Use of Complementary and Alternative Medicine (CAM) in cancer patients: An Italian multicenter survey

Introduction Complementary and Alternative Medicine (CAM) include a wide range of products (herbs, vitamins, minerals, and probiotics) and medical practices, developed outside of the mainstream Western medicine. Patients with cancer are more likely to resort to CAM first or then in their disease history; the potential side effects as well as the costs of such practices are largely underestimated. Patients and method We conducted a descriptive survey in five Italian hospitals involving 468 patients with different malignancies. The survey consisted of a forty-two question questionnaire, patients were eligible if they were Italian-speaking and receiving an anticancer treatment at the time of the survey or had received an anticancer treatment no more than three years before participating in the survey. RESULTS Of our patients, 48.9% said they use or have recently used CAM. The univariate analysis showed that female gender, high education, receiving treatment in a highly specialized institute and receiving chemotherapy are associated with CAM use; at the multivariate analysis high education (Odds Ratio, (OR): 1.96 95% Confidence Interval, CI, 1.27-3.05) and receiving treatment in a specialized cancer center (OR: 2.75 95% CI, 1.53-4.94) were confirmed as risk factors for CAM use. Conclusion Roughly half of our patients receiving treatment for cancer use CAM. It is necessary that health professional explore the use of CAM with their cancer patients, educate them about potentially beneficial therapies in light of the limited available evidence of effectiveness, and work towards an integrated model of health-care provision.


INTRODUCTION
According to the National Center for Complementary and Integrative Health, USA, the Complementary and Alternative Medicine (CAM), include a wide range of products, such as herbs (also known as botanicals), vitamins, minerals, and probiotics, and medical practices, such as acupuncture or magneto-therapy, which are outside of the mainstream Western medicine. Such practices or substances are defined 'alternative' when they are used in place of conventional medicine and 'complementary' when they are used together with conventional medicine [1]. The interpretation provided by Ernst E et al. (UK), is much more complex and defines CAM as 'diagnosis, treatment and/ or prevention which complements mainstream medicine by contributing to a common whole, satisfying a demand not met by orthodoxy, or diversifying the conceptual framework of medicine' [2]. Both the definitions are correct and one does not exclude the other; indeed the use of such unconventional substances/treatments, which has always been very common among the eastern populations and in the developing countries, is acquiring more and more popularity also in the western world for multiple reasons. In the US, the use of CAM is constantly increasing with the largest practice being among the non-hispanic whites [3] and in the rural areas (rate of use up to 63%) compared with the cities [4,5]; in terms of geographic areas the highest rate of use was observed in the Mountain Regions and in New Mexico [6]. A systematic review published in 2013 by Posadzki P et al. highlighted the difficulties of identifying reliable data about the prevalence of CAM due to the poor methodological quality of most studies and the ambiguity of the definition of CAM; however, the authors concluded that the average one-year prevalence is 41.1% and the average lifetime prevalence is 58.1% [7]. A European survey conducted by Molassiotis A. et al., demonstrated that the use of CAM in the EU is about 35.9% of the total patient population. The use of CAM has increased steadily over the past 15 years or so; undoubtedly it has gained medical, economic and social importance [8]. Patients with cancer are more likely to resort to CAM, first or then in their disease history, for a wide range of reasons; firstly, the unfavorable outcome in a relevant percentage of cases leads patients to 'leave no stone unturned'; secondly, the heavy toxicities, often associated with the traditional antineoplastic therapies, induce them to look for something different from the prescribed therapy or more simply for substances presumed to reduce the side effects from such therapies [9]. Nevertheless the literature about CAM prevalence in cancer patients is not particularly rich, especially if we consider only the European papers, and the prevalence is probably underestimated. Many patients do not declare that they engage in this practice, on the one hand because they undervalue the relevance of the products they take, considering them 'natural', unable to interact with the conventional drugs and devoid of side effects, and, on the other hand, because they are somehow reluctant to admit the use of an unconventional treatment, worrying that such behavior may be interpreted as reflecting a loss of trust in their oncologist and the treatment he/she has prescribed. Furthermore most clinicians are unfamiliar with these kinds of treatments [10] and hence do not pay enough attention to this aspect of the anamnesis at the time of the visit; usually they do not explicitly ask about this topic, as they do for all other health matters such as comorbidity or conventional drugs [9,11]. The available studies report that the prevalence of CAM use among cancer patients is in the range of 12.5-73% [12][13][14][15]. This enormous variability is, at least partially, justified by the inconsistent definition of CAM, with some authors including only herbal medications, while some others considering also including dietary supplements and unconventional medical practices (massages, acupuncture). The aim of this study was to assess the use of CAM across a number of Italian cancer hospitals, using the same measurement tool and the same definition of CAM and trace the "identikit" of typical cancer patient "CAM-users".

RESULTS
The five hospitals that participated in the study provided a total of 468 subjects that were eligible for clinical considerations and statistical analysis. Table 1 summarizes the baseline characteristics of the population.

