Physical activity for people living with cancer: Knowledge, attitudes, and practices of general practitioners in Australia.

BACKGROUND: Healthcare professionals' (Oncologists, doctors, and nurses) physical activity (PA) recommendations impact patients living with cancer PA levels. General practitioners (GPs) monitor the overall health of patients living with cancer throughout their treatment journey. This is the first study to explore GP's knowledge, attitudes and practices of PA for patients living with cancer.
METHODS: GPs who see patients living with cancer regularly (n = 111) completed a survey based on The Theory of Planned Behaviour (TPB). Participants (GP's) reported knowledge, attitudes, perceived behaviour control and subjective norms of PA within the cancer population. GP recommendation and referral rates of PA were reported. Principal component analysis was conducted to establish a set of survey items aligned to TPB constructs (attitude, subjective norms, perceived control), and multiple regression analyses characterised associations between these predictor variables and (a) recommendation; and (b) referral-of PA to cancer patients.
RESULTS: GPs (n = 111) recommended PA to 41-60% of their patients and referred 1-20% to PA programs. Multiple regression models significantly predicted the percent of patients recommended PA, p < .0005 adj. R2 = 0.40 and referred PA, p < .0005, adj. R2 = 0.21. GP attitudes and perceived behavioural control and GP's own activity levels were significant predictors of whether patients were recommended and referred for PA, p<0.05.
CONCLUSION: GPs reported positive attitudes and perceptions towards promoting PA for their patients living with cancer. Despite having a positive correlation between PA recommendations and referral rates, a gap was evident between GP's PA beliefs and their individual referral practices. More GP's willing to promote and refer their patients for PA, would improve the physical and mental health outcomes of the cancer population.

Background

Globally, one in three people will to develop cancer during their lifetime [1]. Physical inactivity and poor dietary behaviours are significant contributing factors to cancer-related deaths [2]. Increased levels of physical activity (PA) has been shown to have a positive effect on reducing the development of secondary acute and chronic conditions [3, 4], including: comorbidity development [5]; cognitive impairment [6]); neurological issues [7]; cardiovascular and respiratory conditions [8]; lymphoedema [9]; cancer-related fatigue [10]; decreased bone mineral density [11]; and increased pain levels [12]. Although this evidence exists, PA levels within the cancer population are still sub-optimal, and reduced health outcomes are extremely common [13].

Healthcare professionals (HCPs) (oncologists, doctors and nurses) play a vital role in influencing patients’ overall health and lifestyle behaviours [14]. It has been established that patients who have been told to be active by their HCPs’ have improved exercise and adherence levels [15, 16]. HCPs position of authority creates a sense of respect from their patients and for the advice that they provide, meaning that this conversation about PA could improve the health of their patients.

Several psychosocial factors influence HCP behaviours–for example the Theory of Planned Behaviour (TPB) asserts that attitudes towards a behaviour, subjective appraisal of their group norms about the behaviour, and the perception of having sufficient control/choice to engage in the behaviour all precede intentions and actual behaviour. These variables have been shown to explain 31% of variance in observed behaviour in HCPs: a relatively strong finding compared to other theories [17]. The TPB is composed of three constructs (individuals’ attitude, subjective norms, perceived behavioural control) [18]. The first construct is the individuals’ attitude towards behaviour, in this case PA promotion having a positive or negative contribution to a cancer patients’ care. The second construct is subjective norms which is defined by the perceived belief of what others think about a certain behaviour, in this case what other HCP’s believe about the promotion of PA to their patients living with cancer. The final construct, titled perceived behavioural control, questions beliefs on the difficulty to perform the behaviour, in this case providing PA recommendations [17]. The three constructs are theoretically high predictors of individuals behaviour intentions and in turn lead to behaviour predictions [17].

The Clinical Oncology Society of Australia (COSA) released a position statement in 2018 encouraging all HCPs who are involved in the care of patients living with cancer to discuss and recommend “at least 150 minutes of moderate intensity aerobic exercise and two or three moderate intensity resistance exercise sessions each week” [19]. These recommendations match the guidelines for the general population [20]. In addition, the COSA statement suggests that HCPs should refer their patients for exercise advice to specialists such as accredited exercise physiologists, or physiotherapists with experience in cancer care [19]. PA recommendations from HCPs are appreciated and desired by cancer survivors [21-25] but typically this advice is not being given [16, 25, 26]. Jones, et al. (2002) looked at patient perspectives and reported that 97% of patients living with cancer would like PA to be discussed by their HCPs and 67%, believed that they should be referred for further specialised support, but there is still no best practice evidence within more recent literature about who should have this discussion [27, 28]. PA is a modality of treatment, which cancer survivors can safely control themselves when prescribed [29]. Having the autonomy to control this during the unpredictable treatment period benefits prognosis and can improve patients’ mental and physical health [30].

During treatment in Australia, it is reported that patients contact time is spent predominantly with oncology nurses and less time with their oncologists [31]. Health care professionals report that the discussion of PA is best provided by nurses [32-34]. Primarily in cancer care overall health and well-being is managed by GP’s both during and post treatment [35]. However, the knowledge, attitudes and promotion of PA by GPs has not yet been studied. HCPs attitudes towards PA during the cancer treatment phase is evident [16, 24, 36], however, there is limited understanding of this during the post treatment phase [23, 37–39], GPs views and role within this area has not been explored. PA advice may be better-received post treatment when there is additional time to carry out this discussion with patients and they have had some recovery from the side effects of treatment [35]. GPs are well situated and could play a vital role in influencing cancer patient behaviours during this phase potentially positively impacting quality of life, and reducing co-morbidities and the recurrence of cancer [35].

A study conducted over a ten year period on Australian GPs, which tracked their attitudes and practices towards PA counselling in general population, demonstrated a significant increase in knowledge, confidence and the belief that it is part of their role to promote PA to their patients, however it was shown that GP’s likelihood to promote physical activity was low [40]. Despite this, the TPB is based on the premise that attitudes, subjective norms, and perceived behavioural control affect a person’s behavioural intention and this has a direct effect on behaviour [17]. GP’s intention to engage in recommending PA to patients and referring patients living with cancer to exercise specialists is theoretically influenced by the value of the individuals position on the behaviour (attitudes), how it can be performed, the views of their fellow GP’s (subjective norms) and the perception that the behaviour is within their control [17]. Therefore, the objective of this unique study was to establish predictors and determinants of GPs recommendations and referral behaviours for PA within the cancer population. Utilising the TPB, we hypothesise that; GP attitudes, perceived behavioural control and subjective norms will all be significant predictors of GPs recommendations and referral practices for PA in addition to their personal PA participation.

Materials and methods

Study instrument

The survey was developed based on the Theory of Planned Behaviour (TPB) Survey and items were located and adapted from previous research investigating PA attitudes and practices of other HCPs involved in cancer care [15, 16, 24, 32, 36, 41, 42]. The study instrument was generated on the Qualtrics questionnaire platform [43], and initial validation was conducted on an identical paper based version (S1 Appendix).

