Physical activity and quality of life after colorectal cancer: overview of evidence and future directions

ABSTRACT

Physical activity has been demonstrated to be a safe, feasible and efficacious intervention to improve quality of life in cancer survivors. For colorectal cancer survivors, specifically, the evidence is less clear. Whilst observational studies suggest that physical activity is positively associated with quality of life in this population, the findings from intervention trials have predominantly been null. Methodological limitations of the intervention trials to date may partially explain these findings. Future research will benefit from the use of objective activity monitoring, which will allow researchers to examine the contributions towards quality of life across the full activity spectrum (from sedentary behaviour to vigorous-intensity physical activity). Future research should also consider effects within population subgroups and study underlying biological mechanisms.

KEYWORDS:

• Physical activity
• exercise
• sedentary behaviour
• quality of life
• colorectal cancer
• cancer survivorship

Colorectal cancer: burden of disease

Colorectal cancer is one of the leading causes of morbidity and mortality worldwide. It is the third most common cancer in men, and the second most common cancer diagnosed in women.[1,2] Furthermore, colorectal is the fourth and third leading cause of cancer deaths for men and women, respectively.[1,2] Incidence rates vary widely, with age standardized rates being almost 10 times higher in more developed regions (highest rates in Australia and New Zealand) compared to less developed regions (lowest rates in Western Africa). There is less variability in mortality rates between different geographical regions (sixfold in men, fourfold in women), with the highest mortality rates estimated for Central and Eastern Europe, and the lowest in Western Africa.[1,2]

Five-year survival rates for colorectal cancer have improved over the years, and may now be as high as 65%.[3–5] Ongoing improvements in diagnosis and treatment have resulted in an expected increase in the number of people achieving 5-year survival,[6,7] particularly if worldwide trends in population ageing continue to contribute to a higher incidence of the disease.[2,7]

The ‘burden’ of colorectal cancer survivorship

For the purpose of this review, we use the term ‘cancer survivor’ to refer to any individual diagnosed with cancer, from the time of diagnosis through the rest of life, as proposed by the National Coalition for Cancer Survivorship.[8] Colorectal cancer survivors may encounter considerable physical and psychological morbidity as a result of their diagnosis and subsequent treatment.[9–11] Common psychological problems encountered by cancer survivors include depression and anxiety, and lowered self-esteem and body image.[11,12] Common physical complaints include fatigue, reduced cardiovascular and pulmonary function, muscle weakness and atrophy, pain, ataxia, bowel problems (e.g., diarrhea and/or constipation), sleeping difficulties, nausea and stoma care problems.[13,14] At the same time, the majority of colorectal cancer survivors is suffering from co-existing comorbid conditions,[15] in particular cardiovascular diseases, which will further increase due to the current population ageing.[7] Being diagnosed with cancer may also affect a person’s relationships with family, friends and colleagues.[16] There is an increasing interest from both clinicians and researchers in assessing disease outcomes not only in terms of patient survival but also in terms of quality of life during and following treatment.[14,17,18] Quality of life is a broad, multidimensional concept reflecting patients’ perceptions of both positive and negative aspects of their life with regard to physical health, psychological state, level of independence, social relationships, personal beliefs and their relationships to salient features of their environment.[19,20] Throughout this review, most of the discussion will pertain to health-related quality of life (the elements of quality of life that are affected by health, disease or treatment).[21]

Physical activity and colorectal cancer survivorship

Physical activity is a modifiable health behavior that can produce many health benefits for cancer survivors. The American Cancer Society [22] and the American College of Sports Medicine [23] recommend that, barring some clinical precautions, cancer survivors should engage in the same levels of physical activity that are currently recommended for the general population. Cancer survivors are advised to avoid inactivity and return to normal daily activities as soon as possible; to engage in moderate-to-vigorous physical activity (MVPA; activities that expend ≥3 metabolic equivalents [METs]) [24] at least 150 min/week; and, include strength training on at least two occasions per week. If older age or comorbid chronic conditions limit cancer survivors’ ability to engage in physical activity, it is recommended that survivors are as active as their abilities allow, and that they avoid long periods of inactivity.[22,23] These recommendations were generated in response to epidemiological research demonstrating that physical activity reduces colorectal cancer recurrence and mortality in a dose–response fashion.[25] Physical activity can also improve upper and lower body strength, cardiorespiratory fitness and patient-reported outcomes in colorectal cancer survivors,[26,27] and it may decrease the risk of developing comorbid chronic disease.[28]

Here, we review and synthesize the evidence relating specifically to physical activity and quality of life among colorectal cancer survivors. We discuss existing research in light of contemporary exposure assessment issues and conceptualization of the physical activity spectrum, and highlight key areas to be addressed by future research.

Observational studies of physical activity and quality of life

A marked increase in the number of observational studies investigating physical activity and quality of life outcomes among colorectal cancer survivors can be observed over recent years, with the majority of studies being published in the past decade (Table 1). These studies have been conducted in hospital-based and population-based populations of colorectal cancer survivors from multiple countries over the world, including the United States of America, Canada, Australia and several countries in Western Europe.

Table 1. Summary of observational studies investigating associations of physical activity with quality of life in colorectal cancer survivors.

