Article Tools
Health Care Delivery
Original Contributions
Article Tools
Lifestyle Behaviors in Elderly Cancer Survivors: A Comparison With Middle-Age Cancer Survivors
Abstract
Older cancer survivors reported being less aware of the impact of smoking on their health, more likely to perceive alcohol as beneficial to survival, and less likely to meet exercise goals compared with younger survivors.
Improved cancer screening and treatment have led to a greater focus on cancer survivorship care. Older cancer survivors may be a unique population. We evaluated whether older cancer survivors (age ≥ 65 years) had lifestyle behaviors, attitudes, and knowledge distinct from younger survivors.
Adult cancer survivors with diverse cancer subtypes were recruited from Princess Margaret Cancer Centre (Toronto, Ontario, Canada). Multivariable models evaluated the effect of age on smoking, alcohol, and physical activity habits, attitudes toward and knowledge of these habits on cancer outcomes, and lifestyle information and recommendations received from health care providers, adjusted for sociodemographic and clinicopathologic covariates.
Among the 616 survivors recruited, 23% (n = 139) were older. Median follow-up since diagnosis was 24 months. Older survivors were more likely ex-smokers and less likely current smokers than younger survivors, but they were less likely to know that smoking could affect cancer treatment (adjusted odds ratio [OR], 0.53; P = .007) or prognosis (adjusted OR, 0.53; P = .008). Older survivors were more likely to perceive alcohol as improving overall survival (adjusted OR, 2.39; P = .02). Rates of meeting moderate-to-vigorous physical activity guidelines 1 year before diagnosis (adjusted OR, 0.55; P = .02) and maintaining and improving their exercise levels to meet these guidelines after diagnosis (adjusted OR, 0.48; P = .02) were lower in older survivors. Older and younger cancer survivors reported similar rates of receiving lifestyle behavior information from health care providers (P = .36 to .98).
Older cancer survivors reported being less aware of the impact of smoking on their overall health, more likely perceived alcohol as beneficial to survival, and were less likely to meet exercise goals compared with younger survivors. Survivorship programs need to consider age when counseling on lifestyle behaviors.
In 2013, 71% of all new cancer diagnoses in Canada were made in people age > 60 years, with the median age of diagnosis between 65 and 69 years and an increasing incidence rate of most cancer types with increasing age.1 Along with most industrialized countries, Canada is facing an aging cancer survivor population because of advances in early detection and improved treatment of cancer against a background of increasing life expectancy.1,2 As a result, secondary prevention (including preventing second primary malignancies and recurrence) in the management of cancer and other comorbidities is becoming an emerging priority to improve overall survival and quality of life in this population.
Various secondary prevention strategies, such as smoking cessation, moderation of alcohol consumption, and regular physical activity, have been shown to reduce cancer recurrence and improve quality of life in adult cancer survivors, with similar benefits regardless of age.3–7 In studies including older cancer survivors, these benefits have included reduced risk of contralateral breast cancer with smoking cessation, decreased breast cancer recurrence with alcohol moderation, and decreased rates of colorectal, breast, and prostate cancer–related deaths with increased physical activity.5–14
These lifestyle modifications represent important opportunities for intervention for older cancer survivors, because these survivors are likely to have more medical comorbidities and experience more significant functional limitations as a result of their cancer and cancer-associated treatments when compared with younger patients.15 Thus, any improvement in physical functioning or psychological well-being may make the difference between independent living and otherwise.15 Furthermore, given the advanced age of these survivors, clinicians may be more nihilistic about the value of lifestyle modification and may recommend such behavioral modifications less frequently than in younger survivors.16 Currently, there is limited understanding of the potentially unique needs and lifestyle behaviors of older cancer survivors because of poor representation in clinical trials and study design barriers.17,18 These barriers include multiple comorbidities precluding study enrollment, lack of validated tools or methods to assess treatment tolerability or patient-reported outcomes, and potentially reduced cognitive capacity, comprehension, or ability to provide informed consent.17,18
The study aim was to determine whether differences existed between younger and older patients with cancer with respect to: smoking cessation, alcohol moderation, and regular physical activity; their knowledge of and attitudes toward these behavioral changes; and the frequency of receiving any information or counseling on these behaviors from their health care providers after their cancer diagnosis. The overarching goal was to identify potential areas requiring special considerations for survivorship programming in older patients with cancer.
This cross-sectional study was approved by the research ethics board of the Princess Margaret Cancer Centre–University Health Network (PM), a tertiary cancer center. Patients age ≥ 18 years with a histologic confirmation of malignancy of any stage were included. To include cancer survivors at different stages of their cancer care, there were no restrictions on time since diagnosis or treatment. Cognitive deficits and language barriers preventing the patient understanding or consent during study enrollment were used as exclusion criteria. Patients were recruited from clinic waiting areas by trained research assistants. Patients providing written consent were then asked to complete an in-person questionnaire that assessed sociodemographics, functional status (as measured on the Eastern Cooperative Oncology Group scale and on a separate 5-point Likert scale), smoking and alcohol history (measured in standard number of drinks per week), and physical activity levels (measured in minutes per week) before and after diagnosis using a validated questionnaire.19 Additional questions assessed patient knowledge of and attitudes toward such behaviors in relation to health outcomes. Clinicopathologic information, including date of histologic diagnosis, tumor site, cancer stage, and cancer treatment data (types received to date and treatment intent), was collected through patient medical record review.