The first section comprised seven items relating to GPs’ attitudes about PA during a cancer patient’s treatment period and was depicted on the seven-point Likert scale. The second section contained four items and used a seven-point Likert scale to determine GPs’ attitudes post treatment. The third section was composed of one ordinal item, four yes/no items and two Likert scale items, which assessed GPs’ knowledge of PA for cancer specific and general populations. The fourth section of the survey was based on the promotion of PA. This section included one ordinal question (“what percentage of your patients living with cancer have you recommended PA to?’), and an ordinal question to rank the preference of PA modality. The fifth section of the survey was based on the referral practices of GPs. It comprised two ordinal questions regarding percentage of clients which the GP refers to PA programs and treatment stage of preference to refer. A yes and no item was included to determine if GPs have PA resources readily available to them and a scalar question was used to determine the preferred HCP referral (options were physiotherapist, exercise physiologist, personal trainer, occupational therapist, sports medicine doctor and sport scientist).

GPs’ personal PA levels were assessed with a yes/no question to understand if the GP participates in any regular structured exercise (>2 times per week) and a question to ask what best describes their activity level (seldom, moderate or vigorously active). This question was adapted from a single question study that assessed PA levels and was deemed both valid (weighted kappa = 0.75) and reliable (correlation coefficients r = 0.28 to 0.57) [44]. To enhance response rate, less mentally fatiguing questions were positioned at the end of the survey, i.e. demographics. This included age, gender, years of experience as a GP, and location of current practice (e.g., rural, sub-urban, urban). In addition, an exclusion criterion was introduced to ensure the GP had consulted a cancer patient in the last 12 months.

Initial instrument validation

Prior to validation survey distribution, a total of nine individuals (five academics, two public officials and two accredited exercise physiologists) completed the survey. They were instructed to highlight items of confusion, construction problems, opinion on content and survey flow, and decipher any grammatical concerns. This comprised the content and construct validity portion of the validation process [45, 46]. Irrelevant questions were removed, additional questions of interest based upon expert opinion were added and sections were reordered to maximise response rate (demographic questions were moved to the end of the survey to minimise mental fatigue). Next a validation study was conducted with GPs located in Canberra ACT, Australia to determine the reliability and validity of the survey. This necessary step was identified as part of the systematic literature review conducted in preparing this project (Alderman et al., accepted for publication in October 2020). A minimum of 30 responses were required to satisfy criteria for first response validation [45]. In order to establish test-retest reliability a minimum of 10 retest responses were required. The analyses were conducted with SPSS V25 [47].

Participants

Participants were recruited by private contact via phone and email of general practices. Only GPs were recruited due to the specificity of the target population for the overall research. Participants were recruited via yellow pages email dissemination across all states and territories of Australia (ACT, NT, TAS, WA, SA, NSW, VIC, QLD). In addition, the Australian College of Rural and Remote Medicine forwarded the survey link to their GP contacts.

Study design

The study developed and validated, then distributed as cross-sectional national online survey of GPs who see patients living with cancer on a regular basis, to determine predictors of their recommendation and referral practices. An email was sent to GP clinics (to the practice managers) by the primary investigator stating the nature of the survey and for only GPs to partake in it and included an attached personalised video about the study to assist in increasing recruitment rate. GP’s used the link in the email to access the survey, this link included the participant information, informed consent, and contact information for the primary investigator. If GP’s chose to participate, they were required to tick a box which stated that they agreed to the informed consent prior to accessing the questions, this was an automated process. If they chose to tick the no box, they were not eligible to access the survey. The study was approved by the University of Canberra Human Research Ethics Committee (HREC: 20191802).

Data collection

The online Qualtrics platform was used to disseminate the survey and collect the data for this study. Independent anonymous links were sent via email which directed the participants to the survey. A follow up email was sent two, four and seven weeks after the initial email to remind clinics about the survey, survey cut off was established at eight weeks after initial email.

Data analysis

The analyses were conducted on SPSS V25 [47] on completed surveys. Knowledge of PA was coded “correct” or “incorrect” to determine right PA guideline identification [20]. The conceptual framework of the instruments theoretical pathway is presented in Fig 1. The model was created based on The Theory of Planned Behaviour model [17]. Principal component analysis (PCA) was conducted to determine fit of the conceptual model allowing for item grouping of common themes for overall category score analysis. Bivariate Correlation was conducted to interpret correlation between overall variables and establish relationships. Multiple regression analysis was conducted to determine if components from the PCA and demographics were predictive characteristics of the percentage of patients recommended PA and referred by GPs. Pearson’s product-moment correlation was run to assess the relationship between demographic variables, PCA conceptual components and dependant variables (percentage of patients recommended and referred PA).

Fig 1

Conceptual model.

PA = physical activity. Outlines the conceptual model the generated to construct the survey to align with the Theory of Planned Behaviour. Attitudes = attitudes in respect to PA promotion for the cancer population, Subjective norms = the views of their fellow GP, PBC = the perception that the behaviour is within their control (recommending/ referring PA to cancer patients).

Results

A total of 31 GP responses were collected for the initial validation portion of this study (10 re-tests (of 31 invited to re-test) responses were collected response rate = 28.21%). All negative questions were reversed to establish the positive alternative for comparison purposes. Cronbach’s alpha [45] was used to determine internal consistency (α>0.7) [14], questions assessing similar constructs should have a high level of internal consistency with Cronbach’s alpha. The six questions on GP’s attitudes to PA during treatment demonstrated a high level of internal consistency (α = 0.904). ‘Post treatment’ construct consisted of three questions and had a high internal consistency (α = 0.842). ‘Evidence based practice’ construct consisted of four questions which presented with a low internal consistency (α = 0.530). Upon removing Item 3, an acceptable internal consistence was reached of α = 0.701, however we determined that the removal of this question would impact the response of Item 4. Due to the categorisation of Item 3 and 4, these questions were split into the ‘during’ and ‘post’ treatment constructs, respectively and Cronbach’s alpha was recalculated. ‘During treatment’ construct (7 items) α = 0.839. ‘Post treatment’ construct (4 items), α = 0.749. ‘Evidence based practice’ construct (2 items), α = 0.904. ‘Promotion of physical activity’ construct was not modified and consisted of four items (α = 0.805).

Test-retest results were collected two weeks post-initial response, with 10 participants responding to the second invitation. Each response was individually compared to determine inter-respondent variability and totalled. Variability averaged at 4.57%, hence presenting 95% overall consistency. A Pearson’s product-moment correlation was run to assess the relationship between Likert responses in test-retest comparison [48, 49]. Spearman’s correlation was additionally run [48, 49]. All items presented consistency (p<0.05), there were no significant outliers (S2 Appendix). An exact sign test was used to compare the differences in responses in ‘Yes/ No’ items within the survey [50]. Test-retest responses displayed no statistical difference between responses (0% variation evident- S2 Appendix). Overall, the survey demonstrated sufficient first and secondary response validation protocol [45, 46]. Previously stated amendments were conducted to satisfy validation criteria to prepare tool for national dissemination.

Participant responses

A total of 128 general practitioners completed the survey; 97 online responses and 31 responses were included from the validation study. Only fully completed surveys were included, fifteen responses were excluded due to incomplete/ missing data, leaving a total of 113 completed for analysis. From total responses, 111 met inclusion criteria (having consulted a cancer patient in the last 12 months). The response rate, in relation to the targeted population, is unknown due to additional independent distribution of survey link from further organisations. This achieved sample ‘response representation’ displays a 9.2% margin of error [51]. This percentage is within the accepted margin of error of HCPs surveys (14%) [30], therefore offering an adequate representation of the views and practices of Australian GPs [52]. From 111 responses, 58% of GPs who completed the survey identified as females, examples of exercise intensity were provided [Table 1]. Average age of respondents was 49 (±10.4SD) years old with an average of 19 (±10.4SD) years practicing.

Table 1

Participants’ characteristics.

