data collection in Paper III included both focus group in-terviews and individual interviews. In Paper IV, all interviews were individ-ual. Focus groups are conducted because the group interaction between par-ticipants can improve data collection. However, this strategy requires an ac-tive moderator if the participants does not know each other. A disadvantage of conducting focus group interviews is that participants may feel reluctant to address intimate concerns (156) and, therefore, important information about possible problems may be lost and some people may choose not to participate in the study. To minimise this problem in Paper III, women and men were interviewed in separate groups. Furthermore, it was not feasible to assemble focus groups when interviewing patients before they started adjuvant treat-ment. Instead, individual interviews were conducted with these participants and any benefits from group interactions where therefore lost.

64

In Paper IV, the coaches were interviewed individually and by multiple interviewers. This was a pragmatic solution, and even though an interview guide was used, this could affect dependability negatively because different interviewers may have different interviewing skills and focus on different as-pects during the interview (157).

Reflexivity

“It is the researcher who has to make sense of the data that have been collected by exploring and interpreting them.” Burnard et al., 2008 (120)

Since the researcher more or less is the “instrument” used to conduct the anal-ysis in qualitative studies, previous knowledge and experience, i.e., precon-ception, will influence the research process. Within qualitative research, the concept of reflexivity, i.e., the importance of reflecting on and being aware of the researcher’s own preconceptions, is crucial for establishing rigour (158). For instance, knowledge and experience regarding a subject may be helpful in order to identify new areas of research or research question. However, when conducting an analysis, preconceptions may be too influential, resulting in that the researcher misses or ignores important information that is not in line with previous assumptions, or exaggerates the information that is.

In Paper III, the researchers involved were three nurses, one physiothera-pist, and one psychologist, and the group’s expertise included experience of cancer care, behavioural medicine, and exercise medicine. All researchers but one (AH) had previous experience of conducting qualitative research, such as interview studies. In Paper IV, the researchers consisted of three nurses, two physiotherapists, and a psychologist. Both physiotherapists have been coaches to cancer patients in the Phys-Can intervention study and one of the nurses (AH) has also been working as a substitute coach on a few occasions. One of the senior researchers had developed and taught the coaches in a study-spe-cific course regarding exercise during oncology treatment and behaviour med-icine support.

65

Conclusions

This thesis investigated the effects of exercise intensity and behaviour change support on cancer-related fatigue, explored the safety of exercise during on-cological treatment, and explored the experiences of patients and profession-als in relation to exercise during oncological treatment. The conclusions are:

• Exercise at HI may not provide clinically relevant improvements in CRF in comparison with exercise at LMI in patients undergoing oncological treatment. Furthermore, in a motivated and relatively healthy sample receiving supervised and group-based exercise, ad-ditional behaviour change support is not likely to improve CRF.

• Exercise-related adverse events in persons undergoing oncological

treatment are usually minor, transient, of musculoskeletal origin, and with a similar incidence as in healthy populations. However, a higher risk of minor exercise-related adverse events was reported in HI exercise groups than in LMI groups.

• Patients can experience side effects from oncological treatment as a barrier to engage in physical activity. However, according to pre-vious research, side effects can be alleviated by physical activity. Therefore, engaging in physical activity before the onset of side effects from treatment may be beneficial for patients. Information about physical activity should be given early after diagnosis, as a part of standard care.

• Professionals supervising exercise for persons undergoing oncol-

ogy treatment in a community-based setting may find it highly re-warding, which is promising for implementation in cancer rehabil-itation. However, patients may still receive contradictory infor-mation regarding the safety of exercise from health care staff, which can be difficult for exercise professionals to counteract.

66

Implications and future research

The results from the four papers included in this thesis have several implica-tions both clinical care and cancer rehabilitation in a community-based setting. Firstly, HI exercise may yield small improvements in physical fatigue, but the improvements may not be of clinical relevance. Therefore, patients can be in-formed that exercise at any intensity could be beneficial. Furthermore, there may be a higher risk of having a minor injury when exercising at HI, which should be taken into account when prescribing exercise for patients. However, the incidence of serious adverse events was low and most adverse events were minor, meaning that exercise can be considered safe for patients during cura-tive oncological treatment. Consequently, this should be the main message given to patients. Furthermore, is imperative that future oncology exercise re-search includes safety assessments in a structured and comprehensive fashion and that a consensus regarding what to report is reached.

Furthermore, it is important to talk to patients about physical activity and encourage them to be active. Patients want information regarding physical ac-tivity and what is safe for them to do; providing this support early after diag-nosis may be beneficial. However, providing more behaviour change support to motivated patients engaging in supervised group-based exercise does not seem to provide any additional benefits. Future studies are needed to evaluate behaviour change support in broader clinical populations, for instance among patients not receiving supervised and group-based exercise.

Also, it is important to recognise that knowledge regarding safety and ben-efits of exercise may vary between clinicians meeting patients. Therefore, it may be necessary to address this issue with improved routines and communi-cation, to avoid giving patients contradictory information. More research is needed to find efficient ways to improve the frequency of physical activity communication between health care staff and patients.

If exercise during oncological treatment is to be performed in community-based settings, such as public gyms, personal trainers (or other exercise spe-cialists) who are not experienced with oncology care will need relevant edu-cation in cancer and oncological treatment. It may also be necessary to imple-ment communication routines between the exercise professionals supervising exercise and health care staff.

67

Svensk sammanfattning

Cancer är en folkhälsosjukdom med betydande inverkan på den enskildes liv och hälsa. Ungefär var tredje person i Sverige kommer att få en cancerdiagnos under sin livstid. Moderna cancerbehandlingar och tidigt upptäckt av cancer genom screening (tex. mammografi) har gjort att fler individer överlever sin cancer, men cancerbehandlingarna kan ge biverkningar som kan ha negativa effekter på individens hälsa både på kort och lång sikt. Till exempel rapporte-rar canceröverlevande långtidseffekter av sjukdomen och dess behandling, såsom cancerrelaterad trötthet, försämrad kondition och minskad hälsorelate-rad livskvalité. Cancerrelaterad trötthet definieras som en oroande, ihållande känsla av fysisk, emotionell och/ eller kognitiv trötthet eller utmattning som inte står i proportion till den aktivitet som individen utfört den senaste tiden och som stör den vanliga funktionen. Mellan 30-60% av patienterna upplever måttlig till svår cancer relaterad trötthet. Det är därför nödvändigt att finna och tillhandahålla effektiva behandlingar som kan motverka trötthet.

Historiskt sett har patienter fått råd att vila under behandlingen, men forsk-ning som gjordes under 90- och 00-talet har ifrågasatt detta då man sett att fysisk aktivitet och träning kan minska och ibland till och med förhindra några av de problem som personer med cancer upplever.

Tidigare forskning har visat att fysisk träning, det vill säga fysisk aktivitet som syftar till att upprätthålla eller förbättra fysisk funktion eller kondition, kan minska cancerrelaterad trötthet och förbättra konditionen hos individer även under pågående cancerbehandling. Det saknas däremot kunskap kring om det spelar någon roll hur hårt man tränar (dvs träningsintensiteten) för att minska tröttheten.

Att träna under cancerbehandlingen kan vara en utmaning och patienter med cancer är i genomsnitt mindre fysiskt aktiva än den allmänna befolk-ningen, vilket tyder på att finns hinder som förvårar för patienter att vara fy-siskt aktiva och träna. För att stödja patienter att övervinna sådana hinder kan strategier för beteendeförändring såsom självmonitorering (tex registrera sin träning) och målsättning användas. Men om dessa strategier kan påverka can-cerrelaterad trötthet genom att underlätta träning är oklart.

Tidigare forskning har visat att det är ovanligt med allvarliga komplikat-ioner när man tränar under cancerbehandling. Däremot har risker med att träna under onkologisk behandling inte systematiskt utvärderats och ibland har denna information helt saknats i publicerade studier. Dessutom är det inte

68

klarlagt om det finns särskilda risker med att träna på en hög intensitet under cancerbehandling.

Tidigare forskning visar att träning som leds av tex fysioterapeuter eller personliga tränare kan vara mer effektiv än träning som man gör på egen hand för att förbättra resultaten för patienter som genomgår behandling. Dock har inte erfarenheterna från fysioterapeuter och personliga tränare som ger detta stöd studerats tidigare och kunskap om deras upplevelser kan vara viktigt för framtida utveckling av cancerrehabilitering.

Det övergripande syftet med avhandlingen var att undersöka effekterna av träningsintensitet och stöd för förändring av beteende, om det finns några ris-ker med att träna, samt upplevelser av träning under cancerbehandling för både patienter och coacher.

Metod Phys-Can (Phyical training and Cancer; Fysisk träning och cancer) är ett forskningsprogram som genomförs i tre svenska städer; Linköping, Lund och Uppsala. Phys-Can består av tre studier: en genomförbarhetsstudie, en obser-vationsstudie och en randomiserad kontrollerad studie (RCT). Data från de tre studierna utgör grunden för denna avhandling.

I slutet av 2013 inleddes Phys-Can-forskningsprogram med en genomför-barhetsstudie (delarbete III) i Uppsala. Syftet med denna studie var bland an-nat att identifiera potentiella hindrande och underlättande faktorer för fysisk aktivitet hos patienter som genomgår cancerbehandling. Personer som genom-gick behandling mot bröst-, prostata- eller kolorektalcancer intervjuades om sina upplevelser av fysisk aktivitet under pågående behandling.

Phys-Can-observationsstudie (delarbete II) och RCT (delarbete I och II) genomfördes i Linköping, Lund och Uppsala. Observationsstudien syftade till att beskriva hur sjukdomarna och behandlingarna påverkade det fysiska till-ståndet, psykiskt välbefinnande och hälsorelaterad livskvalité hos personer som får adjuvant cancerbehandling. Deltagarna i observationsstudien fick ingen träningsintervention, och fungerade som en historisk jämförelsegrupp för dem som senare ingick i Phys-Can RCT. Inkluderingen till observations-studien inleddes i september 2014 och slutade 2015, innan RCT inleddes.

Phys-Can RCT (delarbete I och II) var en studie som syftade till att utvär-dera effekterna av träning med låg till måttlig intensitet (LMI) jämfört med hög intensitet (HI), med eller utan extra stöd för beteendeförändring, på can-cerrelaterad trötthet, för personer som genomgick onkologisk behandling. Dessutom utvärderades effekterna av träning på hälsorelaterad livskvalité, kondition, muskelstyrka och andra bieffekter från behandlingen. Inkludering av studiedeltagare till Phys-Can RCT började omedelbart efter avslutad inklu-dering till observationsstudien i mars 2015 och avslutades i april 2018.

69

Resultat Resultaten från delarbete I visar att träning på hög intensitet ger en liten för-bättring av fysisk trötthet i jämförelse med träning på låg-måttlig intensitet hos patienter som genomgår behandling. Dock var skillnaden inte så stor att den var kliniskt betydelsefull. Det fanns inga andra skillnader mellan grup-perna avseende cancerrelaterad trötthet. Det utökade stödet för beteendeför-ändringar förbättrade inte cancerrelaterad trötthet.

Resultaten från delarbete II visade att träningsrelaterade skador hos perso-ner som genomgår onkologisk behandling oftast inte är allvarliga utan består av främst exempelvis muskelsträckningar och vrickningar och förekomsten är likartad som hos friska. Emellertid rapporterades en högre risk för mindre all-varliga träningsrelaterade skador i de grupper som tränade på hög intensitet i jämförelse med grupperna som tränade på låg-måttlig intensitet. Allvarligare händelser var sällsynta varför det verkar säkert att träna även på hög intensitet för dessa patientgrupper.