Cumulative smoking history was calculated through pack-years (number of years of smoking × number of packs of cigarettes smoked per day). Classification of patients by smoking status was similar to that in previous studies by our group.20,21 Ever-smokers were defined as having smoked ≥ 100 cigarettes in their lifetime; others were classified as never-smokers. Among ever-smokers, current smokers smoked within 1 year before their cancer diagnosis; otherwise, they were considered ex-smokers.
Patients were ever-drinkers if they reported consuming at least one standard drink of alcohol based on 13.6 g of ethanol (5 oz of wine, 12 oz of beer, and 1.5 oz of liquor) per month.22 Current drinkers were still consuming alcohol at the time of cancer diagnosis; ex-drinkers had quit before diagnosis. Subsequently, current drinkers were further classified based on their level of alcohol consumption at 1 year after cancer diagnosis (or at follow-up, if < 1-year follow-up to date) as either quitters or those who had cut down on use; others were classified as continued drinkers; ex-drinkers were divided between restarters and abstainers at 1 year after diagnosis.
Physical activity levels were dichotomized based on meeting or not meeting guideline recommendations of at least 150 minutes of moderate-to-vigorous physical activity (MVPA) per week.19,23 Activity levels were compared between 1 year before diagnosis and at follow-up. Change in MVPA was classified as maintaining or improved to recommended levels at follow-up or reduced to or persisting below recommended levels.
For each behavior, patient knowledge of and attitudes about the impact of each habit on different aspects of their quality of life, survival, and treatment was assessed. Specifically for smoking, patients were asked if they felt that smoking could affect their cancer treatment and cancer outcomes. Given that optimal use guidelines are unclear for physical activity and alcohol compared with smoking, the patient's perceived impact of alcohol consumption and physical activity on their quality of life, fatigue, and 5-year overall survival was assessed on a 7-point Likert scale, ranging from “very harmful” to “very beneficial,” which our group developed as a brief exploratory method of assessing these domains.4,11,24,25 These terms were defined by our research assistants to patients during questionnaire completion. Patients were also asked if their family physician or oncologist had ever provided information or counseling on these behaviors after their cancer diagnosis; regarding smoking, only former and current smokers were asked, whereas for alcohol consumption, we focused on current drinkers whose intake exceeded sex-specific guidelines at diagnosis (ie, 14 standard drinks per week for men and nine standard drinks per week for women)26; for physical activity, we focused on all patients. The questionnaire was reviewed for face and content validity by research team members.
All statistical analyses were performed using SAS software (version 9.2; SAS Institute, Cary, NC). Frequencies of sociodemographic and clinicopathologic variables were compared between young (age < 65 years) versus older survivors (age ≥ 65 years) using t and χ2 tests as appropriate. Univariable analysis was applied to assess the effect of dichotomized age and other covariates (Table 1), including pack-years smoked, minutes of physical activity, average drinks per week at diagnosis, and follow-up time since diagnosis, on outcomes using logistic regression. For outcomes where age was significantly associated in univariable analysis (P < .05), multivariable models were then applied to evaluate the independent effect of age on these outcomes. Covariates significant at P < .10 were included in a multivariable model, where a backward selection algorithm was applied, eliminating nonsignificant covariates (P > .05) to identify independent variables for each base model. To each base model, age was then added and tested for its significance using the Wald test. P < .05 was considered significant. In addition, to explore the optimal parameterization of age, for those outcomes where age was significant in multivariable analysis, age was alternatively examined as a continuous variable.
|
| Characteristic | Total, N = 616 (%) | Younger Patients (age < 65 years), n = 477 (%) | Elderly Patients (age ≥ 65 years), n = 139 (%) | P* |
|---|---|---|---|---|
| Total | 100 | 77 | 23 | |
| Female sex | 53 | 56 | 45 | .02 |
| English speaking | 92 | 93 | 89 | .21 |
| Employed or student | 40 | 48 | 15 | < .001 |
| White collar job | 75 | 76 | 71 | .27 |
| White race | 81 | 79 | 89 | .02 |
| Married or common-law marriage | 72 | 71 | 73 | .70 |
| Completed high school | 90 | 93 | 79 | < .001 |
| Annual household income | .17 | |||
| ≥ $80,000 | 37 | 41 | 27 | |
| Unknown | 25 | 23 | 32 | |
| Good/very good/excellent health (self-rated) | 75 | 76 | 71 | .42 |
| ECOG PS 0 or 1 | 83 | 82 | 89 | .05 |
| Disease stage | .43 | |||
| Localized | 35 | 34 | 36 | |
| Metastatic | 33 | 33 | 34 | |
| Hematologic | 23 | 24 | 18 | |
| Received surgery | 67 | 70 | 60 | .02 |
| Received radiotherapy | 46 | 46 | 42 | .39 |
| Received systemic therapy | 64 | 66 | 55 | .01 |
| Treatment intent at diagnosis | .03 | |||
| Curative | 88 | 89 | 84 | |
| Observation/no therapy | 6 | 6 | 7 | |
| Non-curative | 6 | 4 | 10 | |
| Treatment intent at follow-up | .12 | |||
| Curative | 82 | 83 | 78 | |
| No treatment since diagnosis | 6 | 6 | 7 | |
| Non-curative | 11 | 10 | 16 | |
| Disease site | .01 | |||
| Breast | 15 | 17 | 7 | |
| GI | 14 | 12 | 21 | |
| Genitourinary | 13 | 12 | 18 | |
| Gynecologic | 6 | 6 | 6 | |
| Head and neck | 9 | 9 | 12 | |
| Hematologic | 24 | 25 | 19 | |
| Lung | 4 | 3 | 5 | |
| Skin and sarcoma | 8 | 8 | 7 | |
| Thyroid | 6 | 7 | 4 |
Abbreviation: ECOG PS, Eastern Cooperative Oncology Group performance status.
