Physical activity and obesity in Canadian cancer survivors
Population-based estimates from the 2005 Canadian Community Health Survey
Physical inactivity and obesity are associated with poorer disease outcomes in several cancer survivor groups. Few studies, however, have provided population-based estimates of these risk factors in cancer survivors and compared them with individuals without a history of cancer. Here such estimates for the Canadian population are reported.
Data were obtained from the 2005 Canadian Community Health Survey consisting of computer-assisted interviews of 114,355 adults representing an estimated 23,285,548 Canadians. Participants self-reported their cancer history, height, and body weight to calculate body mass index and participation in various leisure-time activities.
Fewer than 22% of Canadian cancer survivors were physically active and over 18% were obese. Few differences were observed between cancer survivors and those without a history of cancer except that: 1) prostate cancer survivors were more likely to be active (adjusted odds ratio [OR] = 1.27; 95% confidence interval [CI] = 1.01–1.59) and less likely to be obese (adjusted OR = 0.71; 95% CI = 0.56–0.90); 2) skin cancer survivors (nonmelanoma and melanoma) were more likely to be active (adjusted OR = 1.33; 95% CI = 1.12–1.59); and 3) obese breast cancer survivors were less likely to be active compared with obese women without a history of cancer (adjusted OR = 0.51; 95% CI = 0.27–0.94).
Canadian cancer survivors have low levels of physical activity and a high prevalence of obesity that, although comparable to the general population, may place them at higher risk for poorer disease outcomes. Population-based interventions to increase physical activity and promote a healthy body weight in Canadian cancer survivors are warranted. Cancer 2008. © 2008 American Cancer Society.
Obesity and physical inactivity are increasingly being recognized as important risk factors for poorer outcomes in several cancer survivor groups.1–5 For example, 3 recent studies of colorectal and breast cancer survivors have reported that higher levels of postdiagnosis physical activity were associated with reduced risks of disease recurrence, cancer-specific mortality, and all-cause mortality.2, 4, 5 Even more research has linked obesity and weight gain to disease outcomes in breast and other cancer survivor groups.1 Finally, a recent meta-analysis of 14 exercise trials in breast cancer survivors concluded that exercise improves quality of life, physical functioning, and reduces fatigue.3 Taken together, these studies suggest that physical activity and body weight may play important roles in disease control and supportive care outcomes in several cancer survivor groups.
Few studies, however, have examined the prevalence of physical activity and obesity in population-based samples of cancer survivors.6 Two US studies used National Health Interview Survey (NHIS) data and reported slightly different results.7, 8 Both studies reported no differences in overweight and obesity levels between cancer survivors and noncancer controls; however, Bellizzi et al.7 reported that cancer survivors were slightly more active than matched controls, whereas Coups and Ostroff8 reported that middle-aged cancer survivors (40–64 years) were slightly less active than controls. The difference in results for activity levels may have resulted from Bellizzi et al.7 adjusting for more variables than Coups and Ostroff,8 including functional limitations.
There are no published representative data on physical activity or obesity levels in Canadian cancer survivors. Our group has conducted several population-based surveys of specific cancer survivor groups in Alberta, including non-Hodgkin lymphoma,9 endometrial,10 bladder,11 multiple myeloma,12 ovarian,13 and colorectal14 but these studies are limited because: 1) they were restricted to Alberta and not representative of the Canadian population; 2) no comparison was made to matched controls; 3) no surveys have been done of breast or prostate cancer survivors, which are the 2 most common cancers in Canada; and 4) the study purpose was transparent, likely resulting in a selection bias of more active survivors.6 The purpose of the present study was to provide population-based estimates of physical activity and obesity levels in Canadian cancer survivors and to compare these prevalences to matched controls without a history of cancer. These data may provide a benchmark for these risk factors that can be used to inform cancer control efforts in Canada.
