Evaluating the role of an exercise intervention for reducing kinesiophobia in cancer patients: A quantitative study

(1)

Department of Public Health and Caring Sciences

Evaluating the role of an exercise intervention for reducing

kinesiophobia in cancer patients:

A quantitative study

Author

Supervisor

Anna-Maria Cahlenstein

Christopher Bean & Päivi Adolfsson

Master thesis in Public Health 30 credits

Examiner

(2)

Acknowledgements

I want to direct my warmest appreciation to my supervisors, Christopher Bean and Päivi Adolfsson, for their professional guidance, valuable encouragement and constructive critiques of this thesis. I would also like to thank Ingrid Demmelmaier for the valuable feedback, and the rest of the Phys-Can staff for providing me with the data. Finally, I wish to thank Oscar, my friends and my family for their ongoing support during my master’s studies.

(3)

SAMMANFATTNING

Bakgrund: Cancer är en av de främsta folkhälsosjukdomarna och är i dagsläget den näst

största dödsorsaken globalt. Individer med cancer är en grupp som riskerar att utveckla

rörelserädsla (dvs. kinesiofobi) på grund av cancerrelaterad fatigue och smärta. Fysisk träning har visat sig ha en minskande effekt på kinesiofobi hos andra patientgrupper, däremot finns det en kunskapslucka beträffande träning och kinesiofobi hos cancerpatienter.

Syfte: Syftet med studien är att undersöka om en träningsintervention kan minska kinesiofobi

hos patienter med bröst-, prostata eller kolorektalcancer. Vidare syftade studien till att undersöka om träningsintensiteten (hög eller låg-måttlig) samt ett beteendeförändringsstöd har någon ytterligare effekt på kinesiofobi.

Metod: Studien baserades på kvantitativa data från ‘Phys-Can’ – en randomiserad

kontrollerad interventionsstudie. Deltagarna (N=577) delades in i fyra grupper: två med träning på hög intensitet och två med låg intensitet; en av varje intensitetsgrupp hade även ett extra beteendeförändringsstöd. Datan samlades in innan interventionen samt 6 månader efter interventionen. Analyser gjordes i SPSS genom univariata- och upprepade generella linjära modeller (ANCOVA).

Resultat: Trots att resultaten visade att det fanns en generellt låg nivå av kinesiofobisymptom

vid förmätningen (M=22.72, SD=4.81), fanns det en statistiskt signifikant minskning av kinesiofobisymptom efter träningsinterventionen (p<.001). Vidare visade resultaten att det inte fanns några skillnader mellan de fyra grupperna (p=.415).

Slutsats: Låg-måttlig samt högintensiv träning kan minska symptom av kinesiofobi hos

patienter med cancer, oberoende av extra beteende förändringsstöd. Det finns ett behov av vidare forskning gällande kinesiofobi och fysisk träning hos patienter med cancer.

(4)

ABSTRACT

Background: Cancer is a prominent public health concern and the second leading cause of

death worldwide. Individuals with cancer have a risk of developing fear of movement (i.e., kinesiophobia) due to cancer-related fatigue and pain. Exercise has been shown to reduce kinesiophobia in other groups of patients, however, there is a lack of knowledge regarding exercise and kinesiophobia in cancer patients.

Objective: The aim of this study is to investigate whether an exercise intervention is

associated with reduced kinesiophobia symptoms in patients with breast, prostate or colorectal cancer. Moreover, the study considers whether the intensity of the exercise (high vs. low-moderate) and the addition of behavior change support has any further effect on

kinesiophobia symptoms.

Method: The study used quantitative data from the ‘Phys-Can’ project – a randomized

controlled trial. Participants (N=577) were allocated to four groups: two high intensity, and two low-moderate intensity; one of each intensity-level group receiving additional behavioral change support. Measures were taken at baseline and the end of a 6 month exercise

intervention. Analyses were conducted in SPSS using univariate and repeated measures general linear models (ANCOVA).

Results: Despite generally low levels of kinesiophobia symptoms at baseline (M=22.72,

SD=4.81), a statistically significant decrease in kinesiophobia symptoms was observed following the exercise intervention (p<.001). There were no differences in the change in the outcome between the four groups (p=.415).

Conclusion: Low-moderate and high intensity exercise may reduce kinesiophobia symptoms

in cancer patients, regardless of additional behavior change support. Further research concerning kinesiophobia and exercise in cancer patients is encouraged.

(5)

1. BACKGROUND

Cancer is a prominent public health concern and is currently the second leading cause of death worldwide (WHO, 2018). In Sweden, approximately 60,000 individuals are diagnosed with cancer each year (Socialstyrelsen, 2019a). In addition, statistics from the National Board of Health and Welfare show that cancer is slightly more common among men than women. In Sweden, the age standardized incidence rate for men is 313.4 per 100,000, while for women the rate is 279.8 per 100,000 (Bray et al., 2018). The most common form of cancer among men is prostate cancer, while for women it is breast cancer (Socialstyrelsen, 2019b). In addition to the deleterious primary effects of the disease, cancer may also cause secondary physical and mental health problems in patients, survivors and their relatives (Bergman & Johansson, 2018).

1.1 Cancer

Cancer is a group of approximately 200 diseases, which all have in common an abnormal, uncontrolled growth of cells, that occurs somewhere in the body. It is the interaction between an individual’s genetics and external agents, such as physical carcinogens, chemical

carcinogens and biological carcinogens, which result in the transformation from a normal cell to a cancerous cell. These cells continue to grow until they form a tumor. In some cases, the tumor cells spread to other organs and form metastases. This process varies in speed

depending on the nature of the tumor (WHO, 2018). The metastases are a major cause of death from cancer (Berman & Johansson, 2018).

Lung cancer is the most frequently diagnosed cancer worldwide, and the leading cause of cancer mortality (Ferlay et al., 2015). Other common types of cancer are prostate cancer, breast cancer, skin cancer, and colorectal cancer (Berman & Johansson, 2018; Ferlay et al., 2015). Low- and middle-income countries have a higher prevalence of cancer, and this is where 70% of all cancer-related deaths occur (WHO, 2018). In those countries, a major risk factor for cancer is chronic infections such as human papillomavirus, hepatitis B and C (Plummer et al., 2016).

When studying cancer and physical activity it is important to control for variables such as gender and age since these variables may have a confounding effect. For example, men and

(8)

2

women appear to have different experiences of cancer, and for both men and women their levels of physical activity typically decrease with age (Pudrovska, 2010; Dolezal et al., 2017).

1.1.1 Causes

While there are several factors associated with an increased risk of developing cancer, ageing is one fundamental risk factor. The high prevalence of cancer might therefore be a result of an increasingly ageing population (Berman & Johansson, 2018). However, there are still other risk factor that contribute to a high number of diagnosed cancers, for example unhealthy lifestyles (Grosso et al., 2017). As much as one third of all cancer incidence may be

attributable to unhealthy lifestyle factors, which could be prevented (Berman & Johansson, 2018). The increased risk associated with lifestyle factors may be reduced by limiting tobacco use, lower consumption of alcohol, red and processed meat, sugar sweetened drinks,

processed foods high in fat, starches or sugars, less sedentary time and more physical activity (World Cancer Research Fund & American Institute for Cancer Research, 2018).