Resultaten från delarbete III visade att patienter kan uppleva biverkningar från cancerbehandling såsom extrem trötthet, smärta eller tarmrelaterade bi-verkningar och osäkerhet om hur de kommer att må under behandlingen som hinder för fysisk aktivitet. Deltagarna utryckte att det var hjälpsamt att få in-formation från sjukvårdspersonal om fysisk aktivitet, då detta gjorde dem mer motiverade att röra på sig.

Resultaten från delarbete IV visade att fysioterapeuter och personliga trä-nare upplever det som mycket givande att instruera träning för personer som genomgår cancerbehandling, vilket är lovande för implementering av träning som en del av cancerrehabiliteringen. Patienter under pågående behandling kan dock fortfarande få motstridig information om huruvida det är säkert att träna eller inte från hälso- och sjukvårdspersonal.

Slutsatser Högintensivträning kan ge små förbättringar i trötthet, muskelstyrka och kon-dition jämfört med låg-måttligt intensiv träning, men förbättringarna verkar inte vara av klinisk betydelse när det kommer till cancer relaterad trötthet och kondition. Därför kan patienter informeras om att träning på både låg-måttlig och hög intensitet kan vara fördelaktigt och de bör informeras om att de kan träna på den intensitet de föredrar.

Att ge mer stöd till beteendeförändring till motiverade patienter som deltar i gruppbaserad träning som leds av en coach verkar dock inte ge några extra fördelar. Framtida studier behövs för att utvärdera stöd för beteendeföränd-ringar i bredare kliniska populationer, till exempel bland patienter som inte får coachning och gruppbaserad träning.

70

Det finnas en högre risk för att få mindre skador såsom tex muskelsträck-ningar om man tränar på hög intensitet i jämförelse med träning på låg-måttlig intensitet, vilket bör tas med i övervägandet när man förskriver träning för patienter. Förekomsten av allvarliga händelser eller skador var dock låg, vilket innebär att träning, även på hög intensitet, kan anses vara säker för den här patientgruppen och därför bör detta vara det huvudsakliga budskapet som ges till patienter.

Vidare är det viktigt att prata med patienter om fysisk aktivitet och upp-muntra dem att röra på sig och träna. Patienter vill ha information om fysisk aktivitet och vad som är säkert för dem att göra; att ge detta stöd tidigt efter diagnos kan vara fördelaktigt.

Det är också viktigt att uppmärksamma att kunskap om säkerhet och förde-lar med träning kan variera mellan hälso- och sjukvårdspersonal som möter patienter. Därför kan det vara nödvändigt att ta itu med detta problem med förbättrade rutiner, kommunikation och utbildning, för att undvika att ge pati-enter motstridig information. Mer forskning behövs för att hitta effektiva sätt att förbättra kommunikation om fysisk aktivitet mellan vårdpersonal och pa-tienter.

Om träning under onkologisk behandling ska utföras utanför sjukhusen, såsom på allmänna gym, kommer personliga tränare (eller andra träningsspe-cialister) som inte har erfarenhet av cancervård, behöva relevant utbildning i cancer och onkologisk behandling. Det kan också vara nödvändigt att imple-mentera rutiner för kommunikation mellan träningspersonal och sjukvårdsper-sonal.

71

Tack

Det finns mycket att vara tacksam för och många människor att tacka för det stöd och den kunskap ni givit mig under min tid som doktorand.

Jag vill börja med att tacka alla personer som deltagit i Phys-Can, för det är naturligtvis så att utan att ni ställt upp så blir det ingen forskning och vården kan inte utvecklas.

Jag vill särskilt tacka min huvudhandledare Karin Nordin för den här möj-ligheten och din uppmuntran och ditt stöd. Vidare vill jag även tacka mina biträdande handledare Birgitta Johansson, Helena Igelström och Calin Radu för all uppmuntran, vägledning och hjälpsamma kommentarer.

Ingrid Demmelmaier, medförfattare och biträdande forskargruppsledare vill jag också särskilt tacka för att du styr upp och alltid är så peppande.

Jag vill också rikta ett stort tack till mina fantastiska doktorandkollegor (före detta och nuvarande), Anne-Sophie Mazzoni, Helena Tegler, Anna-Ka-rin Ax, Anna Johnsson, Ann Christin Helgesen Björke, Mona Pettersson, Ma-rie Kirsebom, Jennifer Viberg Johansson, Helena Wandin, Petronella Bjur-ling-Sjöberg, Åsa Andersén, Marie Höjer Lundh, Susanne Hellerstedt Börjes-son, Stina Isacsson, Ritva Rissanen, Annika Terner, Jenny Stern, Maria Gran-dahl, Maria Ander Gottvall, Susanne Mattsson, Elenore Kaminsky, Tove Godskesen, Camilla Andersson, Sölvi Vejby, Marina Forslund och Anna Hauffman, för att ni lärt mig en massa saker och fått mig skratta även under riktigt gråmulna dagar.

Till min forskargrupp (nuvarande och före detta medlemmar), ovan nämnda samt Hannah Brooke, Emelie Strandberg, Cecilia Arving, Karianne Vassbakk Brovold, Maria Carlsson, Christopher Bean, Erik Berglund, Ingrid Anderzén, Jenny Halvardsson, Noora Berg och Annika Lundquist, vill jag bara säga att jag är väldigt glad och tacksam att jag fått ingå i denna fantastiska grupp. Jag vill även tacka alla som ingått i Phys-Can projektgrupp, tidigare nämnda, samt Sveinung Berntsen Stölevik, Ronnie Pingel, Pernilla Åsenlöf, Katarina Sjövall, Maria Hellbom och Truls Raastad, att få möjlighet att lära av er och delta i ett multidisciplinärt projekt som detta har varit fantastiskt! Dessutom vill jag tacka alla som på ett eller annat sätt arbetat med träningen och testningen i Phys-Can, alla coacher, all personal på Friskis och Svettis, Roger Olsson och Mia Berglund på ”labbet”.

Även ett varmt tack till alla mina lärarkollegor och administrationen som är helt avgörande för att institutionen för folkhälso- och vårdvetenskap ska fungera.

72

Jag vill även tacka läkare och sjuksköterskor samt annan personal på onko-logmottagningen, onkologisk dagvårdsavdelning och bröstmottagningen (Sa-mariterhemmet) vid Akademiska sjukhuset i Uppsala. Er hjälp och upp-muntran när jag informerat och tillfrågat patienter har varit ovärderlig. Jag vill passa på att särskilt tacka kontaktsjuksköterskorna Ann Eriksson, Elisabet Ax-elson och Jenny Furubrand vid onkologmottagningen på Akademiska sjukhu-set samt Lena Asp och Barbro Jung på Samariterhemmet för all hjälp under den här tiden.

Jag vill också tacka mina fina vänner, Kalle, Erica och Elin för att ni finns och alltid är så stöttande!

Tills sist men inte minst vill jag tacka min familj, mamma och pappa för att ni lärt mig vikten av arbetsglädje och att ni alltid låtit mig gå min egen väg, min bror Örjan för att du har vett att ifrågasätta allt och min syster Katarina som jag inte vet vad jag skulle göra utan ditt “ground support” och humor.

73

References

1. The National Board of Health and Welfare. Statistics on Cancer Incidence 2018 [Internet]. 2019 Dec. Report No.: Art.no: 2019-12-6525. Available from: https://www.socialstyrelsen.se/globalassets/sharepoint-dokument/artikelkatalog/ statistik/2019-12-6525.pdf

2. Gegechkori N, Haines L, Lin JJ. Long-Term and Latent Side Effects of Specific Cancer Types. Med Clin North Am. 2017 Nov;101(6):1053–73.

3. Schmitz KH, Courneya KS, Matthews C, Demark-Wahnefried W, GalvãO DA, Pinto BM, et al. American College of Sports Medicine Roundtable on Exercise Guidelines for Cancer Survivors: Med Sci Sports Exerc. 2010 Jul;42(7):1409–26.

4. Campbell KL, Winters-Stone KM, Wiskemann J, May AM, Schwartz AL, Cour-neya KS, et al. Exercise Guidelines for Cancer Survivors: Consensus Statement from International Multidisciplinary Roundtable. Med Sci Sports Exerc. 2019 Nov;51(11):2375–90.

5. Mctiernan A, Friedenreich CM, Katzmarzyk PT, Powell KE, Macko R, Buchner D, et al. Physical Activity in Cancer Prevention and Survival: A Systematic Re-view. Med Sci Sports Exerc. 2019 Jun;51(6):1252–61.

6. Hajdu SI, Darvishian F. A note from history: Landmarks in history of cancer, part 5: Landmarks in History of Cancer, Part 5. Cancer. 2013 Apr 15;119(8):1450–66.

7. Cardoso F, Kyriakides S, Ohno S, Penault-Llorca F, Poortmans P, Rubio IT, et al. Early breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treat-ment and follow-up. Ann Oncol. 2019 Aug;30(8):1194–220.

8. Harbeck N, Gnant M. Breast cancer. The Lancet. 2017 Mar;389(10074):1134–50.

9. Bower JE. Cancer-related fatigue—mechanisms, risk factors, and treatments. Nat Rev Clin Oncol. 2014 Aug 12;11(10):597–609.

10. Bower JE. The role of neuro immune interactions in cancer related fatigue: Bi-obehavioral risk factors and mechanisms. Cancer. 2019 Feb;125(3):353–64.

11. Ahlberg K, Ekman T, Gaston-Johansson F, Mock V. Assessment and manage-ment of cancer-related fatigue in adults. The Lancet. 2003 Aug;362(9384):640–50.

12. Peel AB, Thomas SM, Dittus K, Jones LW, Lakoski SG. Cardiorespiratory Fit-ness in Breast Cancer Patients: A Call for Normative Values. J Am Heart Assoc [Internet]. 2014 Jan 27 [cited 2020 Mar 22];3(1). Available from: https://www.ahajournals.org/doi/10.1161/JAHA.113.000432

13. Buffart LM, Ros WJG, Chinapaw MJM, Brug J, Knol DL, Korstjens I, et al. Me-diators of physical exercise for improvement in cancer survivors’ quality of life: Mediators of exercise in cancer. Psychooncology. 2014 Mar;23(3):330–8.

14. Bozzetti F. Forcing the vicious circle: sarcopenia increases toxicity, decreases response to chemotherapy and worsens with chemotherapy. Ann Oncol. 2017 Sep;28(9):2107–18.

74

15. Krebber AMH, Buffart LM, Kleijn G, Riepma IC, de Bree R, Leemans CR, et al. Prevalence of depression in cancer patients: a meta-analysis of diagnostic inter-views and self-report instruments: Prevalence of depression in cancer patients. Psychooncology. 2014 Feb;23(2):121–30.

16. Mitchell AJ, Chan M, Bhatti H, Halton M, Grassi L, Johansen C, et al. Prevalence of depression, anxiety, and adjustment disorder in oncological, haematological, and palliative-care settings: a meta-analysis of 94 interview-based studies. Lancet Oncol. 2011 Feb;12(2):160–74.

17. Irwin MR. Depression and Insomnia in Cancer: Prevalence, Risk Factors, and Effects on Cancer Outcomes. Curr Psychiatry Rep [Internet]. 2013 Nov [cited 2018 Apr 11];15(11). Available from: http://link.springer.com/10.1007/s11920-013-0404-1

18. Costa AR, Fontes F, Pereira S, Gonçalves M, Azevedo A, Lunet N. Impact of breast cancer treatments on sleep disturbances – A systematic review. The Breast. 2014 Dec;23(6):697–709.