Characteristics	Number (%)
Sex
Male	47(42.34)
Female	64(57.66)
Age (years)
26–35	13(11.71)
36–45	28(25.23)
46–55	38(34.23)
56–65	25(22.52)
Over 65	7 (6.31)
Participate in regular PA
Yes	77 (69.37)
No	64(30.63)
Description of activity level
Vigorously active	5(4.50)
Moderately active	55(49.55)
Seldomly active	51(45.95)
How many years practicing as a GP (years)
0–10	33(29.73)
11–20	26(23.42)
21–30	38(34.23)
31–40	12(10.81)
41+	2(1.80)
Location of practice
Urban	58(52.25)
Sub-urban	33(29.73)
Rural	20(18.02)

Principal component analysis

Principal component analysis (PCA) was run on survey items due to lack of validation measure of TPB components towards GP’s recommendations of PA to patients living with cancer, hence examination of the conceptual model fitting was crucial [Table 2]. Review of the correlation matrix eliminated three questions that did not have at least one correlation coefficient greater than 0.3. The overall Kaiser-Meyer-Olkin (KMO) [53] was 0.831 with all individual KMO values greater than 0.6. Results of Barlett’s test of sphericity were statistically significant (p<0.0005). PCA produced five components with values greater than one. Three components met interpretability criteria and were retained due to component loading. A three-component solution explained 59.8% of total variance. A Varimax orthogonal rotation was employed to aid interpretability. The rotated component matrix was consistent with component loadings [Table 2]. In relation to the Theory of Planned Behavior that was used to construct the scales: component one was consistent with GP attitudes, component two was consistent with perceived behavioral control, and component three with subjective norms. Where cross loading was identified, items were either assigned to the component category to which they showed the highest loading [17].

Table 2

Rotated structure matrix for PCA with varimax rotation of a three component questionnaire.

	Rotated Component Coefficients
Items	C1-Attitude	C2- perceived behavioural control	C3-Norm	Communalities
Current evidence suggests regular physical activity is associated with reduced negative side effects of cancer treatment	0.807			0.713
Current evidence suggests that regular physical activity can improve quality of life of patients living with cancer	0.729			0.671
Most of my patients are capable of participating in physical activity during cancer treatment	0.723			0.754
Physical activity is beneficial during cancer treatment	0.684			0.822
Physical activity is important during cancer treatment	0.657			0.849
Patients would follow my advice, if I provided physical activity recommendations	0.548			0.691
My patients are amenable to receiving advice on the importance of increasing their physical activity levels	0.533			0.770
I feel confident in giving general advice to patients living with cancer about PA		0.838		0.798
Discussing physical activity with patients living with cancer is part of my role as a general practitioner		0.755		0.749
For me, providing a recommendation is easy		0.674		0.581
Physical activity is safe during cancer treatment	0.502	0.518		0.716
Most patients believe they should be physically activity during cancer treatment			0.867	0.777
Fellow general practitioners think patients should participate in PA during cancer treatment			0.770	0.741
Other general practitioners believe it is part of their role to discuss physical activity with their patients			0.687	0.757

Rotation Method: Varimax with Kaiser Normalization.

Rotation converged in 7 iterations.

Note. Major loadings for each item are in bold, only values < .5 are shown, cross loading was satisfied with largest number selection.

Preliminary analyses showed a linear relationship with percentage recommended PA with; percentage of patients referred PA, participation in structured PA, gender, attitudes, perceived behavioural control and norms as assessed by Shapiro-Wilk’s test (p>0.05)–test of normality of data, and there were no outliers [Table 3]. Percentage of patients recommended PA presented a linear relationship with; participation in structured PA, age, gender, attitudes, and perceived behavioural control (p>0.05), with no outlier’s present.

Table 3

Bivariate correlation matrix.

Variable	1	2	3	4	5	6	7	8	9	10	11
1. % Patients recommend PA to	-
2. % Patients refer PA to	.550**	-
3. Participation in structured PA	.328**	.285**	-
4. PA level	.063	.063	-.268**	-
5. Age	.087	-.268**	.012	-.111	-
6. Gender	.264**	.238*	.261**	-.048	-.12	-
7. Years of experience	.159	.038	.063	-.037	.896**	-.055	-
8. Location	-.027	-.052	.086	.015	.104	.069	.218*	-
9. Attitudes	.529**	.188*	.250**	-.025	.247**	.192*	.256**	.027	-
10. PERCEIVED BEHAVIOURAL CONTROLPERCEIVED BEHAVIOURAL CONTROL	.594**	.374**	.260**	.01	.220*	.17	.236*	-.04	.682**	-
11. Norms	.222*	.085	.119	-.027	.14	-.032	.13	-.123	.424**	.404**	-

PA- Physical activity, PERCEIVED BEHAVIOURAL CONTROL- Perceived behavioural control.

** Correlation is significant at the 0.01 level (2-tailed)

* Correlation is significant at the 0.05 level (2-tailed).

Multiple regression analysis

Multiple regression analyses were run to evaluate the relative contributions of attitudes, perceived behavioural control, subjective norms and other predictor variables on self-reported PA recommendation and referral behaviours of GPs, working with cancer patient’s post-treatment. Gender, years practicing as a GP, practice location, personal activity levels, perceived behavioural control, attitudes of PA and subjective norms were assessed as predictor variables [Table 1]. Linearity assumptions were met via partial regression plots of studentized residuals against the predicted values. The data recommended PA and ‘percentage of cases referred for PA’ displayed an independence of residuals, assessed by a Durbin-Watson statistic of 1.907 and 1.842 respectively. Homoscedasticity was observed under visual inspection of plotted studentized residuals versus unstandardized predicted values. No multicollinear results were present (tolerance values greater than 0.1). One studentized deleted residual was presented greater than ±3 (item number 42 = 3.1) but this did not impact additional results and so was retained. No leverage values greater than 0.2, and values for Cook’s distance above 1 were evident. The multiple regression models significantly predicted the percent of patients living with cancer recommended PA F(9,101) = 9.129, P < .0005 adj. R2 = 0.40 and referred to PA F(9,101) = 4.255, P < .0005, adj. R2 = 0.21. Similarly, both items presented three variables that added statistical significance to the prediction, p<0.05. Regression coefficients and standard error can be seen in Table 4.

Table 4

Summary of multiple regression analysis.

	% Recommended PA			% Referred to PA
Variable	B	SE	Beta	B	SE	Beta
Constant	-1.524	1.291		.668	1.274
Gender	.305	.235	.103	.333	.232	.131
Years practicing as a GP	.033	.025	.237	.031	.025	.259
Location	-.142	.151	-.074	-.159	.149	-.097
Description of PA level	.249	.200	.098	.423	.197	.194
Participation in structured exercise	.540	.263	.170*	.639	.260	.234*
Attitudes	.059	.029	.221*	-.032	.028	-.137*
Perceived behavioural control	.148	.039	.394*	.128	.039	.396*
Norms	-.023	.036	-.053	-.015	.035	-.041

Note. B = unstandardized regression coefficient; SE = Standard error of the coefficient; Beta = Standardised coefficient

* = P<0.05.

