Health behaviors during and after a cancer diagnosis

Authors

  • Bernardine M. Pinto Ph.D.,

    Corresponding author
    1. Centers for Behavioral and Preventive Medicine, Brown Medical School and Miriam Hospital, Lifespan Academic Medical Center, Providence, Rhode Island
    • Centers for Behavioral and Preventive Medicine, Brown Medical School and Miriam Hospital, Lifespan Academic Medical Center, One Hoppin Street, Coro Building, Suite 500, Providence, RI 02903
    Search for more papers by this author
    • Fax: (401) 793-8078

  • Joseph J. Trunzo Ph.D.

    1. Centers for Behavioral and Preventive Medicine, Brown Medical School and Miriam Hospital, Lifespan Academic Medical Center, Providence, Rhode Island
    2. Department of Applied Psychology, Bryant College, Bryant University, Smithfield, Rhode Island
    Search for more papers by this author

  • Presented at Cancer Survivorship: Resilience Across the Lifespan, Washington, DC, June 2–4, 2002.

Abstract

With improvements in cancer survival rates, more patients with cancer are living longer, and hence, cancer is becoming viewed as a chronic illness requiring long-term management. An important aspect of patient care during and after cancer treatment is patient health behaviors. For example, sequelae from various cancer treatments, such as chemotherapy and radiotherapy (RT), can compromise health in a variety ways, including decreased immune functioning, cardiotoxic effects of chemotherapy and/or RT, and weight gain. In addition, the stress caused by a cancer diagnosis and its treatment can disrupt existing health behaviors or exacerbate unhealthy behaviors. Continued smoking or alcohol use can complicate treatment and increase risk for further malignancy. Furthermore, decreased physical activity and poor nutrition can cause weight gain, which may contribute to secondary health problems such as cardiovascular disease and diabetes. The authors reviewed the extant literature on four key health behaviors among patients with cancer and survivors: healthy diet, reduced tobacco use, reduced alcohol use, and increased physical activity. They described the prevalence of these behaviors, reviewed the effects of interventions designed to alter unhealthy behaviors, and discussed the implications and future directions for this emerging area of research. Cancer 2005. © 2005 American Cancer Society.

With primary prevention efforts, improved detection, and treatment, cancer incidence rates have decreased among men (Surveillance, Epidemiology, and End Results program and National Program of Cancer Registries data, 1992–1998) and overall mortality rates have decreased in both men and women.1 According to the Office of Cancer Survivorship,2 9.6 million individuals in the U.S. have been diagnosed with cancer (a majority with breast, prostate, and colorectal cancers): 39% were diagnosed 0 to < 5 years ago, 47% were diagnosed 5 to < 20 years ago, and 14% were diagnosed > 20 years previously.

With improvements in survival and the possibility that cancer may be viewed as a chronic disease, the reduction of risk behaviors such as smoking, sedentary behavior, and unhealthy diets or, conversely, the adoption and maintenance of healthy behaviors, becomes more salient. Cancer treatments are known to disrupt health behaviors. For example, retrospective reports show that patients report reduction in exercise behavior during the phase of active treatment3 and changes in taste sensitivity associated with chemotherapy alter food preferences and nutritional intake.4 Furthermore, the stress associated with a cancer diagnosis may disrupt health behaviors and potentially increase unhealthy behaviors. In some instances, continued unhealthy behaviors such as smoking can adversely affect responsiveness to RT (e.g., increased side effects such as oral mucositis)5 and chemotherapy (e.g., increased side effects such as long-term pulmonary toxicity),6 increase postsurgery complications (e.g., impaired wound healing), and increase the risk of disease recurrence or of developing a second malignancy7–9 (see reviews10, 11). In addition, sedentary behavior and unhealthy diets place patients with cancer and survivors at risk for chronic diseases such as cardiovascular disease, diabetes, and osteoporosis.

The current review is based on our previous article that reviewed health behaviors after cancer.12 We focus on four health behaviors: smoking cessation, limitation of alcohol use, healthy diet, and regular exercise among patients with cancer receiving adjuvant treatments (chemotherapy and/or RT) and survivors (defined here as those who have completed surgery, and chemotherapy and/or RT). We describe the prevalence of these behaviors and review the effects of interventions promoting these behaviors. Finally, we present directions for future research.