1.1.2 Cancer treatments and cancer-related symptoms

Cancer treatment options depend on the type, size and characteristics of the cancer. The primary goal of cancer treatment is to remove the tumor and eliminate the cancer cells. This is typically attempted with some combination of surgery, chemotherapy and radiation therapy. Chemotherapy and radiation therapy can also be used as adjuvant treatment, to kill any remaining cancer cells or to decrease the risk that the cancer will recur. Hormone therapy is another type of adjuvant treatment that may be beneficial. If none of the primary and adjuvant treatment methods manage to remove the cancer cells, palliative treatment remains. The aim of palliative treatment is to decrease the side-effects or symptoms from the treatment or the cancer. For example, pain is a symptom that may be relieved with the right palliative treatment (Bergman & Johansson, 2018).

Cancer survivors may experience depression, anxiety and poor health-related quality of life (Jarrett et al., 2013). Moreover, cancer-related fatigue and pain are frequently reported

symptoms in cancer patients and survivors (Lawrence et al., 2004; Glare et al., 2014; Hofman et al., 2007) and might therefore cause a low level of physical activity, high sedentary time and excessive sleep.

(9)

3

1.2 Physical activity and exercise

Physical activity is defined as any bodily movement produced by skeletal muscles that require energy expenditure (Hu, 2013). When the physical activity is planned, structured, repetitive and aims to improve or maintain one or more components of physical fitness, it is called physical exercise (Hu, 2013). The recommendations from the World Health Organization regarding physical activity for the general population (ages 18-64 years) are: at least 150 minutes of moderate-intensity or 75 minutes of vigorous-intensity aerobic physical activity throughout the week. In addition, muscle-strengthening activities should be done two to three times a week. When expressing the level of physical activity, Metabolic Equivalents (METs) are often used to value the intensity. It is the ratio of the relation between the working

metabolic rate and the resting metabolic rate. Moderate intense physical activity, such as brisk walking, dancing, gardening or housework, noticeably accelerates the heartrate and are

approximately 3-6 METs. High intense physical activity, such as running, fast cycling, aerobics or fast swimming, significantly increases the heartrate and are approximately 6 METs or higher (World Health Organization, 2020).

It is increasingly evident that physical activity has numerous beneficial effects for health. It is associated with a reduced risk of different types of cancer, such as breast cancer and colon cancer (Warburton & Bredin, 2017). Regular physical activity also reduces the risk of other prominent diseases, such as type 2 diabetes, gallstones, ischemic heart disease, and ischemic stroke (Warburton & Bredin, 2017). On the other hand, insufficient physical activity and a sedentary lifestyle increase the risk of these diseases (Warburton & Bredin, 2017). Several studies suggest that physical activity has a pain reducing effect and physical exercise is a recognized treatment for a variety of pain conditions (Sluka, Frey-Law & Hoeger, 2018). Individuals who are physically active have a lower risk of developing chronic pain, while chronic pain has been associated with sedentary behavior (Landmark et al., 2011). Insufficient physical activity might result in increased pain, which may in turn increase the experience of kinesiophobia (i.e., fear of movement) (Vlaeyen & Linton, 2012). It might, therefore, be important to maintain a sufficient level of physical activity and reduce sedentary time, when experiencing some form of pain condition.

1.2.1 Physical activity and cancer

The recommendations regarding physical activity for patients with cancer are the same as for the general population (ages 18-64 years) (Campbell et al., 2019). Nonetheless, the physical

(10)

4

activity should be adapted according to the patient’s general condition (Segal et al., 2017). Cancer patients and survivors may have significant benefits to their health from physical exercise. Poor health-related quality of life, low cardiorespiratory fitness and limited physical functioning are common symptoms in cancer patients and survivors (McNeely et al., 2006; Mishra et al., 2012). These symptoms could improve with physical exercise and therefore physical activity may be beneficial to recovery in this group (McNeely et al., 2006; Mishra et al., 2012).

1.3 Behavior change techniques

Changing behavior, especially in meaningful ways and for prolonged period of times, is often difficult and complex due to the involvement of several components. Interventions regarding behavior change are based on theories of behavior change models and techniques. The aim of behavior change techniques is to influence a specific psychological determinant. A

determinant may be an attitude, a habit, or even self-efficacy (i.e., a person’s own belief in their ability to change). Behavioral change techniques are often used in interventions, with the purpose to prevent a specific disease or health condition. There are several behavior change techniques, which are based on different health beliefs models (Ogden, 2007). For example, Michie et al. (2013) developed the behavior change technique taxonomy (Michie et al., 2013). The Behavioral Change Technique Taxonomy (BCTT) is an extensive hierarchically

organized taxonomy of 93 behavioral change techniques. The 93 behavioral change techniques were divided into 16 clusters; Scheduled consequences, Reward and threat, Repetition and substitution, Antecedents, Associations, Covert learning, Natural

consequences, Feedback and monitoring, Goals and planning, Social support, Comparison of behavior, Self-belief, Comparison of outcomes, Identity, Shaping knowledge and Regulation. The BCTT was developed by experts in psychology, behavioral medicine, and health

promotion in seven countries. It is a useful method when identifying, evaluating and implementing different behavioral change interventions (Michie et al., 2013).

1.3.1 Behavioral change techniques and physical activity

Physical activity and exercise play a central role in health promotion and are often studied in interventions regarding behavior change. The determinants of physical exercise can be either social/political or individual, or a combination of these. Interventions that focus on the social/political factors attempt to improve public health at a population level, for example, by communicating the importance of exercise through public awareness campaigns and

(11)

5

subsidized access to training equipment and services. The individual-level determinants are categorized into modifiable and non-modifiable predictors. Examples of the latter include age, prior education, and smoking history. The modifiable predictors include: BMI,

self-motivation, childhood exercise (i.e., creating heathy habits in childhood for better outcomes in adulthood), positive self-image and knowledge. Individuals who are physically active during their childhood and have a positive self-image, are more likely to exercise as adults (Ogden, 2007). However, good knowledge about the health benefits of physical exercise does not necessarily predict exercise behavior (Ogden, 2007).

1.3.2 Behavioral change techniques, physical activity and cancer

Cancer patients and survivors may receive significant health benefits from regular physical activity (Pudkasam et al., 2018). However, due to pain and cancer-related fatigue, cancer patient and survivors might develop fear and avoidance behaviors towards physical movement and, as a result, have elevated sedentary time. To assist these individuals to

participate in regular physical activity, it may be valuable to use behavioral change techniques or models. With the help of behavioral change techniques, key determinants and barriers may be identified and addressed. An earlier study identified perceived control, self-regulation and self-efficacy as key-determinants for physical activity engagement in patients with breast cancer (Pudkasam et al., 2018).

1.4 Kinesiophobia

Individuals who experience a strong fear of physical movement and activity may be described as having “kinesiophobia”. The term was introduced in 1990 by Kori, Miller and Todd and has since been more clearly defined as “an excessive, irrational, and debilitating fear of physical movement and activity resulting from a feeling of vulnerability to painful injury or reinjury” (Kori, Miller & Toss, 1991; Vlaeyen et al., 1995, p. 240). Kinesiophobia is

commonly measured using the Tampa Scale for Kinesiophobia (TSK) – a description of which is provided in Section 2.4. (Lundberg, Styf, & Carlsson, 2004).