19. Savard J, Ivers H, Villa J, Caplette-Gingras A, Morin CM. Natural Course of Insomnia Comorbid With Cancer: An 18-Month Longitudinal Study. J Clin On-col. 2011 Sep 10;29(26):3580–6.

20. Litwin MS, Tan H-J. The Diagnosis and Treatment of Prostate Cancer: A Review. JAMA. 2017 Jun 27;317(24):2532.

21. Seretny M, Currie GL, Sena ES, Ramnarine S, Grant R, MacLeod MR, et al. Incidence, prevalence, and predictors of chemotherapy-induced peripheral neu-ropathy: A systematic review and meta-analysis: Pain. 2014 Dec;155(12):2461–70.

22. Rivera DR, Ganz PA, Weyrich MS, Bandos H, Melnikow J. Chemotherapy-As-sociated Peripheral Neuropathy in Patients With Early-Stage Breast Cancer: A Systematic Review. JNCI J Natl Cancer Inst [Internet]. 2018 Feb [cited 2017 Dec 4];110(2). Available from: http://academic.oup.com/jnci/article/doi/10.1093/ jnci/djx140/4093778/ChemotherapyAssociated-Peripheral-Neuropathy-in

23. Beckwée D, Leysen L, Meuwis K, Adriaenssens N. Prevalence of aromatase in-hibitor-induced arthralgia in breast cancer: a systematic review and meta-analy-sis. Support Care Cancer. 2017 May;25(5):1673–86.

24. Escobar Y, Cajaraville G, Virizuela JA, Álvarez R, Muñoz A, Olariaga O, et al. Incidence of chemotherapy-induced nausea and vomiting with moderately emeto-genic chemotherapy: ADVICE (Actual Data of Vomiting Incidence by Chemo-therapy Evaluation) study. Support Care Cancer. 2015 Sep;23(9):2833–40.

25. Trotti A, Colevas A, Setser A, Rusch V, Jaques D, Budach V, et al. CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol. 2003 Jul;13(3):176–81.

26. Hellbom M, Bergelt C, Bergenmar M, Gijsen B, Loge JH, Rautalathi M, et al. Cancer rehabilitation: A Nordic and European perspective. Acta Oncol. 2011 Feb;50(2):179–86.

27. Felicetti F, Fortunati N, Brignardello E. Cancer survivors: An expanding popula-tion with an increased cardiometabolic risk. Diabetes Res Clin Pract [Internet]. 2018 Feb [cited 2018 Mar 19]; Available from: http://linkinghub.elsevier.com/re-trieve/pii/S0168822717318399

28. Ho PJ, Hartman M, Gernaat SAM, Cook AR, Lee SC, Hupkens L, et al. Associ-ations between workability and patient-reported physical, psychological and so-cial outcomes in breast cancer survivors: a cross-sectional study. Support Care Cancer [Internet]. 2018 Mar 6 [cited 2018 Mar 18]; Available from: http://link.springer.com/10.1007/s00520-018-4132-2

75

29. Lee MK, Yun YH. Working situation of cancer survivors versus the general pop-ulation. J Cancer Surviv. 2015 Jun;9(2):349–60.

30. World Health Organization, World Bank, editors. World report on disability. Ge-neva, Switzerland: World Health Organization; 2011. 325 p.

31. Nordic Cancer Union. From needs to offers Rehabilitation of cancer patients [In-ternet]. Available from: http://www.ncu.nu/Default.aspx?ID=21

32. Ansvarigt Regionalt cancercentrum: Regionalt cancercentrum Syd. Cancerreha-bilitering. Nationellt vårdprogram. Regionala cancercentrum i samverkan; 2019.

33. Sverige, Utredningen en nationell cancerstrategi. En nationell cancerstrategi för framtiden: betänkande. Stockholm: Fritze; 2009.

34. Caspersen CJ, Powell KE, Christenson GM. Physical activity, exercise, and phys-ical fitness: definitions and distinctions for health-related research. Public Health Rep Wash DC 1974. 1985 Apr;100(2):126–31.

35. Segal R, Zwaal C, Green E, Tomasone JR, Loblaw A, Petrella T, et al. Exercise for people with cancer: a clinical practice guideline. Curr Oncol. 2017 Feb 28;24(1):40.

36. Segal R, Zwaal C, Green E, Tomasone JR, Loblaw A, Petrella T, et al. Exercise for people with cancer: a systematic review. Curr Oncol Tor Ont. 2017 Aug;24(4):e290–315.

37. Mustian KM, Alfano CM, Heckler C, Kleckner AS, Kleckner IR, Leach CR, et al. Comparison of Pharmaceutical, Psychological, and Exercise Treatments for Cancer-Related Fatigue: A Meta-analysis. JAMA Oncol. 2017 Jul 1;3(7):961.

38. Stout NL, Baima J, Swisher AK, Winters-Stone KM, Welsh J. A Systematic Re-view of Exercise Systematic Reviews in the Cancer Literature (2005-2017). PM&R. 2017 Sep;9(9):S347–84.

39. van Rooijen SJ, Engelen MA, Scheede-Bergdahl C, Carli F, Roumen RMH, Slooter GD, et al. Systematic review of exercise training in colorectal cancer pa-tients during treatment. Scand J Med Sci Sports. 2018 Feb;28(2):360–70.

40. Courneya KS, McKenzie DC, Mackey JR, Gelmon K, Friedenreich CM, Yasui Y, et al. Effects of Exercise Dose and Type During Breast Cancer Chemotherapy: Multicenter Randomized Trial. JNCI J Natl Cancer Inst. 2013 Dec 4;105(23):1821–32.

41. Kleckner IR, Kamen C, Gewandter JS, Mohile NA, Heckler CE, Culakova E, et al. Effects of exercise during chemotherapy on chemotherapy-induced peripheral neuropathy: a multicenter, randomized controlled trial. Support Care Cancer. 2018 Apr;26(4):1019–28.

42. van Waart H, Stuiver MM, van Harten WH, Geleijn E, Kieffer JM, Buffart LM, et al. Effect of Low-Intensity Physical Activity and Moderate- to High-Intensity Physical Exercise During Adjuvant Chemotherapy on Physical Fitness, Fatigue, and Chemotherapy Completion Rates: Results of the PACES Randomized Clini-cal Trial. J Clin Oncol [Internet]. 2015 Apr 27 [cited 2015 May 26]; Available from: http://jco.ascopubs.org/cgi/doi/10.1200/JCO.2014.59.1081

43. Kapur G, Windsor PM, Mc Cowan C. The effect of aerobic exercise on treatment-related acute toxicity in men receiving radical external beam radiotherapy for lo-calised prostate cancer: Effect of aerobic exercise on radiation-induced acute tox-icity in prostate cancer. Eur J Cancer Care (Engl). 2010 Sep;19(5):643–7.

44. Kampshoff CS, Chinapaw MJM, Brug J, Twisk JWR, Schep G, Nijziel MR, et al. Randomized controlled trial of the effects of high intensity and low-to-mod-erate intensity exercise on physical fitness and fatigue in cancer survivors: results of the Resistance and Endurance exercise After ChemoTherapy (REACT) study. BMC Med. 2015 Dec;13(1):275.

76

45. Schmitt J, Lindner N, Reuss-Borst M, Holmberg H-C, Sperlich B. A 3-week mul-timodal intervention involving high-intensity interval training in female cancer survivors: a randomized controlled trial. Physiol Rep. 2016 Feb;4(3):e12693.

46. Courneya KS, Rogers LQ, Campbell KL, Vallance JK, Friedenreich CM. Top 10 Research Questions Related to Physical Activity and Cancer Survivorship. Res Q Exerc Sport. 2015 Jan 28;1–10.

47. Gray R, Bradley R, Braybrooke J, Liu Z, Peto R, Davies L, et al. Increasing the dose intensity of chemotherapy by more frequent administration or sequential scheduling: a patient-level meta-analysis of 37 298 women with early breast can-cer in 26 randomised trials. The Lancet. 2019 Apr;393(10179):1440–52.

48. Chirivella I, Bermejo B, Insa A, Pérez-Fidalgo A, Magro A, Rosello S, et al. Op-timal delivery of anthracycline-based chemotherapy in the adjuvant setting im-proves outcome of breast cancer patients. Breast Cancer Res Treat. 2009 Apr;114(3):479–84.

49. Zhang L, Yu Q, Wu X-C, Hsieh M-C, Loch M, Chen VW, et al. Impact of chem-otherapy relative dose intensity on cause-specific and overall survival for stage I–III breast cancer: ER+/PR+, HER2- vs. triple-negative. Breast Cancer Res Treat [Internet]. 2018 Jan 24 [cited 2018 Feb 24]; Available from: http://link.springer.com/10.1007/s10549-017-4646-1

50. Lyman GH, Dale DC, Tomita D, Whittaker S, Crawford J. A retrospective eval-uation of chemotherapy dose intensity and supportive care for early-stage breast cancer in a curative setting. Breast Cancer Res Treat. 2013 Jun;139(3):863–72.

51. Weycker D, Barron R, Edelsberg J, Kartashov A, Lyman GH. Incidence of re-duced chemotherapy relative dose intensity among women with early stage breast cancer in US clinical practice. Breast Cancer Res Treat. 2012 May;133(1):301–10.

52. Courneya KS, Segal RJ, Mackey JR, Gelmon K, Reid RD, Friedenreich CM, et al. Effects of Aerobic and Resistance Exercise in Breast Cancer Patients Receiv-ing Adjuvant Chemotherapy: A Multicenter Randomized Controlled Trial. J Clin Oncol. 2007 Oct;25(28):4396–404.

53. Bland KA, Zadravec K, Landry T, Weller S, Meyers L, Campbell KL. Impact of exercise on chemotherapy completion rate: A systematic review of the evidence and recommendations for future exercise oncology research. Crit Rev Oncol He-matol. 2019 Apr;136:79–85.

54. Wang Z, McLoone P, Morrison DS. Diet, exercise, obesity, smoking and alcohol consumption in cancer survivors and the general population: a comparative study of 16 282 individuals. Br J Cancer. 2015 Feb;112(3):572–5.

55. Huy C, Schmidt ME, Vrieling A, Chang-Claude J, Steindorf K. Physical activity in a German breast cancer patient cohort: One-year trends and characteristics as-sociated with change in activity level. Eur J Cancer. 2012 Feb;48(3):297–304.

56. Littman AJ, Tang M-T, Rossing MA. Longitudinal study of recreational physical activity in breast cancer survivors. J Cancer Surviv. 2010 Jun;4(2):119–27.

57. Galvão DA, Newton RU, Gardiner RA, Girgis A, Lepore SJ, Stiller A, et al. Com-pliance to exercise-oncology guidelines in prostate cancer survivors and associa-tions with psychological distress, unmet supportive care needs, and quality of life: Prostate cancer survivors and exercise. Psychooncology. 2015 Oct;24(10):1241–9.

58. Shi Z, Rundle A, Genkinger JM, Cheung YK, Ergas IJ, Roh JM, et al. Distinct trajectories of moderate to vigorous physical activity and sedentary behavior fol-lowing a breast cancer diagnosis: the Pathways Study. J Cancer Surviv [Internet]. 2020 Mar 4 [cited 2020 Mar 8]; Available from: http://link.springer.com/10.1007/s11764-020-00856-0

77

59. Lucas AR, Levine BJ, Avis NE. Posttreatment trajectories of physical activity in breast cancer survivors: MVPA Trajectories in Survivors of BC. Cancer. 2017 Jul 15;123(14):2773–80.

60. Bulmer SM, Howell J, Ackerman L, Fedric R. Women’s Perceived Benefits of Exercise During and After Breast Cancer Treatment. Women Health. 2012 Nov;52(8):771–87.