METHODS

We searched MEDLINE and PsychInfo computerized databases for studies reporting prevalence and interventions targeting smoking, alcohol use, diet, and exercise published in English since 1980. Key words included “cancer” and “cancer survivors, ” “smoking” and “smoking cessation, ” “alcohol” and “alcohol intervention, ” “diet, ” “nutrition, ” “dietary intervention, ” “exercise, ” and “physical activity. ” In an attempt to be broadly inclusive, intervention studies were deemed acceptable for review if they included, at a minimum, a pre-post design. For dietary interventions, we selected studies that aimed to alter overall dietary patterns, rather than adding single nutrients or dietary supplements. Regarding exercise, we excluded cross-sectional studies such as those evaluating changes in exercise tolerance among pediatric cancer cases versus healthy controls, and studies that offered physical therapy to patients with cancer.

Prevalence of Health Behaviors

Adult patients with cancer

In a survey of 1667 survivors of breast and prostate cancer from the Duke Cancer Registry (patients diagnosed up to 6 years previously), Demark-Wahnefried et al.13 found that among respondents (a 58% response rate, 531 with early-stage breast cancer and 447 with prostate cancer), 8% were current smokers (25% were preparing to quit in the next 30 days), 42% reported eating ≥ 5 fruits and vegetables daily, 69% reported a daily dietary fat intake < 30%, and 58% reported regular exercise (defined as ≥ 30 mins on ≥ 3 days per week). These statistics reflect a high prevalence of healthy behaviors among the respondents. Surveys on smaller samples of survivors of breast cancer also tend to support these conclusions.14, 15 In the Duke survey, 80% were interested in health education programs and 57% wanted such information at diagnosis or soon after. Interest in specific programs of diet, exercise, and smoking cessation was high (50–60%). However, younger survivors (age < 65 yrs) were significantly more interested in such programs than older respondents (age ≥ 65 yrs) and survivors of breast cancer were more interested in the programs than those who had survived prostate carcinoma. The sample reported low to moderate confidence in their ability to make lifestyle changes. Although these data suggest a high prevalence of healthy behaviors among those treated for breast and prostate cancer, and a high level of interest in health promotion, patients who volunteer to participate in such surveys are likely to be biased towards reporting healthy behaviors, and as with all surveys, the data are limited by bias of self-report.

Survivors of childhood cancers

Little attention has been paid to health behaviors among survivors of childhood cancers. Mulhern et al.16 surveyed 110 parents of long-term survivors ranging in age from 11 years to 17 years and 40 adult long-term survivors of childhood cancer ranging in age from 18 years to 29 years. Response rates were 57–58%. The prevalence of tobacco and alcohol use was < 10% among the younger age group (age < 18 yrs) as reported by their parents, and 6.4% reported < 1 hour of weekly exercise. In the adult group, 17.5% were currently smoking, 72.5% reported alcohol use (12.5% reported binge drinking, 2.5% reported heavy drinking), and 17.5% reported < 1 hour of weekly exercise. The researchers appropriately concluded that the young adult survivors appeared to practice healthy lifestyles when compared with age peers in the general population. Surprisingly, demographic factors were not associated with healthy behaviors. A majority of both parents and adult survivors believed that it was important for the former patient to remain healthy compared with most other people, suggesting that this group of survivors may be receptive to adopting and maintaining healthy behaviors.

In sum, the prevalence of healthy behaviors has been studied in relatively small, select samples of survivors of pediatric and adult cancers. There is a need to evaluate these behaviors in larger, more representative samples (of varying ages and cultural backgrounds), particularly those with cancers other than those of the breast and prostate.