Author (year)	Country	Study design	Population	PA assessment	QoL assessment	Covariates in multivariable analysis	Main results
Buffart (2012) [29]	The Netherlands	Cross-sectional	Population-based sample of 1371 stage I–IV CRC survivors; mean age 70 y; 56% men; average 3.9 y post-diagnosis.	EPIC questionnaire; min/day in MVPA were calculated.	SF-36: physical and mental QoL. Hospital Anxiety and Depression Scale: distress. Fatigue Assessment Scale: fatigue.	Sociodemographic: age, gender, having partner, employment, education level. CRC-related: time since diagnosis, tumor subsite, stage, treatment with chemotherapy. Other: smoking, BMI, comorbidity.	More time spent in MVPA was associated with higher physical QoL, and lower fatigue and distress.
Courneya (1997) [30]	Canada	Cross-sectional	Population-based sample of 130 stage I–IV CRC survivors who had received adjuvant therapy (currently post-treatment); mean age 62 y; 65% men; average 27 m post-diagnosis.	Leisure Score Index of the Godin Leisure Time Exercise Questionnaire; frequency of mild, moderate and strenuous exercise (at least 15 min duration) at 3 time points was enquired (retrospectively): pre-diagnosis, during treatment and post-treatment; total weighted activity score was calculated.	FACT-C: overall CRC-related QoL^a, physical, functional, emotional, and social well-being, and CRC-specific concerns. Satisfaction With Life Scale: overall life satisfaction score.	No multivariable adjustment was performed.	CRC survivors who reported to be active (≥16 metabolic equivalents of total exercise per week) pre-diagnosis, but inactive during treatment and inactive post-treatment reported lower physical and functioning well-being, CRC-specific concerns (lower score indicates more concerns), and overall CRC-related QoL, than the other participants.
Courneya (1999) [31]	Canada	Prospective; two measurements at baseline (~2 m post-surgery) and 4 m afterwards	Hospital-based sample of 53 stage I–IV post-surgery CRC patients; mean age 61 y; 60% men.	Leisure Score Index of the Godin Leisure Time Exercise Questionnaire; the frequency of mild, moderate and strenuous exercise (at least 15 min duration) was measured pre-diagnosis (retrospectively at baseline) and monthly in the period of 4 months after baseline; 4 measurements were averaged for post-surgery activity level.	FACT-C: overall CRC-related QoL^a, physical, functional, emotional, and social well-being, and CRC-specific concerns. Satisfaction With Life Scale: overall life satisfaction score.	No multivariable adjustment was performed.	Change in mild exercise from pre-diagnosis to post-surgery was favorably correlated with change in physical and functional well-being, CRC-specific concerns, and overall CRC-related QoL from baseline to 4 months afterwards.
Grimmett (2011) [32]	United Kingdom	Cross-sectional	Hospital-based sample of 479 stage I–III CRC survivors; mean age 68 y; 59% men; average 2 y post-diagnosis.	Godin Leisure Time Exercise Questionnaire; dichotomized into active (≥5 bouts of ≥15 min of MVPA) vs inactive (<5 bouts).	EORTC QLQ-C30: global QoL, physical, role, emotional, cognitive and social functioning, fatigue, pain, nausea/vomiting, dyspnea, appetite loss, insomnia, constipation and diarrhea.	Sociodemographic: age, gender, socioeconomic status. CRC-related: recurrence, current treatment, time since diagnosis. Other: number of comorbidities.	Compared to inactive CRC survivors, active survivors had higher global QoL, physical, role, cognitive, and social functioning, and lower levels of fatigue. Active CRC survivors reported less often pain and insomnia than inactive survivors.
Husson^b (2015a) [33] and Husson (2015b) [34]	The Netherlands	Prospective; three measurements with 1 y intervals	Population-based sample of 1739/1734 stage I–IV CRC survivors^c; mean age 68 y; 57% men; average 5 y post-diagnosis.	EPIC questionnaire; hours/week of MVPA were calculated and dichotomized into meeting vs not meeting guideline (≥150 min/week of MVPA).	EORTC QLQ-C30: global QoL, physical, role, emotional, cognitive, and social functioning. Fatigue Assessment Scale: fatigue.	Sociodemographic: age, gender, education level, marital status. CRC-related: time since diagnosis, stage, received (chemotherapy) treatment. Other: number of comorbidities, time of assessment.	Meeting vs not meeting guidelines was associated with better global quality of life, physical, role, cognitive, emotional and social functioning; and higher levels of MVPA with lower fatigue (inter-individual effects). Going from not meeting to meeting guidelines was associated with better global quality of life, physical, and role functioning; and increases in MVPA levels with lower fatigue (intra-individual effects).
Johnson^b (2009) [35] and Thraen-Borowski (2013) [36]	United States of America	Cross-sectional	Population-based sample of 843/832 stage I–IV CRC survivors^d; mean age 82 y; 48% men; average 8 y post-diagnosis.	CHAMPS physical activity questionnaire; MET-hours/week for total PA, MVPA and LPA were calculated and dichotomized into meeting vs not meeting guidelines (150 min/week of MVPA).	SF-36: physical function, role-physical, vitality, social function, role-emotional, mental health, bodily pain, general health.	Sociodemographic: age, gender, education level; social participation. Other: alcohol consumption, BMI, pain, hospitalizations, falls, number of comorbidities, and LPA and MVPA levels for the analyses of MVPA and LPA, respectively.	Total PA and MVPA were associated with higher physical functioning. Meeting PA guidelines was associated with better physical function, role-physical, bodily pain, general health, vitality and social function subscale scores.