MATERIALS AND METHODS
Our analysis is based on the Canadian Community Health Survey (CCHS), Cycle 3.1 (2005) conducted by Statistics Canada.15 One of the main objectives of the CCHS survey is to provide information about the health determinants of Canadians 12 years of age and older with the exception of persons living on Indian Reserves or Crown Lands, those residing in institutions, full-time members of the Canadian Forces, and residents of certain remote regions. The 2005 cycle of the CCHS covered approximately 98% of the Canadian population aged 12 and older. Collection for CCHS Cycle 3.1 took place from January to December 2005. Over the collection period a total of 132,947 valid interviews were conducted using computer-assisted interviewing. Approximately half the interviews were conducted in person and the other half were conducted by telephone.
In total, 168,464 households were selected to participate in the CCHS Cycle 3.1. Of these, responses were obtained from 143,076 households resulting in an overall household-level response rate of 84.9%. From the responding households, 143,076 individuals (1 person per household) were selected to participate in the CCHS Cycle 3.1, out of which a response was obtained from 132,947 individuals, resulting in an overall person-level response rate of 92.9%. At the Canadian population level, this yields a combined response rate of 78.9% for the CCHS Cycle 3.1. The present analysis was restricted to 114,355 respondents aged 18 years or older that had data for the variables of interest, representing an estimated 23,285,548 Canadians.
Cancer Survivor Status
The questions “Do you have cancer?” and “Have you ever been diagnosed with cancer?” were used to establish a history of cancer. An affirmative response to either question was used to indicate a positive history of cancer while a negative response to both questions indicated an absence of cancer history. Although it would have been possible to analyze these items separately as a crude indicator of current disease and/or treatment status, we chose not to conduct such an analysis because of concerns of the validity of such items and the reduced sample sizes. The type of cancer that the respondent had was ascertained from 1 of 2 questions based on sex. Females were asked “What type of cancer do/did you have: breast, colorectal, skin melanoma, skin nonmelanoma, other?” and males were asked the same question with “breast” replaced by “prostate.”
Body Mass Index and Physical Activity Levels
Self-reported height and body weight were used to calculate body mass index (BMI; kg/m2). Participants were classified as healthy weight (including underweight; BMI < 25 kg/m2), overweight (25 kg/m2 to <30 kg/m2), or obese (BMI ≥ 30 kg/m2) according to Health Canada guidelines.16 Physical activity level was assessed using a modification of the Minnesota Leisure-time Physical Activity Questionnaire.17 Respondents were asked if they had participated in any of the following activities during their leisure time in the past 3 months: walking for exercise, gardening or yard work, swimming, bicycling, popular or social dance, home exercises, ice hockey, ice skating, in-line skating or rollerblading, jogging or running, golfing, exercise class or aerobics, downhill skiing or snowboarding, bowling, baseball or softball, tennis, weight-training, fishing, volleyball, basketball, soccer, and any additional physical activities not specified by the interviewer. They were then asked the number of times they engaged in the activity and the average duration per session. These data were weighted by the metabolic equivalent task (MET) value associated with each activity to arrive at an energy expenditure value expressed in kilocalories per kilogram of body weight per day (kcal/kg/day). Physical activity level was categorized as: inactive (<1.5 kcal/kg/day), moderately active (1.5–2.99 kcal/kg/day), or active (≥3.0 kcal/kg/day). Expending 1.5 kcal/kg/day is roughly equivalent to walking 30 minutes/day whereas expending 3.0 kcal/kg/day is roughly equivalent to walking 1 hour/day.18 These cutpoints correspond roughly to the ‘minimal’ and ‘desirable’ public health guidelines, respectively, proposed by most federal agencies including Health Canada,19 the American College of Sports Medicine and the American Heart Association,20 the Institute of Medicine,21 and the American Cancer Society.22
Age, sex, ethnicity, and annual income were used as covariates in the statistical analyses. Age was calculated from birth and observation dates, while sex was coded as a dichotomous variable. The CCHS questionnaire asks questions about ethnic origin and allows for multiple responses. Because of sample size limitations, however, ethnicity was coded as a dichotomous variable (1 = white, 2 = nonwhite). To adjust for discrepancies in the income levels of respondents, the income decile of the respondent was included in the statistical models. This derived variable represents a distribution of household income which is relative to the low-income cutoff for their respective household and community size. The deciles only contain data from respondents that offered valid responses to income questions; therefore, answers of “not stated” or “refused” resulted in the respondents being excluded from the analyses. We also adjusted for mode of data collection (in-person vs telephone interview) and found no effects on our odds ratios (ORs) to the second decimal point (data not shown).