1.4.1 Predictors and associations of kinesiophobia

Besides pain and catastrophizing, little else is known about predictors and factors that may increase the risk of experiencing kinesiophobia. Brännström & Fahlström (2008) observed a difference regarding kinesiophobia among men and women with musculoskeletal pain. It seemed that men rated higher scores of kinesiophobia than women (Brännström & Fahlström,

(12)

6

2008). In patients with anterior cruciate ligament reconstruction, predictors for high levels of kinesiophobia were noted to be the timing (i.e., prolonged time from injury-to-surgery), high preoperative pain levels, male sex and low body mass index (Theunissen et al., 2019). Lower levels of education, and higher levels of depression and anxiety, may also predict

kinesiophobia in individuals with chronic neck and back pain (Bilgin et al., 2019).

Kinesiophobia is well documented among patients with musculoskeletal pain, such as pain in the lower back, neck, shoulders and upper extremities (Brännström & Fahlström, 2008). Brännström & Fahlström (2008) observed that kinesiophobia was found in 56% of patients with musculoskeletal pain. Furthermore, kinesiophobia and avoidance behavior toward physical activity is a common problem among patients experiencing chronic-fatigue

syndrome and fibromyalgia (Nijs et al., 2013). Patients who experience kinesiophobia or fear and avoidance behavior may experience psychological changes such as depression and disability (Vlaeyen et al., 1995; Oskay et al., 2017; Gencay Can, 2019). Kinesiophobia has also been correlated with lower levels of physical activity among persons with arterial hypertension (Kocjan, 2015).

A study from Sweden found a positive association between fear of movement and sedentary time in overweight persons with obstructive sleep apnea (Igelström, Emtner, Lindberg, & Åsenlöf, 2013). Moreover, lower levels of kinesiophobia are associated with higher levels of physical activity among older adults with chronic pain, which indicates that higher levels of kinesiophobia may cause lower levels of physical activity (Larsson, Ekvall Hansson,

Sundquist, & Jakobsson, 2016).

1.4.2 Cancer and kinesiophobia

There are only a small number of studies that have considered kinesiophobia in patients with cancer. In one such study from the Netherlands, the modified TSK – Fatigue (TSK-F) was validated to assess fear of movement in cancer survivors with fatigue (Velthuis et al., 2012). Another study investigated the impact of kinesiophobia on lymphedema, upper extremity function, depression and quality of life in Turkish breast cancer survivors (Gencay Can, 2019). The results showed a correlation between high scores of kinesiophobia and severity of lymphedema. Moreover, Van der Gucht and colleagues (2020) observed that kinesiophobia may be the main contributor to self-reported pain-related disability among women with breast cancer. While it is conceivable that cancer patients are likely to experience kinesiophobia, due

(13)

7

to cancer-related pain, this topic has received limited empirical attention and is in need of further investigation.

1.5 Theoretical framework

1.5.1 Fear-avoidance behavior

Vlaeyen et al. (1995) presented an influential model based on some possible factors that could contribute to the state of chronic pain. The model was defined as the ‘fear-avoidance model’ and aims to describe a person’s behavior and experiences of the pain intensity. Avoidance behaviors typically appear during states of acute pain and in such cases may be favorable for the healing process, since such behaviors initially protect the injury. However, when an individual develops a prolonged avoidance behavior towards physical activity or physical movement, the avoidance behavior goes from having a healing effect to having a pain stimulating effect (Vlaeyen et al., 1995).

The model (see Figure 1) emphasizes the way a person’s thoughts influence the intensity of the pain. Negative thinking, and especially catastrophic thinking (i.e., imagining the worst possible outcomes), regarding the pain often plays a role in heightening the intensity of the pain. In most cases, when an injury occurs, the painful experience does not cause any sustained fear of movement or reinjury, since the injured person has strategies to cope with the fear of movement/reinjury. As a result, the injury has the opportunity to recover. Experiencing catastrophic thoughts is a strong predictor for developing fear and avoidance behaviors. In some cases, the experience of the pain, accompanied by catastrophic thinking, creates fear of movement and/or reinjury. As a result of not confronting the fear, the

avoidance behaviors occur and may manifest to greater symptoms such as disability, disuse (which can cause atrophy and disability), and depression, which may all increase the pain experience (Vlaeyen et al., 1995). Since the model was first presented, it has developed and the current version also considers other factors that may affect the fear and avoidance behavior, such as self-efficacy, hypervigilance and information about the illness that may cause negative thinking (Vlaeyen & Linton, 2012).

(14)

8

Figure 1. Illustration of the Fear-Avoidance model reproduced from Vlaeyen & Linton (2012) Note: ‘Injury’ is placed at the top and center of the model. The individual’s appraisal of the painful experiences results in either ‘no fear’, which leads to confrontation of the pain and eventual recovery. If ‘pain catastrophizing’ occurs, this can lead to a feedback loop resulting in more negative experiences of the pain.

1.6 Rationale

Cancer is a highly salient public health concern and a major cause of death worldwide. Cancer patients may experience cancer-related fatigue, pain and mental health conditions such as depression and anxiety. Since fear-avoidance behaviors have been associated with these symptoms, it is also likely that cancer patients may experience kinesiophobia symptoms. However, few previous studies have investigated the associations between kinesiophobia symptoms and an exercise intervention in cancer patients. Hence, there is little known on the subject. The main hypothesis of the current study is that exposure to an exercise intervention may result in a decreased level of kinesiophobia symptoms in cancer patients.

(15)

9

1.7 Study aim and objectives

The aim of the study was to investigate whether participation in an exercise intervention (i.e., exposure) is associated with reduced kinesiophobia symptoms in cancer patients. Further analyses were conducted to consider:

i. If there is a dose-response relationship between the level of exercise intensity and any reduction in the kinesiophobia symptoms, and

ii. whether any such relationship is moderated by the addition of behavior change support.

2. METHOD

2.1 The ‘Phys-Can’ study

The present study used data from a larger project: the ‘Phys-Can’ study, which is a randomized controlled multicenter trial designed to test the effect of physical activity

intervention (low-to-moderate or high intensity), with or without additional behavior change support techniques, on cancer-related fatigue and health-related quality of life. The study did not include a control group. The sample was stratified and included participants from three hospitals – one in Lund/Malmö, Uppsala and Linköping. These hospitals had oncology departments, which had the possibility to cooperate with Uppsala University in the research and were therefore included in the sample. There were nine stratified groups, since each hospital comprised three groups. The groups were stratified according to the three different cancer diagnoses: breast cancer, prostate cancer, and colorectal cancer.

2.2 Study design

The present study assessed whether exposure to an exercise intervention reduced fear of movement (i.e., kinesiophobia symptoms) in cancer patients. The data was measured with quantitative methods, such as standardized inventories, therefore a quantitative design was conducted. Quantitative research is useful when studying variables in numeric form and analyzing the data statistically (Creswell, 2012). The key variable was kinesiophobia symptoms and these were measured at baseline (T0) and post intervention; 6 months from baseline (T3). Kinesiophobia was measured with a Modified Swedish version of the Tampa

(16)

10

Scale for Kinesiophobia. Other data was collected at intervening time points (T1 and T2), but that data was not relevant to the current study.

