Abstract
Purpose. The purpose of the study was to determine the prevalence of complementary and alternative medicine (CAM) use among US cancer survivors; examine whether use varies by underweight/normal weight, overweight, and obese body mass index status; determine reasons for use; and document disclosure rates of CAM use to medical professionals. Methods. Data for 1785 cancer survivors were obtained from the 2007 National Health Interview Survey and CAM supplement. The prevalence and associations of CAM use in the previous 12 months were compared among underweight/normal weight, overweight, and obese adult cancer survivors. Results. Nearly 90% of cancer survivors used at least one type of CAM therapy in the 12 months preceding the survey. Those who were overweight, but not obese, were more likely to use a CAM modality compared to normal/underweight respondents. Over two thirds (71%) reported using CAM therapy for general health and wellness and 39.3% used CAM because a health care provider recommended it. Disclosure rates of CAM use to conventional medical professionals varied widely by CAM modality. Conclusions. An overwhelming majority of US cancer survivors use CAM for a variety of reasons. Overweight cancer survivors may be more likely to use CAM than those who are underweight, normal weight, or obese. Cancer survivors should be screened by medical providers for the use of CAM therapies; furthermore, prospective clinical research evaluating the efficacy and safety of biologically based CAM therapies, often used by cancer survivors, is important and necessary for the well-being of this population.
Introduction
The American Cancer Society estimated that 1.6 million new cases of cancer would be diagnosed in the United States in 2014, 1 and approximately 20% of all cases are attributed to weight, weight gain, and obesity. 2 A study conducted by the International Agency for Research on Cancer concluded that obesity was a cause of 11% of colon cancer cases, 9% of postmenopausal breast cancer cases, 39% of endometrial cancer cases, 25% of kidney cancer cases, and 37% of esophageal cancer cases. 3 With obesity a common risk factor for cancer, many cancer patients may be overweight or obese at the time of diagnosis.4,5 Excess weight has been shown to have profound effects on cancer progression and prognosis. For example, obesity is associated with increased risk of positive surgical margins, 6 cancer recurrence,7,8 higher grade and larger tumors, 9 and cancer death. 10 Prior studies have also shown an association between weight status and chemotherapy resistance5,9,11-14 and indicated that obese cancer patients are at increased risk of developing problems following surgery, including wound complication, lymphedema, second cancers, and chronic diseases such as cardiovascular disease and diabetes. 15 Increased illness burden and poor outcomes experienced by overweight and obese cancer survivors may encourage the use of additional health care modalities, such as complementary and alternative medicine (CAM), that are beyond the scope of conventional medicine.16-18
The National Center for Complementary and Alternative Medicine (NCCAM) defines CAM as a group of heterogeneous health care systems, practices, and products that are outside the realm of conventional medicine. 19 Several studies have examined the prevalence, trends, and reasons for CAM use among the US cancer population.18,20-26 A recent meta-analysis of 152 studies from 18 countries concluded that CAM use is highest in the United States and that over the past 30 years, especially after year 2000, the proportion of cancer patients using CAM has increased. 27 Boon et al’s study on trends in CAM use by breast cancer survivors showed that use of CAM therapies increased from 67% to 82% between 2002 and 2005. 28 CAM use among cancer survivors may be motivated by a variety of reasons, 29 including improving overall well-being, preventing complications and illness, managing cancer-related symptoms and late-effects, and promoting a healthier lifestyle.30,31 Moreover, a large proportion of patients may use CAM for weight loss and weight control. 32
Despite extensive research examining CAM use within various cancer subpopulations, there is a knowledge gap on the patterns and prevalence of CAM use in the increasing overweight or obese US cancer survivor population. Information on the prevalence and associations of CAM use among this population can guide clinicians as they provide care to cancer survivors who are using nonconventional therapies and treatments. Moreover, it can guide the development of interventions to promote appropriate and informed use of effective evidence-based CAM practices among this population and may form the basis for much discussion on CAM, especially as it relates to its integration into routine care. Therefore, using data from the 2007 National Health Interview Survey (NHIS) and CAM supplement, this study sought to (a) determine the prevalence and patterns of CAM use among US cancer survivors; (b) examine whether there are differences in CAM use by underweight/normal weight, overweight, and obese body mass index (BMI) status; and (c) describe reasons for CAM use within this population as well as disclosure rates to medical professionals.