61. Maley M, Warren BS, Devine CM. A Second Chance: Meanings of Body Weight, Diet, and Physical Activity to Women Who Have Experienced Cancer. J Nutr Educ Behav. 2013 May;45(3):232–9.

62. Shang J, Wenzel J, Krumm S, Griffith K, Stewart K. Who Will Drop Out and Who Will Drop In: Exercise Adherence in a Randomized Clinical Trial Among Patients Receiving Active Cancer Treatment. Cancer Nurs. 2012;35(4):312–22.

63. Courneya KS, Segal RJ, Gelmon K, Reid RD, Mackey JR, Friedenreich CM, et al. Predictors of Supervised Exercise Adherence during Breast Cancer Chemo-therapy: Med Sci Sports Exerc. 2008 Jun;40(6):1180–7.

64. Bourke L, Homer KE, Thaha MA, Steed L, Rosario DJ, Robb KA, et al. Inter-ventions for promoting habitual exercise in people living with and beyond cancer. In: The Cochrane Collaboration, editor. Cochrane Database of Systematic Re-views [Internet]. Chichester, UK: John Wiley & Sons, Ltd; 2013 [cited 2014 Nov 6]. Available from: http://doi.wiley.com/10.1002/14651858.CD010192.pub2

65. Turner RR, Steed L, Quirk H, Greasley RU, Saxton JM, Taylor SJ, et al. Inter-ventions for promoting habitual exercise in people living with and beyond cancer. Cochrane Gynaecological, Neuro-oncology and Orphan Cancer Group, editor. Cochrane Database Syst Rev [Internet]. 2018 Sep 19 [cited 2019 Mar 5]; Avail-able from: http://doi.wiley.com/10.1002/14651858.CD010192.pub3

66. Hallward L, Patel N, Duncan LR. Behaviour change techniques in physical ac-tivity interventions for men with prostate cancer: A systematic review. J Health Psychol. 2020 Jan;25(1):105–22.

67. Bandura A. Health Promotion by Social Cognitive Means. Health Educ Behav. 2004 Apr;31(2):143–64.

68. Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychol Rev. 1977 Mar;84(2):191–215.

69. Bartholomew LK, editor. Planning health promotion programs: an intervention mapping approach. 3rd ed. San Francisco, CA: Jossey-Bass; 2011. 745 p.

70. Nadler M, Bainbridge D, Tomasone J, Cheifetz O, Juergens RA, Sussman J. On-cology care provider perspectives on exercise promotion in people with cancer: an examination of knowledge, practices, barriers, and facilitators. Support Care Cancer [Internet]. 2017 Mar 3 [cited 2017 Mar 7]; Available from: http://link.springer.com/10.1007/s00520-017-3640-9

71. Santa Mina D, Petrella A, Currie KL, Bietola K, Alibhai SMH, Trachtenberg J, et al. Enablers and barriers in delivery of a cancer exercise program: the Canadian experience. Curr Oncol. 2015 Sep 11;22(6):374.

72. Lavallée JF, Abdin S, Faulkner J, Husted M. Barriers and facilitators to partici-pating in physical activity for adults with breast cancer receiving adjuvant treat-ment: A qualitative metasynthesis. Psychooncology [Internet]. 2019 Jan 24 [cited 2019 Jan 30]; Available from: http://doi.wiley.com/10.1002/pon.4980

73. Maltser S, Cristian A, Silver JK, Morris GS, Stout NL. A Focused Review of Safety Considerations in Cancer Rehabilitation. PM&R. 2017 Sep;9:S415–28.

74. Cave J, Paschalis A, Huang CY, West M, Copson E, Jack S, et al. A systematic review of the safety and efficacy of aerobic exercise during cytotoxic chemother-apy treatment. Support Care Cancer. 2018 Oct;26(10):3337–51.

78

75. Niemeijer A, Lund H, Stafne SN, Ipsen T, Goldschmidt CL, Jørgensen CT, et al. Adverse events of exercise therapy in randomised controlled trials: a systematic review and meta-analysis. Br J Sports Med. 2019 Sep 28;bjsports-2018-100461.

76. Hootman JM. Association among Physical Activity Level, Cardiorespiratory Fit-ness, and Risk of Musculoskeletal Injury. Am J Epidemiol. 2001 Aug 1;154(3):251–8.

77. Francis P, Whatman C, Sheerin K, Hume P, Johnson MI. The Proportion of Lower Limb Running Injuries by Gender, Anatomical Location and Specific Pa-thology: A Systematic Review. J Sports Sci Med. 2019;18(1):21–31.

78. Taxbro K, Hammarskjöld F, Thelin B, Lewin F, Hagman H, Hanberger H, et al. Clinical impact of peripherally inserted central catheters vs implanted port cath-eters in patients with cancer: an open-label, randomised, two-centre trial. Br J Anaesth. 2019 Jun;122(6):734–41.

79. Hootman JM, Macera CA, Ainsworth BE, Addy CL, Martin M, Blair SN. Epide-miology of musculoskeletal injuries among sedentary and physically active adults: Med Sci Sports Exerc. 2002 May;34(5):838–44.

80. Chopra V, Anand S, Hickner A, Buist M, Rogers MA, Saint S, et al. Risk of venous thromboembolism associated with peripherally inserted central catheters: a systematic review and meta-analysis. The Lancet. 2013 Jul;382(9889):311–25.

81. Chopra V, O’Horo JC, Rogers MAM, Maki DG, Safdar N. The Risk of Blood-stream Infection Associated with Peripherally Inserted Central Catheters Com-pared with Central Venous Catheters in Adults: A Systematic Review and Meta-Analysis. Infect Control Hosp Epidemiol. 2013 Sep;34(9):908–18.

82. Kang J, Chen W, Sun W, Ge R, Li H, Ma E, et al. Peripherally inserted central catheter-related complications in cancer patients: a prospective study of over 50,000 catheter days. J Vasc Access. 2017 Mar;18(2):153–7.

83. Gao Y, Liu Y, Chen W, Wei L, Song L, Ma X. Peripherally inserted central cath-eter thrombosis incidence and risk factors in cancer patients: a double-center pro-spective investigation. Ther Clin Risk Manag. 2015 Jan;153.

84. Fallouh N, McGuirk HM, Flanders SA, Chopra V. Peripherally Inserted Central Catheter-associated Deep Vein Thrombosis: A Narrative Review. Am J Med. 2015 Jul;128(7):722–38.

85. Leung A, Heal C, Perera M, Pretorius C. A systematic review of patient-related risk factors for catheter-related thrombosis. J Thromb Thrombolysis. 2015 Oct;40(3):363–73.

86. Liem TK, Yanit KE, Moseley SE, Landry GJ, DeLoughery TG, Rumwell CA, et al. Peripherally inserted central catheter usage patterns and associated sympto-matic upper extremity venous thrombosis. J Vasc Surg. 2012 Mar;55(3):761–7.

87. Jones LW. Evidence-based risk assessment and recommendations for physical activity clearance: cancer. Appl Physiol Nutr Metab. 2011 Jul;36(S1):S101–12.

88. Doyle C, Kushi LH, Byers T, Courneya KS, Demark-Wahnefried W, Grant B, et al. Nutrition and physical activity during and after cancer treatment: an American Cancer Society guide for informed choices. CA Cancer J Clin. 2006 Dec;56(6):323–53.

89. McNeely ML, Campbell K, Ospina M, Rowe BH, Dabbs K, Klassen TP, et al. Exercise interventions for upper-limb dysfunction due to breast cancer treatment. Cochrane Breast Cancer Group, editor. Cochrane Database Syst Rev [Internet]. 2010 Jun 16 [cited 2018 Mar 20]; Available from: http://doi.wiley.com/10.1002/14651858.CD005211.pub2

90. Yi X, Chen J, Li J, Feng L, Wang Y, Zhu J-A, et al. Risk factors associated with PICC-related upper extremity venous thrombosis in cancer patients. J Clin Nurs. 2014 Mar;23(5–6):837–43.

79

91. Mijwel S, Backman M, Bolam KA, Jervaeus A, Sundberg CJ, Margolin S, et al. Adding high-intensity interval training to conventional training modalities: opti-mizing health-related outcomes during chemotherapy for breast cancer: the OptiTrain randomized controlled trial. Breast Cancer Res Treat. 2018 Feb;168(1):79–93.

92. DiSipio T, Rye S, Newman B, Hayes S. Incidence of unilateral arm lymphoedema after breast cancer: a systematic review and meta-analysis. Lancet Oncol. 2013 May;14(6):500–15.

93. Brennan MJ. Lymphedema following the surgical treatment of breast cancer: A review of pathophysiology and treatment. J Pain Symptom Manage. 1992 Feb;7(2):110–6.

94. Stuiver MM, ten Tusscher MR, Agasi-Idenburg CS, Lucas C, Aaronson NK, Bos-suyt PM. Conservative interventions for preventing clinically detectable upper-limb lymphoedema in patients who are at risk of developing lymphoedema after breast cancer therapy. In: The Cochrane Collaboration, editor. Cochrane Data-base of Systematic Reviews [Internet]. Chichester, UK: John Wiley & Sons, Ltd; 2015 [cited 2015 May 26]. Available from: http://doi.wiley.com/10.1002/14651858.CD009765.pub2

95. Wanchai A, Armer JM. Effects of weight-lifting or resistance exercise on breast cancer-related lymphedema: A systematic review. Int J Nurs Sci. 2019 Jan;6(1):92–8.

96. Nelson NL. Breast Cancer-Related Lymphedema and Resistance Exercise: A Systematic Review. J Strength Cond Res. 2016 Sep;30(9):2656–65.

97. Sweegers MG, Altenburg TM, Chinapaw MJ, Kalter J, Verdonck-de Leeuw IM, Courneya KS, et al. Which exercise prescriptions improve quality of life and physical function in patients with cancer during and following treatment? A sys-tematic review and meta-analysis of randomised controlled trials. Br J Sports Med. 2017 Sep 27;bjsports-2017-097891.

98. Clifford BK, Mizrahi D, Sandler CX, Barry BK, Simar D, Wakefield CE, et al. Barriers and facilitators of exercise experienced by cancer survivors: a mixed methods systematic review. Support Care Cancer. 2018 Mar;26(3):685–700.

99. Covington KR, Hidde MC, Pergolotti M, Leach HJ. Community-based exercise programs for cancer survivors: a scoping review of practice-based evidence. Sup-port Care Cancer. 2019 Dec;27(12):4435–50.

100.Berntsen S, Aaronson NK, Buffart L, Börjeson S, Demmelmaier I, Hellbom M, et al. Design of a randomized controlled trial of physical training and cancer (Phys-Can) – the impact of exercise intensity on cancer related fatigue, quality of life and disease outcome. BMC Cancer [Internet]. 2017 Dec [cited 2018 Mar 23];17(1). Available from: http://bmccancer.biomedcentral.com/articles/ 10.1186/s12885-017-3197-5

101.Petrie A, Sabin C. Medical statistics at a glance. 2nd ed. Malden, Mass: Black-well; 2005. 157 p.

102.Robertson RJ, Goss FL, Dube J, Rutkowski J, Dupain M, Brennan C, et al. Vali-dation of the adult OMNI scale of perceived exertion for cycle ergometer exer-cise. Med Sci Sports Exerc. 2004 Jan;36(1):102–8.

103.Borg G. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med. 1970;2(2):92–8.

104.Lestuzzi C, Vaccher E, Talamini R, Lleshi A, Meneguzzo N, Viel E, et al. Effort myocardial ischemia during chemotherapy with 5-fluorouracil: an underesti-mated risk. Ann Oncol. 2014 May;25(5):1059–64.

