A large body of evidence supports the positive effects of physical activity (PA) and exercise for cancer patients all along the cancer trajectory (Fong et al., 2012; Furmaniak et al., 2016; Christensen et al., 2018). Exercise has shown to increase physical fitness (Strasser et al., 2013; Scott et al., 2018; Sweegers et al., 2019), health related quality of life (Mishra et al., 2012; Buffart et al., 2017) and improvement in cancer-related fatigue (CRF; Mustian et al., 2017; Van Vulpen et al., 2020). Prior to initial treatment, exercise can improve the fitness of cancer patients and reduce the likelihood of certain complications (Vermillion et al., 2018). During treatment, exercise has shown preventive health effects on cardio- and neurotoxicity (Chen et al., 2017; Kleckner et al., 2018); furthermore, it is closely associated with reducing cancer mortality and the probability of recurrence (Cormie et al., 2017; Friedenreich et al., 2020). Despite the benefits of regular PA, cancer patients tend to reduce their PA-levels after their diagnosis (Mason et al., 2013; Pugh et al., 2020) and only a minority of cancer survivors meet the PA guidelines (Blanchard et al., 2008; Lin et al., 2019). This can be contributed to particular internal as well as external barriers (e.g., the availability of programs and qualified facilities) cancer patients have to overcome, such as the frequently reported internal barrier of CRF (Brunet et al., 2013; Hardcastle et al., 2018).

CRF is one of the most common and distressing symptoms reported by cancer patients, produced by the cancer as well as its treatment (Bower, 2014; Berger et al., 2015). It is defined as a feeling intensive physical, emotional and cognitive exhaustion, without getting relieve after resting or sleeping (Weis, 2011). By the time cancer is diagnosed, 40% of patients already report signs of CRF (Rüffer, 2019). This proportion goes up to 60–100% during cancer treatments like chemotherapy and radiotherapy (Weis, 2011). CRF may even continue to impact the lives of long-term survivors for up to 10 years, as noted in 25–33% of cases (Bower, 2014). As of right now, there is no promising pharmacological treatment for CRF (Bower, 2014), with the most effective approach to mitigate CRF being PA (Mustian et al., 2017). This means cancer patients are often trapped in a vicious circle, since CRF prevents PA and thus increases CRF symptoms.

Patients need certain skills, knowledge and motivation (Schmid et al., 2020), i.e., they need competence, in order to positively engage in an active lifestyle. Thus, Pfeifer et al. (2013) developed the model of domain-specific physical activity-related health competence (PAHCO), inspired by the health model of Lenartz (2012), which focuses on promoting person-oriented competencies that enable a person to integrate an effective PA level into daily life. It contains three sub-competences (movement competence, control competence, and PA-specific self-regulation competence). Movement competence is a necessary condition that enables a person to realize the health benefits of exercise and everyday PA. Basically, movement competence comprises of motor skills and abilities. Control competence describes the degree a person is capable of utilizing their knowledge, based on the positive effects of exercise, how to implement health enhancing PA in their daily life, and how to control physical load in regards to their body signals, such as breathing, perceived exertion or heart rate. The third sub-competence, PA-specific self-regulation competence, encompasses the motivational and volitional abilities of a person to engage in health enhancing PA and implementing it into daily life (Sudeck and Pfeifer, 2016). Each of these sub-competencies contributes to regular health-promoting activity behavior. Sudeck and Pfeifer proposed a questionnaire to operationalize PAHCO, which predicts the PA-levels of patients undergoing medical rehabilitation (Sudeck and Pfeifer, 2016; Carl et al., 2020a).

A typical characteristic of CRF is its impairing effect on motivation (Brunet et al., 2013; Hardcastle et al., 2018); furthermore, CRF shows a positive relationship with depression (Bower et al., 2000; Brown et al., 2013). Considering that motivation and volition help using PA as constituents of the PAHCO model, PAHCO can play an important role in the correlation between CRF and LTPA. Since physical fatigue (PF) shows a certain sensitivity to PA (Van Vulpen et al., 2016), we would expect the relationships to be the strongest for this fatigue dimension. Because we were interested in the voluntary engagement of PA and exercise, in the hypotheses we focused on leisure time physical activity (LTPA). We therefore derived following five hypotheses.

We first investigated the factorial validity of the PAHCO questionnaire in an US based sample, since the PAHCO questionnaire has thus far only been investigated in German speaking populations (Sudeck and Pfeifer, 2016), afterwards we tested the hypotheses in a structural model.