*P values represent comparisons of variables between older and younger cancer survivors using Fisher's exact, Pearson's χ2, or t test, where appropriate.
Overall effective response rate was 65% (n = 616): 551 participants (89%) completed the smoking behavior section, 560 (91%) completed the physical activity section, and 548 (89%) completed the alcohol consumption section. Response rate was 68% for younger cancer survivors and 61% for older cancer survivors. Baseline sociodemographic characteristics are listed in Table 1. Median age at diagnosis was 54 years (range, 18 to 85 years; interquartile range, 43 to 63 years). Median follow-up time since diagnosis (± standard deviation) was 27 ± 54 months. There was broad representation of disease sites among respondents, and 23% of respondents were older patients. Most patients were English speaking, white, married or in a common-law marriage, high school graduates, employed in white collar jobs, and relatively functionally asymptomatic of cancer; most had localized tumors and had received curative-intent therapy. Older patients were more likely to be men, retired or unemployed, and white; they were less likely to have completed high school, received surgery or chemotherapy, and been initially treated with curative intent.
Smoking variables are listed in Table 2; only one sixth (17%) of patients surveyed were smoking at diagnosis, and approximately half (54%) were lifetime nonsmokers. Overall, age was not associated with ever having smoked (P = .23) or with quitting after cancer diagnosis (P = .60). No differences in rates of smoking cessation counseling were observed by age group from either the family physician (P = .61) or oncologist (P = .78). In multivariable analysis, older survivors were twice as likely to be ex-smokers as never-smokers (adjusted odds ratio [OR], 2.10; 95% CI, 1.31 to 3.36; P < .001) and 3× less likely to be current smokers as never-smokers, when compared with younger patients (adjusted OR, 0.35; 95% CI, 0.16 to 0.78; P < .001; Table 3). When queried about attitude and knowledge, older patients were half as likely to feel that exposure to smoking during cancer treatment could affect their response to treatment (adjusted OR, 0.53; 95% CI, 0.33 to 0.84; P = .007) or their overall prognosis (adjusted OR, 0.53; 95% CI, 0.34 to 0.85; P = .008; Table 3). As a continuous variable, age was similarly significantly associated with being an ex-smoker instead of a never-smoker, and a nonsignificant trend of age being inversely associated with knowledge of the impact of smoking on prognosis or treatment was identified (Table 3).
|
| Variable | Younger Patients (age < 65 years), n = 477 (%) | Older Patients (age > 65 years), n = 139 (%) | P* |
|---|---|---|---|
| Smoking | |||
| Baseline smoking status | < .001 | ||
| Current smoker | 19 | 9 | |
| Ex-smoker | 24 | 46 | |
| Never-smoker | 57 | 44 | |
| Change in smoking status after diagnosis† | .60 | ||
| Continued | 45 | 36 | |
| Quit | 55 | 64 | |
| Belief continued that smoking can affect treatment | .02 | ||
| Yes | 59 | 46 | |
| No or don't know | 41 | 54 | |
| Belief continued that smoking can affect prognosis | .02 | ||
| Yes | 60 | 47 | |
| No or don't know | 40 | 53 | |
| Received oncologist counseling | 55 | 60 | .78 |
| Received physician counseling | 25 | 28 | .61 |
| Alcohol | |||
| Alcohol use status at diagnosis | .17 | ||
| Current drinker | 60 | 67 | |
| Ex-drinker | 18 | 18 | |
| Never-drinker | 22 | 15 | |
| Change in alcohol use for current drinkers‡ | .73 | ||
| Quit or cut down | 56 | 58 | |
| Continued | 44 | 42 | |
| Change in alcohol use for ex-drinkers§ | .64 | ||
| Remained abstinent | 85 | 81 | |
| Restarted | 15 | 19 | |
| Perception of alcohol effect on quality of life | .06 | ||
| Improves outcome | 10 | 19 | |
| No effect | 42 | 40 | |
| Worsens outcome | 48 | 42 | |
| Perception of alcohol effect on fatigue | .61 | ||
| Improves outcome | 6 | 9 | |
| No effect | 39 | 41 | |
| Worsens outcome | 55 | 50 | |
| Perception of alcohol effect on overall survival | .02 | ||
| Improves outcome | 8 | 17 | |
| No effect | 45 | 45 | |
| Worsens outcome | 47 | 39 | |
| Received information from oncologist | 12 | 11 | .98 |
| Received information from family physician | 8 | 0 | ‖ |
| Physical Activity | |||
| Exercising (at any level) 1 year before diagnosis | 53 | 40 | .01 |
| Exercising at MVPA levels before diagnosis | 34 | 22 | .01 |
| Maintained or improved to meet MVPA levels after cancer diagnosis | 30 | 16 | .03 |
| Perception of exercise effect on quality of life | .38 | ||
| Improves outcome | 93 | 90 | |
| No effect | 5 | 8 | |
| Worsens outcome | 2 | 3 | |
| Perception of exercise effect on fatigue | .11 | ||
| Improves outcome | 75 | 68 | |
| No effect | 13 | 12 | |
| Worsens outcome | 12 | 20 | |
| Perception of exercise effect on overall survival | .19 | ||
| Improves outcome | 90 | 85 | |
| No effect | 8 | 14 | |
| Worsens outcome | 1 | 2 | |
| Received information from oncologist | 7 | 7 | .80 |
| Received information from family physician | 6 | 4 | .36 |
Abbreviation: MVPA, moderate-to-vigorous physical activity.