Lewis (2014) [37]	United States of America	Prospective; three measurements at diagnosis, 12 m and 24 m post-diagnosis	Population-based sample of 453 stage II colon cancer patients^e; mean age 63 y; 40% men.	Self-developed and validated questionnaire, including recreational activities (walking, weight lifting and yoga), and light, mild and vigorous exercise at baseline. At follow-up measurements changes in physical activity were enquired (increase, no change or decrease).	FACT-C: overall CRC-related QoL^a, overall generic QoL (FACT-G)^f, Trial Outcome Index (TOI) – CRC^g, CRC-specific concerns. SF-12: physical and mental functioning.	Sociodemographic: age, gender, race, education level, ratio of income to poverty. CRC-related: chemotherapy treatment. Other: BMI, smoking, alcohol consumption, fruit and vegetable consumption, fat consumption, percentage energy from fat, supplement use.	Increasing physical activity over 24 m period post-treatment was associated with better overall CRC-related and generic QoL, TOI-CRC and physical functioning (SF-12).
Lynch (2008) [38]	Australia	Prospective; three measurements at 6, 12 and 24 m post-diagnosis	Population-based sample of 1966 CRC survivors (Duke’s stage A to D)^h; age range 20–80 y; 60% men.	Active Australia Survey; categorized into inactive (0 min/week MVPA), insufficiently active (0–149 min/week MVPA) and sufficiently active (≥150 min/week of MVPA).	FACT-C: overall CRC-related QoL^a.	Sociodemographic: age, gender, education level, marital status. CRC-related: tumor subsite, cancer stage, presence of stoma, surgery only vs surgery and adjuvant therapy. Other: BMI, fatigue, nausea, fecal control problems, smoking, time of assessment.	Both insufficiently and sufficiently active vs inactive were associated with better overall CRC-related QoL at given time point (overall effect). Higher category of MVPA (e.g., sufficiently active vs insufficiently active) was associated with better overall CRC-related QoL (inter-individual effects). Increasing one category of MVPA was associated with better overall CRC-related QoL (intra-individual effects).
Peddle (2008) [39]	Canada	Cross-sectional	Population-based sample of 413 stage I–IV CRC survivors; mean age 61 y; 55% men; average 62 m since diagnosis.	Modified Leisure Score Index from the Godin Leisure Time Exercise Questionnaire; dichotomized into meeting vs not meeting guideline (≥60 min/week of vigorous PA or ≥150 min/week of MVPA).	FACT-C: overall CRC-related QoL^a, Trial Outcome Index (TOI) – CRC^g, physical, social, emotional, and functional well-being, CRC-specific concerns fatigue (FACT-F), and TOI – fatigue^i.	Sociodemographic: education level, employment status, marital status. CRC-related: tumor subsite, local recurrence, metastatic recurrence, treatments, time since diagnosis.	Meeting vs not meeting guidelines was favorably associated with overall CRC-related QoL, TOI – CRC, physical and functional well-being, fatigue and TOI – fatigue.
Schlesinger (2014) [40]	Germany	Cross-sectional	Population-based sample of 1389 stage I–IV CRC survivors; mean age 69 y; 56% men; average 6 y since diagnosis.	EPIC questionnaire; dichotomized into above or below median hours/week of recreational activity.	EORTC QLQ-C30: global QoL; dichotomized on gender-specific median into high vs low.	Sociodemographic: age, gender, education level, living arrangement. CRC-related: presence of stoma. Other: alcohol consumption, BMI, dietary index (based on intake of fruits, vegetables, whole-grain bread and red meat), smoking status.	Higher recreational activity level (above median) was associated with a better global QoL score.
Vallance^b (2014) [41] and Vallance (2015) [42]	Canada and Australia	Cross-sectional	Population-based sample of 178/180 stage I–III colon cancer survivors^j; mean age 64 y; 56% men; average 19 m post-diagnosis.	Tri-axial Actigraph GT3X1 accelerometer; worn on hip for 7 days during waking hours to measure min/day in MVPA, which was categorized into quartiles and dichotomized into meeting vs not meeting guidelines (≥150 min/day of total MVPA).	FACT-C: overall CRC-specific QoL^a, Trial Outcome Index (TOI) – CRC^g, CRC-specific concerns, fatigue (FACT-F). Patient Health Questionnaire-9: depression symptoms Spielberger’s State Anxiety Inventory: anxiety symptoms Satisfaction with Life Scale: overall satisfaction with life.	Sociodemographic: age, gender. CRC-related: time since diagnosis, chemotherapy treatment, cancer stage. Other: smoking status, study site (Canada/Australia), BMI, sedentary time, comorbidity.	More time spent in MVPA was favorably associated with overall CRC-related QoL, TOI-CRC score, CRC-specific concerns, fatigue and satisfaction with life. Meeting MVPA guidelines was favorably associated with overall CRC-related QoL, TOI-CRC score, CRC-specific concerns, fatigue, anxiety symptoms and satisfaction with life.
van Roekel (2015) [43]	The Netherlands	Cross-sectional	Hospital-based sample of 151 stage I–III CRC survivors; mean age 70 y; 62% men; average 6 y post-diagnosis.	SQUASH; min/week in LPA and MVPA were calculated and categorized into quartiles.	EORTC QLQ-C30: global QoL, physical, role and social functioning. 12-item WHODAS II: disability. Checklist Individual Strength: fatigue. Hospital Anxiety and Depression Scale: distress.	Sociodemographic: age, gender, education level. CRC-related: time since diagnosis, cancer stage, chemotherapy treatment, Other: number of comorbidities, smoking status, BMI, and min/week in LPA and MVPA for the analyses of MVPA and LPA, respectively.	More time spent in MVPA and LPA was associated with better physical functioning; higher LPA was also associated with better role functioning and lower disability.