2.3 Sample selection

The sample comprise pre-existing data from the Phys-Can study. The original sampling and data collection processes involved random selection from three different Swedish university hospitals in Uppsala, Linköping and Lund/Malmö, and was stratified by cancer type. The Phys-Can study included approximately 600 patients diagnosed with breast, colorectal or prostate cancer. Individuals with cognitive disorders or severe emotional instability, those who were not able to execute basic daily activities, and those with other disabling co-morbid conditions which could contraindicate physical exercise, were not included in the study (Berntsen et al., 2017). Of the 600 participants originally recruited, 577 were randomized and included in the present study (Figure 2). Further information about the characteristics of the participants are provided in Table 2.

2.4 Data collection methods

The Phys-Can publication board provided permission and access to use the data for the present study. In the present study, kinesiophobia symptoms were evaluated using a 14-item Swedish version of the Tampa Scale for Kinesiophobia (TSK-SV-14), which is a shortened and translated version of the original TSK (Miller, Kori & Todd, 1991). The 14-item questionnaire asks respondents to rate their level of agreement to 14 statements about their attitudes to physical activity (see Appendices A and B, for the Swedish version of the

questionnaire and an English translation, respectively). Each statement (e.g., ‘I cannot do the same things as others because there is too much risk of being injured’) is answered with a 4-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree). Four items (3, 6, 9 and 13) are reverse-worded statements. Total scores can range from 14 to 56. The TSK-SV-14 is a shortened version of the previously validated TSK-SV-17 (Lundberg, Styf & Carlsson, 2004). For the TSK-SV-17, scores greater than 37 indicate a high degree of kinesiophobia (Vlaeyen et al., 1995). The TSK-SV-14 has three fewer items that the 17-item version, which suggests that a score of 25 may be indicative of a high degree of kinesiophobia when the 14-item version of the measure is used. Regarding the 17-item version, other cut-off scores have also been proposed, including 35.5 and 40 (Barke et al., 2012; Vlaeyen et al., 1999). Neblett et al. (2015) suggested that four severity levels could provide better guidelines

(17)

11

for interpreting the kinesiophobia score, instead of a cut-off score. In the present study, reliability analysis of the TSK-SV-14 suggests internal consistency was generally acceptable at both T0 (Cronbach’s alpha = .63) and T3 (Cronbach’s alpha = .61).

Figure 2. Diagram of participant flow from enrolment to follow-up measurement. Adapted from Berntsen et al., (2017). Note: H = High intensity training; H+BCS = High intensity training plus additional behavioral change support; LM = Low-moderate intensity training; LM+BCS = Low-moderate intensity training plus additional behavioral change support.

2.5 Procedure

The data was collected from newly diagnosed cancer patients in three university hospitals in Sweden. Initially patients received written information about the Phys-Can study. They were informed that they did not show any signs of medical contraindications concerning their participation in the Phys-Can study and those who were eligible for the study were given

(18)

12

detailed written and verbal information from the research staff. Thereafter, potential

participants had a chance to ask questions, consent to participate in the study and the baseline assessments were made.

As outlined in Figure 2, participants were randomized into four intervention groups: high intensity training with additional behavior change support (H+BCS), high intensity training without additional behavior change support (H), low to moderate intensity training with additional behavior change support (LM+BCS), and low to moderate intensity training without additional behavior change support (LM). The exercise intervention period lasted for 6 months and included cardiovascular endurance training and resistance training two times a week. The cardiovascular endurance training was performed at home. The resistance training was performed in a public gym and supervised by physiotherapists and personal trainers. All participants were provided with behavior change support based on the behavior change technique taxonomy developed by Michie et al. (2013). As summarized in Table 1, the behavioral change techniques included social support from coaches and peers, structuring the physical environment with scheduled resistance training sessions and feedback based on monitoring heart rate, cardiopulmonary exercise tests and physical strength tests. In addition, two of the intervention groups were also provided with additional behavior change support, comprising self-monitoring with training logs, individual goal setting, action planning with an initial interview regarding exercise habits and planning, and problem solving.

Table 1. Summary of Behavior Change Support in the Phys-Can study

BCTs to all four groups

Social support – from coaches, and peers (i.e., other group members). Structuring the environment – scheduled exercise sessions.

Feedback – fitness tests and heart rate monitor. Self-monitoring – training logs.

Additional BCTs to the Behavior Change Support (+BCS) groups

Individual goal setting and review of goals. Exercise planning.

(19)

13

Measurement of kinesiophobia symptoms was conducted twice (at T0 and T3). Participants completed their adjuvant oncology treatment before the exercise intervention concluded and the measure for T3 was conducted (Berntsen et al., 2017).

2.6 Ethical considerations

The Phys-Can study was approved by the Regional Ethical Review Board in Uppsala, Sweden (DNR 2014/249). It was also registered in ClinicalTrials.gov (TNR NCT02473003) in

October 2014. Throughout the whole study process, ethical considerations have been applied according to the Swedish Research Council (Vetenskapsrådet, 2017). The data was analyzed securely in Uppsala and handled with confidentiality.

2.7 Data analysis

The present study assessed kinesiophobia symptom scores from T0 and T3. Initially, demographic data was analyzed to describe the sample. Frequency and percentage

distributions were calculated for variables regarding gender, diagnosis, treatment and BMI. Calculation of the age variable provided a mean and standard deviation. Univariate and repeated measures general linear models (one-way analysis of covariance [ANCOVA]) were used to identify possible cross-sectional and prospective associations to determine if

participation in the exercise intervention was associated with reduced symptoms of kinesiophobia.

Further, the analyses considered the potential effect of the exercise intensity (high vs. low-moderate intensity training) and behavioral change support (yes vs. no). The results provided descriptive statistics such as means and standard deviations and associations with significance levels set at .05 (p), parameter estimates (b) with 95% confidence intervals (95% CI).

All analyses were conducted in IBM SPSS version 26. To improve estimate precision and address potential confounding, final models were adjusted for age and sex.

2.8 Variables

Intervention group

The data was numeric and on a nominal level. The values ranged from 1-4 in the following order and were labelled as: ‘1’ = HIGH intensity training WITH additional behavioral

(20)

LOW-14

MODERATE intensity training WITH additional behavioral support; and ‘4’ = LOW-MODERATE intensity training WITHOUT additional behavioral support.

Kinesiophobia symptoms

The scores for kinesiophobia symptoms were measured twice (T0 and T3). The data for the variables was numeric and on a scale level. In the analysis the variables for kinesiophobia symptoms were the dependent variables. Higher scores indicate higher kinesiophobia symptoms and lower scores indicate lower symptoms of kinesiophobia.

Confounding variables

Age and sex could have a confounding effect on the outcome and were therefore used as control variables. Age (years) and sex were also included to describe the participant

characteristics. The data for age and sex were both collected at T0. The variable for age was treated as a continuous variable, while the data for sex was categorical and on a nominal level.

Demographic variables

The demographic variables were collected at T0 and used to identify the participant characteristics.

For the variable concerning BMI, the variable was measured with objective measurements. In the primary data set, the variable was numeric, on a scale level and named “T0_bmi”.

However, a second variable for the T0 measure BMI was made to categorize the values. The values ranged from 1-4 and were labelled as: ‘1’ = Below 18.5kg/m2 (Underweight); ‘2’ = 18.5–25kg/m2 (Normal); ‘3’ = 25–30kg/m2 (Overweight) and ‘4.’ = Above 30kg/m2 (Obese).