80

105.Lundh Hagelin C, Wengström Y, Runesdotter S, Johan Fürst C. The psychomet-ric properties of the Swedish Multidimensional Fatigue Inventory MFI-20 in four different populations. Acta Oncol. 2007 Jan;46(1):97–104.

106.Smets EM, Garssen B, Bonke B, De Haes JC. The Multidimensional Fatigue In-ventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psy-chosom Res. 1995 Apr;39(3):315–25.

107.Purcell A, Fleming J, Bennett S, Burmeister B, Haines T. Determining the mini-mal clinically important difference criteria for the Multidimensional Fatigue In-ventory in a radiotherapy population. Support Care Cancer. 2010 Mar;18(3):307–15.

108.McGlothlin AE, Lewis RJ. Minimal Clinically Important Difference: Defining What Really Matters to Patients. JAMA. 2014 Oct 1;312(13):1342.

109.Cella D, Eton DT, Lai J-S, Peterman AH, Merkel DE. Combining Anchor and Distribution-Based Methods to Derive Minimal Clinically Important Differences on the Functional Assessment of Cancer Therapy (FACT) Anemia and Fatigue Scales. J Pain Symptom Manage. 2002 Dec;24(6):547–61.

110.Yellen SB, Cella DF, Webster K, Blendowski C, Kaplan E. Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage. 1997 Feb;13(2):63–74.

111.Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: A Quality-of-Life Instrument for Use in International Clinical Trials in Oncology. JNCI J Natl Cancer Inst. 1993 Mar 3;85(5):365–76.

112.Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psy-chiatr Scand. 1983 Jun;67(6):361–70.

113.Federici S, Bracalenti M, Meloni F, Luciano JV. World Health Organization dis-ability assessment schedule 2.0: An international systematic review. Disabil Re-habil. 2017 Nov 6;39(23):2347–80.

114.El Fakir S, Abda N, Bendahhou K, Zidouh A, Bennani M, Errihani H, et al. The european organization for research and treatment of cancer quality of life ques-tionnaire-BR 23 breast cancer-specific quality of life questionnaire: psychometric properties in a Moroccan sample of breast cancer patients. BMC Res Notes. 2014 Dec;7(1):53.

115.Üstün TB, editor. Measuring health and disability: manual for WHO Disability Assessment Schedule WHODAS 2.0. Geneva: World Health Organization; 2010. 90 p.

116.Longo DL, Duffey PL, DeVita VT, Wesley MN, Hubbard SM, Young RC. The calculation of actual or received dose intensity: a comparison of published meth-ods. J Clin Oncol. 1991 Nov;9(11):2042–51.

117.Nordin Å, Taft C, Lundgren-Nilsson Å, Dencker A. Minimal important differ-ences for fatigue patient reported outcome measures—a systematic review. BMC Med Res Methodol. 2016 Dec;16(1):62.

118.Cocks K, King MT, Velikova G, de Castro G, Martyn St-James M, Fayers PM, et al. Evidence-based guidelines for interpreting change scores for the European Organisation for the Research and Treatment of Cancer Quality of Life Question-naire Core 30. Eur J Cancer. 2012 Jul;48(11):1713–21.

119.Graneheim UH, Lindgren B-M, Lundman B. Methodological challenges in qual-itative content analysis: A discussion paper. Nurse Educ Today. 2017 Sep;56:29–34.

120.Burnard P, Gill P, Stewart K, Treasure E, Chadwick B. Analysing and presenting qualitative data. BDJ. 2008 Apr 26;204(8):429–32.

81

121.Graneheim UH, Lundman B. Qualitative content analysis in nursing research: concepts, procedures and measures to achieve trustworthiness. Nurse Educ To-day. 2004 Feb;24(2):105–12.

122.Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006 Jan;3(2):77–101.

123.Clarke V, Braun V. Thematic analysis. J Posit Psychol. 2017 May 4;12(3):297–8.

124.lag-2003460-om-etikprovning-av-forskning-som_sfs-2003-460 [Internet]. [cited 2020 Mar 1]. Available from: https://www.riksdagen.se/sv/dokument-lagar/dokument/svensk-forfattningssamling/lag-2003460-om-etikprovning-av-forskning-som_sfs-2003-460

125.World Medical Association Declaration of Helsinki: Ethical Principles for Med-ical Research Involving Human Subjects. JAMA. 2013 Nov 27;310(20):2191.

126.Catt S, Sheward J, Sheward E, Harder H. Cancer survivors’ experiences of a com-munity-based cancer-specific exercise programme: results of an exploratory sur-vey. Support Care Cancer Off J Multinatl Assoc Support Care Cancer. 2018 Sep;26(9):3209–16.

127.Mazzoni A-S, Carlsson M, Berntsen S, Nordin K, Demmelmaier I. “Finding my own motivation” — A Mixed Methods Study of Exercise and Behaviour Change Support During Oncological Treatment. Int J Behav Med. 2019 Oct;26(5):499–511.

128.Kraemer WJ, Ratamess NA. Fundamentals of Resistance Training: Progression and Exercise Prescription: Med Sci Sports Exerc. 2004 Apr;36(4):674–88.

129.Bjørke ACH, Sweegers MG, Buffart LM, Raastad T, Nygren P, Berntsen S. Which exercise prescriptions optimize V̇O 2 max during cancer treatment?—A systematic review and meta analysis. Scand J Med Sci Sports. 2019 Sep;29(9):1274–87.

130.Fairman C, Focht B, Lucas A, Lustberg M. Effects of exercise interventions dur-ing different treatments in breast cancer. J Community Support Oncol. 2016 May;14(5):200–9.

131.Mijwel S, Bolam KA, Gerrevall J, Foukakis T, Wengström Y, Rundqvist H. Ef-fects of Exercise on Chemotherapy Completion and Hospitalization Rates: The OptiTrain Breast Cancer Trial. The Oncologist. 2020 Jan;25(1):23–32.

132.Van Vulpen JK, Velthuis MJ, Steins Bisschop CN, Travier N, Van Den Buijs BJW, Backx FJG, et al. Effects of an Exercise Program in Colon Cancer Patients undergoing Chemotherapy: Med Sci Sports Exerc. 2016 May;48(5):767–75.

133.Neil-Sztramko SE, Winters-Stone KM, Bland KA, Campbell KL. Updated sys-tematic review of exercise studies in breast cancer survivors: attention to the prin-ciples of exercise training. Br J Sports Med. 2017 Nov 21;bjsports-2017-098389.

134.Neil-Sztramko SE, Medysky ME, Campbell KL, Bland KA, Winters-Stone KM. Attention to the principles of exercise training in exercise studies on prostate can-cer survivors: a systematic review. BMC Cancer. 2019 Dec;19(1):321.

135.Kirkham AA, Bland KA, Zucker DS, Bovard J, Shenkier T, Mckenzie DC, et al. “Chemotherapy-periodized” Exercise to Accommodate for Cyclical Variation in Fatigue: Med Sci Sports Exerc. 2020 Feb;52(2):278–86.

136.Van Vulpen JK, Sweegers MG, Peeters PHM, Courneya KS, Newton RU, Aaronson NK, et al. Moderators of Exercise Effects on Cancer-related Fatigue: A Meta-analysis of Individual Patient Data. Med Sci Sports Exerc. 2020 Feb;52(2):303–14.

137.Schmitz KH, Ahmed RL, Troxel A, Cheville A, Smith R, Lewis-Grant L, et al. Weight Lifting in Women with Breast-Cancer–Related Lymphedema. N Engl J Med. 2009 Aug 13;361(7):664–73.

82

138.Clemons M, Stober C, Kehoe A, Bedard D, MacDonald F, Brunet M-C, et al. A randomized trial comparing vascular access strategies for patients receiving chemotherapy with trastuzumab for early-stage breast cancer. Support Care Can-cer [Internet]. 2020 Jan 30 [cited 2020 Feb 5]; Available from: http://link.springer.com/10.1007/s00520-020-05326-y

139.Qiu X, Guo Y, Fan H, Shao J, Zhang X. Incidence, risk factors and clinical out-comes of peripherally inserted central catheter spontaneous dislodgment in on-cology patients: A prospective cohort study. Int J Nurs Stud. 2014 Jul;51(7):955–63.

140.Bertoglio S, Faccini B, Lalli L, Cafiero F, Bruzzi P. Peripherally inserted central catheters (PICCs) in cancer patients under chemotherapy: A prospective study on the incidence of complications and overall failures: PICC Venous Access for Chemotherapy Delivery. J Surg Oncol. 2016 May;113(6):708–14.

141.Robinson A, Souied O, Bota AB, Levasseur N, Stober C, Hilton J, et al. Optimal vascular access strategies for patients receiving chemotherapy for early-stage breast cancer: a systematic review. Breast Cancer Res Treat. 2018 Oct;171(3):607–20.

142.van der Leeden M, Huijsmans RJ, Geleijn E, de Rooij M, Konings IR, Buffart LM, et al. Tailoring exercise interventions to comorbidities and treatment-in-duced adverse effects in patients with early stage breast cancer undergoing chem-otherapy: a framework to support clinical decisions. Disabil Rehabil. 2017 Jan 5;1–11.

143.Halverson JL, Martinez-Donate AP, Palta M, Leal T, Lubner S, Walsh MC, et al. Health Literacy and Health-Related Quality of Life Among a Population-Based Sample of Cancer Patients. J Health Commun. 2015 Nov 2;20(11):1320–9.

144.Plummer LC, Chalmers KA. Health literacy and physical activity in women di-agnosed with breast cancer: HEALTH LITERACY AND PHYSICAL ACTIV-ITY. Psychooncology. 2017 Oct;26(10):1478–83.

145.Schmitz KH, Campbell AM, Stuiver MM, Pinto BM, Schwartz AL, Morris GS, et al. Exercise is medicine in oncology: Engaging clinicians to help patients move through cancer. CA Cancer J Clin. 2019 Nov;69(6):468–84.

146.Park J-H, Lee J, Oh M, Park H, Chae J, Kim D-I, et al. The effect of oncologists’ exercise recommendations on the level of exercise and quality of life in survivors of breast and colorectal cancer: A randomized controlled trial: Exercise Recom-mendation in Breast Cancer and CRC. Cancer. 2015 May;n/a-n/a.

147.Browall M, Mijwel S, Rundqvist H, Wengström Y. Physical Activity During and After Adjuvant Treatment for Breast Cancer: An Integrative Review of Women’s Experiences. Integr Cancer Ther. 2018 Mar;17(1):16–30.

148.Loprinzi PD, Cardinal BJ, Si Q, Bennett JA, Winters-Stone KM. Theory-based predictors of follow-up exercise behavior after a supervised exercise intervention in older breast cancer survivors. Support Care Cancer. 2012 Oct;20(10):2511–21.

149.Courneya KS, Segal RJ, Reid RD, Jones LW, Malone SC, Venner PM, et al. Three independent factors predicted adherence in a randomized controlled trial of resistance exercise training among prostate cancer survivors. J Clin Epidemiol. 2004 Jun;57(6):571–9.

150.Husebø AML, Dyrstad SM, Søreide JA, Bru E. Predicting exercise adherence in cancer patients and survivors: a systematic review and meta-analysis of motiva-tional and behavioural factors. J Clin Nurs. 2013 Jan;22(1–2):4–21.

151.Kampshoff CS, Jansen F, van Mechelen W, May AM, Brug J, Chinapaw MJ, et al. Predictors of physical activity in colorectal cancer survivors after participation in a telephone-delivered multiple health behavior change intervention. Int J Be-hav Nutr Phys Act. 2014;11(1):80.