BMI: body mass index; CHAMPS: Community Healthy Activities Model Program for Seniors; CRC: colorectal cancer; EORTC QLQ-C30: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire – Core 30; EPIC: European Investigation into Cancer; FACT-C: Functional Assessment of Cancer Therapy-Colorectal; LPA: light physical activity; min: minutes; m: months; MVPA: moderate-to-vigorous physical activity; PA: physical activity; QoL: quality of life; SF-12: Short Form-12; SF-36: Short Form-36; SQUASH: Short Questionnaire to ASsess Health-enhancing physical activity; WHODAS II: World Health Organization Disability Assessment Schedule II.

^a FACT-C score: composite score of physical, social, emotional, and functional well-being, and CRC-specific concerns subscales.

^b Two publications arising from one study were included in one row in the table. The sets of confounders adjusted for in multivariable models slightly differed between the two studies; all used covariates in at least one of these studies are included in the table.

^c Individuals who completed ≥2 questionnaires, which could be included in multivariable longitudinal analyses of the two papers of Husson et al. (2015a) [33] and (2015b) [34].

^d The study of Johnson et al. included a total of 843 individuals,[35] and the study of Thraen-Borowski et al. 832.[36]

^e Number of individuals included at diagnosis, of whom 325 participants had data at 24 m and were included in longitudinal analysis.

^f FACT-G score: composite score of physical, social, emotional and functional well-being.

^g Composite score for physical and functional well-being, and CRC-specific concerns subscales.

^h Number of individuals with data at 6 m and included in multivariable analysis; of these 1657 had data at 12 m, and 1488 at 24 m.

ⁱ Composite scores for physical and functional well-being, and fatigue subscales.

^j The study of Vallance et al. (2014) included 178 colon cancer survivors (56% men), and the study of Vallance et al. (2015) 180 colon cancer survivors (55% men).

Table 2. Summary of intervention trials investigating the impact of physical activity on quality of life in colorectal cancer survivors.

Author (year)	Sample size	Mean age (years)	Timing	Intervention	Outcome measures	Main results
Bourke (2011)[105]	n = 18 (two arms)	Ix: 68 Control: 70	Post-treatment	Ix: supervised and home-based aerobic and resistance exercise, combined with dietary advice; 12 weeks, 30 min 3 times per week. Control: standard care (12 weeks)	1. FACT-C^a 2. FACT-F^b	1. No difference 2. No difference
Courneya (2003) [67]	n = 102 (two arms)	Ix: 60 Control: 61	Post-surgery	Ix: home-based aerobic exercise; 16 weeks, 20–30 min 3–5 times per week. Control: standard care (16 weeks)	1. FACT-G^c 2. FACT-C^a 3. FACT-F^b	1. No difference 2. No difference 3. No difference
Gillis (2014) [69]	n = 77 (two arms)	Prehab: 66 Rehab: 66	Preoperative Postoperative	Prehab and rehab groups received identical Ix. Ix: home-based exercise alternating aerobic and resistance training; 50 min 3 times per week Mean days of prehab = 24.5 days, both groups continued for 8 weeks after surgery	1. SF-36^d	1. No difference
Hawkes^e (2013) [70] and (2014) [71]	n = 410 (2 groups)	Ix: 65 Control: 68	Within 12 months after diagnosis	Ix: telephone-delivered health coaching, participant handbook and pedometer; 12-month intervention, 11 telephone health coaching sessions delivered over first 6 months Control: freely available health education brochures	1. Short SF-36^d 2. FACT-C^a	1. Between-group difference observed for physical function only 2. No difference
Pinto (2013) [68]	n = 46 (2 groups)	Ix: 60 Control: 56	Post-treatment	Ix: home-based exercise; 12 weeks, 30–50 min 2–5 times per week. Control: standard care (12 weeks)	1. FACT-C^a 2. FACT-F^b	1. No difference 2. No difference

Ix: intervention; Prehab: prehabilitation; Rehab: rehabilitation.

^a FACT-C score: composite score of physical, social, emotional, and functional well-being, and CRC-specific concerns subscales.

^b FACT-F score: composite score of physical, social, emotional, and functional well-being, and fatigue-specific concerns subscales.

^c FACT-G score: composite score of physical, social, emotional and functional well-being.

^d SF-36: physical function, role-physical, vitality, social function, role-emotional, mental health, bodily pain, general health.

^e Two publications arising from one study were included in one row in the table.

Nearly all of the observational studies to date have used self-report measures to assess physical activity levels. Measures utilized by these studies include the European Investigation into Cancer (EPIC) physical activity questionnaire,[44] Godin Leisure Time Exercise Questionnaire,[45] Community Healthy Activities Model Program for Seniors (CHAMPS) questionnaire,[46] Active Australia Survey,[47] the Short QUestionnaire to ASsess Health-enhancing physical activity (SQUASH),[48] and a self-developed physical activity questionnaire[49] (see Table 1). To date, only one published study in colorectal cancer survivors has used accelerometers (Actigraph GT3X+) to derive a more objective measurement of physical activity levels.[41] As physical activity guidelines around the world have typically focused on MVPA, most studies have examined time spent in MVPA per week, or adherence to MVPA guidelines (achieving 150 min/week; yes/no).[29,32–36,38,39,41–43] Some studies examined total physical activity (incorporating all intensities),[35,37] light physical activity [35,43] or specific types of physical activity (e.g., exercise [30,31,37] and/or recreational activities).[37,40]

A variety of quality of life outcomes have been assessed across different studies, including overall/global quality of life (general health),[30–33,36–41,43] physical functioning,[29–33,35–37,39,43] role functioning,[32,33,36,43] social functioning,[30–33,36,39,43] emotional functioning/mental health (distress: depression/anxiety),[29–33,36,37,39,42,43] cognitive functioning,[32,33] fatigue,[29,32,34,39,41,43] colorectal cancer-specific concerns,[30,31,37,39,41] symptoms,[32,36] disability [43] and satisfaction with life [30,31,42] (Table 1). Both generic (Short Form-36 [SF-36] [50] and Short Form-12 [SF-12]) [51] and colorectal cancer-specific quality of life questionnaires (Functional Assessment of Cancer Therapy-Colorectal [FACT-C] [52] and European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire – Core 30 [EORTC QLQ-C30]) [53] were applied to assess these outcomes. In addition, specific questionnaires were used to measure fatigue (Fatigue Assessment Scale,[54] Functional Assessment of Cancer Therapy-Fatigue [FACT-F] [55] and Checklist Individual Strength),[56] distress (Hospital Anxiety and Depression Scale,[57] and the Patient Health Questionnaire-9 [58] and Spielberger’s State Anxiety Inventory [59] to measure depression and anxiety symptoms, respectively), satisfaction with life (Satisfaction With Life Scale) [60] and disability (12-item World Health Organization Disability Assessment Schedule II).[61]