The variable concerning education was numeric and on a nominal level. The values ranged from 1-4 and were labelled as:

‘1’ = Elementary school; ‘2’ = High school; ‘3’ = University; and ‘4’ = Other.

The cancer diagnosis variable was numeric and on a nominal level. The values ranged from 1-3 and were labelled as: ‘1’ = Breast cancer; ‘2’ = Colorectal cancer; and ‘1-3’ = Prostate cancer.

The variable concerning the cancer treatment was numeric and on a nominal level. The values range from 1-6 and were labelled as: ‘1’ = Intravenous cytostatic; ‘2’ = Peroral cytostatic; ‘3’

(21)

15

= Intravenous + peroral cytostatic; ‘4’ = Adjuvant radio therapy; ‘5’ = Curative radio therapy (Prostate cancer); and ‘6’ = Endocrine treatment.

3. RESULTS

3.1 Demographic data

Demographic data was collected at T0 and was analyzed to describe the characteristics of the participants. As outlined in Table 2, the majority of the sample were women (80.6%). More than half (58%) had a degree from university/college. The most frequent cancer diagnosis was breast cancer (79.2%) and the most common treatment method was intravenous cytostatic. Further, the majority of the participants had a BMI of 18.5–25kg/m2, which indicated a normal BMI. There were no significant differences between the four groups in terms of their demographic profiles. Further characteristics are presented in Table 2, stratified by training group.

(22)

16

Table 2. Characteristics of the participants at baseline.

Note: a19 missing values; b5 missing values; c35 missing values

Whole sample (n = 577) H + BCS (n = 144) H (n = 144) LM + BCS (n = 145) LM (n = 144) Sex, n (%)

Men 112 (19.4) 29 (20.1) 28 (19.4) 27 (18.6) 28 (19.4) Women 465 (80.6) 115 (79.9) 116 (80.6) 118 (81.4) 116 (80.6) Age (years) mean (SD) 58.73 (11.9) 59.30 (12.9) 58.09 (11.35) 57.97 (11.63) 59.58 (11.79) Education, n (%)a Elementary school 62 (10.7) 16 (11.1) 14 (9.7) 12 (8.3) 20 (13.9) High school 136 (23.6) 40 (27.8) 37 (25.7) 28 (19.3) 31 (21.5) University/College 336 (58.2) 79 (54.9) 84 (58.3) 92 (63.4) 81 (56.3) Other 24 (4.2) 6 (4.2) 3 (2.1) 8 (5.5) 7 (4.9) Diagnosis, n (%) Breast cancer 457 (79.2) 113 (78.5) 115 (79.9) 116 (80) 113 (78.5) Colo-rectal cancer 23 (4.0) 5 (3.5) 6 (4.2) 6 (4.1) 6 (4.2) Prostate cancer 97 (16.8) 26 (18.1) 23 (16) 23 (15.9) 25 (17.4) Treatment, n (%)b Intravenous cytostatic 285 (49.4) 70 (48.6) 68 (47.2) 74 (51) 73 (50.7) Peroral cytostatic 9 (1.6) 1 (0.7) 3 (2.1) 2 (1.4) 3 (2.1) Intravenous + peroral cytostatic 14 (2.4) 5 (3.5) 4 (2.8) 3 (2.1) 2 (1.4) Adjuvant radio therapy 153 (26.5) 38 (26.4) 42 (29.2) 37 (25.5) 36 (25.0) Curative radio therapy (Prostate

cancer) 52 (9.0) 15 (10.4) 12 (8.3) 13 (9) 12 (8.3) Endocrine treatment 59 (10.2) 14 (9.7) 14 (9.7) 15 (10.3) 16 (11.1) BMI (kg/m2), n (%)c Under 18.5 8 (1.4) 2 (1.4) 3 (2.1) 1 (0.7) 2 (1.4) 18.5-25 259 (44.9) 58 (40.3) 69 (47.9) 65 (44.8) 67 (46.5) 25-30 188 (32.6) 44 (30.6) 46 (31.9) 54 (37.2) 44 (30.6) >30 87 (15.1) 28 (19.4) 20 (13.9) 14 (9.7) 25 (17.4)

(23)

17

3.2 Levels of kinesiophobia symptoms

Univariate general linear model analysis (ANCOVA) was conducted to determine the symptoms of kinesiophobia among the participants at baseline(T0). The variable for

kinesiophobia (“newT0_tampa_sum”) had a mean of 22.72 (min: 14, max: 37) and a standard deviation of 4.81; indicating that the majority of the participants had low kinesiophobia symptoms at baseline (Table 2). A higher score on the TSK indicated a higher level of kinesiophobia symptoms. There were 77 missing values at baseline measures for kinesiophobia symptoms.

The training group variable was added to the univariate general linear model, to determine whether there were any differences between the four groups (H+BCS; H; LM+BCS; LM) regarding the levels of kinesiophobia at baseline (T0). Table 3 provides means, Ns and standard deviations for the four different groups, regarding kinesiophobia, measured at baseline. The majority of the participants in the four groups had low symptoms of kinesiophobia at baseline. The means showed a slight tendency to a difference regarding kinesiophobia symptoms in the group allocated to high intensity training without additional behavioral support (23.25), since the mean value was a slightly higher than in the rest of the groups. However, the results of the between-subjects effects indicated that the level of kinesiophobia symptoms at baseline was not statistically significantly different between the four interventions groups, F(3, 496) = .705, p = .549. The between-subjects effects of the analysis are provided in Table 4. The first model displays the crude data and the second model is adjusted for sex and age. In addition, the parameter estimates also indicated that there were no significant differences between the four groups and the level of kinesiophobia symptoms at baseline. Table 5 provides values for the parameter estimates (b) and 95% confidence intervals (95% CI) at baseline.

(24)

18

Table 3. Descriptive statistics of kinesiophobia symptoms among the whole sample and the four different training groups at baseline.

Note: a77 missing value

Mean Std. Deviation N

HIGH intensity training WITH additional behavioral support (H+BCS)

22.65 4.98 125

HIGH intensity training WITHOUT additional behavioral support (H)

23.25 5.53 129

LOW-MODERATE intensity training WITH additional behavioral support (LM+BCS)

22.47 4.71 124

LOW-MODERATE intensity training WITHOUT additional behavioral support (LM)

22.52 5.05 122

(25)

19

Table 4. Between-subjects effects from the univariate general linear model to assess differences regarding kinesiophobia symptoms between training groups at baseline.

Sum of

Squares df Mean Square F p ηp²

Model 1 (Crude) Intercept 258040.64 1 258040.64 11114.07 .000 .957 Training group 49.14 3 16.38 .705 .549 .004 Error 11515.87 496 23.22 Model 2 (Adjusted) Intercept 3286.04 1 3286.04 142.14 .000 .223 Training group 49.33 3 16.44 .711 .546 .004 Error 11420.52 494 23.19

(26)

20

Table 5. Crude and adjusted parameter estimates from univariate the general linear model to assess differences regarding kinesiophobia symptoms between training groups at baseline

Main effects

Model 1 (Crude) Model 2 (Adjusted)

b [95% CI] b [95% CI]

T0 measures

Intercept 22.53 [21.67, 23.38] 22.97 [19.06, 26.88]

HIGH intensity training WITH additional behavioral

support (H+BCS) .12 [-1.08, 1.33] .15 [-1.05, 1.35]

HIGH intensity training WITHOUT additional behavioral

support (H) .72 [-.47, 1.91] .71 [-.49, 1.90]

LOW-MODERATE intensity training WITH additional

behavioral support (LM+BCS) -.06 [-1.26, 1.15]

-.09 [-1.30, 1.11] LOW-MODERATE intensity training WITHOUT

additional behavioral support (LM) (Reference) -- --

(27)

21

3.3 Effect of the exercise intervention on kinesiophobia symptoms

Repeated measures general linear model analysis (ANCOVA) was conducted to evaluate the effect of the exercise intervention on kinesiophobia symptoms in cancer patients. The

available data for the kinesiophobia variables included data from 385 participants, with a total of 192 missing values. Baseline (T0) measures were compared to follow-up (T3) measures, which indicated a decrease in the mean value from T0 (22.40) to T3 (21.26). The means and standard deviations for the whole sample are presented in Table 6.