Methods
Data Source
The data for this study were obtained from the 2007 NHIS, an annual cross-sectional survey of a representative sample of the civilian, noninstitutionalized US household population. The NHIS is conducted by the Centers for Disease Control and Prevention’s National Center for Health Statistics and is composed of questions from core and supplemental questionnaires. The supplemental questionnaire queries about the 36 types of CAM therapies commonly used in the United States. 33 NHIS data are collected through personal interviews conducted in respondent’s home and uses a multistage sampling design. The response rate for the sample adult component of the NHIS in 2007 was 87.1%, 33 and details of the survey procedures and methodology can be found elsewhere. 34
Measures
Cancer Survivorship Status
In this study, cancer survivors include persons with a history of cancer diagnosis (ie, respondents who answered “yes” to the question, “Have you ever been told by a doctor or other health professional that you had cancer or a malignancy of any kind?”), regardless of whether or not they had symptoms of cancer at the time of the survey. The current cancer status (ie, active disease or remission) was not assessed in the survey. If the respondents reported a history of cancer, they were asked the cancer site (ie, “What kind of cancer was it?”). There were 23 393 (30.9%) respondents who answered the question, “Have you ever been told by a doctor or other health professional that you had cancer or a malignancy of any kind?” Of these, 1785 (7.6%) respondents diagnosed with cancer were included in this study. In order to compare with prior research on CAM use 22 and to maintain a sufficient sample size, those reporting non–melanoma skin cancer were included in our final sample.
Body Mass Index
We defined BMI as weight (kg)/(height (m))2 and categorized weight status using the following categories: underweight/normal weight (BMI <25), overweight (BMI ≥25 and <30), and obese (BMI ≥30). Self-reports of respondents’ weight and height were used to calculate BMI. We combined respondents with normal and underweight BMI into one category to maintain an adequate sample size in the control (referent) group for statistical analysis.
Complementary and Alternative Medicine Use
The NHIS CAM supplement included questions pertaining to the use of 36 different CAM therapies used within the last 12 months. For these analyses, the therapies were collapsed into the following 5 CAM modalities as recognized by the NCCAM 19 : manipulative and body-based therapies (ie, chiropractic or osteopathic manipulation, massage, movement techniques), biologically based therapies (ie, chelation therapies, herbs, vitamins and minerals, special diets), mind–body interventions (ie, biofeedback, hypnosis, yoga, tai chi, qi gong, and relaxation techniques such as meditation, guided imagery, progressive relaxation, deep breathing exercises, support group, stress management class), energy healing (eg, reiki), and alternative medical systems (ie, acupuncture, ayurveda, naturopathy, homeopathy, traditional healers). Traditional healers include curanderos, espiritistas, hierberos, shamans, botanicas, Native American healers, and sobadors. Special diets include vegetarian, macrobiotic, Atkins, Pritikin, Ornish, Zone, and South Beach. Movement-based therapies include Feldenkreis, Alexander technique, Pilates, and Trager. Respondents were classified as using any CAM therapy if they answered “yes” to any of the listed therapies. Consistent with previous research, 33 we created categories of overall CAM use, and use of mind–body interventions without prayer.35,36
Reasons for CAM Use and Disclosure of CAM Use to Medical Professionals
Factors highly associated with CAM use were assessed using standardized questions.
37
Respondents who gave affirmative responses for using a particular CAM therapy were then asked to indicate their reasons for using it:
Did you choose [insert CAM therapy] for any of the following reasons? For cancer; for weight problem; to enhance immune function; for general wellness or disease prevention; a conventional medical professional suggested you try [insert CAM therapy]; conventional medical treatments would not help you; it was recommended by friend, family, or coworkers; or conventional treatments were too expensive?”
Disclosure of CAM use to medical professionals was ascertained by the question, “During the past 12 months, did you let any of the conventional medical professionals know about the use of [insert CAM therapy]?” Respondents could choose either “yes” or “no.”