Participant PA levels

General Practitioners’ participation in structured PA was statistically significant to the prediction of percentage of patients living with cancer that were both recommended and referred PA (p<0.05). Descriptive statistics suggest 69.4% (n = 77) of GPs reported that they participated in structured PA, with the majority rating their PA levels at ‘moderately active at least 30 minutes three times per week’ (49.6%). In relation to knowledge of PA, 61.3% (n = 68) of GPs correctly identified the correct PA guidelines for general population. Only 3.6% noted that they had received PA training during their studies, 25.2% (n = 28) conducted additional training regarding PA knowledge of patients living with cancer, and 19.8% (n = 22) were aware of the Clinical Oncology Society of Australia (COSA) guidelines. In addition, only 31.5% (n = 35) of GPs had access to resources or were aware of PA services for patients living with cancer. Yet, 94.6% (n = 105) of GPs allocated exercise physiologists (59.5%) and physiotherapists (35.1%) as their HCP of choice to deliver PA guidance and services.

Participant attitudes toward PA

General practitioners’ attitudes towards PA for patients living with cancer were a significant predictor of the percentage of patients living with cancer recommended and referred for PA (p<0.05). Within this component, a high percentage of GPs had acknowledged that current evidence suggests that regular PA can both improve quality of life (92.8%) and reduce negative side effects of cancer specific treatment (89.2%). A large percentage of GPs recognised that PA is both important (91.0%) and beneficial (91.0%) to a cancer patient’s journey. Seventy-two percent of GPs believed their patients to be capable of participating in PA and 79.3% stated that they believe their patients to be amenable to receiving PA advice. In addition, 64% of GPs believe their patients would follow their advice if provided. Still, only 27% reported that their patients asked them about PA.

Participant perceptions

The perceived ability to perform the behaviour was a significant predictor of both percentages of patients recommended and referred PA (p<0.05). Within this component, it was evident that 90.1% of GPs believe it to be safe for patients living with cancer to participate in PA. Sixty-seven percent of GPs believed providing PA recommendations was easy, and 77.5% were confident in giving recommendations (the highest modality of PA recommended being walking—49.5%). In addition to this, a high percentage of GPs believed that it was their role to discuss PA with their patients living with cancer (53.2%).

Discussion

This is the first study of its kind to assess the knowledge, attitudes, and practices of Australian GPs regarding PA for the cancer population. Results offered some support for the TPB, but not all principles were supported. As hypothesised, both perceived behavioural control and attitudes were significant predictors of intentions to both recommend PA and refer patients to PA programs or for exercise specialist advice. Subjective norms, however, were shown to have no correlation with either behaviour, contradicting the hypotheses derived from the TPB. In addition, GPs personal PA participation predicted both the percentage of patients recommended to do PA and referred to PA programs or for further support by an exercise specialist.

The findings of this study suggested that the GPs were recommending PA to 41–60% of their patients living with cancer, indicating that 40–59% of patients were not receiving any discussion about PA. This number further decreases with GPs only referring 1–20% of patients living with cancer to an exercise program or to receive exercise specialist PA advice. These findings are consistent with similar studies looking at other HCPs (such as oncologists, oncology nurses and specialists) involved in cancer care [16, 32, 45, 54–57].

A large proportion of the surveyed GP population reported that they were, themselves, physically active (~70% and meeting the PA guidelines [Table 1]. This is substantially higher than reported in the general population in Australia (55.4%) [58]. In addition, similar studies that investigated oncology HCPs showed that >60% of practitioners were not meeting PA guidelines [16, 24, 27, 33, 36, 41, 59, 60]. This may suggest that GPs who responded to this survey report being more physically active than other HCPs report or perhaps those who responded were particularly interested in this topic because they were already regular exercises; those who didn’t think it was important most likely didn’t respond. This study builds upon the existing evidence showing that overall GP’s personal PA levels played a significant role in the likelihood that they would recommend PA to their patients living with cancer [61] [Table 4]. Nonetheless, the current study is the only research conducted on GP’s that has demonstrated a positive association between personal PA levels and PA referral practices [32, 36, 55, 59].

A positive relationship was also observed between PA knowledge and the promotion of PA [41], with >60% of GPs in the current sample correctly identifying the PA guidelines for the general population. This proportion is significantly higher in comparison to other HCPs involved in cancer care (<50%) [16, 57]. Accurate PA knowledge is arguably essential for informing best-practice. Accurate, in-depth PA recall to patients is positively associated with uptake and adherence levels of HCPs patients living with cancer [15, 16]. Despite 61% of GPs correctly identifying the general population PA guidelines, this study identified that GPs received minimal education on PA for the cancer population, both during university-training and post-qualification professional education. The effect of PA education among GPs has not been thoroughly investigated. Identification that a minimal percentage of GPs were provided with PA education suggested that additional research should be conducted investigating the knowledge and education of GPs involved in cancer care in order to implement intervention-based studies at both an undergraduate and professional development level. This could assist in altering the training and applied practice of PA among GPs which would improve cancer patient outcomes.

In line with the TPB, the identified correlation between GPs attitudes and their PA recommendations and referral practices is consistent with existing research on oncologists in this area [41]. The current study shows that overall GPs attitudes of PA for patients living with cancer are significantly higher than other HCPs involved in cancer care. A high majority HCPs (oncologists, oncology nurses and specialists) involved in cancer care understand that PA improves the quality of life of their patients living with cancer [33, 34, 41]. Conversely, on average only 50% of HCPs recognise that regular PA is associated with a reduction in the negative side effects resulting from cancer treatment [34], compared to 89% of the GPs in the current study. In comparison to oncology HCPs, greater portions of GPs understand that PA is beneficial and important for patients living with cancer. General practitioners also recognise that patients living with cancer are capable of participating in PA, which had previously been noted as a sizeable barrier to PA recommendations in previous oncology HCP studies [24, 54, 55].

The perceived ability to control the implementation, perceived behavioural control of PA, as defined as a concept of TPB was a significant predictor of practice outcomes [Table 4]. These results are consistent with an international study investigating the knowledge, attitudes and practices of oncologists carried out in 2018 [41]. Further investigation on the individual items of the component showed a similar disconnection between the perceived behavioural control of GPs and other oncology HCPs. The current study identified that GPs have a significant understanding that PA is safe for patients living with cancer and that it is part of their role to provide these recommendations. Yet, a smaller percentage of GPs have the confidence to provide these recommendations to patients living with cancer and even fewer find it easy in comparison to percentage that thought it was their role [33, 62]. This percentage gap could be due to lack of specific GP education on PA recommendations for the cancer population.

In the current study, only 32% of GPs reported having access to resources about PA for patients living with cancer or specific PA programs. Studies conducted investigating patients living with cancer or PA programs indicated lack of referrals, with the average program occupancy rate approximately 70% across locations [63]. The difference between GPs access to resources and referrals and the researched occupancy rate for PA programs for cancer survivors implies a lack of awareness or additional barriers stopping them from referring. Further research should be conducted to see if the implementation of education tools within this population could increase cancer patients’ participation in PA. A similar small but promising study was conducted on oncologists, they were provided with a 30-second education tool about PA recommendations. Results reported that patients PA increased by 3.4 MET hours per week in comparison to the control group [64].

As previously discussed, a larger scale study conducted over 1997-2007on Australian GPs, displayed a consistent incline in the knowledge, confidence, and perception that it was their role was to discuss PA. Despite this natural incline, no significant increase or correlation was seen in the promotion and practices of PA among participants [40]. Unlike this study, the current study demonstrated that the attitudes and perceived behavioural control of GPs were significant predictors of PA recommendations and referrals. The current study specifically looked at GPs knowledge, attitudes, and practices of PA specifically within the cancer population. Despite the predictive correlation between the TPB components and GPs practices, on average greater than 80% of GPs have positive attitudes and perceived behavioural control towards PA in the cancer population. However, the average percentage of patients living with cancer recommended and referred to PA is substantially lower; 41–60% and 1–20% respectively. This inconsistency could reflect that the current study is potentially capturing some psychological predictors, however no other systemic, relational, resource and time-constraint predictors which could be of importance. Therefore, interventions targeting solely the GPs psychological predictors may not create a significant change. Thus, assessing the system holistically for future studies should be considered.