On the basis of the 2005 sample the weighted prevalences and number of men and women in Canada reporting a cancer history were computed. In addition, the proportions of men and women in each of the categories of leisure-time physical activity and BMI with a history of cancer were calculated and compared with the proportions of respondents not reporting a history of cancer. The adjusted ORs for the prediction of physical activity (≥3.0 kcal/kg/day) and obesity (BMI ≥ 30 kg/m2) were calculated using logistic regression models. For each estimate, 95% confidence intervals (CIs) were computed using boot-strapping techniques and the SAS “Bootvar” macro using bootstrap weights supplied by Statistics Canada. Records with missing or unknown values for any of the required variables were excluded from the analysis. All data analyses were conducted using SAS v. 9.1 (Cary, NC).
Table 1 presents the weighted population prevalences and the number of Canadian adults reporting the various types of cancer recorded in the CCHS. Approximately 3.4% of men and 5.6% of women reported being diagnosed with a cancer other than skin cancer. These prevalence rates translate into more than 1,000,000 Canadian survivors of ‘nonskin’ cancers including about 223,000 breast, 147,000 prostate, and 100,000 colorectal cancer survivors. Table 2 presents the descriptive demographic characteristics for cancer survivors and those without a history of cancer and shows that cancer survivors were more likely to be female, older, and white.
Table 1. Prevalence (%) and Number of Adults 18+ Years of Age in the 2005 Canadian Community Health Survey (Cycle 3.1) Reporting a History of Cancer by Site
Table 2. Demographic Profile of Study Participants Overall and by Cancer Survivor Status
| Decile 1||9.6||10.5||9.6|
| Decile 2||9.9||13.2||9.7|
| Decile 3||9.8||12.2||9.7|
| Decile 4||9.8||10.7||9.7|
| Decile 5||9.9||9.2||10.0|
| Decile 6||10.2||9.8||10.2|
| Decile 7||10.1||8.8||10.2|
| Decile 8||10.3||8.1||10.4|
| Decile 9||10.0||8.6||10.1|
| Decile 10||10.5||9.1||10.6|
Table 3 presents the weighted population prevalences for physical activity and obesity among cancer survivors and those without a history of cancer. Fewer than 22% of cancer survivors reported being active, with the lowest rates reported by female colorectal cancer survivors (13.8%), breast cancer survivors (16.6%), female melanoma skin cancer survivors (19.1%), and male colorectal cancer survivors (20.1%). Overall, female cancer survivors were less likely to be active than male cancer survivors (19.6% vs 24.1%). An additional 25% of cancer survivors reported being moderately active. Just over 18% of cancer survivors reported being obese and another 34% reported being overweight, with little variation among the cancer survivor groups. Male cancer survivors, however, were less likely to be healthy weight than female cancer survivors (38.3% vs 53.3%) and more likely to be overweight (43.5% vs 28.2%).