The results of the repeated measures general linear model analysis (ANCOVA) showed that there was a statistically significant time effect of the exercise intervention in relation to kinesiophobia symptoms. Participants had a significant decrease in kinesiophobia symptoms after the intervention at T3 (i.e., 6 months after baseline measures), F(1,384)=20.32, p<.001. The results are illustrated in Figure 3.

Figure 3. Estimated marginal means for kinesiophobia symptoms in whole sample, measured at baseline (T0) and follow-up (T3). Note: Adjusted for sex and age.

21.00 21.20 21.40 21.60 21.80 22.00 22.20 22.40 22.60 T0 T3 E sti m ate d M ar gi na l M ea ns Measure

(28)

22

The training group variable was included to assess whether there were any differences in the change in kinesiophobia symptoms between the groups in relation to exercise intensity and the additional behavioral change support. The means and standard deviations, regarding kinesiophobia, between the four groups are presented in Table 6.

When including the training group variable in the general linear model analysis (one-way repeated measures ANCOVA) there was not a statistically significant result F(3,481)=.954, p=.415. These results indicate that there was no difference between the groups, regarding the effect of the intensity of the exercise nor the additional behavior change support, on the outcome. As such, all of the groups had a decrease in the level of kinesiophobia symptoms after the intervention, regardless of the exercise intensity or the additional behavior change support. The results are presented in Figure 4 and Tables 7 and 8.

(29)

23

Table 6. Means and standard deviations for kinesiophobia at T0 and T3 measures, among the four different training groups and whole sample

Mean Std. Deviation N

T0 measures

HIGH intensity training WITH additional

behavioral support (H+BCS) 22.44 4.85 96

HIGH intensity training WITHOUT additional

behavioral support (H) 22.97 4.56 96

LOW-MODERATE intensity training WITH

additional behavioral support (LM+BCS) 22.09 4.58 94

LOW-MODERATE intensity training

WITHOUT additional behavioral support (LM) 22.11 4.80 99

Whole sample 22.40 4.70 385

T3 measures

HIGH intensity training WITH additional

behavioral support (H+BCS) 21.35 4.64 96

HIGH intensity training WITHOUT additional

behavioral support (H) 21.74 4.42 96

LOW-MODERATE intensity training WITH

additional behavioral support (LM+BCS) 20.79 4.17 94

WITHOUT additional behavioral support (LM) 21.17 5.25 99

Whole sample 21.26 4.64 385

(30)

24

Figure 4. Estimated marginal means at baseline (T0) and follow-up (T3) for kinesiophobia symptoms between the four training groups. Note: Adjusted for sex and age.

20 20 21 21 22 22 23 23 24 T0 T3 E st im at ed M ar gi n al M ea n s Measure

Effect of exercise intervention on kinesiophobia symptoms, between groups

H+BCS H

LM+BCS LM

(31)

25

Table 7. Between-subjects effects from repeated measures general linear model to assess differences between training groups in kinesiophobia symptoms after exercise intervention

Sum of

Squares df Mean Square F p ηp²

Model 1 (Crude) Intercept 366881.04 1 366881.04 11695.15 .000 .968 Training group 89.78 3 29.93 .954 .415 .005 Error 11952.11 481 31.37 Model 2 (Adjusted) Intercept 5025.25 1 5025.25 159.67 .000 .296 Training group 90.02 3 30.01 .953 .415 .007 Error 11928.30 379 31.47

(32)

26

Table 8. Crude and adjusted parameter estimates from repeated measures general linear models to assess differences between training groups regarding kinesiophobia symptoms after exercise intervention

Main effects

Model 1 (Crude) Model 2 (Adjusted)

b [95% CI] b [95% CI]

T0 measures

Intercept

22.11 [21.18, 23.04] 22.11 [17.69, 26.52]

HIGH intensity training WITH

additional behavioral support. .33 [-1.00, 1.65] .34 [-.99, 1.66] HIGH intensity training WITHOUT

additional behavioral support .86 [-.47, 2.18] .84 [-.48, 2.17]

WITH additional behavioral support _{-.03 [-1.36, 1.31]} _{-.05 [-1.38, 1.29]}

LOW-MODERATE intensity training WITHOUT additional behavioral support (Reference)

- -

T3 measures

Intercept 21.17 [20.25, 22.09] 24.32 [19.98, 28.67]

HIGH intensity training WITH

additional behavioral support. .18 [-1.13, 1.49] .18 [-1.12, 1.49] HIGH intensity training WITHOUT

additional behavioral support _{.57 [-.74, 1.88]} _{.56 [-.75, 1.86]}

WITH additional behavioral support -.38 [-1.70, .93] -.40 [-1.13, .92]

LOW-MODERATE intensity training WITHOUT additional behavioral support (Reference)

- -

(33)

27

4. DISCUSSION

4. 1 Main findings

The main objective of this study was to explore if an exercise intervention had an effect on kinesiophobia symptoms in cancer patients. Moreover, it aimed to explore if the level of intensity and the addition of behavioral change support could have a further effect on the outcome. The results of the present study showed that there were low kinesiophobia

symptoms among the participants at baseline. Further, the results suggested that the exercise intervention was associated with a decrease in the level of kinesiophobia symptoms in cancer patients. However, the results indicated that the level of exercise intensity, as well as the additional behavioral change support, did not have any notable influence on the outcome.

4.1.1 Kinesiophobia symptoms in cancer patients

Kinesiophobia symptoms have previously been reported in patients with pain symptoms, as well as in patients with chronic fatigue syndrome (Bilgin et al., 2019; Brännström &

Fahlström, 2008; Nijs et al., 2013; Larsson et al., 2016). In the present study, both baseline and post intervention measures indicated low levels of kinesiophobia symptoms in the sample. Even though the result in the present study did not indicate that cancer patients experience high or severe symptoms of kinesiophobia, it is still possible that it might be an issue among other groups of cancer patients, due to the risk of cancer-related fatigue and cancer-related pain. In the present study, the experience of pain symptoms and cancer-related fatigue in the participants were not analyzed. More information about the experiences of pain and cancer-related fatigue among the participants might have provided additional context for understanding the results.