Conclusion

It is well understood by GPs that it is part of their role to be promoting PA to their clients within the cancer population (>85%) and the general population [40]. With the consistent growth of research proving the benefits of PA for patients living with cancer with a reduction of adverse effects from cancer treatments, an emphasis needs to be placed on GPs to promote this message to their patients. Evidence suggests that one of the greatest limitations to increase the evidence in this space is the current lack of referrals from oncology HCPs and GP’s into PA programs run by exercise specialists [29]. The current study suggests that enhancing the psychological aspects of GPs attitudes and perceived behavioural control towards PA could create a positive impact on the percentage of patients living with cancer both recommended and referred for PA. In addition to this, GPs are in a highly influential position to enhance PA referral rates due to their positive attitudes and knowledge of the important role of PA for the cancer population. GP’s in Australia have the power to activate government funded (Medicare) ‘rebate-able’ programs with exercise specialists for patients living with cancer, meaning more patients could potentially have access to programs enhancing physical and mental health. Future studies and patient care policies should look at ways to integrate this information to enhance GP’s involvement in the care of patients living with cancer throughout the cancer treatment and survivorship continuum, through specific cancer care education and training.

Limitations

Despite the study findings, limitations of this study must be taken into consideration. The authors acknowledge that due to the study design consisting of a self-reported survey, a bias likely evident as a result of potential over reporting of personal PA levels by the GP’s. In addition, response rate was limited and may not fully represent the overall Australian GP population. Assessing demographic data of the sample displayed a significant proportion of GPs reporting themselves to be meeting the PA guidelines. A bias is also likely due to the percentage of GPs self-reporting that they are significantly more active than the Australian population. Perhaps those who responded to the survey were particularly interested in this topic because they were already regular exercises; those who didn’t think it was important may not have responded. Strengths of the study include, its integration of the conceptual model generated around the theory of planned behaviour, principal component analysis to test the fitting of the conceptual model, the survey validation and the systematic literature review conducted to highlight the gap in the literature. This was also the first study of its kind to look at GP’s knowledge and practices of PA within the cancer population.

(XLSX)

Click here for additional data file.

Survey

(DOCX)

Click here for additional data file.

Test-retest analysis procedure.

(DOCX)

Click here for additional data file.

PCA total variance SPSS output.

(DOCX)

Click here for additional data file.

17 Jul 2020

PONE-D-20-10478

Physical Activity for Cancer Patients: Knowledge, Attitudes, and Practices of General Practitioners in Australia

PLOS ONE

Dear Dr. Toohey,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

This is an important and timely paper.  The reviewers have made a number of suggestions, particularly around including more information in the analysis section, as well as explaining the TPB more clearly.

Please submit your revised manuscript by Aug 31 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Adam Todd, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

3. In your Methods section, please provide additional information about the participant recruitment method and the demographic details of your participants. Please ensure you have provided sufficient details to replicate the analyses such as: a) the recruitment date range (month and year) and b) a description of any inclusion/exclusion criteria that were applied to participant recruitment.

4. To comply with PLOS ONE submission guidelines, in your Methods section, please provide additional information regarding your statistical analyses. For more information on PLOS ONE's expectations for statistical reporting, please see https://journals.plos.org/plosone/s/submission-guidelines.#loc-statistical-reporting.

5. We ask that you please remove citations for unavailable and unpublished work, including manuscripts that have been submitted but not yet accepted (e.g., “in submission,” “data not shown”). Instead, include those data as supplementary material or deposit the data in a publicly available database.

6. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

7. Your ethics statement must appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please also ensure that your ethics statement is included in your manuscript, as the ethics section of your online submission will not be published alongside your manuscript.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: PLOS ONE Review:

Physical Activity for Cancer Patients: Knowledge, Attitudes, and Practices of General Practitioners in Australia.

Abstract:

Line 69-70: was this in relation to the behaviour of promoting exercise or their actual exercise behaviour?

Line 71-72: be more specific. Fit of TPB to what? Associations of what?

Background:

Line 84: avoid ambiguous terms “are likely”…change to “will” if this is what the research is showing. Also, can you contextualize this geographically. Do you mean globally?

Line 89-90: reference this statement.

Line 94: add ‘exercise’ before ‘levels’

Line 99-100: not sure what you mean by “playing a strong predictive power for HCP’s compared to…” please clarify.

Line 100-101: reference this sentence “TPB is composed of…” (Ajzen article)

Line 101: change ‘constraint’ to ‘construct’

Line 103-104: wording needs updating grammatically

Line 112: Specify COSA acronym in line 109

Line 119: remove ‘of’

Line 115: Jones article is quite old, will likely be able to supplement with something more recent. Much has changed since 2002 in regard to oncology rehab, that is why I suggest this.

Line 124: change ‘reported’ to ‘report’

Line 124-125: again, can you contextualize this; where does that happen? This is not always the case in North America.

Line 126: change ‘to’ to ‘towards’

Line 129: missing word after ‘treatment (32)’

Line 133: missing a word after ‘decade’

Line 136-137: I would not say that the TPB is based on the premise that HCPs…The TPB is based on the premise that attitudes, subjective norms, and PBC affect a person’s behavioural intention and this has a direct effect on behaviour.

Line 138: Change to GP as you are using elsewhere in intro.

Overall, I think the TPB needs to be explained more clearly; specifically, the ‘intention’ component was largely missing and is very important to this theory.

Good ideas highlighted in the intro however.

Materials and Methods:

Missing information between ‘participant recruitment’ and ‘data collection’: Line 152-163

- More details are needed on the process after the recruitment email was sent; did those interested respond via email to XX.

- Who then emailed the survey to the GP’s??

- Why does the data collection section say it was also practice managers and clinics that were emailed? More information needed. Also in this section should reference ethics received.

Line 162: Can you give a few survey question examples within the manuscript? It would be interesting to see how your survey addressed each of the TPB components; perhaps a table with one column: TPB construct, another column “survey question(s) related to this construct”. I see you have It as an appendices but would be good to summarize within the written document.

Line 162: give more information in regard to the consent process. Did they have to sign this form prior to completing the survey, etc.

Line 162-163: confirm why follow up email was sent…if they did not complete the survey within two weeks. How many follow up emails were sent? Over what time frame? How long did respondents have to complete the survey?

Figure 1:

- Note: this conceptual model does not align. The overall behaviour is PA promotion but you are describing the attitude and subjective norm towards PA in general. These are two different behaviours. In the text you describe subjective norms related to PA promotion, but not in the figure. The attitude should be “attitudes in respect to PA promotion for the cancer population”…do they think it is important / helpful, etc. Must all link together and make sense.

Good description of initial validation.

Line 212-215 should go in ‘recruitment’ section. It is confusing how it is laid out…I think you are describing two separate studies. The methodology section should only be in reference to this study.

Paragraph Lines 219-231: you are presenting results here. Should go in results section. Organize methods / results more clearly to delineate the steps that were taken. It is all important, just needs to be organized more appropriately. Same thing for paragraph starting Line 233.