Table 3. Proportions [95% Confidence Intervals] of Canadian Adults 18+ Years of Age in Each Physical Activity and Body Mass Index (BMI) Category by Cancer Survivor Status in the 2005 Canadian Community Health Survey (Cycle 3.1)
| Active||23.9 [20.0–27.8]||20.1 [11.5–28.6]||31.5 [24.4–38.5]||26.2 [19.8–32.5]||20.3 [17.0–23.6]||24.1 [21.6–26.6]||27.4 [26.8–28.0]|
| Moderate||24.4 [20.8–28.1]||24.3 [15.5–33.0]||23.0 [17.7–28.2]||31.9 [25.4–38.5]||21.4 [17.7–25.1]||23.4 [21.1–25.6]||25.0 [24.4–25.6]|
| Inactive||51.7 [47.2–56.2]||55.7 [45.9–65.4]||45.5 [38.3–52.8]||41.9 [35.2–48.7]||58.3 [54.1–62.5]||52.5 [49.9–55.2]||47.6 [46.9–48.3]|
| Healthy weight||36.1 [32.0–40.2]||43.7 [32.1–55.3]||36.9 [29.5–44.4]||35.6 [29.1–42.2]||38.3 [34.0–42.6]||38.3 [35.7–40.9]||42.0 [41.3–42.6]|
| Overweight||48.3 [44.1–52.6]||38.9 [29.5–48.4]||41.0 [33.9–48.0]||46.2 [39.1–53.3]||41.6 [36.8–46.4]||43.5 [41.0–46.1]||41.1 [40.5–41.8]|
| Obese||15.6 [12.8–18.4]||17.3 [11.4–23.3]||22.1 [15.6–28.6]||18.2 [12.8–23.5]||20.2 [16.5–23.8]||18.1 [16.2–20.0]||16.9 [16.4–17.4]|
|Breast||Colorectal||Skin melanoma||Skin nonmelanoma||Other||Any cancer||No cancer|
| Active||16.6 [13.8–19.5]||13.8 [09.0–18.5]||19.1 [14.3–23.8]||21.4 [16.4–26.5]||21.2 [18.2–24.1]||19.6 [17.8–21.5]||22.7 [22.1–23.2]|
| Moderate||26.8 [23.3–30.3]||22.6 [16.1–29.0]||27.3 [21.3–33.2]||27.0 [20.8–33.2]||23.4 [20.5–26.3]||25.2 [23.3–27.1]||26.1 [25.6–26.7]|
| Inactive||56.6 [52.6–60.5]||63.7 [56.6–70.7]||53.7 [47.2–60.1]||51.6 [44.9–58.3]||55.4 [52.2–58.6]||55.2 [53.1–57.3]||51.2 [50.6–51.9]|
| Healthy weight||51.4 [47.7–55.0]||54.1 [46.6–61.6]||51.3 [44.8–57.8]||54.6 [48.1–61.2]||54.4 [51.4–57.4]||53.3 [51.2–55.3]||58.6 [58.0–59.2]|
| Overweight||30.1 [26.9–33.2]||27.5 [20.6–34.5]||30.7 [25.4–36.0]||24.8 [19.6–30.0]||27.2 [24.7–29.7]||28.2 [26.5–29.9]||27.0 [26.4–27.5]|
| Obese||18.6 [15.8–21.4]||18.4 [12.4–24.4]||18.0 [13.4–22.6]||20.6 [15.7–25.4]||18.4 [16.1–20.7]||18.5 [17.0–20.1]||14.4 [14.0–14.8]|
| || ||Both Sexes|
|Colorectal||Skin melanoma||Skin nonmelanoma||Other||Any cancer||No cancer|
| Active|| ||17.1 [12.0–22.2]||25.5 [21.1–30.0]||23.6 [19.6–27.6]||20.9 [18.7–23.1]||21.4 [20.0–22.9]||25.1 [24.7–25.5]|
| Moderate|| ||23.5 [17.9–29.1]||25.0 [21.0–29.0]||29.2 [24.8–33.6]||22.7 [20.5–25.0]||24.5 [23.0–25.9]||25.6 [25.2–26.0]|
| Inactive|| ||59.4 [53.2–65.5]||49.4 [44.6–54.3]||47.2 [42.3–52.1]||56.4 [53.9–59.0]||54.1 [52.5–55.7]||49.4 [48.9–49.8]|
| Healthy weight|| ||48.6 [41.4–55.4]||43.8 [38.8–48.7]||46.0 [41.1–50.9]||48.8 [46.2–51.4]||47.2 [45.5–48.9]||50.2 [49.7–50.6]|
| Overweight|| ||33.6 [27.7–39.6]||36.1 [31.7–40.5]||34.5 [29.6–39.4]||32.2 [29.8–34.5]||34.4 [32.9–35.9]||34.1 [33.7–34.6]|
| Obese|| ||17.8 [13.6–22.0]||20.1 [16.2–24.1]||19.5 [16.0–22.9]||19.0 [17.0–21.0]||18.4 [17.1–19.6]||15.7 [15.4–16.0]|