*t test was used to compare proportions between older and younger cancer survivors.
†Change in smoking status was assessed only for current smokers; no ex-smokers or never-smokers started smoking after diagnosis (younger patients, n = 81; older patients, n = 11).
‡Younger patients, n = 244; older patients, n = 78.
§Younger patients, n = 74; older patients, n = 21.
‖Could not assess for association; no older cancer survivors received information from their family physician.
|
| Outcome | Multivariable Analysis (older v younger cancer survivors)* | Adjustment Variables | |||||
|---|---|---|---|---|---|---|---|
| Age As Dichotomized Variable | Age As Continuous Variable | ||||||
| Adjusted OR | 95% CI | P | Adjusted OR† | 95% CI | P | ||
| Smoking | |||||||
| Baseline smoking status | < .001 | < .001 | Ethnicity, education, sex, ECOG PS | ||||
| Current v never-smoker | 0.35 | 0.16 to 0.78 | 0.79 | 0.94 to 1.11 | |||
| Ex-smoker v never-smoker | 2.10 | 1.31 to 3.36 | 1.48 | 1.26 to 1.74 | |||
| Belief continued that smoking can affect treatment (yes v no or unaware) | 0.53 | 0.33 to 0.84 | .007 | 0.89 | 0.79 to 1.02 | .09 | Treatment intent at follow-up |
| Belief continued that smoking can affect prognosis (yes v no or unaware) | 0.53 | 0.34 to 0.85 | .008 | 0.90 | 0.79 to 1.02 | .11 | Treatment intent at follow-up |
| Alcohol | |||||||
| Perception of alcohol effect on overall survival (improves v worsens or has no effect) | 2.39 | 1.16 to 4.91 | .02 | 1.26 | 0.99 to 1.61 | .06 | Income |
| Physical Activity | |||||||
| Exercising (at any level) 1 year before diagnosis (yes v no) | 0.61 | 0.40 to 0.95 | .03 | 0.79 | 0.69 to 0.89 | < .001 | Ethnicity, education, employment type |
| Exercising at MVPA levels before diagnosis (yes v no) | 0.55 | 0.34 to 0.89 | .02 | 0.77 | 0.67 to 0.87 | < .001 | Ethnicity, education, received chemotherapy |
| Exercise after cancer diagnosis (maintained or improved to v worsened or persisted below MVPA levels) | 0.48 | 0.27 to 0.87 | .02 | 0.79 | 0.68 to 0.91 | < .001 | Ethnicity, ECOG PS, marital status |
NOTE. Backward selection of covariates, set at P < .05, was performed in our multivariable base model selection for each outcome, with all covariates that were significantly associated with that outcome (P < .10) in univariable analysis. Age was evaluated in this final base model. Each final base model of outcome was adjusted for variables in the far-right column.
Abbreviations: ECOG PS, Eastern Cooperative Oncology Group performance status; MVPA, moderate-to-vigorous physical activity; OR, odds ratio.
*Age ≥ 65 v < 65 years.
†Per 10-year age increase.
Alcohol consumption summary statistics are listed in Table 2; of the three fifths of patients surveyed who drank regularly at diagnosis, almost 60% quit or cut down after their diagnosis. Among ex-drinkers, one sixth restarted. There were no significant differences in baseline drinking status (P = .17) or changes in drinking status after diagnosis (current drinkers, P = .73; ex-drinkers, P = .64) between younger and older patients. Although > 80% of patients in both age groups felt that alcohol would either have no effect on or worsen their quality of life, fatigue, and survival, older patients were more likely to perceive that alcohol could improve 5-year overall survival (adjusted OR, 2.39; 95% CI, 1.16 to 4.91; P = .02; Table 3). A similar nonsignificant trend with overall survival was also identified when age was analyzed as a continuous variable (Table 3). There was also a nonsignificant univariable association of older patients being more likely to perceive that alcohol could improve quality of life (P = .06). No differences were found between older and younger patients with respect to their perception of alcohol on fatigue (P = .61). Among current drinkers exceeding sex-specific guidelines, few reported receiving information or guidance on alcohol consumption (< 15% from oncologist; < 10% from family physician), regardless of patient age. There was no statistically significant difference between age groups for receiving information from their oncologist (P = .98); comparisons could not be made for family physician, because no older cancer survivors had received information from their family physician.