Results:

Line 248: how much needed to be complete to be included? All questions? Confirm in methods.

Line 255: add ‘exercise’ before intensity. Also, I think you are missing a statement about % who take part in exercise prior to this sentence.

Line 280-281; this should go in the methods. Use sub-headings in the results section to make clear to readers.

Line 293-295; can you describe this further…for example, what Gender? How many years practicing? This information will be interesting for readers but isn’t stated anywhere.

Line 311-312; what there a cut off value for this? How much did practitioners have to exercise to be more likely to predict…any?

Discussion:

Line 362-363: you should link this finding to limitations also; perhaps those who responded to your recruitment email were particularly interested in this topic because they were already regular exercises; those who didn’t think it was important didn’t respond. Add into limitations also.

Paragraph starting line 369. I think it was also interesting that only 19% were aware of the new COSA guidelines. This was a major publication in exercise oncology globally! So, the fact that GP’s within Australia are not aware of it is surprising.

Line 418 and throughout discussion: do not need to reference tables in discussion sections

Throughout discussion section you may want to refer to other studies looking at barriers/facilitators to exercise promotion for this population. What gets in the way and how does this relate to the TPB components? How can this be addressed?

Summary:

Overall, an interesting topic. Thank you for taking the time to conduct research in this area. Needs some edits for clarity before publication.

I am concerned around how the TPB was described and how the conceptual model include two different behaviours (PA and PA promotion)...this needs to be clarified.

More detailed and clear information needed in methods; results were presented in methods as well.

Increase clarify of results (give more specific information).

Reviewer #2: Thanks for the opportunity to review this paper, it was interesting to read. Please see comments which are a mix of relevant points and pedantry.

Abstract

I wonder if the term 'cancer patients' is appropriate these are people with a diagnosis of cancer and that label really does jar when I read it. We probably don't label patients as 'diabetics' or schizophrenics' any linger so maybe the term 'patients with cancer' or perhaps 'patients living with cancer', although not exactly mellifluous. would be more appropriate? Same comment applies to 'the cancer population' and other such terms

Background

Line 85 33% of cancer related deaths are attributable to physical inactivity....' I wonder is this statement too strong. I agree that physical inactivity and poor diet are contributory factors but as an absolute causal link it's very difficult to be that categorical. Could you say that they are significant contributory factors?

Line 93 -'It has been established that patients who have been told to be active by their HCPs’ have improved levels and adherence levels' this doesn't make sense to me improved levels and adherence level ? Im presuming you mean persistence with the exercise programme? Maybe reword ?

Line 103 - 'in this case what other HCP’s physical activity believe and promotion of PA to their cancer patients...' consider reword doesn't make sense should that beliefs are?

Line 105 repetition of 'perceived' not sure second one is necessary and removing would improve sentence flow

Line 106 should read 'individuals''?

Line 109 - define acronym if you are going to use it in the rest of the document Clinical Oncology Society of Australia (COSA)....

Line 112 should read 'the general population'

Line 118 '...treatment, which cancer...'

Line 122 - could these two sentence be combined, second sentence is partially repeating the first.

Line 126 is a possessive so should read 'HCPs'"

Line 133 should read '...Australian GPs. which ...'

Line 138 you have used GPs elsewhere, why write out in full here?

Methods

I'm not convinced that you need to specify whether data is ordinal, nominal or scalar. Its repetitive and doesn't help the flow of the text. I would argue anyone reading this is capable of that distinction and the nature of the data in statistical terms is immaterial, it is what it is.

Results

Line 258 - surprising that you only had people identify as male or female and there were no other options. Did you provide options for gender or was your question about sex ie male/female as options? If so this should read sex.

Line 296 should read recommended 'PA' ....

Line 311 - personal preference but as a reader where percentages are quoted I often like to see the number, just for context as I read so 69% (n=22) or 22 GPs (x%), whichever works best for you.

Discussion I would argue that Table 5 should be in the results section and discussed in the context of other literature in the discussion. If you feel uncomfortable with the results of the literature review from other HCPs then remove this and discussion in the discussion section. Feels unusual to see a table presented like this in the discussion.

Line 396 should read 'Conversely, on average....'

Line 390 should read ...in PA, which....'

Line 404 should read 'In the current study, only 32%....'

Line 430 is adverse effects a better term?

We would normally expect to see a conclusion in a study of this type, I think it would add clarity to the final part of this paper, which does wander a little bit.

Reviewer #3: I will focus on methods and reporting. Statistical analyses are appropriate.

Major

1) State clearly all information in the data analysis section, all used variables. How was correct and incorrect PA knowledge determined? Clarify what the outcome is and how it is recorded.

Minor

1) Abstract: more clarity on regression modelling, no information on covariates of interest

2) the survey is potential problematic, as the authors acknowledge. the generalisability of the survey is questionable.

3) in table 4 better to report CIs rather than SEs

4) Consider other graphical outputs to present your results and make them more accessible

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

20 Sep 2020

20th September, 2020

Dear Reviewers,

Thank you very much for kindly taking the time to provide very helpful feedback on the manuscript. All suggested changes from the reviewers have been carefully considered and added to the manuscript. Below is the list of suggested changes along with the responses added to the manuscript (I have also included where these changes can be found in the manuscript).

Yours sincerely,

Dr Kellie Toohey (corresponding author)

Response to reviewers

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

Reviewer #3: Yes

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: PLOS ONE Review:

Physical Activity for Cancer Patients: Knowledge, Attitudes, and Practices of General Practitioners in Australia.

Abstract:

Line 69-70: was this in relation to the behaviour of promoting exercise or their actual exercise behaviour?

• The behaviour of promoting exercise – we have made the following change to make it clearer: Participants (GP’s) reported knowledge, attitudes, perceived behaviour control and subjective norms of PA within the cancer population.

Line 71-72: be more specific. Fit of TPB to what? Associations of what?

• Thank you this has been corrected and now reads:

Principal component analysis was conducted to establish a set of survey items aligned to TPB constructs (attitude, subjective norms, perceived control), and multiple regression analyses characterised associations between these predictor variables and (a) recommendation; and (b) referral – of PA to cancer patients.

Background:

Line 84: avoid ambiguous terms “are likely”…change to “will” if this is what the research is showing. Also, can you contextualize this geographically. Do you mean globally?

• Changed to: Globally, one in three people will develop cancer during their lifetime

Line 89-90: reference this statement.

• Added: {McTiernan, 2019 #3501} Physical Activity in Cancer Prevention and Survival: A Systematic Review

Line 94: add ‘exercise’ before ‘levels’

• Added: y their HCPs’ have improved exercise and adherence levels

Line 99-100: not sure what you mean by “playing a strong predictive power for HCP’s compared to…” please clarify.

• This has been re-worded to: Several psychosocial factors influence HCP behaviours – for example the Theory of Planned Behaviour (TPB) asserts that attitudes towards a behaviour, subjective appraisal of their group norms about the behaviour, and the perception of having sufficient control/choice to engage in the behaviour all precede intentions and actual behaviour. These variables have been shown to explain 31% of variance in observed behaviour in HCPs: a relatively strong finding compared to other theories (17)

Line 100-101: reference this sentence “TPB is composed of…” (Ajzen article)

• Added

Line 101: change ‘constraint’ to ‘construct’

• Changed

Line 103-104: wording needs updating grammatically

• This has been corrected and now reads: subjective norms which is defined by the perceived belief of what others think about a certain behaviour, in this case what other HCP’s believe about the promotion of PA to their cancer patients.