83

152.Robertson L, Richards R, Egan R, Szymlek-Gay EA. Promotion and support of physical activity among cancer survivors: a service provider perspective: Physi-cal activity among cancer survivors: a provider perspective. Psychooncology. 2012 Jan;n/a-n/a.

153.Dalzell MA, Smirnow N, Sateren W, Sintharaphone A, Ibrahim M, Mastroianni L, et al. Rehabilitation and exercise oncology program: translating research into a model of care. Curr Oncol. 2017 Jun 28;24(3):191.

154.Spencer JC, Wheeler SB. A systematic review of Motivational Interviewing in-terventions in cancer patients and survivors. Patient Educ Couns. 2016 Jul;99(7):1099–105.

155.Prochaska JO, Velicer WF. The Transtheoretical Model of Health Behavior Change. Am J Health Promot. 1997 Sep;12(1):38–48.

156.McLafferty I. Focus group interviews as a data collecting strategy. J Adv Nurs. 2004 Oct;48(2):187–94.

157.Polit DF, Beck CT. Nursing research: generating and assessing evidence for nurs-ing practice. 8th ed. Philadelphia: Wolters Kluwer Health/lippincott Williams & Wilkins; 2008. 796 p.

158.Darawsheh W, Stanley M. Reflexivity in research: Promoting rigour, reliability and validity in qualitative research. Int J Ther Rehabil. 2014 Dec 2;21(12):560–8.

84

Appendix

Appendix 1. Overview of inclusion and exclusion criteria Paper I-III

Inclusion criteria Breast cancer Prostate cancer Colorectal

cancer

Age over 18 yrs X X X

Understands written and spoken Swedish X X X

Neo-adjuvant chemotherapy X

Adjuvant chemotherapy X X

Adjuvant radiation therapy X

Curative radiation therapy X

Adjuvant endocrine therapy X X

Neo- and/or adjuvant endocrine therapy1 X

Exclusion criteria

Inability to perform basic activities in daily life X X X

Cognitive inability (i.e., dementia) X X X

Serious psychiatric disease X X X

Impairments that can prevent PA2 X X X

Cardiovascular disease that can prevent PA3 X X X

Lung disease that can prevent PA4 X X X

Completed treatment for other cancer disease < 12

months ago or not recovered from previous cancer treatment

X X X

Body mass index < 18.5 X X X

Fibromyalgia X X X

Pregnancy X X

Breast cancer stage IIIb–IV X 1Men with high risk prostate cancer receive neo-adjuvant endocrine treatment during three months before curative radiation treat-ment and then adjuvant endocrine treatment for 2 years after. 2 For example, orthopaedic and neurological diseases causing a need for walking aids.3 For example, unstable angina pectoralis, severe heart failure. 4 For example, late-stage chronic obstructive pulmo-nary disease. PA = physical activity

85

Appendix 2. Disease stage and treatment characteristics Paper I HI BCS

n (%) HI n (%) LMI BCS

n (%) LMI n (%)

Breast cancer 113 115 116 113 Stage

In situ 3 (3.0) 2 (2.0) 5 (4.9) 2 (2.0) I (T1N0M0) 60 (60.6) 55 (55.0) 47 (46.1) 57 (58.2) II (T1-2N1M0, T2-3N0M0) 34 (34.3) 40 (40.0) 44 (43.1) 34 (34.7) III (T1-2N2M0,T3N1-2M0) 2 (2.0) 3 (3.0) 6 (5.9) 5 (5.1)

Chemotherapy 70 (67.3) 66 (61.1) 69 (64.5) 71 (67.0) Adjuvant 60 (85.7) 52 (78.8) 58 (84.1) 61 (85.9) Neoadjuvant 10 (14.3) 14 (21.2) 11 (15.9) 10 (14.1) Type

F/E75C600 23 (33.9) 23 (34.8) 17 (24.6) 28 (39.4) F/E90-100C500-600 38 (54.3) 37 (56.1) 47 (68.1) 38 (53.5) Docetaxel75-80 29 (41.4) 25 (37.9) 22(31.9) 36 (50.7) Docetaxel90-100 30 (42.9) 30 (45.5) 42 (60.9) 29 (40.8) Paclitaxel 10 (14.5) 8 (12.1) 6 (8.7) 3 (4.3) Docetaxel Capecitabine 6 (8.6) 3 (4.5) 2 (2.9) 1 (1.4) CEX 5 (7.1) 3 (4.5) 2 (2.9) 0

Antibody treatment 20 (28.6) 19 (28.8) 17 (24.6) 23 (32.4) Radiotherapy 84 (80.8) 88 (81.5) 85 (79.4) 92 (86.8)

Type Breast/chest wall only 49 (58.3) 53 (60.2) 48 (56.5) 58 (63.0) Axillary only 3 (3.6) 3 (3.4) 2 (2.4) 0 Breast and axillary 32 (38.1) 32 (36.4) 35 (41.2) 34 (37.0)

Endocrine treatment 72 (69.2) 75 (69.4) 84 (78.5) 79 (74.5) Prostate cancer 26 23 23 25 Stage

T1aN0M0 11 (47.8) 12 (52.2) 9 (39.1) 9 (39.1) T2N0M0 9 (39.1) 7 (30.4) 8 (38.1) 9 (39.1) T3N0M0 1 (4.3) 2 (8.7) 1 (4.8) 3 (13.0) T4N0M0 1 (4.3) 1 (4.3) 0 1 (4.3) T1c-4N1M0 1 (4.3) 0 3 (14.3) 1 (4.3)

Gleason Score ≤7 14 (58.3) 15 (65.2) 16 (76.2) 18 (78.3) >7 10 (41.7) 8 (34.8) 5 (23.8) 5 (21.7)

PSA, ng/ml, mean (SD) 13.1 (18.6) 14.5 (32.4) 9.7 (8.3) 15.9 (28.7) Radiotherapy 25 (100.0) 20 (100.0) 22 (100.0) 25 (100.0)

Type Brachy only 0 0 0 2 (8.0) External only 14 (56.0) 5 (25.0) 10 (45.5) 16 (64.0) Brachy and external 11 (44.0) 15 (75.0) 12 (54.5) 7 (28.0)

Endocrine treatment 13 (50) 13 (56.5) 10 (43.5) 15 (60) Colorectal cancer 5 6 6 5 Stage

II (T3-4N0M0) 0 2 (33.3) 1 (16.7) 2 (40.0) III (T2-4N1-2M0) 3 (75.0) 4 (66.7) 5 (83.3) 3 (60.0) IV (T3N1M1), radically removed distant metastasis

1 (25.0) 0 0 0

Chemotherapy type CAPOX 3 (60.0) 4 (66.7) 4 (66.7) 2 (40.0) Capecitabine single 2 (40.0) 2 (33.3) 2 (33.3) 3 (60.0)

86

Radiotherapy, pre-operative 1 (20.0) 2 (33.3) 0 0 n vary due to missing data. Data are number (%), unless specified. Percentage is of those with available data. HI BCS= high-intensity with behaviour change support, HI= high-intensity without behaviour change support, LMI BCS=Low-moderate intensity with behaviour change support, LMI=Low-moder-ate intensity. PSA; Prostate-Specific Antigen. TNM classification according to American Joint Com-mittee on Cancer 8 2017, T: tumour size N: lymph node status M: distant metastasis. F/EC: Fluoroura-cil 500-600mg/m2 and/or only Epirubicin 75-100mg/m2 –Cyclophosphamide 500-600mg/m2; CEX: Epirubicin-Cyclophosphamide-Capecitabine; CAPOX: Capecitabine-Oxaliplatin. Patients could have more than one type of treatment.

87

Appendix 3. Resistance training PHYS-CAN (Swedish)

88

Träning för bål, rygg och bäckenbotten

89

Appendix 4. Age, disease stage and treatment characteristics Paper II Care as usual HI LMI Total, n 90 288 289 Age, years, mean (sd) [min–max] 57.7 (11.2) [28–80] 58.7 (12.2) [22–85] 58.8 (11.7) [30–85] Breast cancer, total n (% of group) 86 (84.3) 228 (79.2) 229 (79.2) Oncological treatment breast cancer, n (%)

70 (81.4) 212 (93.0) 213 (93.0)

Chemotherapy total 29 (41.4) 136 (59.6) 140 (61.1) Adjuvant chemotherapy1 28 (96.5) 112 (82.4) 119 (85.0) Neo-adjuvant chemotherapy1 1 (0.3) 24 (17.6) 21 (15.0)

Chemotherapy type1 F/E75C600 15 (51.7) 46 (33.8) 45 (32.1) F/E90-100C500-600 14 (42.3) 75 (55.1) 85 (60.7) Docetaxcel75-80 18 (62.0) 54 (39.7) 58 (41.4) Docetaxcel90-100 10 (34.4) 60 (44.1) 71 (50.7) Paclitaxcel 0 (0.0) 18 (13.2) 9 (6.4) Docetaxcel Capacetabine 1 (0.3) 9 (6.6) 3 (2.1) CEX 1 (0.3) 8 (5.9) 2 (1.4)

Antibody treatment1 6 (8.5) 39 (28.7) 40 (28.6) Radiotherapy 51 (59.3) 172 (74.6) 177 (77.3)

Endocrine treatment 54 (77.1) 147 (64.5) 163 (71.2) Missing 16 (18.6) 16 (7.0) 16 (7.0) Surgery breast cancer, total n 69 204 202 Breast conserving treatment 38 (55.0) 136 (66.7) 129 (63.8)

Axillary node dissection only 4 (5.8) 24 (11.8) 25 (12.4) Sentinel node only 54 (78.3) 148 (72.5) 137 (67.8) Both 10 (14.5) 26 (12.7) 32 (15.8)

Missing 17 (19.8) 24 (10.5) 27 (11.8) Prostate cancer, total n (% of group) 49 (17.0) 48 (16.6) Oncological treatment prostate cancer, n (%)

45 (91.8) 46 (95.8)

Brachytherapy only 0 (0.0) 2 (4.4) External therapy only 18 (40.9) 24 (53.3) Brachy- and external therapy comb. 26 (59.1) 19 (42.2) Neo- and/or adjuvant endocrine treat-ment

26 (53.0) 25 (52.0)

Missing, n (%) 4 (8.2) 2 (4.2) Colorectal cancer, total n (% of group) 4 (3.9) 11 (3.8) 12 (4.2) Oncological treatment colorectal cancer 4 (100.0) 11 (100.0) 11 (91.7) Chemotherapy type, n (%) CAPOX 1 (25.0) 7 (63.6) 6 (50.0) Capecitabine single 3 (75.0) 4 (36.4) 5 (41.6) Pre-operative radiotherapy, n (%) 0 (0.0) 3 (27.3) 0 (0.0) Missing 0 (0.0) 0 (0.0) 1 (8.3) Surgery colorectal cancer, total n (%) 4 (100.0) 11 (100.0) 11 (100.0) Right hemicolectomy 2 (50.0) 3 (27.3) 9 (81.8) Left hemicolectomy 0 (0.0) 2 (18.2) 0 (0.0) Sigmoid resection 2 (50.0) 1 (9.0) 1 (9.9) Rectal excision 0 (0.0) 2 (18.2) 0 (0.0) Frontal resection of rectum 0 (0.0) 2 (18.2) 1 (9.9) Missing 0 (0.0) 0 (0.0) 0 (0.0) Central venous access2 Type of access n (%) PICC 30 (100.0) 132 (93,6) 141 (97,9) VAP 0 (0.0) 7 (5,0) 2 (1,4) CVC 0 (0.0) 2 (1,4) 1 (0,7) Days inserted, mean (SD) [Median] 97.72 (38.57)

[110.0] 101.88 (33.44) [107.0]

98,48 (29.78) [106.0]

HI = high-intensity exercise group. LMI = Low-moderate intensity exercise group.1n (%) of participants with breast cancer receiving chemotherapy. F/EC: Fluorouracil 500–600 mg/m2 and/or only Epirubicin 75–100 mg/m2 -- cyclophosphamide 500–600 mg/m2. CEX: Epirubicin-Cyclophosphamide-Capecitabine.