Table 4 presents the odds of being active or obese depending on whether or not the participant reported a history of cancer. The primary differences that emerged between cancer survivors and those without a history of cancer were: 1) prostate cancer survivors were more likely to be active (adjusted OR = 1.27; 95% CI = 1.01–1.59) and less likely to be obese (adjusted OR = 0.71; 95% CI = 0.56–0.90), and 2) skin cancer survivors (melanoma and nonmelanoma combined) were more likely to be active (adjusted OR = 1.33; 95% CI = 1.12–1.59), although this association appeared to be restricted primarily to men. Table 5 reports the results of the logistic regression models predicting either obesity or physical activity within stratified categories of physical activity or BMI. Results showed that obese breast cancer survivors were less likely to be active than obese women without a history of cancer (adjusted OR = 0.51; 95% CI = 0.27–0.94).
Table 4. Odds Ratios [95% Confidence Intervals] for Being Physically Active or Obese Based on Cancer Survivor Status in Men and Women 18+ Years of Age in the 2005 Canadian Community Health Survey (Cycle 3.1)
| Yes||1.14 [0.67–1.94]||0.86 [0.55–1.35]||0.95 [0.60–1.50]||0.88 [0.58–1.34]||1.07 [0.74–1.55]||0.86 [0.63–1.17]|
| Yes||1.27 [1.01–1.59]||0.71 [0.56–0.90]||—||—||—||—|
| Yes||—||—||1.01 [0.80–1.06]||1.06 [0.86–1.31]||—||—|
| Yes||0.89 [0.72–1.10]||1.06 [0.81–1.38]||1.15 [0.94–1.40]||1.12 [0.94–1.34]||1.04 [0.90–1.21]||1.10 [0.95–1.28]|
| Yes||1.62 [1.14–2.30]||1.34 [0.90–1.99]||1.04 [0.73–1.48]||1.06 [0.77–1.47]||1.35 [1.05–1.74]||1.21 [0.94–1.56]|
| Yes||1.27 [0.90–1.80]||0.98 [0.66–1.48]||1.26 [0.91–1.74]||1.19 [0.87–1.63]||1.27 [1.00–1.60]||1.08 [0.85–1.37]|
| Yes||1.49 [1.16–1.90]||1.11 [0.83–1.49]||1.16 [0.91–1.48]||1.10 [0.88–1.38]||1.33 [1.12–1.59]||1.10 [0.92–1.32]|
| Yes||1.07 [0.91–1.26]||0.87 [0.73–1.04]||1.11 [0.95–1.29]||1.10 [0.95–1.26]||1.09 [0.97–1.22]||1.00 [0.89–1.12]|
Table 5. Odds Ratios [95% Confidence Intervals] for Being Physically Active or Obese Based on Cancer Survivor Status in Men and Women 18+ Years of Age in the 2005 Canadian Community Health Survey (Cycle 3.1), Stratified by Body Mass Index and Physical Activity Categories
| Yes||1.22 [0.89–1.68]||0.85 [0.60–1.21]||0.51 [0.27–0.94]||1.14 [0.87–1.48]||1.16 [0.77–1.75]||0.53 [0.29–0.98]|
| Yes||1.13 [0.77–1.66]||1.24 [0.88–1.73]||1.61 [0.90–2.89]||0.69 [0.49–0.97]||0.72 [0.44–1.17]||0.82 [0.46–1.45]|
| Yes||1.25 [0.70–2.25]||0.80 [0.48–1.33]||0.97 [0.50–1.89]||0.72 [0.48–1.10]||1.36 [0.76–2.42]||0.81 [0.39–1.68]|
| Yes||1.04 [0.70–1.55]||1.98 [1.40–2.79]||1.24 [0.57–2.69]||1.32 [0.94–1.86]||1.22 [0.76–1.97]||1.10 [0.51–2.35]|
| Yes||1.14 [0.81–1.61]||1.42 [0.95–2.13]||1.36 [0.73–2.51]||1.17 [0.84–1.63]||0.99 [0.61–1.61]||1.20 [0.61–2.38]|
| Yes||1.02 [0.82–1.27]||0.95 [0.77–1.19]||1.35 [0.88–2.06]||0.94 [0.79–1.13]||1.32 [0.94–1.86]||1.49 [0.99–2.22]|
| Yes||1.10 [0.84–1.44]||1.71 [1.31–2.24]||1.35 [0.85–2.16]||1.21 [0.96–1.53]||1.03 [0.72–1.47]||1.14 [0.70–1.84]|
| Yes||1.13 [0.96–1.33]||0.98 [0.83–1.16]||1.18 [0.85–1.63]||0.93 [0.81–1.06]||1.14 [0.90–1.46]||1.10 [0.80–1.50]|