Line 112: Specify COSA acronym in line 109

• This has been added to line 109 and now reads: The Clinical Oncology Society of Australia (COSA) released

Line 119: remove ‘of’

• Removed

Line 115: Jones article is quite old, will likely be able to supplement with something more recent. Much has changed since 2002 in regard to oncology rehab, that is why I suggest this.

• An updated references had been added and a bit more information has been added to the sentence to update the context: Jones, et al. (2002) looked at patient perspectives and reported that 97% of cancer patients would like PA to be discussed by their HCPs and 67%, believed that they should be referred for further specialised support, but there is still no best practice evidence within more recent literature about who should have this discussion

Line 124: change ‘reported’ to ‘report’

• Changed

Line 124-125: again, can you contextualize this; where does that happen? This is not always the case in North America.

• This has been changed to give more context: During treatment in Australia, it is reported that patients contact time is spent predominantly with oncology nurses and less time with their oncologists

Line 126: change ‘to’ to ‘towards’

• Changed

Line 129: missing word after ‘treatment (32)’

• This has been changed: PA advice may be better-received post treatment when there is additional time to carry out this discussion with patients and they have had some recovery from the side effects of treatment

Line 133: missing a word after ‘decade’

• Has been changed to: A study conducted over a ten-year period on Australian GPs

Line 136-137: I would not say that the TPB is based on the premise that HCPs…The TPB is based on the premise that attitudes, subjective norms, and PBC affect a person’s behavioural intention and this has a direct effect on behaviour.

• This has been changed – thank you for the wording it sounds a lot better.

Line 138: Change to GP as you are using elsewhere in intro.

• This has been changed

Overall, I think the TPB needs to be explained more clearly; specifically, the ‘intention’ component was largely missing and is very important to this theory.

• More detail has been added to explain the TBC more clearly.

Good ideas highlighted in the intro however.

Materials and Methods:

• The authors have re-arranged the materials and methods to improve flow of the manuscript.

Missing information between ‘participant recruitment’ and ‘data collection’: Line 152-163

• We have added study design and re-arranged the materials and methods section line 191

- More details are needed on the process after the recruitment email was sent; did those interested respond via email to XX.

• This has now been added with further details.

- Who then emailed the survey to the GP’s??

• This has been added in to line 194: . An email was sent to GP practices by the primary investigator stating the nature of the survey and for only GPs to partake in it and included an attached personalised video about the study to assist in increasing recruitment rate.

- Why does the data collection section say it was also practice managers and clinics that were emailed? More information needed.

• This was how we gained access to the GP’s – the practice Managers then sent this on to the GP’s.

Also in this section should reference ethics received.

• This has been moved to study design section

Line 162: Can you give a few survey question examples within the manuscript? It would be interesting to see how your survey addressed each of the TPB components; perhaps a table with one column: TPB construct, another column “survey question(s) related to this construct”. I see you have It as an appendices but would be good to summarize within the written document.

• Thank you for the comment – table two has the rotation of a three-component questionnaire in regard to the TBC and the items (questions) and the weight of the TBC that they relate to.

Line 162: give more information regarding the consent process. Did they have to sign this form prior to completing the survey, etc.

• More details about this has been added to the study design line 190

Study Design

The study developed and validated, then distributed as cross-sectional national online survey of GPs who see patients living with cancer on a regular basis, to determine predictors of their recommendation and referral practices. An email was sent to GP clinics (to the practice managers) by the primary investigator stating the nature of the survey and for only GPs to partake in it and included an attached personalised video about the study to assist in increasing recruitment rate. GP’s used the link of the email to access the survey, this link included the participant information, informed consent, and contact information for the primary investigator. If GP’s chose to participate, they were required to tick a box which stated that they agreed to the informed consent prior to accessing the questions, this was an automated process. If they chose to tick the no box, they were not eligible to access the survey. The study was approved by the University of Canberra Human Research Ethics Committee (HREC: 20191802).

Line 162-163: confirm why follow up email was sent…if they did not complete the survey within two weeks. How many follow up emails were sent? Over what time frame? How long did respondents have to complete the survey?

• Thank you, more detail has been added to data collection

Data collection

The online Qualtrics platform was used to disseminate the survey and collect the data for this study. Independent anonymous links were sent via email which directed the participants to the survey. A follow up email was sent two, four and seven weeks after the initial email to remind clinics about the survey, survey cut off was established at eight weeks after initial email.

Figure 1:

- Note: this conceptual model does not align. The overall behaviour is PA promotion but you are describing the attitude and subjective norm towards PA in general. These are two different behaviours. In the text you describe subjective norms related to PA promotion, but not in the figure. The attitude should be “attitudes in respect to PA promotion for the cancer population”…do they think it is important / helpful, etc. Must all link together and make sense.

• This has been re-worded and we hope it addresses your comments.

Good description of initial validation.

• Thank you

Line 212-215 should go in ‘recruitment’ section.

• This has been relocated.

It is confusing how it is laid out…I think you are describing two separate studies. The methodology section should only be in reference to this study.

• This has been

Paragraph Lines 219-231: you are presenting results here. Should go in results section.

• Moved to the results section

Same thing for paragraph starting Line 233.

• Moved to results section

Organize methods / results more clearly to delineate the steps that were taken. It is all important, just needs to be organized more appropriately.

• This section has been arranged so that the validation of the study instrument is at the start of both the methods and the results section. This should ease the flow of the paper for the reader.

Results:

Line 248: how much needed to be complete to be included? All questions? Confirm in methods.

• This has been added to the results: Only fully completed surveys were included, fifteen responses were excluded due to incomplete/ missing data, leaving a total of 113 completed for analysis. From total responses, 111 met inclusion criteria (having consulted a cancer patient in the last 12 months).

• This has been added to the methods: The analyses were conducted on SPSS V25 (43) on completed surveys.

Line 255: add ‘exercise’ before intensity. Also, I think you are missing a statement about % who take part in exercise prior to this sentence.

• Added

Line 280-281; this should go in the methods. Use sub-headings in the results section to make clear to readers.

• This has been moved

• Headings have been added to the results section

Line 293-295; can you describe this further…for example, what Gender? How many years practicing? This information will be interesting for readers but isn’t stated anywhere.

• This information can be found in table 1 – this has now been added so that the reader can find it easier.

Line 311-312; what there a cut off value for this? How much did practitioners have to exercise to be more likely to predict…any?

• Over 150 minutes of exercise was used as a cut off for moderately active participants, this has been changed in the manuscript to reflect this

Discussion:

Line 362-363: you should link this finding to limitations also; perhaps those who responded to your recruitment email were particularly interested in this topic because they were already regular exercises; those who didn’t think it was important didn’t respond. Add into limitations also.

• This has been added to both the discussion and limitations as requested

Paragraph starting line 369. I think it was also interesting that only 19% were aware of the new COSA guidelines. This was a major publication in exercise oncology globally! So, the fact that GP’s within Australia are not aware of it is surprising.

• Yes we also found this surprising however, we so also see this clinically.

Line 418 and throughout discussion: do not need to reference tables in discussion sections

Throughout discussion section you may want to refer to other studies looking at barriers/facilitators to exercise promotion for this population. What gets in the way and how does this relate to the TPB components? How can this be addressed?

• Reference has been made to a number of other studies as suggested see line 465, 471, 495, 499, 500, 504, 508, 514.

• Table 5 has been removed and references to tables have been removed – further details and references have been added to discuss findings from other studies.