The reported levels of physical activity in colorectal cancer survivors across studies vary widely. For example, studies in the US,[36] Canadian [39] and Australian [38] colorectal cancer survivors observed a range in self-reported adherence to MVPA guidelines of 26–52%. Two studies that have been conducted in Dutch colorectal cancer survivors found a much higher percentage adherence to MVPA guidelines ranging from 71 to 82%.[33,43] Accelerometer data collected from Australian and Canadian colorectal cancer survivors demonstrated that 53% of study participants accrued at least 150 min/week in MVPA (this estimate included all time spent at that intensity; a smaller proportion [16%] met this threshold when only MVPA accrued in bouts of 10 min or more was included).[41] These differences are likely due to differences in the instruments applied to measure physical activity (e.g., questionnaire applied, and activities included in the questionnaire), as well as differences between study populations that may influence physical activity levels. Indeed, several sociodemographic and health-related correlates of lower physical activity have been identified in colorectal cancer survivors, including older age,[29,62,63] female gender,[29,63] lower education level,[62,64] higher body mass index[29,63] and more comorbidities.[29,62]

In spite of the differences in characteristics of study populations and methods applied to measure physical activity and quality of life, studies have consistently found higher physical activity levels to be associated with better quality of life outcomes in colorectal cancer survivors (Table 1). To our knowledge, Courneya et al. were the first to study associations of physical activity with quality of life in colorectal cancer survivors.[30] In a cross-sectional study, they observed that colorectal cancer survivors (on average 27 months post-diagnosis) who were active pre-diagnosis, but inactive during treatment and post-treatment, had lower physical and functional well-being, lower overall quality of life and more colorectal cancer-specific concerns in comparison to other participants.[30] In a subsequent prospective study in postsurgical colorectal cancers survivors, Courneya et al. found that an increase in mild exercise from pre-diagnosis (retrospectively assessed, post-surgery) to post-surgery (average of monthly measurements up to 4 months after baseline) was favorably associated with changes in physical and functional well-being, overall quality of life and colorectal cancer-specific concerns in the period from baseline to 4 months afterwards.[31] No significant associations with emotional and social functioning were observed in either of these Canadian studies.[30,31] While the analyses in these studies did not adjust for confounding factors, they were the first to indicate that physical activity may be associated with quality of life in this population.

There are now numerous cross-sectional studies that support the preliminary findings of Courneya et al., observing favorable associations of physical activity with a range of quality of life outcomes. Specifically, higher levels of MVPA and/or adherence to MVPA guidelines have been associated with better overall quality of life/general health,[32,36,39,40] physical functioning,[29,32,35,39,43] role functioning,[32,36] social functioning,[32,36] emotional functioning (lower distress),[29] cognitive functioning [32] and lower fatigue levels.[29,32,39] These findings have been replicated by Vallance et al. in a cross-sectional study with accelerometer data. Vallance et al. found that colon cancer survivors who accrued greater volumes of MVPA reported better overall quality of life and satisfaction with life,[42] and decreased colorectal cancer-specific concerns, fatigue and anxiety symptoms.[41,42] Importantly, two large prospective studies in Australian (6–24 months post-diagnosis) and Dutch colorectal cancer survivors (average 5 months post-diagnosis) have found that increases in self-reported MVPA levels were associated with increases in overall quality of life,[33,38] physical functioning,[33] and role functioning,[33] and reductions in fatigue levels.[34] Another prospective study in US colorectal cancer survivors found that increasing total physical activity in the period from diagnosis to 2 years afterwards was associated with better overall quality of life and physical functioning.[37]

Overall, the findings from observational studies are consistent, although some differences can be observed for separate quality of life outcomes. In particular, some studies have found significant associations with social functioning,[32,33,36] while others did not.[39,43] In addition, most studies did not demonstrate an association with emotional functioning/mental health,[32,36,37,39,43] but one prospective study found an inter-individual effect of meeting MVPA guidelines with better emotional functioning [33] and two cross-sectional studies reported a significant association of higher MVPA levels with lower distress [29] and lower anxiety symptoms.[42] These differences may be caused by differences across study populations (e.g., by age and time since diagnosis), or by differences in methods of assessing social and emotional functioning (which may be more subjective than physical functioning scales). Subgroup analyses have rarely been carried out to explore possible heterogeneity across sociodemographic or disease-specific characteristics.

The findings with regard to light-intensity physical activity (i.e. activities with an intensity of 1.6–2.9 metabolic equivalents [METs] [24]) and quality of life in colorectal cancer survivors are also inconsistent. One cross-sectional study in US colorectal cancer survivors did not find an association between light-intensity physical activity levels (MET-hours/week) and physical functioning, after adjustment for MVPA.[35] However, a recent study in Dutch colorectal cancer survivors observed that more time spent in light-intensity physical activity (minutes/week) was associated with better physical and role functioning and lower disability, independent from MVPA levels.[43] These different findings may again have been due to differences across study populations, or methods of light-intensity physical activity assessment.