Smoking

Prevalence

Tobacco use in patients with cancer is a serious health problem with significant implications for morbidity and mortality.11 Among smoking-related cancers (e.g., lung, head and neck cancer), 95–99% of patients have a smoking history.17, 18 Although earlier studies indicate that self-reported smoking cessation rates are fairly high, ranging from 40% to 70%,19–25 this remains a high-risk group for continued smoking and concomitant poor health outcomes.11 For example, in an epidemiologic study investigating the risk of second primary cancers in patients diagnosed with head and neck cancer, smoking and alcohol use were found to significantly increase the risk of a second diagnosis. Odds ratios indicate that smoking ≥ 40 cigarettes daily compared with light smokers or nonsmokers increased the risk of upper aerodigestive tract cancers 3-fold.26

Several studies have investigated smoking rates among adult survivors of childhood and adolescent cancer.27–31 This is an important area because these individuals may be at particular risk for tobacco-related problems due to reduced pulmonary function32 and the cardiotoxic effects of certain chemotherapy agents and RT.33 Most of these studies report little difference between smoking rates among adult survivors and their siblings or other age and gender-matched controls. Generally, nonsignificant patterns indicate that adult survivors of childhood cancer are less likely to smoke, but if they do smoke, they are less likely to quit.28, 31 However, a more recent study contradicts these findings. Emmons et al.34 found that the smoking rates of survivors of childhood cancer (17%) were lower than the general population and that once smoking was initiated, survivors of cancer were more likely to quit. Factors independently associated with a statistically significant relative risk for smoking initiation in this population included older age at diagnosis, lower household income, less education, not receiving pulmonary-related cancer treatment, and not receiving brain RT. In addition, those age < 3 years old when diagnosed were significantly more likely to be ex-smokers. Further research is needed to clarify the discrepancies among these studies and clearly identify risk factors associated with smoking among adult survivors of childhood cancer.

Factors associated with smoking after a cancer diagnosis

Compared with those who abstained from tobacco use, patients with head and neck cancer who continued to use tobacco were less likely to have received postoperative RT, had less extensive disease, had oral cavity disease, and had higher levels of education.35 There is evidence that patients with lung cancer undergoing lung resection who quit smoking preoperatively have better long-term abstinence rates than those who quit postoperatively.36 Schnoll et al.37 also found that having completed treatment increased the likelihood of smoking among those with head and neck or lung cancer, which has implications for the timing of smoking cessation interventions.

Psychologic and physiologic factors also play an important role in predicting tobacco use. Schnoll et al.37 found that continued smoking in patients with head and neck and lung cancer is related to a lower readiness to quit, having relatives who smoke at home, greater time between diagnosis and assessment, greater nicotine dependence, lower self-efficacy, lower risk perception, fewer identified pros and greater identified cons to quitting, more fatalistic beliefs, and higher emotional distress. Raichle et al.38 found that low levels of public self-consciousness (not being concerned over one's presentation to the public) were associated with continued smoking in patients with head and neck cancer. Furthermore, patients with head and neck cancer who attribute the cause of their cancer to past substance abuse (tobacco and alcohol use) and also expect that their future cancer-related health is contingent on their behavior have a lower likelihood of continued smoking.39 Experiencing physical barriers to smoking due to treatment and receiving counseling at the time of diagnosis are predictive of quitting in patients with head and neck cancer. Heavy alcohol use is a negative predictor of smoking cessation among these patients.40

It is clear that multiple factors play a role in predicting who will continue to use tobacco after a cancer diagnosis. Future research should focus on determining the most salient demographic, psychological, and medical (disease and treatment) variables to continued smoking. This will afford the best chance of predicting which patients are at highest risk for continued tobacco use and will maximize intervention effectiveness by targeting the most salient factors at the most appropriate time.

Smoking cessation interventions and effects

Although it is encouraging that a majority of those diagnosed with cancer seem to be capable of quitting without the benefit of intervention, a significant number of patients are unable to quit without professional support. Our review found four randomized clinical trials with patients of varying cancer diagnoses (head and neck, breast, gynecologic, urologic, gastrointestinal). Each compared an intervention with usual care.41–45 Unfortunately, with the exception of the study conducted by Gritz et al. (n = 186),41, 42 most suffer from small sample sizes of ≤ 30 participants. The length of follow-up ranged from 5 weeks to 12 months and abstinence was generally measured over 7 days, however, continuous abstinence was measured in 1 study.41 Interventions included inpatient counseling delivered by nurses or surgeons and outpatient or telephone booster sessions. The content of the interventions included advice to quit, explanation of the benefits of quitting, setting a quit date, overcoming barriers to quitting, and follow-up support. None of these studies reported significant intervention effects. This is possibly due to the small sample size in three of the studies. In the 1 study with a larger sample,41 the 1-year continuous abstinence rate of 64.6% included those who had quit previous to the intervention, which may have diluted the intervention effect.