4.1.2 Physical exercise and kinesiophobia symptoms

The result of the present study showed that participation in a physical exercise intervention was associated with a decreased level of kinesiophobia symptoms in cancer patients. Kinesiophobia has previously been explored among other groups of patients, such as

musculoskeletal pain patients (Bilgin et al., 2019; Brännström & Fahlström, 2008). One such study, including patients with chronic nonspecific lower back pain, had similar results to the present study, and explored how an exercise intervention, based on pilates training, decreased the level of kinesiophobia (Cruz-Diaz et al., 2018). Even though the participants’

(34)

28

the exercise intervention lasted for 6 months, similar to the present study. These findings are in agreement with Vlaeyen et al. (1995), who suggest that one important factor in eliminating or reducing the fear of movement/reinjury may be exposure to physical activity.

4.1.3 Physical exercise intensity and behavior change support

In the present study, no significant difference in the outcome was observed between the four different intervention groups. One previous trial added cognitive-behavioral physiotherapy (i.e., interactive behavioral modification therapy) to physical exercise, to explore if the intervention improved the outcome in chronic neck pain patients (Thompson et al., 2016). The results showed a statistically significant difference between the intervention group and the control group, which indicated that the additional cognitive-behavioral physiotherapy could be beneficial in decreasing kinesiophobia. In the present study, the exercise

interventions with the additional behavioral change support did not provide any different result from the other groups. Nonetheless, in the present study the characteristics of the participants differed from Thompson et al. (2016) in several aspects. The present study

observed patients with cancer, whereas Thompson et al. (2016) observed patients with chronic neck pain. It is possible that behavior change support combined with an exercise intervention could have the same beneficial effects in cancer patients as in chronic neck pain, therefore, future studies should consider this aspect.

No significant difference was observed between the two levels of exercise intensity in the present study. No previous study has explored the dose-response relationship between exercise intensity and the effect on kinesiophobia symptoms. However, some studies have explored the dose-response relationship between exercise intensity and other physiological and psychological outcomes in cancer patients (Buffart et al., 2014). One such study explored the effect of low and moderate intensity exercise on several physical and psychological outcomes, such as aerobic capacity, body composition, quality of life, fatigue and anxiety. The results were in line with the results in the present study and indicated that there were no significant differences between the two groups (Burnham & Wilcox, 2002). However, Kampshoff et al. (2015) observed some differences between low-moderate intensity exercise and high intensity exercise in cancer patients. The group with high intensity exercise seemed to improve the global quality of life and anxiety, whereas both of the groups improved physical functioning compared to a waiting-list control group.

(35)

29

4.1.4 Kinesiophobia symptoms and exercise intervention in patients with cancer

Few studies have investigated kinesiophobia in cancer patients. One previous study evaluated the effect of a comprehensive rehabilitation program on kinesiophobia for post-treatment patients with cancer (Hanssens et al., 2011). The rehabilitation program included physical exercise (60 minutes/three times a week), psychoeducation and individual counseling and lasted for 12 weeks. The results of the study did not show any statistically significant effect on kinesiophobia, in contrast to the present study. However, the exercise intervention in the present study lasted for 6 months and included a larger sample size which might contribute to the different result. Nevertheless, there is a need for further studies to further explore these issues.

4.1.5 Fear-Avoidance and kinesiophobia

Despite the fact that the majority of the participants in the present study did not show severe symptoms of kinesiophobia, patients with cancer might be in risk of experiencing fear-avoidance behavior and developing kinesiophobia. Unlike previous studies, the TSK-SV-14 was used for measuring kinesiophobia in the present study. Using other versions of the TSK-SV, such as the 11, 13 or 17-item versions, could possibly have contributed to different results. While there are recommended cut-off scores for the TSK, it has been suggested that it would be better to treat the scores for kinesiophobia symptoms on severity levels rather than using arbitrary cut-off scores (Neblett et al., 2015). Another suggestion is to include other measures for fear-avoidance behaviors, such as the ‘Fear-Avoidance Belief Questionnaire’, when exploring kinesiophobia in cancer patients. Including more than one measure and treating the kinesiophobia scores according to a severity scale, might improve the interpretation of the results, since there is a limited knowledge on this subject. Regular physical activity and exercise are important factors in maintaining, and improving, the health status of cancer patients (Buffart et al., 2014). Since cancer is a prominent public health concern, preventing kinesiophobia and fear-avoidance behaviors in cancer patients should be a matter of priority in the public health area.

The participants in the present study did not seem to experience severe symptoms of kinesiophobia. When interpreting these results according to the Fear-Avoidance model by Vlayen & Linton (2012), it is possible that most of the participants in this study, had high self-efficacy, did not have catastrophic thoughts and a good mental health and wellbeing.

(36)

30

These factors could have enabled them to confront the pain that they might have experienced, resulting in low levels of kinesiophobia symptoms.

4.2 Methodological discussion

4.2.1 Strengths

The main strength of this study is its large sample size and the randomization from the larger Phys-Can project. The original sample included 600 participants, of which 577 were

randomized and included in the present study. A large sample and randomization increase the trustworthiness of a study. A randomized and controlled study design is the preferred design when exploring the effect of an intervention and therefore, it is often called ‘the gold

standard’ (Hariton & Locascio, 2018).

Regarding the exercise duration, the intervention period lasted for 6 months. Such a duration has previously been shown to be appropriate in exercise interventions to improve several physiological and psychological factors for cancer patients (Buffart et al., 2014). As such, the 6 month exercise duration was a strength in the present study.

4.2.2 Limitations

Like other studies, the present study has some limitations. In the sample of those who were randomized and included in the present study (n=577), there was a majority of women (80.6%). Since men and women appear to have different experiences of both cancer and kinesiophobia, it would have been better to have a more equal proportion of men and women in the sample. This would also have allowed for evaluating potential differences between men and women in regard to the outcome. In the present study, sex was controlled for in the adjusted analyses (Model 2). Further, there was a majority of participants with breast cancer (79.2%). Due to the limited knowledge about kinesiophobia in cancer patients, it is unknown if the type of cancer could affect the experience of kinesiophobia symptoms in relation to physical activity.

In the present study, more than half of the participants had a degree from university or college. Individuals with higher education are more prone to take participation in physical activities (Stockholms läns landsting & Karolinska institutet, 2011). Unlike sex and age, education was not controlled for in the adjusted analyses. Therefore, education level could have had a confounding effect on the results. To address potential confounding, education

(37)

31

level should be controlled for in future studies.

The data regarding kinesiophobia was measured with the TSK-SV-14. Like all questionnaires, there are some limitations and weaknesses in the obtained data. First, the questionnaire is providing data that is self-reported from the participants, which makes the data less objective. The results might be affected by response bias, in which the responding participant might be overly negative or positive in his or her responses, in the moment when answering the questionnaire. To prevent response bias from affecting the data, the participants should answer the same questionnaire more than once, in a close time interval to the first measure. Their responses should provide similar answers at each point of measure, if the answers differ, there may be bias (Creswell, 2012).

Another limitation is that the sample was gathered from three different university hospitals in four larger cities – Uppsala, Lund/Malmö and Linköping. These cities are geographically oriented in the south and middle parts of Sweden and, therefore, the sample does not include participants from the northern parts of the country. It is possible that different results would have been obtained if the study also represented the north of Sweden, since there might be some differences regarding demography, health, wellbeing and socio-economic status between different regions in Sweden. Further, the sample only included patients with breast, prostate, or colorectal cancer and excluded those with cognitive disorders or severe emotional instability, those who were not able to execute basic daily activities, and those with other disabling co-morbid conditions which could contraindicate physical exercise. Therefore, the results can only be applied on a rather narrow population, which affects the generalizability of the study. Breast and prostate cancer are two of the most common cancers and also two forms of cancer with best prognosis. It is likely that the inclusion of participants with other forms of cancer could contribute to different results.