Sociodemographic, Behavioral, and Clinical Factors
Several potential confounders associated with obesity, cancer, and CAM use were considered. Among them were sociodemographic factors such as age, gender, race, nativity (ie, US-born), household income, educational attainment, marital status, and region of residence. Marital status was recoded into the following variables: married, widowed, divorced or separated, and single. “Married” was recoded from the categories of married-spouse in household, married-spouse not in household, and living with a partner. Those who were classified under the categories of unknown marital status or single were categorized as “single.” Health insurance status was recoded as a dichotomous variable (ie, insured/not insured). Respondents were classified as having a comorbidity if they reported having cardiovascular disease (ie, stroke, high blood pressure, coronary heart disease) or respiratory disease (eg, emphysema or asthma) in the 12 months preceding the survey. Those who reported ever having a musculoskeletal disorder (ie, arthritis, neck pain, back pain) or diabetes were also included. Cancer type was recoded into the following categories: male-specific cancer (ie, prostate cancer and testicular cancer), female-specific cancer (ie, ovarian, cervical, breast cancer), and non-gender-specific cancer (eg, lung cancer, colorectal cancer). As a measure of respondent’s health behavior, we included data on smoking status and alcohol use (ie, current, former, never). Similar to prior studies, 35 we calculated time since cancer diagnosis from the question “How long have you had cancer?” using 4 categories: within the past year, 2 to 5 years, 6 to 10 years, >11 years.
Statistical Analysis
Data were weighted and analyzed using STATA 12 for Windows (Stata Corp Inc, College Station, TX) to account for the complex sampling design and weighting procedures. We computed frequencies and percentages for categorical variables (see Table 1), reason for CAM use (see Figure 1), and disclosure of respondent’s CAM use to conventional medical providers (see Figure 2). The prevalence and weighted percentage of CAM use across all categorical variables were computed. Additionally, a multivariable logistic regression model was generated to examine the associations between BMI status and overall CAM use while adjusting for sociodemographic, clinical, and behavioral variables. Adjusted odds ratios with 95% confidence intervals and P values for the associations between categorical variables and overall CAM use were reported (see Table 2). A χ2 test was performed to explore differences in the respondent’s use of CAM. All proportions and population counts presented are weighted to provide national estimates.
Sample Characteristics (N = 1785).
Abbreviation: BMI, body mass index.
Reasons for complementary and alternative medicine (CAM) use.
Disclosure of complementary and alternative medicine (CAM) use to conventional medical provider.
Characteristics Associated With CAM Use Among Cancer Survivors.
Abbreviations: CAM, complementary and alternative medicine; OR, odds ratio; CI, confidence interval; BMI, body mass index.
P < .05. **P < .01. ***P < .001.
Results
Sample Characteristics
Of the 1785 respondents included in our sample, 60.5% were female, 87.7% were white, and 45.2% were between the ages of 55 and 74 years (see Table 1). Most of the respondents were born in the United States (91.5%), had health insurance (94.1%), and had a high school diploma (54.4%). Nearly half (49.2%) of the cancer survivor population were married, over one third resided in the southern region (37.7%), and more than two fifths had an annual household income of <$35 000 (43.1%). Majority of respondents had a BMI classified as overweight or obese (59.4%), were diagnosed with a non-gender-specific cancer (61%), and had a musculoskeletal comorbidity (61.7%). More than half of our sample had never smoked cigarettes (53.2%).
Use of Complementary and Alternative Medicine
Overall, nearly 90% of the cancer survivors used at least one type of CAM product in the 12 months preceding the survey (see Table 2). CAM use was highest among respondents who were overweight (92.6%), followed by those categorized as obese (89.4%). The most commonly used CAM modalities include mind–body interventions (70.7%) and biologically based approaches (70.3%). A smaller percentage used manipulative and body-based approaches (14.5%), alternative medical systems (3.0%), and energy healing (0.8%).