This analysis was able to confirm the factorial validity for the American-English version of the PAHCO, involving US-American cancer patients and survivors. The loadings and correlations between the latent PAHCO variables in our model roughly coincides with the correlational structure of the German original publication, reporting slightly lower loadings on MR and CC as well as slightly higher loadings on SC. Furthermore, we identified a moderate positive association between PAHCO-dimensions and LTPA, which conforms to observations made in other populations and also confirms H₂ (Sudeck and Pfeifer, 2016; Carl et al., 2020a). Of all three PAHCO dimensions, MR and LTPA showed the strongest association, while the association between LTPA and the other two remaining PAHCO dimensions had slightly smaller correlation coefficients. Ostensibly, the strong correlation between MR and LTPA can be attributed to the items loading on MR, addressing depression, stress and inner tension, which are elevated for cancer patients in comparison to the general public or even patients with other chronic diseases (Gil et al., 2012; Hartung et al., 2017; Rao et al., 2019). This may suggest that physical activity has an instrumental functionality when it comes to cancer patients since it helps them cope with psychological symptoms. It also goes in line with the recently published ACSM Roundtable on Physical Activity Guidelines for Cancer Patients (Campbell et al., 2019) which confirms a positive effect of exercise on depressive symptoms and anxiety. Nevertheless, the three correlation coefficients between LTPA and the PAHCO dimensions do not differ significantly from each other, considering their 95% confidence intervals, hence, conclusions must be drawn with caution and can be premature. According to H₃, all PAHCO-dimensions displayed a negative relationship to the CRF-dimensions, however, MR showed the weakest correlation to either of the CRF dimensions, which are usually linked to mood disturbances and have a strong relation with depression (Bower et al., 2000; Brown et al., 2013).

A relationship between the CRF dimensions and LTPA has not been observed, which opposes the literature and H₁, that exhibits a small to moderate negative associations between CRF and PA levels (Kummer et al., 2013; Galiano-Castillo et al., 2014; Romero et al., 2018). This could be partially explained by a possible selection bias due to a lower response rate to the SQUASH compared to the other questionnaires (about 59%). Pertaining the questionnaires, participants who chose not to answer the SQUASH scored significantly lower in three of the four items loading on MR, significantly higher in three of seven items loading on PF, and significantly higher on two of four items loading on CF. Thus, the patients who completed the SQUASH and were investigated for the final analysis presented a higher PAHCO and lower CRF than the others. Bearing in mind that the SQUASH items are more arduous to answer than the Likert-scaled questions of the other questionnaires, it appeared that people with a positive attitude toward PA in general and LTPA in particular would invest more effort into completing the survey.

We could not observe a ceiling effect or lack of variance in the LTPA-Data; though, a plausible correlation between the PAHCO-dimensions and PA as well as between the PAHCO dimensions and EORTC FA12 were established. Accordingly, we should have been able to identify a relevant association, despite the likelihood of selection bias.

A conceivable mediation for the effect of CRF on LTPA via PAHCO renders obsolete (H₄ and H₅) given the lack of association between the two variables. Considering the cross-sectional character of the study, derived conclusions should not indicate causation, and further investigations need to examine relationship between CRF, LTPA and the role of PAHCO in a longitudinal manner. Subsequently, the time-lagged cross-correlations between CRF and LTPA would reveal further information about the relationship between these variables along the cancer trajectory. Due to the decline in PA-levels patients experience following cancer diagnosis and treatment, finding the causes as well as potential interventions are of great interests. Besides its explanatory value, increasing PAHCO in patients could potentially increase adherence to given exercise programs and an enhanced active lifestyle. One of the most common reported intrinsic barriers of cancer patients toward LTPA is the lack of self-efficacy (Brunet et al., 2013; Phillips and McAuley, 2013; Ungar et al., 2016; Hardcastle et al., 2018). In a systematic review the perception of control over ones well-being was one of the most common facilitators of exercise (Clifford et al., 2018). Another review has shown a strong prediction of instrumental attitude and planning on LTPA levels (Vallance et al., 2008). In fact, all these constructs can be considered constituents of PAHCO.

For patient care, PAHCO could be used to assess the level of support patients require to initiate an exercise routine and LTPA. By stratifying patients in accordance with their PAHCO levels, exercise therapists would be able to provide educational instructions tailored particularly to every patient's needs. By increasing PAHCO, patients would presumably feel empowered to engage in exercise and LTPA independently. In regards to health literacy, cancer patients with lower levels of knowledge seemed to necessitate expert support more than patients with higher levels, who were also more likely to perform research themselves (Morris et al., 2013). Additionally, the stratification process may prevent the misallocation of resources by providing especially patients in need with the highest level of guidance. Consequently, PAHCO is a promising concept to promote PA in cancer patients, which would result in vital lifestyle changes and increased participation in exercise programs.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by PennState College of Medicine internal review board (Study ID: HRP-591-PEXO, Clincial Trials.gov: NCT04328038). The patients/participants provided their written informed consent to participate in this study.

MK: data collection, conducting the statistical analysis, and writing the manuscript. CK: data entry, descriptive statistics, and writing the manuscript. RW and KS: study design, coordination of data collection, and critical review of the manuscript. JW: study design, conducting the statistical analysis, and critical review of the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer SS declared a shared affiliation, with no collaboration, with the authors (CK and JW) to the handling editor at the time of review.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.