Given the high risk for increased morbidity and mortality associated with continued tobacco use, further research is needed to determine the efficacy and efficiency of smoking cessation interventions for patients with cancer and survivors. Research should focus on the effectiveness of routine use of the Agency of Healthcare and Research Quality guidelines for smoking cessation46 and who would be the most effective healthcare professional (e.g., physicians, nurses, psychologists) to deliver such interventions. In addition, psychopathology is an important factor to consider in smoking cessation, as depression before quit attempts is predictive of early relapse in many smoking cessation treatments.47, 48 Patients with cancer who continue to smoke may respond to more intensive interventions, such as extended counseling, support groups, nicotine replacement therapy (e.g., patch, gum, inhalers), and medication (e.g., buproprion, fluoxetine). There is scope to evaluate the effects of these intensive interventions among those who are not able to quit smoking on their own.

Alcohol Use

Alcohol use after a cancer diagnosis is less well studied than smoking and tobacco use. Although research indicates an association between alcohol use and increased risk of developing certain kinds of cancer,26, 49, 50 to our knowledge, few studies to date have investigated alcohol use and its risks after a diagnosis. Most of this work has focused on patients with head and neck cancer, the most clearly recognized at-risk group among patients with cancer. There is evidence that heavy drinking, independent of smoking, can cause a 5.8-fold increase in the risk of head and neck neoplasm.51 The risk of second primary tumors in patients with head and neck cancer is increased by 3.8 for those drinking ≥ 15 drinks of beer per week compared with lighter drinkers.26 Deleyiannis et al.50 investigated 649 patients with head and neck cancer and categorized them according to severity of premorbid alcohol use. Those in the most severe category of alcoholic use with a history of other systemic alcohol-related health problems (e.g., liver disease) had a relative risk of death in 5 years that was 2.76 times greater than abstainers or those consuming < 1 drink per week. These risks were independent of age, site of cancer, anatomic stage, histopathologic grade, smoking, and type of treatment. Patients in the two highest categories of alcohol use had an increased risk of dying from not only head and neck cancer, but also of cardiovascular, pulmonary, and other alcohol-related diseases. Our review did not find any intervention studies for alcohol use in a cancer population, nor could we identify any studies that specifically investigated patterns and predictors of alcohol use among survivors of cancer.

It is clear that further research into alcohol use among those diagnosed with cancer, especially head and neck cancer, is warranted. Comprehensive, standardized methods of assessing alcohol use and determining what demographic, psychologic, and medical (disease and treatment) variables are predictive of relapse and abstinence are needed. Generally, the research of alcohol use in patients with cancer and survivors is far behind that of smoking, but these variables have considerable overlap and use of one tends to affect use of the other.40 Therefore, these behaviors should be studied in conjunction with each another. Specifically, factors that are predictive of continued alcohol use need to be identified so that interventions of appropriate duration and intensity can be developed and implemented at a time to maximize effectiveness, thus mitigating short and long-term morbidity and mortality.

Dietary Changes

Prevalence

There is an extensive body of literature on dietary changes to prevent cancer, but less is known about the diets of patients with known disease. In a review of evidence relevant to breast cancer and obesity, Chlebowski et al.52 concluded that women who are overweight or gain weight after diagnosis are at greater risk for breast cancer recurrence and death compared with women who are not overweight. Baseline data from dietary intervention studies indicate that, in one study, women reported dietary intakes of total energy, fat protein, and carbohydrates that were comparable to the dietary patterns described in the U.S. population survey,53 whereas in the other study, the women had a higher percentage of calories from fat.54 Recently, Maunsell et al.55 completed a 1-year follow-up of 250 women with newly diagnosed nonmetastatic breast cancer enrolled in a randomized trial of a psychological screening program and counseling. At 12 months, 41% reported dietary changes after diagnoses with a decrease in meat (77%) and an increase in fruit and vegetable intake (72%) being the most frequent dietary change. Women who reported changes were younger, more distressed at baseline, and had obtained mental health services over the previous year. The high frequency of dietary changes after a breast cancer diagnosis is consistent with those reported by Demark-Wahnefried et al.13