As summarized in Table 2, approximately one third (31%) of patients met MVPA guidelines 1 year before diagnosis, falling to one quarter (27%) at follow-up. At follow-up, only 48% of patients who met MVPA guidelines at diagnosis had maintained those levels after diagnosis; 19% of those not meeting MVPA guidelines at diagnosis had improved their physical activity levels to meet guidelines after diagnosis. One year before diagnosis, older patients were 40% less likely to be exercising regardless of level (adjusted OR, 0.61; 95% CI, 0.40 to 0.95; P = .03) or meeting or exceeding MVPA guidelines (adjusted OR, 0.55; 95% CI, 0.34 to 0.89; P = .02) compared with younger adults (Table 3). Older patients were also half as likely as their younger counterparts to maintain or improve their physical activity levels to meet MVPA guidelines after cancer diagnosis (adjusted OR, 0.48; 95% CI, 0.27 to 0.87; P = .02; Table 3). Similar associations with age as a continuous predictor were found with exercising or meeting MVPA guidelines before diagnosis and meeting MVPA guidelines after diagnosis (Table 3).
Older patients held perceptions similar to those of younger patients with respect to impact of exercise on quality of life (P = .38), fatigue (P = .11), and overall survival (P = .19). Fewer than 10% of both age groups reported ever having received information regarding physical activity from either their oncologist or family physician. There were no statistically significant differences between age groups with respect to receiving information on physical activity from their oncologist (P = .80) or family physician (P = .36).
Patients with cancer are now living longer; thus, survivorship care is becoming increasingly important.9 Older patients make up a significant proportion of cancer survivors, and special considerations may be required for their care. In our study, we found that when compared with younger cancer survivors, older cancer survivors were less likely to believe in a negative impact of smoking on cancer outcomes, more likely to perceive alcohol as being beneficial to their survival, and had lower rates of exercising even before their cancer diagnosis. They were also less likely to improve their physical activity levels after cancer diagnosis to meet guidelines.
Previous studies have examined the effect of some of these behaviors in either heterogeneous populations composed of both older and younger cancer survivors or solely older cancer survivors and found prognostic and/or quality-of-life benefits for these behaviors.10–14,27 One study reported that older cancer survivors had reduced risk perceptions with respect to cancer recurrence and its consequences compared with younger cancer survivors and that these risk perceptions were correlated with behavior change.28 However, to our knowledge, direct comparison of older and younger cancer survivors with respect to their perceptions or knowledge of potential lifestyle behaviors for cancer survivorship has not been previously studied.
Differences by age groups on perceptions of benefits and harms of continued smoking and alcohol consumption are likely multifactorial. Older cancer survivors may not perceive the importance of these lifestyle behaviors, because they have lessened risk perceptions and reduced perceived severity of cancer recurrence.28 Older patients may be negatively influenced by their own self-perceptions, which can reduce their self-efficacy for change and will to live. In turn, these factors may affect perceptions of possible benefits from these lifestyle behaviors.16 Older patients with cancer also may not recognize the need for lifestyle behavior modification.11 Specifically for alcohol, despite well-documented scientific evidence to the contrary regarding cancer-related outcomes, older patients often believe alcohol to be beneficial, possibly because of publicized cardiovascular benefits of moderate alcohol consumption.4,13,14
Not surprisingly, older patients were less likely to be physically active compared with younger cancer survivors, before and after cancer diagnosis. Reasons may relate to medical comorbidities, functional decline, and accessibility to exercise programs.29,30 However, other contributing factors may include a lack of awareness that physical activity guidelines apply to older patients and to cancer survivors equally. The extremely low rates of health care provider counseling may reflect biases or lack of knowledge among the providers, but they may also reflect safety concerns of both patients and physicians.4,24 Programs targeting older cancer survivors have been shown to help maintain physical activity levels after a cancer diagnosis.31
Counseling rates by oncologists for smoking were relatively high (> 50%), likely influenced by a local smoking cessation initiative at our institution. Counseling rates for alcohol consumption and physical activity, both dismally low, were likely reflective of a lack of systematic approach at our center, but they may also have resulted from the lack of widely adopted guidelines for these two behaviors in the setting of cancer survivorship, as compared with smoking.4,11,24,25 This is a missed opportunity, because oncologists can serve as catalysts for lifestyle behavior change.4 The reduced motivation, reduced perception of benefit, and low rates of counseling may impede older cancer survivors from adopting such lifestyle behaviors.32
Furthermore, behavioral modification may be a time-sensitive phenomenon in patients with cancer. Receiving a diagnosis of cancer has been documented as a teachable moment in cancer care for patients to change lifestyle behaviors.20,21,33,34 With improvements in cure rates and patients with metastatic disease living longer, the definition of when a patient with cancer transitions to a cancer survivor has blurred. Because modification of these behaviors may also reduce treatment toxicity, improve treatment outcomes, and improve quality of life during treatment,27,35–38 lifestyle modification is as important in the acute treatment phase as in the post-therapy follow-up period.