Randomized trials of physical activity and quality of life

Relatively few randomized controlled trials with colorectal cancer survivors have examined the impact of physical activity on quality of life (Table 2). The most recent systematic review and meta-analysis on this topic was published in 2013, and reported five randomized controlled trials with a total of 238 participants.[65] Of these studies, only three evaluated quality of life.[66–68] Each of the three trials delivered home-based exercise programs, ranging from 12 to 16 weeks in duration, and each trial contained a standard care/control group. Sample sizes were small and ranged from 18 to 102 participants (mean = 54 participants). All three trials used the FACT-C [52] and the FACT-F [55] to measure disease-specific quality of life and fatigue, respectively. Overall, no evidence was found for short-term or long-term effects on quality of life or fatigue.[66–68] Since the publication of this systematic review and meta-analysis, two trials examining the effect of physical activity interventions on quality of life in colorectal cancer survivors have been published. One study examined prehabilitation (preoperative) compared to rehabilitation (postoperative) [69] and the other study examined the efficacy of a post-treatment multiple health behavior (including physical activity) intervention.[70,71]

Gillis et al. compared the effects of prehabilitation to rehabilitation exercise training among patients receiving resection for colorectal cancer.[69] Prehabilitation is based on the premise that physical activity prior to treatment may facilitate patient outcomes both during and after treatment. Early prehabilitation studies in the cancer context were conducted in lung cancer survivors and involved structured exercise training for a period of time before surgical resection.[72] In this single-blind, parallel-arm randomized controlled trial, a total of 77 colorectal cancer patients were randomized to receive either prehabilitation (n = 38) or rehabilitation (n = 39). The intervention for both groups consisted of unsupervised, moderate aerobic and resistance training up to 50 min on at least 3 days/week. Participants also received a nutrition intervention from a registered dietitian and a relaxation intervention delivered by a trained psychologist. Patients in the prehabilitation group began the intervention immediately upon randomization, while patients in the rehabilitation group started the intervention within 1 week of their surgery. The primary outcome of the intervention was functional walking capacity 8 weeks post-surgery; quality of life was measured using the SF-36.[50] Overall, there were no significant differences observed between the two groups on any of the SF-36 subscales.

In the largest randomized controlled trial to date, Hawkes et al. examined the effects of a 6-month telephone-delivered lifestyle intervention (targeted toward physical activity, weight management, dietary habits, alcohol and smoking behaviors) among 410 colorectal cancer survivors who were within 12 months of their diagnosis.[70,71] Participants were randomized to the lifestyle intervention or a usual care group. Participants in the lifestyle intervention received 11 telephone-delivered health-coaching sessions over a 6-month period. These sessions were supported by educational materials, postcard prompts, a step pedometer and a study newsletter. Quality of life and cancer-related fatigue were primary outcomes of this study; quality of life was assessed by the SF-36.[50] In the main outcomes paper, there were no between-group differences at 6 (post-intervention) or 12 months (primary end point) for any quality of life subscale (physical component score, mental component score, cancer-related fatigue).[70] A subsequent publication from this group examined the intervention effect on quality of life, as assessed by the FACT-C:[52] no significant between-groups differences emerged on any of the FACT subscales, except for physical functioning at 6 and 12 months.[71]

Overall, the evidence from randomized controlled trials for the role of physical activity in enhancing quality of life in colorectal cancer survivors appears weak. This is contrary to the evidence that exists for cancer survivors overall, which suggests that physical activity has a beneficial effect on quality of life.[18] There are, however, several plausible explanations for the lack of observed effects in studies conducted among colorectal cancer survivors. These include the small sample sizes (studies may not have been adequately powered to detect significant differences in quality of life between intervention and control groups), and the home-based, unsupervised nature of the physical activity interventions (which may have compromised intervention adherence and thus reduced the capacity of the interventions to produce meaningful increases in physical activity and thereby quality of life).

Despite the limited success of the aforementioned trials, several larger trials examining physical activity and quality of life in colorectal cancer survivors are ongoing. Ho et al. are undertaking a 3-year randomized controlled trial designed to increase physical activity and multiple health outcomes, including quality of life, in a sample of 222 colorectal cancer survivors.[73] The intervention is distance-based and includes telephone calls, stage-matched print materials and quarterly group meetings. Physical activity is being assessed via accelerometer, and quality of life is being assessed via both generic (SF-12) [51] and disease-specific measures (FACT-C).[52] Courneya et al. are currently conducting the Colon Health and Life-Long Exercise Change (CHALLENGE) trial, designed to determine the effects of a structured physical activity intervention on a range of health outcomes, including quality of life, among stage II or III colon cancer survivors who have recently completed adjuvant therapy.[74] In this international trial (involving participants from Canada, Australia, the United States, France, Israel and Korea, plus the United Kingdom in 2016), 962 survivors will be randomized to the physical activity intervention or to a control group. Intervention participants receive a structured and supervised 3-year physical activity program with behavioral support (face-to-face and telephone). The primary end point is disease-free survival; quality of life is a secondary end point, and will be assessed using the SF-36,[50] FACT-G [75] and FACT-F.[55] These ongoing intervention trials will provide more insight into the potential role of physical activity as a means to improve quality of life among colorectal cancer survivors.

Sedentary behavior and quality of life

Sedentary behavior (commonly conceptualized as sitting time) is defined by low energy expenditure (≤1.5 METs) and a sitting or lying posture during waking hours.[76] It is a health-related risk factor that is additional to, and distinct from, the hazards of insufficient MVPA: high levels of MVPA and sedentary behavior can coexist.[77] A growing body of evidence has associated sedentary behavior with all-cause and cardiovascular mortality, type 2 diabetes and cardiovascular disease.[77,78]

Among colorectal cancer survivors, self-reported sedentary behavior has been associated with poorer survival,[79,80] increased risk of heart disease [28] and an increase in BMI.[81] Only two studies to date have examined the association between sedentary behavior and quality of life (both using the FACT-C) [52] in this population. In a prospective study in colorectal cancer survivors, Lynch et al. found that survivors who watched television (a common leisure-time sedentary behavior) for more than 5 h/day had a 16% lower total quality of life score than survivors watching less than 2 h/day. Furthermore, survivors who increased their television viewing from one category to the next (e.g., moving from less than 2 h/day to between 3 and 4 h/day) had significantly lower quality of life than those who maintained or decreased their viewing time. Deleterious associations of television viewing time were found with quality of life, overall and across all subscales (physical well-being, social well-being, emotional well-being, functional well-being and colorectal cancer-specific concerns).[82] Conversely, Vallance et al. found that accelerometer-assessed sedentary time (minutes/day) was not associated with quality of life in colon cancer survivors. This cross-sectional study considered total sedentary time, and time accrued in prolonged bouts of 30 min or more; no meaningful relationship with quality of life emerged for either variable.[41] In another publication from the same study, Vallance et al. observed no association of accelerometer-assessed sedentary time with symptoms of depression or anxiety, or with satisfaction with life.[42]