Dietary interventions

There are two ongoing major multicenter randomized, controlled trials of dietary interventions funded by the National Cancer Institute: the Women's Intervention Nutrition Study (WINS) and the Women's Healthy Eating and Living Study (WHELS). WHELS is a part of the Women's Health Initiative. The WINS study is a multicenter, randomized, controlled trial investigating the effects of reducing dietary fat intake with adjuvant systemic therapy on cancer recurrence rates in postmenopausal women with early stage, surgically treated breast cancer.56 The primary aim of the WHELS trial is to evaluate the effects of a high-vegetable, low fat diet in reducing breast cancer recurrence and mortality. Participants (n = 3088) were randomly assigned to an intensive diet intervention or comparison group and will be followed through 2006.57 While awaiting the results of these trials, our review includes the pilot studies on which these major trials were based.

Our review found five randomized clinical trials delivering dietary interventions to patients with cancer and survivors. Sample sizes ranged from 93 to 240 subjects and the intervention goals included reduction of fat intake to 15–20% of total energy,53, 54, 58 increasing vegetable and fiber intake to reduce the risk of disease recurrence,59 and the adoption of a low fat, high fiber diet.60 All the studies used self reported data, but two added biochemical and physiologic indices, such as serum lipid levels54 and circulating carotenoid concentrations.59 All 5 studies included relatively long follow up periods, ranging from 1 to 2 years after baseline assessment. Intervention features typically included individualized dietary counseling focusing on low fat, high vegetable, and high fiber food preparation and intake. With the exception of one study that used volunteer dieticians and nurses,53 the other four studies used trained dieticians to deliver the intervention in individual54, 58 or group sessions.53, 60 In one study, the intervention was delivered via telephone contact.59 Control groups received mindfulness stress reduction in one study60 and minimal printed materials or no intervention in the other studies. The dietary interventions were offered to patients in treatment54, 58 and to those within 4 years of diagnosis.59 It is interesting to note that intervention group attrition rates were as high as 50% when participants were recruited shortly after diagnosis.54, 58 Conversely, retention rates were high (75–80%) in studies when participants were recruited ≥ 1 year since diagnosis.53, 59

Effects of dietary interventions

The dietary interventions produced generally positive results with reductions in total fat intake, increases in fruit and vegetable intake, or increases in carbohydrates. For example, Nordevang et al.58 achieved a reduction in total energy from fat intake from 36% to 23% in their intervention group versus from 37% to 34% in the control group over the course of 1 year. In the WINS pilot,54 the intervention group reported reducing their fat intake to 20.3% versus 31.5% in the control group, but these changes were not reflected in group differences in lipid profiles. The intervention group in the WHELS pilot59 achieved significant changes through the 1-year follow-up in daily consumption of vegetables (+4.6 servings per day), fruit (+0.7 servings per day), and fiber (+6.4 g/kcal) and significantly reduced fat intake (−9.9% of total energy) versus controls. Circulating concentrations of carotenoids also increased in the intervention group. Kristal et al.53 reported a significant decrease in fat intake in the intervention group to 20% of total energy and increased carbohydrates to 62.6% of total energy versus 28.8% and 55.7% for controls, respectively. Finally, in the fifth study,60 women who received the nutrition intervention reported a large reduction in fat consumption at 4 months (a 5.8% decrease) and also at 1 year (a 4.1% decrease) versus the comparison groups. Similarly, they showed a significant reduction in body mass at 4 months (a loss of 1.3 kg). These results were not maintained at 12 months. Although weight loss was not a goal for the programs, significant weight losses in the intervention groups were noted in three of the studies.53, 54, 60

The strengths of these studies include strong research designs with extended follow-ups lasting ≤ 2 years. They also investigated different and, perhaps, more cost-efficient modes of intervention delivery via telephone counseling59 and the use of volunteer staff.53 Two studies also corroborated self-reported data with biologic and physiologic indices.54, 59 Limitations include relatively homogeneous samples consisting mostly of middle-aged white participants. In addition, the role of exercise was not adequately evaluated in any of the studies and although significant changes in diet did occur, it is unclear if these improvements last beyond the follow-up periods of the studies. More importantly, the association between these dietary changes and actual risk for cancer recurrence has not been adequately studied to date, but major trials such as the WINS and WHELS are expected to yield valuable information to address this and related questions.