At the bivariate level, BMI was significantly associated with CAM use (P < .05). Those who did or did not use CAM were as follows, respectively, overweight (92.6% vs 7.4%), obese (89.4% vs 10.6%), and normal/underweight (88.4% vs 11.6%). There was no significant difference in CAM use among cancer survivors with other comorbidities. However, in our study, we observed a significant association (P < .05) between cancer type and CAM use in the bivariate analysis, although the association became insignificant at the multivariate level. Cancer survivors with a male-specific cancer reported a higher level of CAM use compared with a female-specific cancer (92.2% vs 83.7%, respectively). Also, CAM use was 88% among persons with a non-gender-specific cancer. Race and alcohol use were not significantly associated with CAM use. See Table 2.
In multivariable logistic regression, overweight BMI status was significantly associated with CAM use (P < .01) such that the odds for overweight respondents were nearly twice that of underweight/normal weight respondents. In addition, there was significant association between female cancer survivors and CAM use (P < .01). Female respondents had about 2.5 times the odds of CAM use compared to their male counterparts (odds ratio [OR] = 2.54, 95% confidence interval [CI] = 1.59-4.07, P < .001). Moreover, respondents who resided in the western region of the United States (OR = 1.86, 95% CI = 1.03-3.33, P < .05), were married (P < .05), or had a college degree (OR = 1.90, 95% CI = 1.02-3.54, P < .05) had a higher odds of using CAM. Former (OR = 0.96, 95% CI = 0.68-1.37, P > .05) and current (OR = 0.12, 95% CI = 0.03-0.46, P < .01) smokers were less likely to use CAM compared to nonsmokers. Those who were diagnosed 2 to 5 years prior were more than 6 times as likely than those diagnosed a year or less prior to the survey (OR = 6.23, 95% CI = 1.29-30.03, P < .05). See Table 2.
Reasons for CAM Use and Disclosure of CAM Use to Health Care Providers
Reasons for using CAM varied widely among cancer survivors (see Figure 1). Seventy-one percent of the population reported using CAM “for general health and wellness,” and nearly two fifths (39.3%) reported that a “health care provider recommended it.” The least common reported reasons for CAM use were to “enhance energy” (13.7%), “for cancer” (4%), “conventional medicine is too expensive” (3.9%), and “weight problems” (0.2%). See Figure 1.
The disclosure of CAM use to conventional medical professionals varied widely by CAM modality (see Figure 2). The most highly disclosed CAM therapies were biologically based approaches (53.2%). Relatively few respondents revealed their use of mind–body interventions (8.2%), manipulative and body-based therapies (5.3%), alternative medical systems (2.0%), and/or energy healing (0.5%) to their medical practitioner.
Discussion
Complementary and alternative medicine use has been studied extensively within various US cancer populations.38-42 However, to our knowledge this is the first study to use nationally representative data to examine whether there are differences in CAM use by BMI status.
Of the 1785 cancer survivors in our sample, a high proportion (89%) used a CAM modality within the 12 months preceding the 2007 survey. Interestingly, prevalence of CAM use in the sample was high compared to findings from previous studies analyzing the 2007 NHIS. This may be due to differences in study design and analytical methods. Previously, Mao et al found 43.3% of the cancer population used CAM within the previous 12 months; however, a notable difference is that those who were diagnosed with non–melanoma skin cancer were excluded from the study. 35 Furthermore, while most studies reported categorical and narrower definitions of CAM use,43-45 our study reports CAM use from a more extensive list of CAM therapies, as well as the aggregate rate of CAM use. Notably, Mao et al excluded the use of vitamins/minerals from biologically based CAM. 35 Furthermore, Anderson and Taylor reported vitamin/mineral and herb use of 76% and 32% by cancer population, respectively; however, those were the only CAM therapies classified as biologically based modality 22 in the analysis of the 2007 survey. It is important to note that our study includes the use of vitamins/minerals along with other therapies considered biologically based CAM consistent with NCCAM’s definition of CAM, which may be a factor for the higher prevalence of CAM use reported in the present study.
Although obesity was associated with CAM use, obese respondents had 13% higher odds of CAM use compared to underweight/normal weight respondents. Previously, Bertisch et al found that obese respondents were less likely to use CAM compared to those who were normal weight in an analysis of the general US population in the 2002 NHIS. 46 The varying reports may be due to differences in the rates of CAM use for each year the survey was conducted, the unit of analysis, and the scope of CAM therapies analyzed in each study. Research by Puhl and colleagues indicates that some CAM modalities may be more difficult to perform by obese adults, and therefore may create feelings of self-doubt, discomfort, and embarrassment.47,48 This may explain the lower likelihood of use of some CAM modalities among the obese respondents compared with overweight respondents in our study.