Summary:

Overall, an interesting topic. Thank you for taking the time to conduct research in this area. Needs some edits for clarity before publication.

I am concerned around how the TPB was described and how the conceptual model include two different behaviours (PA and PA promotion)...this needs to be clarified.

More detailed and clear information needed in methods; results were presented in methods as well.

Increase clarify of results (give more specific information).

• Thank you for your amazing feedback, it has improved the manuscript considerably – the authors hope that with the suggested additions that you will be happy with the changes.

Reviewer #2: Thanks for the opportunity to review this paper, it was interesting to read. Please see comments which are a mix of relevant points and pedantry.

Abstract

I wonder if the term 'cancer patients' is appropriate these are people with a diagnosis of cancer and that label really does jar when I read it. We probably don't label patients as 'diabetics' or schizophrenics' any linger so maybe the term 'patients with cancer' or perhaps 'patients living with cancer', although not exactly mellifluous. would be more appropriate? Same comment applies to 'the cancer population' and other such terms

• Thank you for your very thoughtful comment, this has been changed as suggested

Background

Line 85 33% of cancer related deaths are attributable to physical inactivity....' I wonder is this statement too strong. I agree that physical inactivity and poor diet are contributory factors but as an absolute causal link it's very difficult to be that categorical. Could you say that they are significant contributory factors?

• This has been changed and now reads: Physical inactivity and poor dietary behaviours are significant contributing factors to cancer-related deaths

Line 93 -'It has been established that patients who have been told to be active by their HCPs’ have improved levels and adherence levels' this doesn't make sense to me improved levels and adherence level ? Im presuming you mean persistence with the exercise programme? Maybe reword ?

• This has been changed and now reads: It has been established that patients who have been told to be active by their HCPs’ have improved exercise and adherence levels

Line 103 - 'in this case what other HCP’s physical activity believe and promotion of PA to their cancer patients...' consider reword doesn't make sense should that beliefs are?

• This has been changed to: The second construct is subjective norms which is defined by the perceived belief of what others think about a certain behaviour, in this case what other HCP’s believe about the promotion of PA to their patients living with cancer

Line 105 repetition of 'perceived' not sure second one is necessary and removing would improve sentence flow

• Thank you, that has been changed to: The final construct, titled perceived behavioural control, questions beliefs on the difficulty to perform the behaviour, in this case providing PA recommendations

Line 106 should read 'individuals''?

• Thank you, that has been corrected

Line 109 - define acronym if you are going to use it in the rest of the document Clinical Oncology Society of Australia (COSA)....

• This has now been added

Line 112 should read 'the general population'

• This has been added

Line 118 '...treatment, which cancer...'

• Added

Line 122 - could these two sentence be combined, second sentence is partially repeating the first.

• The first sentence was deleted with more information added to the second sentence, it now reads: During treatment in Australia, it is reported that patients contact time is spent predominantly with oncology nurses and less time with their oncologists

Line 126 is a possessive so should read 'HCPs'"

• This sentence and the sentence below has been changed and now reads: HCPs attitudes towards PA during the cancer treatment phase is evident (16, 24, 36), however, there is limited understanding of this during the post treatment phase (23, 37-39), GPs views and role within this area has not been explored.

Line 133 should read '...Australian GPs. which ...'

• A comma has been added as I think that is what you meant

Line 138 you have used GPs elsewhere, why write out in full here?

• This has been changed

Methods

I'm not convinced that you need to specify whether data is ordinal, nominal or scalar. Its repetitive and doesn't help the flow of the text. I would argue anyone reading this is capable of that distinction and the nature of the data in statistical terms is immaterial, it is what it is.

• Thank you, this has been removed.

Results

Line 258 - surprising that you only had people identify as male or female and there were no other options. Did you provide options for gender or was your question about sex ie male/female as options? If so this should read sex.

• Thank you, this has been changed

Line 296 should read recommended 'PA' ....

• Changed

Line 311 - personal preference but as a reader where percentages are quoted I often like to see the number, just for context as I read so 69% (n=22) or 22 GPs (x%), whichever works best for you.

• This has been added

Discussion I would argue that Table 5 should be in the results section and discussed in the context of other literature in the discussion. If you feel uncomfortable with the results of the literature review from other HCPs then remove this and discussion in the discussion section. Feels unusual to see a table presented like this in the discussion.

• This has been removed from the manuscript

Line 396 should read 'Conversely, on average....'

• This has been added.

Line 390 should read ...in PA, which....'

• This has been added

Line 404 should read 'In the current study, only 32%....'

• This has been added

Line 430 is adverse effects a better term?

• This has been changed as suggested

We would normally expect to see a conclusion in a study of this type, I think it would add clarity to the final part of this paper, which does wander a little bit.

• This has been added

Thank you for your very thoughtful comments, they have made a big difference to the paper.

Reviewer #3: I will focus on methods and reporting. Statistical analyses are appropriate.

Major

1) State clearly all information in the data analysis section, all used variables. How was correct and incorrect PA knowledge determined? Clarify what the outcome is and how it is recorded.

• This has been changed throughout the manuscript, we hope that they changes improve the clarity and flow of the paper.

Minor

1) Abstract: more clarity on regression modelling, no information on covariates of interest

• This has been changed and added to the abstract

2) the survey is potential problematic, as the authors acknowledge. the generalisability of the survey is questionable.

• Thank you – yes we have added this to our limitations

3) in table 4 better to report CIs rather than SEs

• The authors believe that due to the categorical nature of the analysis it would be difficult to report the range of CI values. SE was used to measure the accuracy of the estimates equal to the SD of the distribution which is more beneficial than proposing a range with the large number of values for each parameter. An example is provided below for the gender variable in % recommended PA:

• 1 = male, 2 = female

What gender do you identify as? Statistic Std. Error

What percentage of your cancer patients have you recommended physical activity to? 1 Mean 3.60 .199

95% Confidence Interval for Mean Lower Bound 3.20

Upper Bound 4.00

5% Trimmed Mean 3.61

Median 4.00

Variance 1.855

Std. Deviation 1.362

Minimum 1

Maximum 6

Range 5

Interquartile Range 2

Skewness -.238 .347

Kurtosis -.604 .681

2 Mean 4.38 .183

95% Confidence Interval for Mean Lower Bound 4.01

Upper Bound 4.74

5% Trimmed Mean 4.43

Median 5.00

Variance 2.143

Std. Deviation 1.464

Minimum 1

Maximum 6

Range 5

Interquartile Range 3

Skewness -.529 .299

Kurtosis -.956 .590

4) Consider other graphical outputs to present your results and make them more accessible

• Thank you, we have added sub-headings to assist with the flow of the results section, it is now much easier to follow

________________________________________

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Response to Reviewers - GP Paper - Final.docx

Click here for additional data file.

20 Oct 2020

Physical Activity for People Living with Cancer: Knowledge, Attitudes, and Practices of General Practitioners in Australia

PONE-D-20-10478R1

Dear Dr. Toohey,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Adam Todd, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #3: The authors have responded to the points raised satisfactorily. My last point in the previous iteration related to thinking about graphs to present the findings. Currently the paper is lacking in that area. The authors misunderstood that point last time.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

26 Oct 2020

PONE-D-20-10478R1

Physical Activity for People Living with Cancer: Knowledge, Attitudes, and Practices of General Practitioners in Australia

Dear Dr. Toohey:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Adam Todd

Academic Editor

PLOS ONE