Regular Exercise

Prevalence

Cancer diagnosis and treatment are associated with increased distress (anxiety and depressed mood), reduced physical functioning, increased fatigue, changes in body image, and weight loss/gain.61–66 In the past, patients with cancer were generally advised to reduce activity. However, prolonged sedentariness can precipitate a disuse syndrome including decreased muscle mass and strength, poorer cardiovascular functioning, limited range of motion, an increased sense of disability and increased risk of injury. Cross-sectional studies in small and larger samples of survivors of cancer (a majority being survivors of breast cancer) suggest that a high number of respondents report regular exercise.13–15 For example, Demark-Wahnefried et al.13 found that 58% of their respondents reported regular exercise (defined as 30 mins of exercise on ≥ 3 days per week). Retrospective studies among patients with breast and colorectal cancer have shown reduced participation in exercise after a cancer diagnosis and then an increase in participation after treatment has been completed.3, 67 However, posttreatment levels of exercise are lower than prediagnosis levels. Among patients with breast cancer, exercise during active treatment has been associated with higher quality and satisfaction with life. Maintenance of exercise across time has been optimally associated with enhanced quality and satisfaction with life.3

In a small sample of survivors of breast cancer followed prospectively in the first year after treatment completion, Pinto et al.68 found that 58% of the survivors did not exercise regularly at quarterly assessments over 12 months. The major goals of the study were to evaluate mood and coping in the first year after treatment. Hence, overestimation of exercise behavior may have been less likely. Analyses of longitudinal data revealed nonsignificant increases in the minutes of moderate and vigorous exercise over the year. It is interesting to note that women who adopted vigorous exercise were more likely to be married/live with a partner, younger, had completed treatment earlier, had greater social support, and were more depressed at baseline. Exercise participation was predictive of physical functioning but not of cancer-related symptoms or mood.68 Results from this longitudinal study (in contrast to cross-sectional data) suggest that survivors of breast cancer do not, spontaneously, become more physically active after treatment. In addition, weekly exercise was below national recommendations69 among those who did become active.68

Exercise interventions

There has been growing interest in evaluating the effects of exercise participation in intervention trials among patients with cancer and survivors.70, 71 These interventions were offered chiefly to middle-aged patients with early-stage breast cancer, who were receiving chemotherapy72–74 or RT.75 A few programs have been offered to those who have completed medical treatment for cancer.73, 76–78 Moderate aerobic exercise (60–85% of maximum heart rate) has been promoted in these interventions that have focused on short-term adoption (< 6 mos) of physical activity.75, 78, 79–86 Strength-training components were included in two studies.76, 77 On-site supervised exercise has been the norm for patients who were receiving chemotherapy or RT,74, 84 although a few researchers have studied the effects of home-based programs.74, 75, 83 Nearly all the interventions were offered independent of the healthcare provider with the exception of on-site physician supervision of exercise among patients who had completed bone marrow transplants.79–82

Effects of exercise interventions

As reflected in previous reviews,12, 70, 71 participants in the exercise arms of these studies have shown improvements in functional capacity as demonstrated in performance on exercise stress tests72, 80, 81, 84, 86 or field tests of fitness.75, 83 For example, MacVicar and Winningham72 found 20% improvement in maximal oxygen uptake among patients who participated in supervised exercise versus controls. Other benefits include reduction of fatigue,75, 82, 87 prevention of weight gain,85 attenuation of cancer treatment sequelae such as neutropenia (low leukocyte count),80 pain,80 and sleep problems.75 When psychological outcomes are considered, women who exercise during or after cancer treatments report reductions in anxiety75, 88 and depression,88 and improvements in body esteem.89 The programs have documented that it is safe and feasible for patients with cancer to participate in supervised exercise programs (see guidelines4, 90). In addition, retention rates in the exercise conditions have been impressive (73–100%).