Cancer survivors were more likely to use CAM if they had a college education, married, female, and resided in the western region of the United States. These findings are consistent with those of prior studies among the general and cancer population.33,49 Interestingly, cancer survivors who reported being diagnosed 2 to 5 years prior to the survey were nearly 6 times more likely to use CAM compared with those diagnosed within a year prior to the survey. It is likely that the amount and types of CAM therapies used by cancer survivors change as time since diagnosis increases and as their health care needs change. For example, it is possible that cancer survivors who are diagnosed within the past 2 to 5 years use CAM as a means to help prevent cancer recurrence or progression following their primary cancer treatment, as supported by findings in the literature. 49 This is suggested by research which shows that cancer patients are more likely to use a nontraditional therapy if they had a history of cancer recurrence or metastasis. 50 In addition, cancer patients may be more likely to use CAM after cancer treatment, as supported by a 2006 study that showed that 54% of cancer respondents used CAM after a diagnosis of cancer and use was more likely for those who had a history of chemotherapy or enrollment in clinical trials. 51
Consistent with prior research on CAM use among cancer survivors, 52 biologically based therapies (70.3%) and mind–body interventions (70.7%) were the most commonly used modalities that may be explained by various factors. First, biologically based and mind–body CAM therapies can be acquired in a market offering inexpensive alternatives to conventional medicines. Herbals and supplements have been shown to be highly accessible via stores, the Internet, and street vendors.42,53 Additionally, mind–body interventions, such as meditation and deep breathing exercises, also offer a low-cost, nonintensive, and flexible option for cancer survivors.22,25,54 A recent longitudinal study demonstrated that cancer survivors participating in a 7-week yoga program reported a decrease in mood disturbance, stress symptoms, and improved health-related quality of life. 55 Furthermore, special diets consumed over a long period of time provide a specific range of nutrients that are speculated to have chemotherapeutic or chemopreventative effects; such diets may be of interest to cancer survivors who are interested in ways to prevent new cancers and late effects and maintain their health. 19 Moreover, it has been established that between 30% and 60% of cancer survivors have reported consumption of a “healthier diet” with decreased consumption of meat and increased consumption of vegetables and fruits. This, among other research, suggests that a significant percentage of those who have survived cancer may opt for lifestyle modifications that include CAM in the forms of special diets or mind–body interventions.56-59
A greater illness burden (eg, poor prognosis and comorbid disease) has been shown to be associated with increased CAM use.24,60 Furthermore, obesity has been associated with later-stage diagnosis of cancer61-63; as such, it may be possible that those cancer survivors who are overweight or obese use CAM to cope with the secondary sequelae and late effects associated with their cancer diagnosis.4,11,64 For example, advanced-stage breast cancer patients reported CAM use to strengthen the immune system (40%), treat cancer (32%), relieve the side effects and treatment (21%). 65 Although our study does not echo prior studies that show CAM use for the treatment of cancer, a high prevalence of use for health maintenance and disease prevention (72%) indicates a strong desire to mitigate the effects of illness in the cancer survivor population.
The 2007 NHIS survey did not specifically query for the use of CAM for weight loss; however, our analyses of the data revealed that less than 1% used CAM for a weight problem. Despite our results, it is worth noting that other reports have shown that certain biologically based modalities for weight loss can increase the risk for cancer and other serious health conditions. For example, in Belgium, cases of women with extensive interstitial fibrosis of the kidneys following use of a Chinese herbal weight-loss regimen were observed. 66 Findings from a population-based case-controlled study from 1997 to 2002 conducted in Taiwan showed that the use of herbal medicines known to contain aristolochic acid was associated with increased risk of cancer of the urinary tract. 67 Cases of probable aristolochic acid–induced renal failure have been reported in the United States, 68 and despite warnings by the US Food and Drug Administration and other similar actions by regulatory authorities in several countries, herbal products that likely contain aristolochic acid could still be easily purchased on the Internet at least as recently as 2012. 69 Unregulated use of herbals and other biologically based modalities pose a risk to populations with a high prevalence of overweight and obesity, as well as cancer survivors who may seek a means for weight loss.