Barriers to improving lifestyle behavior changes in older cancer survivors include the following: (1) ageism may play a role on the part of both patients and providers (both can have misperceptions that older patients cannot be taught new ways or benefit from lifestyle modifications, despite evidence to the contrary when improved communication strategies are implemented16,31); (2) accessibility and traveling to such behavior modification programs can be difficult for older cancer survivors, and home-based programs may help with the dissemination of these strategies30; (3) concerns about intervening because of multiple comorbidities (ie, osteoporosis, cardiovascular disease, diabetes, functional decline) in older cancer survivors. Nonetheless, modification of these behaviors can actually help these comorbidities.27,33 To help reduce some of these barriers, models of shared care between geriatricians and oncologists may be of value and should be studied further.18
There are limitations to this study. First, within the older survey participants, 78% were age 65 to 74 years, 21% were age 75 to 84 years, and only 1% were age ≥ 85 years. The latter two age groups were under-represented, making it difficult to determine if there are age-based differences within the oldest adult cancer survivor populations. However, additional sensitivity analyses restricting the younger age group to those ages 50 to 64 yielded similar significant results or results with nonsignificant trends with the same directionality (Appendix Table A1, online only). Second, a longitudinal design with longer follow-up is needed to assess sustained changes in lifestyle behaviors in patients with cancer. Recall errors and biases of social desirability are standard concerns in all surveys, because participants may report their health habits in a positive manner to investigators. In addition, our assessment tools for perceptions were exploratory in nature, and future validation of their psychometric properties is warranted. Also, although functional status was analyzed as a covariate in our analyses, our study did not include medical comorbidities, symptoms including pain, or quality-of-life measures, which may also influence changes in and perceptions of these lifestyle behaviors, especially given the short follow-up period. Finally, our patients were recruited from a single large North American regional cancer center, and the results may not be generalizable to other patient populations.
In summary, this study has demonstrated that older and younger cancer survivors have significantly different beliefs and perceptions of the impact of smoking and alcohol use and different levels of physical activity. Older patients may also have additional specific barriers and require specific considerations when implementing behavioral modification strategies. More attention should be paid to ensure that the older patients receive adequate and specific education on the impact of their health habits and are encouraged on their postcancer journey toward a healthier lifestyle.
Acknowledgment
Supported in part by the Alan B. Brown Chair in Molecular Genomics, the Cancer Care Ontario Chair in Experimental Therapeutics and Population Studies, and the Posluns Family Foundation. C.N.H. and L.E. contributed equally to this work as first authors; G.L. and S.M.H.A. contributed equally to this work as senior authors. Presented as a poster at the 49th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 4, 2013, and orally at the 33rd Annual Scientific Meeting of the Canadian Geriatrics Society, Toronto, Ontario, Canada, April 18-20, 2013.
Disclosures provided by the authors are available with this article at jop.ascopubs.org.
Conception and design: Chongya Niu, Lawson Eng, M. Catherine Brown, Peter Selby, Doris Howell, Jennifer M. Jones, Geoffrey Liu, Shabbir M.H. Alibhai
Administrative support: Mary Mahler, M. Catherine Brown
Collection and assembly of data: Chongya Niu, Lawson Eng, Dan Pringle, Mary Mahler, Oleksandr Halytskyy, Rebecca Charow, Christine Lam, Ravi M. Shani, Jodie Villeneuve, Kyoko Tiessen, M. Catherine Brown
Data analysis and interpretation: Chongya Niu, Lawson Eng, Xin Qiu, Xiaowei Shen, Osvaldo Espin-Garcia, Yuyao Song, M. Catherine Brown, Peter Selby, Wei Xu, Geoffrey Liu, Shabbir M.H. Alibhai
Manuscript writing: All authors
Final approval of manuscript: All authors
| 1. | Canadian Cancer Statistics 2013 2013 Canadian Cancer Society Advisory Committee on Cancer Statistics Toronto, Ontario, Canada Canadian Cancer Society Google Scholar |
| 2. | LF Ellison, K Wilkins : Canadian trends in cancer prevalence Health Rep 23: 7– 16,2012 Medline, Google Scholar |
| 3. | CM Blanchard, KS Courneya, K Stein , etal: Cancer survivors' adherence to lifestyle behavior recommendations and associations with health-related quality of life: Results from the American Cancer Society's SCS-II J Clin Oncol 26: 2198– 2204,2008 Link, Google Scholar |