Future directions

Unfortunately, not much is known about exercise participation among those diagnosed with cancer other than that of the breast, and it is timely to evaluate the potential effects of regular exercise among these individuals. Among the cancer groups that have been studied, attention should be paid to the characteristics predicting which patients are more likely to adopt and maintain physical activity. Researchers should also recognize that the type of exercise and the exercise “dose” has varied considerably across studies, making it necessary to identify a minimum threshold for benefits. In addition, investigators should focus on the type of benefits (e.g., symptom management, physical functioning, quality of life, mood and mental health benefits) that would be most relevant to patients in specific phases of disease diagnosis and treatment (i.e., at diagnosis, during treatment, posttreatment, and at extended periods after treatment). Aerobic exercise has been evaluated in the studies reviewed, but it is worthwhile to assess the feasibility and effects of strength training among patients with cancer and survivors. For practical reasons such as the expenses associated with on-site supervised programs, the role of home-based programs should also be considered with the caveat that it is important to document exercise adherence (frequency, duration, and intensity) and any side effects. Finally, as this field develops and the effects of exercise programs become clearer, it would then be timely to evaluate the maintenance of exercise and its associated effects.

Before we conclude the current review, which focused on the efforts to alter a single risk behavior, there have been efforts to target more than one risk behavior as reflected in weight management interventions for patients with cancer. One of these was a single group study (n = 55) among newly diagnosed patients with breast cancer91 in which weight gain was successfully prevented (70% met weight goals) and there are other ongoing randomized trials among larger samples that aim at modifying both diet and sedentary behavior.92, 93 Although additional studies aimed at weight management, they focused on dietary changes alone.60, 94–98

SUMMARY

With the growing number of survivors of cancer, the overall health and well-being of these individuals will require attention. As our review indicates, interest in assessing and changing health risk behaviors among patients with cancer and survivors has a relatively short history and, hence, there are several avenues for growth in this field.

It is clear that women diagnosed with breast cancer have been extensively involved in studies on dietary changes and exercise, and patients with head and neck and lung cancers have been the focus of smoking cessation efforts. The relevance of these behaviors to those diagnosed with other cancers and the effects of these behavior changes on symptom management, QOL, and overall risk reduction are worthwhile avenues to be pursued. The National Cancer Institute has recognized that the burden of cancer is disproportionately found among racial and ethnic minorities,99 yet their representation is extremely limited in studies of health behaviors.

As the field grows, it is important to address the limitations of research design and delivery among the studies that have been reviewed in the current article. In intervention studies, the role of theory in guiding the interventions has not been as well integrated as one might expect. With the increasing number of interventions, particularly those promoting exercise adoption, one might get a clearer understanding of the most opportune time to intervene after a diagnosis. It is plausible that healthcare providers can play a critical role in interventions offered to patients in treatment and those in follow-up care, but their roles in facilitating behavior change have been underutilized. The interventions reviewed, for the most part, involved face-to-face contact with a healthcare professional. There is ample scope for exploring the role of computer technologies for behavior change in this population.100, 101 It is clear that patients with cancer, like others, vary in their readiness to make and sustain behavior changes. There are opportunities to identify and intervene with patients and survivors who are more ready to make behavior changes, and those at risk for continued unhealthy behaviors. In noncancer populations, there is increasing attention to making changes in more than one health behavior and the potential role for change in one risk behavior (e.g., quitting smoking) to function as a gateway to making changes in others.102–104 To date, most studies have focused on changing a single risk behavior among patients with cancer and survivors. A few multiple-risk factor interventions have been offered in the context of weight management.91–93 Finally, the studies reviewed did not assess the cost-effectiveness of their interventions, which is understandable given that they tested efficacy. It is timely for investigations to include cost-effectiveness measures in intervention research, keeping in mind that if the interventions are shown to be efficacious, their potential for dissemination will need to be evaluated.

Ancillary