Interestingly, our analyses of the 2007 NHIS data reveals that conventional health care providers may be the source for encouraging CAM use among cancer survivors. Nearly 40% of cancer survivors used a CAM modality because a health care provider recommended it. A recent analysis of the 2007 NHIS by Mao et al shows that CAM use is more likely among cancer survivors versus the noncancer population if a health care provider has recommended it. 35 This suggests that there may be some integration of CAM therapies in conventional cancer care settings. Thus, more research should be conducted on the best practices for the safe use and incorporation of CAM therapies in the practice of conventional medicine.
Rates of disclosure of CAM use to medical providers varied significantly by CAM modality. More than half (53.2%) of the cancer population apprised conventional medical providers of their CAM use. Previous studies reported lower disclosure rates by the cancer and general populations (22.7% and 40%, respectively) compared with the current study.35,70,71 It may be speculated that this variation may be due to the specific CAM modalities assessed by each study (eg, CAM use considered was focused on provider-based therapies vs non-provider-based therapies). Other possible explanations are the differences in the population samples analyzed as a result of varying exclusion criteria. Among those who reported using biologically based CAM, we found that disclosure of CAM use to conventional medical providers is less than reported usage of such products (53.2% vs 70.3%, respectively). A limited amount of studies have been conducted to evaluate the safety and efficacy of botanicals or supplements on symptoms related to cancer. These CAM products can potentially interact with treatments prescribed by conventional health care providers. With over two thirds of our sample population being overweight or obese, it must be considered that adverse effects from conventional chemotherapeutics may be exacerbated by certain biologically based CAMs that potentially could be used for weight control. Clearly, health care providers should screen for the use of biologically based therapies in all patients being treated with conventional chemotherapy; however, for those with certain comorbidities (eg, overweight and obese), it is especially necessary in order to enhance patient safety and well-being.
This study has several limitations. First, the cross-sectional nature of the study precludes definitive conclusions on causality. Second, the data are self-reported, which may be subject to recall and social desirability bias. Third, there may be a misrepresentation of those diagnosed with cancer since primary sources were not accessed as part of the NHIS survey. We speculate that this may have resulted in an underrepresentation of cancer survivors in our study.35,46 Finally, since the beginning of this study, the National Center for Health Statistics has released more recent data using a CAM supplement in the 2012 NHIS survey. The CAM supplement contains modifications in the types of CAMs queried between 2007 and 2012. This may limit analysis on the trends of CAM use for the obese and overweight cancer population. 72 Establishing the trajectory of its use will help inform health care providers and policy makers of appropriate health interventions and programs when addressing the health care needs of overweight and obese cancer survivors.
Notwithstanding these limitations, this study uses a large, diverse, and national data set to examine the prevalence and associations of CAM use among US cancer survivors, showing that respondents who are overweight, but not obese, were more likely to use a CAM compared with normal/underweight respondents. These findings form a basis for researchers and medical professionals working to address the needs of a cancer survivorship population with increasing adiposity who may seek nonconventional means to improve and maintain health. Further research is needed on the biological, medical, clinical, and psychological effects of CAM use among cancer survivors. Prospective studies analyzing CAM use among overweight and obese cancer survivors before and after primary cancer diagnosis could help inform health care providers on the specific types, patterns, and reasons of use within this population. In addition, with increased understanding of the biological effects and toxicity of various CAM therapies, health care professionals can more efficiently screen for CAM use and better guide patient practices to reduce morbidity and mortality in the cancer survivor population, as well as in those being actively treated with chemotherapy. Finally, additional queries assessing attitudes, beliefs, and uses of CAM in regard to overweight, obesity, and cancer will help inform health care providers on best practices when attempting to treat cancer survivors who use CAM therapies.
Footnotes
Acknowledgements
The authors would like thank Dr. Jeffrey D.White for editing the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this manuscript was provided by the National Cancer Institute (NCI). The NCI did not participate in the design, analysis, interpretation of data and writing of manuscript.