For sedentary behavior, context may be extremely important when considering associations with quality of life, and may help to explain the disparate findings across these studies. Some sedentary behaviors, such as television viewing, may remove individuals from social interactions and thereby compromise social and emotional components of quality of life. Other sedentary behaviors, such as having coffee with friends or participating in online chat, may promote social interaction, and have the opposite effect.[83] Another explanation for the difference in findings from the two sedentary behavior and quality of life studies may be the use of a hip-worn accelerometer by Vallance et al., which poorly distinguishes sitting from standing.[84]

Self-report versus objective activity monitoring: the importance of exposure assessment

Nearly all of the research pertaining to physical activity and quality of life among colorectal cancer survivors has relied on self-reported estimates of physical activity (and sedentary behavior). Physical activity questionnaires have focused predominantly on MVPA, and sedentary behavior has mostly been quantified by a simple item assessing television viewing time. Such measures capture only a small fraction of an individual’s total activity.[85] Contemporary research,[86,87] using accelerometers, indicates that adults spend the vast proportion of their day sitting or in light-intensity physical activity; these behaviors are difficult to measure accurately and reliably by questionnaire.[88] Further, self-report measures of physical activity are prone to random and systematic error; the latter, in particular, may change the mean value of a score across a study sample and the strength and/or direction of associations of physical activity with other variables,[89] such as quality of life.

A recent study by Boyle et al. highlights the considerable exposure misclassification bias that can occur in studies using self-reported estimates of MVPA and sedentary behavior in cancer survivors. Boyle et al. compared the accelerometer-based (hip-worn Actigraph GT3X+) and self-reported estimates of MVPA (Godin Leisure Time Exercise Questionnaire) [45] and sedentary behavior (Marshall Sitting Time Questionnaire) [90] of 176 colon cancer survivors from Australia and Canada.[91] As the MVPA self-report questionnaire asked respondents to only record physical activity performed for at least 10 min, the accelerometer-derived estimate was similarly comprised of MVPA time accrued in bouts of at least 10 min (allowing for an interruption of up to 2 min). There was a large difference in estimates of MVPA derived from different methods: the mean minutes per day spent in MVPA was 12 min based on accelerometer data and 26 min based on self-reported data. Correlation between the methods was fair (ρ = 0.51); however, agreement was poor (ICC = 0.33; 95% CI: 0.18, 0.46). Mean daily time spent sedentary per day was similar for both methods (approximately 8.5 h/day); however, both correlation (ρ = 0.19) and agreement were poor (ICC = 0.16; 95% CI: 0.01, 0.30).

Objective monitoring devices, such as accelerometers, provide detailed data on the frequency, intensity and duration of physical activities performed across the day, and how these activities are accumulated over the course of the day and across the week. These electronic devices are small and light-weight, and are usually worn on a waist band, wrist strap or attached with sticking plaster to the thigh. Their relatively low cost has resulted in rapid adoption as a method for assessing physical activity in free-living participants.[92] However, they do not accurately capture workloads pertaining to certain activities such as stationary biking, swimming, weight training. Some types of accelerometers do not capture data relative to body positioning (e.g., sitting, standing, lying down) well, which can limit their ability to measure sedentary behavior and/or stationary standing.[84] Further, there are limitations to using ‘cutpoints’ to interpret accelerometer data, but there is considerable work being done to refine machine learning algorithms that improve the processing and interpretation of accelerometer data.[92] While it cannot identify the settings in which physical activity occurs, nor other important contextual data, objective activity monitoring nevertheless reduces the measurement error that has typically been associated with self-report physical activity.[93] Studies that have incorporated objective activity monitoring into their data collection schedules are now beginning to deliver important scientific information on relationships of these behaviors with health outcomes.

Expert commentary

Summary of research findings to date: physical activity and quality of life

A consistent body of evidence from observational studies has emerged, demonstrating favorable associations of higher levels of MVPA with multiple quality of life outcomes among colorectal cancer survivors. Most of these studies have been cross-sectional; only four prospective studies have been conducted to date. Some of the observational studies have recruited large, representative samples, which has allowed researchers to examine associations of MVPA with multiple quality of life outcomes. Emerging evidence also suggests that light physical activity may be associated with quality of life, although few studies have been conducted to date.

Only five randomized, controlled trials have examined the effect of physical activity interventions on quality of life in colorectal cancer survivors. None have demonstrated a significant effect on quality of life outcomes, except for one large trial observing an association with physical well-being. That said, no adverse effects on health-related outcomes have been reported from these trials. A key limitation of trials to date is the questionable fidelity to home-based intervention protocols and lack of a meaningful increase in physical activity levels. There are at least two ongoing randomized trials that will provide higher quality data that will enable a more rigorous assessment of the feasibility of using physical activity as an intervention to improve quality of life in colorectal cancer survivors.

Finally, only two studies have examined sedentary behavior and its association with quality of life in this population, and their results were inconsistent. The prospective study employing self-reported television viewing time as a proxy for sedentary behavior found significant, unfavorable associations with total quality of life. Conversely, the cross-sectional study employing hip-worn accelerometers to assess sedentary time found no association with several quality of life outcomes.