90

CAPOX: Capecitabine-Oxaliplatin. PICC: peripherally inserted central catheter, VAP: venous access port or subcutaneous infusion port, CVC: central venous catheter.

91

Appendix 5. Alcohol, tobacco, BMI, Physical activity and exercise adher-ence characteristics Paper II Care as usual HI LMI Total, n 90 288 289 Alcohol, n 32 276 275 Never or less than one standard glass/week, n (%)

16 (50.0) 91 (33.0) 116 (42.2)

1–4 standard glasses/week, n (%) 11 (34.4) 128 (46.4) 112 (40.7) 5–9 or more standard glasses/week, n (%)

5 (15.6) 57 (20.7) 47 (17.0)

Missing, n (%) 58 (64.4) 14 (4.9) 18 (6.2) Smoking and snuff use, n 28 254 258 Never, n (%) 21 (75.0) 142 (55.9) 146 (56.6) Previously or Less than daily, n (%) 4 (14.3) 97 (38.2) 94 (36.4) Daily, n (%) 3 (10.7) 15 (5.9) 18 (7.0) Missing, n (%) 62 (68.9) 34 (11.8) 31 (10.7) Body mass index, n (sd) BMI ≤18.4 kg/m2 0 (0.0) 5 (1.7) 2 (0.7) BMI 18.5–24.9 kg/m2 31 (34.4) 124 (43.1) 129 (44.6) BMI 25–29.9 kg/m2 30 (33.3) 90 (31.3) 97(33.6) BMI ≥ 30 kg/m2 10 (11.1) 48 (16.7) 39 (13.5) Missing, n (%) 19 (21.1) 21 (7.3) 22 (7.6) PA measured with SenseWearTM, mean (sd) MVPA hrs/day 1.1 (0.56) 1.2 (0.87) 1.2 (0.78) MET 1.25 (0.16) 1.29 (0.21) 1.27 (0.19) Missing 20 (22.2) 34 (11.8) 25 (8.7) Adherence to exercise intervention (randomised intensity), n = 469 Resistance traning2 % (SD) 57.9 (23.7) 62.8 (22.3) Endurance traning2 % (SD) 38.3 (32.9) 51.0 (38.6) Combined % (SD) 52.7 (24.2) 63.3 (23.5) HI = high-intensity exercise group. LMI = Low-moderate intensity exercise group.1n (%) of participants with breast cancer receiving chemotherapy. BMI= Body Mass Index. PA = physical activity. MVPA: mod-erate-to-vigorous intensity physical activity. MET: Metabolic equivalent ((mlO2/kg/ml) x 3.5). 2 Percent of performed session/number of prescribed sessions.

92

Appendix 6. Characteristics participants paper III

Focus group in-terviews, Women n= = 10

Focus group in-terviews, Men n= = 8

Individual in-terviews, Women n= = 5

Age, years mean (min-–max) 59.5 (42-–74) 65.5 (61-–70) 56.6 (49-–64)

Type of cancer Breast Prostate Colorectal

9

1

8

5

Ongoing treatment Radiation/endocrine Chemotherapy Endocrine

1 9

3

5

4* 1*

Level of education Primary education Secondary/high school education College/university education

1 2 7

1 4 3

3 2

physical activity in the last 6 months Endurance exercise† Not at all < 150 mins /week ≥ 150 mins/week Resistance training Not at all < twice a week ≥ twice a week

missing: 2

2 3 3

missing: 2 8

4 4

4 2 2

1 3 1

5

Possibility to be physically active, VAS mm median (min–max)

81.5 (49–100)

71.5 (28–94)

84.0 (68–97)

Motivation to be physically active, VAS mm median (min–max)

93.0 (49–100)

69.0 (43–91)

80.0 (49–98)

*= planned treatment † at least medium intensity.

93

Appendix 7. Main effects of exercise intensity, additional behaviour change support and interaction for secondary outcomes Paper I

Secondary out-comes

Exercise inten-sity AMD (95%CI)

p- value

BCS AMD (95%CI)

p- value

Interaction AMD (95%CI)

p- value

EORTC QLQ-C30 Summary Score

-0.71 (-2.48 to 1.06)

0.431 -0.77 (-2.54 to 0.99)

0.390 -2.77 (-4.53 to -1.02)

0.002

HADS Depression 0.03 (-0.44 to 0.51)

0.895 -0.25 (-0.75 to 0.25)

0.332 0.17 (-0.32 to 0.66)

0.502

HADS Anxiety 0.16 (-0.38 to 0.71)

0.555 0.23 (-0.35 to 0.80)

0.434 0.52 (-0.04 to 1.09)

0.071

WHODAS Work subscale1

0.45 (-0.90 to 1.80)

0.507 0.57 (-0.67 to 1.81)

0.362 0.78 (-0.60 to 2.17)

0.263

WHODAS Social Participation sub-scale

0.12 (-0.67 to 0.90)

0.771 0.40 (-0.41 to 1.21)

0.337 0.50 (-0.28 to 1.29)

0.207

Average 1RM left & right leg, kg2

3.98 (0.65 to 7.30)

0.019 2.83 (-0.63 to 6.30)

0.108 -0.56 (-3.77 to 2.66)

0.733

1RM chest press, kg2

0.39 (-1.66 to 2.44)

0.709 1.41 (-0.62 to 3.44)

0.172 -1.32 (-3.30 to 0.67)

0.193

VO2max, ml/kg/min2

1.60 (0.12 to 3.07)

0.034 0.76 (-0.67 to 2.18)

0.297 -1.24 (-2.67 to 0.19)

0.089

Sleep, hrs/day2 0.10 (-0.14 to 0.34)

0.404 -0.02 (-0.26 to 0.21)

0.842 -0.10 (-0.33 to 0.13)

0.388

Sedentary time, hrs/day2

-0.27 (-0.68 to 0.13)

0.187 -0.14 (-0.55 to 0.26)

0.487 0.25 (-0.15 to 0.65)

0.224

MVPA, hrs/day2 0.06 (-0.13 to 0.26)

0.521 0.07 (-0.13 to 0.26)

0.504 -0.26 (-0.45 to -0.07)

0.008

Relative Dose In-tensity, %3

-1.66 (-4.41 to 1.09)

0.235 -1.92 (-4.67 to 0.84)

0.172 -2.35 (-5.09 to 0.38)

0.092

Included multiple imputation by chained equations to account for missing data, presented as adjusted mean difference and 95% confidence intervals (CI) (n=577). 1For patients who reported working. 2Base-line values not included in analysis due to missing data >10%. 3For patients treated with chemotherapy. Linear regression analyses adjusted for hospital, cancer site, and baseline measure of outcome. Bold indicates p-value<0.05. AMD: Adjusted mean difference; 95%CI: 95% confidence intervals; BCS: Addi-tional behaviour change support; EORTC QLQ-C30: European Organisation for Research and Treatment of Cancer, Quality of life Questionnaire C30; HADS: Hospital Anxiety and Depression scale; WHODAS: World Health Organization Disability Assessment Schedule; 1RM: 1 repetition maximum; VO2max: max-imal volume of oxygen uptake; MVPA: moderate-to-vigorous intensity physical activity. Bold indicates p-value<0.05.

A

ppen

dix

8. M

ain

effe

cts o

f exe

rcis

e in

tens

ity, B

CS

and

inte

ract

ion

on e

ach

subs

cale

of t

he E

OR

TC Q

LQ-C

30

Exer

cise

inte

nsity

AM

D

(95%

CI)

p-

valu

e B

CS

AM

D (9

5%C

I)

p-va

lue

Inte

ract

ion

AM

D (9

5%C

I)

p-va

lue

Phys

ical

func

tioni

ng

0.08

(-1.

96 to

2.1

3)

0.93

5 0.

76 (-

1.39

to 2

.91)

0.

488

-1.7

3 (-

3.77

to 0

.31)

0.09

7

Rol

e fu

nctio

ning

-2

.85

(-6.

80 to

1.1

1)0.

158

-1.9

6 (-

6.14

to 2

.22)

0.

356

-2.6

7 (-

6.62

to 1

.29)

0.18

6

Emot

iona

l fun

ctio

ning

-1

.49

(-4.

59 to

1.6

1)0.

345

-2.4

1 (-

5.70

to 0

.89)

0.

152

-2.8

3 (-

5.96

to 0

.31)

0.07

7

Cog

nitiv

e fu

nctio

ning

0.

48 (-

2.68

to 3

.63)

0.76

7 -1

.43

(-4.

78 to

1.9

1)

0.40

0-4

.67

(-7.

84 to

-1.5

0)0.

004

Soci

al fu

nctio

ning

-1

.95

(-5.

45 to

1.5

5)0.

274

0.62

(-3.

06 to

4.2

9)

0.74

1 -2

.22

(-5.

71 to

1.2

7)0.

212

Fatig

ue

-0.0

7 (-

3.52

to 3

.38)

0.96

7 1.

27 (-

2.16

to 4

.71)

0.

467

2.90

(-0.

51 to

6.3

1)0.

095

Nau

sea

and

vom

iting

0.

45 (-

1.13

to 2

.04)

0.57

4 0.

56 (-

1.11

to 2

.23)

0.

510

1.85

(0.2

6 to

3.4

4)0.

023

Pain

4.

60 (0

.70

to 8

.50)

0.02

1 1.

95 (-

2.05

to 5

.95)

0.

338

3.02

(-0.

91 to

6.9

5)0.

132

Dys

pnoe

a -2

.24

(-5.

70 to

1.2

2)0.

204

0.08

(-3.

58 to

3.7

5)

0.96

4 1.

29 (-

2.20

to 4

.77)

0.46

8

Inso

mni

a 1.

58 (-

3.42

to 6

.57)

0.53

6 0.

41 (-

4.64

to 5

.47)

0.

872

9.27

(4.0

5 to

14.

49)

0.00

1

App

etite

loss

0.

49 (-

1.85

to 2

.83)

0.67

9 -0

.88

(-3.

40 to

1.6

3)

0.49

10.

51 (-

1.83

to 2

.85)

0.66

9

Con

stip

atio

n -2

.24

(-5.

44 to

0.9

6)0.

169

-1.2

0 (-

4.32

to 1

.92)

0.

449

0.50

(-2.

66 to

3.6

6)0.

756

Dia

rrho

ea

-2.0

7 (-

5.44

to 1

.31)

0.22

9 0.

53 (-

3.05

to 4

.11)

0.

770

1.60

(-1.

77 to

4.9

7)0.

352

Fina

ncia

l diff

icul

ties

0.71

(-2.

55 to

3.9

6)0.

669

-1.3

2 (-

4.67

to 2

.03)

0.

438

-0.1

4 (-

3.42

to 3

.14)

0.93

1

Incl

uded

mul

tiple

impu

tatio

n by

cha

ined

equ

atio

ns to

acc

ount

for m

issi

ng d

ata,

pre

sent

ed a

s adj

uste

d m

ean

diff

eren

ce a

nd 9

5% c

onfid

ence

inte

rval

s (C

I)

(n=5

77).