In Canada there were over 1,000,000 survivors of ‘nonskin’ cancers in 2005 including 223,000 breast, 147,000 prostate, and 100,000 colorectal cancer survivors. Fewer than 22% of these survivors were physically active at the desired level, with the lowest activity rates reported by female colorectal cancer survivors (13.8%), breast cancer survivors (16.6%), female melanoma skin cancer survivors (19.1%), and male colorectal cancer survivors (20.1%). Just over 18% of cancer survivors reported being obese, with little variation among the cancer survivor groups. The only differences between cancer survivors and those without a history of cancer were: 1) prostate cancer survivors were more likely to be active and less likely to be obese; 2) skin cancer survivors, especially men, were more likely to be active; and 3) obese breast cancer survivors were less likely to be active than obese women without a history of cancer.
Our data can be compared with the 2 recent US population-based studies using National Health Interview Survey (NHIS) data from 20008 and 1998–2001.7 Our physical activity prevalence of 21.4% is slightly lower than the 24.6% reported by Coups and Ostroff8 and the 29.6% reported by Bellizzi et al.,7 although we used the cutpoint of at least 1 hour/day of walking. Our moderately active cutpoint is more comparable to the US studies and showed that about 46% of Canadian cancer survivors were at least ‘minimally’ active, although the measures in the studies are not directly comparable. Our overweight and obesity rates can be directly compared with the American studies. Our data showed that 18.4% of Canadian cancer survivors were obese and another 34.4% were overweight, which is comparable to the 21.9% obese and 37.1% overweight reported by Bellizzi et al.7 and the 60% overweight/obese combined reported by Coups and Ostroff8 (the obesity category was not reported separately).
Overall, our study showed no differences between cancer survivors as a group and those without a history of cancer in the prevalences of physical activity and obesity. Coups and Ostroff8 did not conduct an overall analysis but did report no differences across 3 age groups in the prevalence of being overweight (obesity rates were not compared) and no differences across 2 age groups in the prevalence of being physically active (younger than 40 years and 65+ years). Middle-aged cancer survivors (40–64 years), however, were less likely to be active. Conversely, although Bellizzi et al.7 also reported no differences between cancer survivors and controls in obesity prevalence, they did report that cancer survivors were actually 9% more likely to be physically active. This estimate is identical to our estimate for the ‘nonskin’ cancers, although Bellizzi et al.7 adjusted for more variables than we did, including functional limitations. Taken together, these 3 population-based studies suggest that the rates of physical activity and overweight/obesity in cancer survivors as a group are comparable to the general US and Canadian populations.