| 4. | KM Bellizzi, JH Rowland, DD Jeffery , etal: Health behaviors of cancer survivors: Examining opportunities for cancer control intervention J Clin Oncol 23: 8884– 8893,2005 Link, Google Scholar |
| 5. | ML Kwan, WY Chen, SW Flatt , etal: Postdiagnosis alcohol consumption and breast cancer prognosis in the after breast cancer pooling project Cancer Epidemiol Biomarkers Prev 22: 32– 41,2013 Crossref, Medline, Google Scholar |
| 6. | AM Haydon, RJ Macinnis, DR English , etal: Effect of physical activity and body size on survival after diagnosis with colorectal cancer Gut 55: 62– 67,2006 Crossref, Medline, Google Scholar |
| 7. | HD Klepin, SG Mohile, S Mihalko : Exercise for older cancer patients: Feasible and helpful? Interdiscip Top Gerontol 38: 146– 157,2013 Crossref, Medline, Google Scholar |
| 8. | CI Li, JR Daling, PL Porter , etal: Relationship between potentially modifiable lifestyle factors and risk of second primary contralateral breast cancer among women diagnosed with estrogen receptor-positive invasive breast cancer J Clin Oncol 27: 5312– 5318,2009 Link, Google Scholar |
| 9. | W Demark-Wahnefried, EC Clipp, MC Morey , etal: Lifestyle intervention development study to improve physical function in older adults with cancer: Outcomes from project LEAD J Clin Oncol 24: 3465– 3473,2006 Link, Google Scholar |
| 10. | BM Pinto, JJ Trunzo : Health behaviors during and after a cancer diagnosis Cancer 104: 2614– 2623,2005 suppl Crossref, Medline, Google Scholar |
| 11. | AV Rao, W Demark-Wahnefried : The older cancer survivor Crit Rev Oncol Hematol 60: 131– 143,2006 Crossref, Medline, Google Scholar |
| 12. | J Ligibel : Lifestyle factors in cancer survivorship J Clin Oncol 30: 3697– 3704,2012 Link, Google Scholar |
| 13. | N Druesne-Pecollo, Y Keita, M Touvier , etal: Alcohol drinking and second primary cancer risk in patients with upper aerodigestive tract cancers: A systematic review and meta-analysis of observational studies Cancer Epidemiol Biomarkers Prev 23: 324– 331,2014 Crossref, Medline, Google Scholar |
| 14. | LS Evangelista, L Sarna, ML Brecht , etal: Health perceptions and risk behaviors of lung cancer survivors Heart Lung 32: 131– 139,2003 Crossref, Medline, Google Scholar |
| 15. | NE Avis, GT Deimling : Cancer survivorship and aging Cancer 113: 3519– 3529,2008 suppl Crossref, Medline, Google Scholar |
| 16. | M Ory, M Kinney Hoffman, M Hawkins , etal: Challenging aging stereotypes: Strategies for creating a more active society Am J Prev Med 25: 164– 171,2003 suppl 2 Crossref, Medline, Google Scholar |
| 17. | C Townsley, GR Pond, B Peloza , etal: Analysis of treatment practices for elderly cancer patients in Ontario, Canada J Clin Oncol 23: 3802– 3810,2005 Link, Google Scholar |
| 18. | KM Bellizzi, KM Mustian, OG Palesh , etal: Cancer survivorship and aging: Moving the science forward Cancer 113: 3530– 3539,2008 suppl Crossref, Medline, Google Scholar |
| 19. | G Godin, RJ Shephard : A simple method to assess exercise behavior in the community Can J Appl Sport Sci 10: 141– 146,1985 Medline, Google Scholar |
| 20. | L Eng, J Su, X Qiu , etal: Second-hand smoke as a predictor of smoking cessation among lung cancer survivors J Clin Oncol 32: 564– 570,2014 Link, Google Scholar |
| 21. | A Kashigar, S Habbous, L Eng , etal: Social environment, secondary smoking exposure, and smoking cessation among head and neck cancer patients Cancer 119: 2701– 2709,2013 Crossref, Medline, Google Scholar |
| 22. | HA Skinner, WJ Sheu : Reliability of alcohol use indices: The lifetime drinking history and the MAST J Stud Alcohol 43: 1157– 1170,1982 Crossref, Medline, Google Scholar |
| 23. | Tips to get active Public Health Agency of Canada http://www.phac-aspc.gc.ca/hp-ps/hl-mvs/pa-ap/04paap-eng.php Google Scholar |
| 24. | SA Sabatino, RJ Coates, RJ Uhler , etal: Provider counseling about health behaviors among cancer survivors in the united states J Clin Oncol 25: 2100– 2106,2007 Link, Google Scholar |
| 25. | CL Rock, C Doyle, W Demark-Wahnefried , etal: Nutrition and physical activity guidelines for cancer survivors CA Cancer J Clin 62: 243– 274,2012 Crossref, Medline, Google Scholar |
| 26. | SJ Bondy, J Rehm, MJ Ashley , etal: Low-risk drinking guidelines: The scientific evidence Can J Public Health 90: 264– 270,1999 Crossref, Medline, Google Scholar |
| 27. | W Demark-Wahnefried, BM Pinto, ER Gritz : Promoting health and physical function among cancer survivors: Potential for prevention and questions that remain J Clin Oncol 24: 5125– 5131,2006 Link, Google Scholar |
| 28. | AB Mullens, KD McCaul, SC Erickson , etal: Coping after cancer: Risk perceptions, worry, and health behaviors among colorectal cancer survivors Psychooncology 13: 367– 376,2004 Crossref, Medline, Google Scholar |