Line

ar re

gres

sion

ana

lyse

s adj

uste

d fo

r hos

pita

l, ca

ncer

site

, and

bas

elin

e m

easu

re o

f out

com

e. B

old

indi

cate

s p-v

alue

<0.0

5. E

OR

TC Q

LQ-C

30: E

uro-

pean

Org

anis

atio

n fo

r Res

earc

h an

d Tr

eatm

ent o

f Can

cer,

Qua

lity

of li

fe Q

uest

ionn

aire

C30

; AM

D: A

djus

ted

mea

n di

ffer

ence

; 95%

CI:

95%

con

fiden

ce in

ter-

vals

; BC

S: B

ehav

iour

cha

n ge

supp

ort.

App

endi

x 9.

Res

pond

er a

naly

sis f

or M

ultid

imen

sion

al fa

tigue

inve

ntor

y, p

hysi

cal f

atig

ue.

HI B

CS

(n =

109

) H

I (n

= 10

6)

LM

I BC

S (n

= 1

02)

LM

I (n

= 11

0)

Cha

nge

in M

FI

PF

n C

umul

a-tiv

e %

Pr

opor

-tio

n by

ca

tego

ry

Cha

nge

in M

FI

PF

n C

umul

a-tiv

e %

Pr

opor

tion

by c

ate-

gory

Cha

nge

in M

FI

PF

nC

umul

a-tiv

e %

Pr

opor

-tio

n by

ca

tego

ry

Cha

nge

in

MFI

PF

nC

umul

a-tiv

e %

Pr

opor

-tio

n by

ca

tego

ry

-14

1 0.

9 -1

3 4

3.8

-13

1 1.

0 -1

4 1

0.9

-13

1 1.

8 -1

2 1

4.7

-12

3 3.

9 -1

2 4

4.5

-12

1 2.

8 -1

1 2

6.6

-11

1 4.

9 -1

0 2

6.4

-10

2 4.

6 -9

4

10.4

-1

0 2

6.9

-9

1 7.

3 -9

2

6.4

-8

5 15

.1

-9

3 9.

8 -8

1

8.2

-8

5 11

.0

-7

4 18

.9

-8

2 11

.8

-7

3 10

.9

-7

9 19

.3

-6

5 23

.6

-7

2 13

.7

-6

6 16

.4

-6

12

30.3

-5

6

29.2

-6

3

16.7

-5

5

20.9

-5

6

35.8

-4

7

35.8

-5

4

20.6

-4

7

27.3

-4

7

42.2

-3

9

44.3

-4

7

27.5

-3

9

35.5

-3

8

49.5

-2

11

54

.7

54.7

-3

9

36.3

-2

13

47

.3

47.3

-2

7

56.0

56

.0

-1

13

67.0

-2

9

45.1

45

.1

-1

9 55

.5

-1

14

68.8

0

13

79.2

-1

12

56

.9

0 12

66

.4

0 9

77.1

1

7 85

.8

31.1

0

14

70.6

1

7 72

.7

25.5

1

4 80

.7

24.8

2

6 91

.5

1 10

80

.4

35.3

2

10

81.8

2

8 88

.1

3 2

93.4

2

7 87

.3

3 4

85.5

3

4 91

.7

4 3

96.2

3

3 90

.2

4 5

90.0

4

1 92

.7

5 2

98.1

4

3 93

.1

5 3

92.7

5

4 96

.3

11

1 99

.1

5 4

97.1

6

2 94

.5

6 2

98.2

13

1

100.

0 14

.2

6 1

98.0

7

1 95

.5

9 2

100.

0 19

.37

1 99

.0

9 2

97.3

8

1 10

0.0

19.6

10

1

98.2

11

1 99

.1

13

1 10

0.0

27.3

MFI

= M

ultid

imen

sion

al fa

tigue

inve

ntor

y. H

I BC

S= h

igh-

inte

nsity

with

beh

avio

ur c

hang

e su

ppor

t, H

I= h

igh-

inte

nsity

with

out b

ehav

iour

cha

nge

supp

ort,

LMI B

CS=

Low

-mod

erat

e in

tens

ity w

ith b

ehav

iour

cha

nge

supp

ort,

LMI=

Low

-mod

erat

e in

tens

ity. G

reen

= im

prov

ed fa

tigue

, Yel

low

= un

chan

ged

fatig

ue, R

ed=

wor

sene

d fa

tigue

. Low

er sc

ore

indi

cate

s les

s fa

tigue

.

App

endi

x 10

. Res

pond

er a

naly

sis f

or F

AC

IT-F

. H

I BC

S H

I LM

I BC

S LM

I C

hang

e in

FA

CIT

-F

n C

umul

a-tiv

e %

Pr

opor

tion

by c

ate-

gory

Cha

nge

in

FAC

IT-F

n

Cum

ula-

tive

%

Prop

ortio

n by

cat

e-go

ry

Cha

nge

in

FAC

IT-F

n

Cum

ula-

tive

%

Prop

ortio

n by

cat

e-go

ry

Cha

nge

in

FAC

IT-F

n

Cum

ula-

tive

%

Prop

ortio

n by

cat

e-go

ry

23

1 0,

9 24

1

0,9

24

1 0,

9 29

1

0,9

20

1 1,

8 22

1

1,9

23

1 1,

9 27

1

1,7

17

1 2,

7 19

1

2,8

21

1 2,

8 23

1

2,6

14

2 4,

5 18

2

4,6

19

1 3,

7 19

1

3,5

11

1 5,

4 16

1

5,6

18

2 5,

6 18

3

6,1

10

1 6,

3 14

1

6,5

16

2 7,

4 17

1

7,0

9 2

8,1

13

1 7,

4 15

4

11,1

16

1

7,8

8 4

11,7

12

2

9,3

13

1 12

,0

14

3 10

,4

7 1

12,6

11

3

12,0

12

1

13,0

13

2

12,2

6

3 15

,3

10

1 13

,0

10

4 16

,7

12

2 13

,9

5 8

22,5

9

1 13

,9

9 3

19,4

11

2

15,7

4

6 27

,9

8 1

14,8

8

1 20

,4

10

1 16

,5

3 7

34,2

34

7

8 22

,2

7 4

24,1

9

3 19

,1

2 4

37,8

6

2 24

,1

6 6

29,6

8

1 20

,0

1 12

48

,6

5 7

30,6

5

3 32

,4

7 4

23,5

0

10

57,7

3

3 33

,3

33

4 4

36,1

6

4 27

,0

-1

8 64

,9

2 12

44

,4

3 6

41,7

42

5

3 29

,6

-2

5 69

,4

35

1 4

48,1

2

7 48

,1

4 5

33,9

-3

7

75,7

0

10

57,4

1

6 53

,7

3 1

34,8

35

-4

8

82,9

-1

9

65,7

0

11

63,9

2

8 41

,7

-5

2 84

,7

-2

6 71

,3

38

-1

5 68

,5

1 8

48,7

-6

2

86,5

-3

6

76,9

-2

4

72,2

30

0

12

59,1

-7

1

87,4

-4

1

77,8

-3

3

75,0

-1

10

67

,8

-8

3 90

,1

-5

4 81

,5

-4

7 81

,5

-2

4 71

,3

37

-9

1 91

,0

-6

4 85

,2

-5

6 87

,0

-3

3 73

,9

-10

2 92

,8

-7

3 88

,0

-6

3 89

,8

-4

2 75

,7

-11

2 94

,6

-8

4 91

,7

-7

2 91

,7

-5

5 80

,0

-12

1 95

,5

-9

2 93

,5

-8

2 93

,5

-6

3 82

,6

-15

1 96

,4

-10

3 96

,3

-9

1 94

,4

-7

2 84

,3

-21

1 97

,3

-11

1 97

,2

-11

1 95

,4

-8

4 87

,8

-22

1 98

,2

-12

1 98

,1

-12

1 96

,3

-9

3 90

,4

-25

1 99

,1

-15

1 99

,1

-13

1 97

,2

-10

5 94

,8

-26

1 10

0,0

31

-39

1 10

0,0

29

-16

1 98

,1

-11

1 95

,7

-20

1 99

,1

-17

1 96

,5

-27

1 10

0,0

28

-19

2 98

,3

-22

1 99

,1

-23

1 10

0,0

29

FAC

IT-F

= Fu

nctio

nal A

sses

smen

t of C

hron

ic Il

lnes

s The

rapy

– F

atig

ue sc

ale.

HI B

CS=

hig

h-in

tens

ity w

ith b

ehav

iour

cha

nge

supp

ort,

HI=

hig

h-in

tens

ity w

ithou

t beh

avio

ur

chan

ge su

ppor

t, LM

I BC

S=Lo

w-m

oder

ate

inte

nsity

with

beh

avio

ur c

hang

e su

ppor

t , L

MI=

Low

-mod

erat

e in

tens

ity. G

reen

= im

prov

ed fa

tigue

, Yel

low

= un

chan

ged

fatig

ue,

Red

= w

orse

ned

fatig

ue. H

ighe

r sco

re in

dica

tes l

ess f

atig

ue.

App

endi

x 11

. Res

pond

er a

naly

sis f

or E

OR

TC Q

LQ C

30 F

atig

ue su

bsca

le.

HI B

CS

HI

L

MI B

CS

L

MI

Cha

nge

in

C30

FA

n C

umul

ativ

e %

Pr

opor

tion

by c

ate-

gory

Cha

nge

in

C30

FA

n C

umul

ativ

e %

Pr

opor

tion

by c

ate-

gory

Cha

nge

in

C30

FA

n C

umul

ativ

e %

Pr

opor

tion

by c

ate-

gory

Cha

nge

in

C30

FA

n C

umul

ativ

e %

c

-56

1 0.

9 -6

7 1

0.9

-67

1 0.

9 -5

6 2

1.7

-44

1 1.

8 -5

6 3

3.7

-56

4 4.

6 -4

4 4

5.2

-33

9 9.

9 -4

4 3

6.5

-44

8 12

.0

-33

6 10

.4

-28

1 10

.8

-33

9 14

.9

-22

14

25.0

-2

2 14

22

.6

-22

5 15

.3

-22

14

28.0

-1

1 13

37

.0

-11

17

37.4

-11

20

33.3

-1

1 12

39

.2

39.2

-6

1

38.0

38

.0

0 27

60

.9

-6

1 34

.2

34.2

0

36

73.0

33

.8

0 38

73

.1

35.1

11

27

84

.3

0 33

64

.0

29.8

11

15

87

.0

11

15

87.0

17

1

85.2

11

21

82.9

22

12

98

.1

17

1 88

.0

22

11

94.8

22

10

92.0

67

1

99.1

22

9

96.3

33

5

99.1

33

5 96

.4

100

1 10

0.0

27.0

33

3

99.0

26

.9

56

1 10

0.0

44

1 97

.3

44

1 10

0.0

56

1 98

.2

67

1 99

.1

78

1 10

0.0

36.0

HI B

CS=

hig

h-in

tens

ity w

ith b

ehav

iour

cha

nge

supp

ort,

HI=

hig

h-in

tens

ity w

ithou

t beh

avio

ur c

hang

e su

ppor

t, LM

I BC

S=Lo

w-m

oder

ate

inte

nsity

with

beh

avio

ur c

hang

e su

ppor

t, LM

I=Lo

w-m

oder

ate

inte

nsity

. Gre

en=

impr

oved

fatig

ue, Y

ello

w=

unch

ange

d fa

tigue

, Red

= w

orse

ned

fatig

ue. C

hang

e sc

ores

wer

e co

llaps

ed to

clo

sest

inte

ger.

Low

er sc

ore

indi

cate

s les

s fat

igue

.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1656

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-406981

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020