Some differences did emerge, however, for specific cancer survivor groups. Our data showed that only 16.6% of breast cancer survivors were active and that obese breast cancer survivors were less likely to be active than obese women without a history of cancer. This finding is consistent with a study by Irwin et al.23 that showed that obese women reported a disproportionate decline in physical activity levels after a breast cancer diagnosis compared with nonobese women. This finding is cause for concern because physical activity may be particularly important for obese breast cancer survivors. Obesity itself is strongly associated with poorer disease outcomes1 and obese breast cancer survivors may particularly benefit from higher physical activity levels in terms of disease outcomes.2
Consistent with Coups and Ostroff8 and Bellizzi et al.,7 we also found prostate cancer survivors to report among the highest rate of physical activity for any cancer survivor group. In our study the physical activity rate was actually significantly higher than matched controls. It is unclear why Canadian prostate cancer survivors would be more active than their noncancer counterparts. The higher activity level is unlikely the result of a selection bias wherein more active men are diagnosed with prostate cancer and simply continue this higher level of activity after their diagnosis. In fact, physical inactivity is likely a risk factor for prostate cancer,24 suggesting that, if anything, we would expect men with newly diagnosed prostate cancer to be slightly less active than their noncancer counterparts at the time of diagnosis. It is possible that a Canadian trial published in 2003 influenced practice in this group.25 In that trial, resistance exercise was shown to significantly improve muscular strength, fatigue, and quality of life in prostate cancer survivors receiving androgen deprivation therapy.25 The finding that prostate cancer survivors were less likely to be obese may be due in part to the higher activity levels but may also be the result of medical treatments known to result in loss of lean body mass.26 Consequently, the lower BMI in prostate cancer survivors may not necessarily be a positive finding if it is primarily the result of muscle loss.
The finding that skin cancer survivors, especially males, were more active than noncancer controls may have resulted from a selection bias. Recent research has shown that physically active people have a higher risk of developing primary skin cancer, likely resulting from greater exposure to the sun during outdoor activities.27, 28 It is possible, therefore, that the higher levels of prediagnosis physical activity simply continue in this group after diagnosis. The finding that skin cancer survivors do not reduce their physical activity levels after their diagnosis is a positive finding but it needs to be accompanied by appropriate sun protection behaviors. Efforts to promote physical activity in all cancer survivor groups should be accompanied by guidelines for appropriate sun protection behaviors.
Finally, consistent with Coups and Ostroff8 and Bellizzi et al.,7 we also found colorectal cancer survivors to report among the lowest rate of physical activity in any cancer survivor group. This finding is particularly concerning given 2 recent studies showing higher physical activity levels may result in better disease control in this group.4, 5
Our study has important strengths and limitations. It provides the first population-based data comparing Canadian cancer survivors to noncancer controls on 2 risk factors that continue to grow in importance. The study did rely on self-report, however, which likely resulted in underestimating BMI and overestimating physical activity levels. Furthermore, we did not validate our self-report measure of cancer status and there is no indication from the CCHS questionnaire of when the respondent was first diagnosed with cancer, and so time since diagnosis could not be analyzed. Perhaps even more important, there was no information on current disease status or cancer treatments, both of which are known to dramatically effect physical activity and body composition levels.29–31 Finally, the CCHS questionnaire did not allow respondents to report other common cancers that make up a large portion of the cancer survivorship community (eg, lymphomas, bladder, gynecologic cancers).
In summary, Canadian cancer survivors have low levels of physical activity and high levels of obesity that may place them at higher risk for poorer disease control and supportive care outcomes. Overall, the majority of Canadian cancer survivors were inactive and about one-fifth were obese. These prevalence estimates were comparable to the general population and suggest that neither cancer survivors nor noncancer controls are obtaining sufficient amounts of physical activity, and a cancer diagnosis does not appear to be prompting significant behavior change. Prostate and skin cancer survivors appeared to be more active, while obese breast cancer survivors appeared to be less active than their matched counterparts. Moreover, cancer survivor groups with the most to gain from physical activity (eg, breast, colorectal) appeared to be the least active. Population-based lifestyle interventions to increase physical activity and promote a healthy body weight in Canadian cancer survivors are warranted.