On the basis of the available evidence (particularly when considering the positive associations of physical activity with survival outcomes) it is reasonable to counsel colorectal cancer survivors to engage in regular physical activity, if feasible. Similarly, although the scant evidence pertaining to sedentary behavior and quality of life in this population is mixed, there are few – if any – contraindications to recommending colorectal cancer survivors spend less time sitting and more time engaged in non-sedentary tasks.[94] Indeed, it has been proposed that reducing sedentary behavior may be a more feasible and appropriate approach for a wider proportion of older adults (particularly clinical populations, and individuals for whom exercise contraindications exist) than focusing solely on physical activity promotion.[95]

Future research directions

To address the significant gaps in knowledge that exist, we suggest that further research, preferably with a prospective design and incorporating objective activity monitoring, is needed to further study associations of all intensities of activity (sedentary behavior, light-intensity physical activity and MVPA) with quality of life in colorectal cancer survivors. Accelerometers are now relatively inexpensive, and methods for data collection and summarization are now well developed (and continually being improved).[96] These data provide detailed information on the volumes and patterns of different activity intensities across the whole day (or week).

Future research should also consider the effects of physical activity on quality of life within population subgroups. To date, research has been conducted in predominantly Caucasian populations (although the ongoing Ho et al. trial has recruited Chinese participants). It is unlikely that physical activity affects quality of life uniformly across all colorectal cancer survivors. These associations may be moderated by particular demographic attributes; site, stage or sub-type of disease; treatment regime; comorbid conditions; or genetic profiles. Understanding any relevant subgroup differences will help design and deliver the most appropriate form of intervention to those who are most likely to benefit. Tailoring interventions in this manner would improve the cost efficiency of such interventions.[93]

Finally, it is important to understand the biological mechanisms that underpin associations between physical activity and quality of life in colorectal cancer survivors.[97] Different pathways are likely implicated for different quality of life-related constructs, such as physical functioning, emotional well-being and fatigue.[98] Future research should consider if and how the effect of physical activity on quality of life is mediated by molecular pathways (e.g., metabolic hormones or inflammatory peptides/cytokines), genetic and epigenetic processes, immune response and the microbiome. Molecular pathologic epidemiology is a multidisciplinary field that considers the interrelationship of exogenous and endogenous factors, and may provide useful insights to these ends.[99] Mechanistic understanding could strengthen causal inferences from epidemiological data, provide insights into gene–environment interactions and identify new (or refine existing) intervention targets.

Five-year view

We anticipate that additional observational studies of physical activity and quality of life among colorectal cancer survivors will be initiated in the near future. Observational studies can recruit large, population-based samples to provide a wide variation of data across both the physical activity and quality of life spectrums. In addition, the representative nature of data from well-designed observational studies allows us to extrapolate results to the ‘true’ population of colorectal cancer survivors. We expect to soon see prospective studies collecting objective data on physical activity and sedentary behavior, as well as biological samples (e.g., blood or saliva) that will enable the study of potential biological mechanisms mediating the physical activity – quality of life pathway. Advancements in statistical theory and methods have greatly improved the capacity for causal inference from observational data.[100] Prospective studies will increasingly employ more complex modeling strategies (e.g., marginal structural models) to estimate causal effects, accounting for variation in physical activity over time (as well as time varying confounding and mediation).[101]

The next 5 years should see results emerge from at least two ongoing trials of physical activity inventions in colorectal cancer survivors.[73,74] Outcomes from these studies will provide more definitive answers regarding the causal effects of physical activity on quality of life outcomes in this population. We also expect the commencement of new physical activity intervention trials, incorporating wearable technology activity monitors, such as ‘fitness bands’. Fitness bands are popular with consumers, and provide automated feedback on time spent in physical activity and sedentary behavior, as well as interactive behavior change tools, via mobile telephone applications or personal computer. The technology is low cost, has wide reach and has broad applications for use in clinical and public health settings. Initial interventions using fitness bands have shown promising improvements in physical activity.[102,103]

Finally, we anticipate that prognostic prediction models of quality of life for colorectal cancer survivors will be a tool employed to personalize interventions toward survivors who would benefit most.[104] Such models will also help clinicians to plan and provide appropriate treatment for their patients, including referral to exercise physiologists or physical therapists for physical activity-related therapies, when appropriate.

Financial & competing interests disclosure

BM Lynch is supported by a fellowship from the National Breast Cancer Foundation (ECF-15-012). EH van Roekel is funded by a grant from the Stichting Alpe d’HuZes within the research program ‘Leven met kanker’ of the Dutch Cancer Society (Grant No. UM-2010-4867), and by the GROW School for Oncology and Developmental Biology. JK Vallance is supported by the Canada Research Chairs program and a Population Health Investigator Award from Alberta Innovates – Health Solutions. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Key issues

• Observational studies have consistently demonstrated a significant, positive association between moderate-to-vigorous physical activity and quality of life in colorectal cancer survivors.

• Emerging evidence also suggests that light physical activity may be associated with quality of life, although few studies have been conducted to date.

• Randomized, controlled trials of physical activity interventions delivered to colorectal cancer survivors have not produced any significant effects, with the exception of a modest improvement in physical well-being in one trial.

• Methodological limitations of the intervention trials to date (e.g., the small sample sizes, and the home-based, unsupervised nature of the physical activity interventions) may at least partially explain the lack of effect on quality of life.

• Sedentary behavior may also affect the quality of life of colorectal cancer survivors. Only two studies have addressed this question, with mixed findings.

• The incorporation of objective activity monitoring into future studies is essential. Use of accelerometers will improve the quality of the physical activity data collected, and allow examination of associations across the whole spectrum of activity (sedentary behavior, light-, moderate- and vigorous-intensity physical activity).

• Future research should consider the effects of physical activity on quality of life within population subgroups, so as to enable tailoring of interventions and delivery to colorectal cancer survivors who would benefit most.

• Research is needed into the biological mechanisms that underpin the associations between physical activity and quality of life in colorectal cancer survivors, to strengthen causal inferences from epidemiological data and identify intervention targets.

• Newer statistical techniques have improved the capacity for causal inference from observational data. Research in this field must keep pace with these developments, in order to fully exploit the data that have been and will be collected.