The largest limitation of the present study is that it does not include a control group. This makes it difficult to assess whether the actual exercise intervention was the factor that decreased the level of kinesiophobia. The decrease in the level of kinesiophobia symptoms could simply be an effect of the time and not an effect of the intervention. These limitations should be considered in future studies.

(38)

32

4.2.3 Ethical considerations

The present study was conducted in accordance to Swedish Research Council

(Vetenskapsrådet, 2017). The participants in the Phys-Can study were given both verbal and written information about the study and provided written consent. Participation in the intervention was voluntary, and the participants had the opportunity to withdraw their participation at any time. The data set that was analyzed in the present study comprised anonymized data to ensure confidentiality. The identity of the participants remained anonymous to the writer. The present study only assessed the relevant components and variables for the study questions of interest. If any differences between the four groups would have been observed, resulting in a statistically significant effect on kinesiophobia, it would have been ethical to consider providing the other groups with the same intervention that demonstrated the best effect.

4.2.4 Clinical implications

The findings of this study suggest that cancer patients could benefit from an exercise

intervention with the aim to reduce kinesiophobia symptoms. Even though the majority of the participants had a low level of kinesiophobia at baseline, there was still a decrease after the intervention. Cancer patients may experience factors that could contribute to kinesiophobia, such as pain, anxiety and caner-related fatigue, and even though the participants in this study had low levels of kinesiophobia, there is still a risk that it might be an issue among other cancer patients. Further studies, which address the limitations noted here, are required to better understand the relationship between kinesiophobia symptoms and physical activity in cancer patients.

4.2.5 Reliability and validity

The reliability and validity of the TSK has been evaluated in several studies, which have mainly focused on the psychometric properties of the TSK (Bunketorp et al., 2005;

Burwinkle, Robinson & Turk, 2005; Lundberg, Styf & Carlsson, 2004; French et al., 2007; Geisser, Haig & Theisen, 2000). The results have shown that the TSK is applicable in different groups of individual, countries, languages, symptoms and diseases, including the Swedish version TSK-SV (Lundberg, Styf & Carlsson, 2004). The TSK-SV is therefore valid and reliable to use in studies and clinically, in a Swedish population (Lundberg, Styf,

Carlsson, 2009). While a higher internal consistency than that reported in the present study may be desirable, psychometric assessments of reliability (e.g., Cronbach’s alpha) are only

(39)

33

one factor that should be used when evaluating the suitability of theoretically-informed measures such as the TSK (Taber, 2018).

Since the present study did not include a control group, the internal validity may be affected. Including a control group would have provided more validity to the study, since it could have contributed more information concerning potentially naturally occurring changes in the outcome variable over time. Regarding the results, it is difficult to explain that the decrease of kinesiophobia symptoms is mainly due to the exercise intervention. It is not certain that that the decreased level of kinesiophobia symptoms could have appeared for other reasons, during the time of the intervention period. However, as mentioned before, several studies indicate that physical activity and exercise interventions do have an effect on reducing kinesiophobia symptoms in other groups of patients with different pain symptoms.

4.3 Conclusion

In line with other studies, participation in the exercise intervention appeared to be associated with reduced symptoms of kinesiophobia over time. The level of exercise intensity did not have any additional effect on the outcome. Further, the outcome was not moderated by the additional behavior change support. Participation in the exercise intervention was associated with reduced symptoms of kinesiophobia regardless of the exercise intensity and the

additional behavior change support. Following an extensive search of the literature, to the author’s knowledge, this is the only study which has focused on the effects of an exercise intervention in relation to kinesiophobia symptoms in cancer patients. The results in the present study suggest that future studies could continue to explore the relationship between the different intensity of exercise-based interventions and behavior change support in relation to kinesiophobia symptoms in cancer patients. Future studies, which should aim to address the limitations noted here – in particular the need to include a control group, could provide more information and be beneficial for informing the treatment and rehabilitation of cancer patients and survivors. Preventing kinesiophobia and fear-avoidance behaviors in cancer patients is an important public health concern and should be addressed in future health promotion projects.

(40)

34

REFERENCES

Bergman, O., & Johansson, E. (2018). Cancer i siffror 2018: Populärvetenskapliga fakta om cancer.

Barke, A., Baudewig, J., Schmidt-Samoa, C., Dechent, P., & Kröner-Herwig, B. (2012). Neural correlates of fear of movement in high and low fear-avoidant chronic low back pain patients: An event-related fMRI study: Pain, 153(3), 540–552.

https://doi.org/10.1016/j.pain.2011.11.012

Berntsen, S., Aaronson, N. K., Buffart, L., Börjeson, S., Demmelmaier, I., Hellbom, M., Hojman, P., Igelström, H., Johansson, B., Pingel, R., Raastad, T., Velikova, G., Åsenlöf, P., & Nordin, K. (2017). 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, 17(1), 1-12.

https://doi.org/10.1186/s12885-017-3197-5

Bilgin, S., Cetin, H., Karakaya, J., & Kose, N. (2019). Multivariate analysis of risk factors predisposing to kinesiophobia in persons with chronic low back and neck pain. Journal of Manipulative and Physiological Therapeutics, 42(8), 565–571.

https://doi.org/10.1016/j.jmpt.2019.02.009

Bray, F., Ferlay, J., Soerjomataram, I., Siegel, R. L., Torre, L. A., & Jemal, A. (2018). Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide

(41)

35

for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians, 68(6), 394–424.

https://doi.org/10.3322/caac.21492

Brännström, H., & Fahlström, M. (2008). Kinesiophobia in patients with chronic

musculoskeletal pain: Differences between men and women. Journal of Rehabilitation Medicine, 40(5), 375–380. https://doi.org/10.2340/16501977-0186

Buffart, L. M., Galvão, D. A., Brug, J., Chinapaw, M. J. M., & Newton, R. U. (2014). Evidence-based physical activity guidelines for cancer survivors: Current guidelines, knowledge gaps and future research directions. Cancer Treatment Reviews, 40(2), 327–340. https://doi.org/10.1016/j.ctrv.2013.06.007

Bunketorp, L., Carlsson, J., Kowalski, J., & Stener-victorin, E. (2005). Evaluating the reliability of multi-item scales: A non-parametric approach to the ordered categorical structure of data collected with the Swedish version of the Tampa Scale for

Kinesiophobia and the Self-Efficacy Scale. Journal of Rehabilitation Medicine, 37(5), 330–334. https://doi.org/10.1080/16501970510036411

Burnham, T. R., & Wilcox, A. (2002). Effects of exercise on physiological and psychological variables in cancer survivors: Medicine & Science in Sports & Exercise, 34(12), 1863– 1867. https://doi.org/10.1097/00005768-200212000-00001

Burwinkle, T., Robinson, J. P., & Turk, D. C. (2005). Fear of movement: Factor structure of the Tampa Scale of Kinesiophobia in patients with Fibromyalgia Syndrome. The Journal of Pain, 6(6), 384–391. https://doi.org/10.1016/j.jpain.2005.01.355

Evaluating the role of an exercise intervention for reducing kinesiophobia in cancer patients: A quantitative study

Department of Public Health and Caring Sciences