| 29. | DC Snyder, MC Morey, R Sloane , etal: Reach out to ENhancE wellness in older cancer survivors (RENEW): Design, methods and recruitment challenges of a home-based exercise and diet intervention to improve physical function among long-term survivors of breast, prostate, and colorectal cancer Psychooncology 18: 429– 439,2009 Crossref, Medline, Google Scholar |
| 30. | VB Stull, DC Snyder, W Demark-Wahnefried : Lifestyle interventions in cancer survivors: Designing programs that meet the needs of this vulnerable and growing population J Nutr 137: 243S– 248S,2007 suppl Crossref, Medline, Google Scholar |
| 31. | W Demark-Wahnefried, MC Morey, R Sloane , etal: Reach out to enhance wellness home-based diet-exercise intervention promotes reproducible and sustainable long-term improvements in health behaviors, body weight, and physical functioning in older, overweight/obese cancer survivors J Clin Oncol 30: 2354– 2361,2012 Link, Google Scholar |
| 32. | IM Rosenstock, VJ Strecher, MH Becker : Social learning theory and the health belief model Health Educ Q 15: 175– 183,1988 Crossref, Medline, Google Scholar |
| 33. | W Demark-Wahnefried, NM Aziz, JH Rowland , etal: Riding the crest of the teachable moment: Promoting long-term health after the diagnosis of cancer J Clin Oncol 23: 5814– 5830,2005 Link, Google Scholar |
| 34. | PJ Lawson, SA Flocke : Teachable moments for health behavior change: A concept analysis Patient Educ Couns 76: 25– 30,2009 Crossref, Medline, Google Scholar |
| 35. | ER Gritz, C Dresler, L Sarna : Smoking, the missing drug interaction in clinical trials: Ignoring the obvious Cancer Epidemiol Biomarkers Prev 14: 2287– 2293,2005 Crossref, Medline, Google Scholar |
| 36. | C Doyle, LH Kushi, T Byers , etal: Nutrition and physical activity during and after cancer treatment: An American Cancer Society guide for informed choices CA Cancer J Clin 56: 323– 353,2006 Crossref, Medline, Google Scholar |
| 37. | GP Browman, G Wong, I Hodson , etal: Influence of cigarette smoking on the efficacy of radiation therapy in head and neck cancer N Engl J Med 328: 159– 163,1993 Crossref, Medline, Google Scholar |
| 38. | MA List, A Siston, D Haraf , etal: Quality of life and performance in advanced head and neck cancer patients on concomitant chemoradiotherapy: A prospective examination J Clin Oncol 17: 1020– 1028,1999 Link, Google Scholar |
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml.
No relationship to disclose
No relationship to disclose
No relationship to disclose
Employment: Hoffmann-La Roche
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
Employment: Fusion Chiropractic and Integrative Health (I)
No relationship to disclose
Consulting or Advisory Role: Pfizer, Nabi Biopharmaceuticals, Mallinckrodt Pharmaceuticals
Research Funding: Pfizer
No relationship to disclose
No relationship to disclose
No relationship to disclose
Consulting or Advisory Role: Novartis, Pfizer
No relationship to disclose
|
| Outcome | Multivariable Analysis (older v younger cancer survivors)* | Adjustment Variables | ||
|---|---|---|---|---|
| Adjusted OR | 95% CI | P | ||
| Smoking | ||||
| Baseline smoking status | < .001 | Ethnicity, education, sex, ECOG PS | ||
| Current v never-smoker | 0.30 | 0.13 to 0.71 | ||
| Ex-smoker v never-smoker | 1.05 | 1.01 to 1.08 | ||
| Belief continued that smoking can affect treatment (yes v no or unaware) | 0.59 | 0.35 to 0.99 | .04 | Treatment intent at follow-up |
| Belief continued that smoking can affect prognosis (yes v no or unaware) | 0.58 | 0.35 to 0.97 | .04 | Treatment intent at follow-up |
| Alcohol | ||||
| Perception of alcohol effect on overall survival (improves v worsens or has no effect) | 2.36 | 1.02 to 5.46 | .05 | Income |
| Physical Activity | ||||
| Exercising (at any level) 1 year before diagnosis (yes v no) | 0.82 | 0.51 to 1.34 | .43 | Ethnicity, education, employment type |
| Exercising at MVPA levels before diagnosis (yes v no) | 0.72 | 0.43 to 1.27 | .28 | Ethnicity, education, received chemotherapy |
| Exercise after cancer diagnosis (maintained or improved to v worsened or persisted below MVPA levels) | 0.68 | 0.35 to 1.32 | .25 | Ethnicity, ECOG PS, marital status |
NOTE. Backward selection of covariates, set at P < .05, was performed in our multivariable base model selection for each outcome, with all covariates that were significantly associated with that outcome (P < .10) in univariable analysis. Age was evaluated in this final base model. Each final base model of outcome was adjusted for variables in far-right column.
Abbreviations: ECOG PS, Eastern Cooperative Oncology Group performance status; MVPA, moderate-to-vigorous physical activity; OR, odds ratio.
*Age ≥ 65 (n = 139) v 50 to 64 years (n = 229); age as dichotomized variable.