Socio-demographic and clinical characteristics of the sample
With respect to demographics, the cohort was well balanced between males and females (48.9% vs. 51.1%), the median age at the time of diagnosis of cancer was 57 years old, while the age at the time of the survey was between 40 and 70 years of age for more than half of the patients. The educational level was low (primary or secondary school) for 38%, high (high school, degree or higher) for 62%; the Italian region of origin was the South or the Islands for  Table 2 shows the answers about the general feeling with respect to conventional medicine and CAM: More than 90% thought they had a good/excellent awareness about the disease and the therapies received. Almost all the interviewed declared that they have trust in conventional medicine and oncological treatments, 94.2% and 95% respectively, though roughly 50% expected to be treated rather than cured and there was a relevant percentage, 32%, admitting to be concerned about financial speculation around anticancer drugs. Of the population surveyed, 75.6% said they knew what CAM was, but only 27.6% were aware of the difference between alternative and complementary medicine. Among the patients aware of what CAM was (354), most had heard about it from the media, friends, or other patients, and only the 5.9% from a doctor (Figure 1). Eight patients (3.5%) reported sideeffects from the CAM therapy assumption. Most seemed to be transient side-effects and they were all related to ingesting herbs or minerals. These side effects included stomach aches (two with aloe, one with vitamin C and two with high consumption of green tea) gastric upset and nausea from using unspecified herbs (two patients), and diarrhoea with aloe (one patient). The most frequent answer to the question 'what CAM is?' was 'medicines to reduce AC toxicities' followed by 'supplements' and 'overall', which meant 'all the above'. Of the subjects, 48.9% (229) answered 'yes' to the direct question 'have you ever used any CAM?'. Table 3 shows the distribution of CAM users by type, prescriber, cost, and satisfaction with the product. Interestingly, the use of CAM mainly derived from an auto-prescription (67%), most patients were not aware of the potential side effects, the substance was bought by the patient himself at an average cost of between 100 and 300 euros, the duration of the treatment was less than one year in 83.8% of the cases. Evaluating the variables associated with CAM use through the univariate analysis, we noticed that it was more popular among females compared with males (56.9 and 44.5%, p= 0.01), and older patients, ≥ 70 years of age, compared with younger, 40-70, (54.6 and 42.4%, p< 0.001). Also, a medium-high educational level seemed more frequently associated with CAM use compared to a low one (74.2 vs. 25.8%, p < 0.001). Looking at the primary disease, it appeared that patients with a more aggressive malignancy, in particular lung cancer, use CAM more frequently compared to the others; furthermore, CAM was resulted more popular among subjects receiving AC alone [139 patients (60.7%)] rather than other treatment modalities. Other characteristics associated with CAM use include anxiety disorders, lack of chronic pain, and metastatic disease (Table 4). Of note is that there was no difference in terms of CAM use based on trust in conventional medicine and oncological treatments and based on the personal awareness about the disease ( Table  5). Considering the 228 patients who did use a CAM: only the 3.9% declared that they had experienced side effects from such therapies [9 patients had diarrhea (G2) during concomitant use of AC and aloe]; 63.1%, on the basis of his/her personal experience do trust these medications; 82% think that the favorable response achieved is due to both the AC and the CAM used; 86.8% admit to having spent between 100 and 300 euros monthly on the chosen CAM. Finally, the multivariate analysis indicated that the only factors which were confirmed to be significantly associated with the use of CAM were a high educational level (Odds Ratio, OR: 1.96 95% Confidence Interval, CI, 1.27-3.05) and receiving treatment in a specialized cancer center rather than in a peripheral hospital/clinic (OR: 2.75 95% CI, 1.53-4.94) ( Table 6).

DISCUSSION
Current international research suggests that 29-91% of cancer patients seek CAM in addition to anticancer therapies [16][17][18][19][20][21][22][23]. Our study demonstrates that CAM use among cancer patients in Italy is fairly widespread, with nearly half of those interviewed (48.9%) reporting an ongoing or recent use of CAM. Patients suffering from cancer are more likely to use CAM at some point in their disease history however, many available publications show that unconventional medicine is routinely used by many other categories of patients: the average lifetime prevalence for UK patients reported by Posadzky P et al. [7] is 51.8%; another work published in 2015 [16] analyzing the data from 263 questionnaires completed by New Mexican patients attending outpatient clinics for different diseases (heart failure, HIV, hepatitis, etc.), found a 56% incidence of CAM use with a significantly higher rate among females with a high educational level, who declared that they use CAM for general wellness and disease prevention. The study by Zhang Y et al. evaluating a large population of adults in the US (34.525) also concluded that CAM use was more popular among females with a higher educational level; in that work, the main problems leading to CAM were back pain, arthralgia, and migraine [24]. Looking at the literature about the www.impactjournals.com/oncotarget  [15] reported the data of a survey carried out among 699 cancer patients who were receiving AC, 33% declared that they use complementary therapies. The characteristics associated with CAM use were: being female, having breast cancer and high level of education; the most frequent practices were exercise, herbal therapy, and spiritual healing; there were no differences with respect to reported depressive symptomatology or spirituality. Nightingale G et al. [25] explored the prevalence of CAM and factors influencing CAM use in a secondary analysis of 248 senior adults with cancer (mean age 79.9) who received an initial comprehensive geriatric oncology assessment; 234 subjects were included in the final analysis. In that study, the prevalence of CAM use was much lower compared to our finding, 26.5%; polypharmacy, vision impairment and urologic comorbidities were found associated with CAM use. Such a difference in terms of prevalence of CAM use is in part justified by the direct involvement of a clinical pharmacist who went through the complete list of medications of each patient and actively helped to define what CAM was; also, vitamins were excluded. In regards to patient demographics, a higher rate of CAM use was observed in women (19% vs. 8%) though the population was not well balanced (64% females, 36% males); the educational level was not mentioned. An interesting study involving 294 prostate cancer patients reported a 25% rate of CAM use, mainly vitamins, low-fat diets, lycopene and green tea. The multivariate analyses revealed no differences in mental health scores between users and non-users; users     (24) has corroborated some findings already observed in previous papers. Firstly, the higher prevalence of CAM use among females and, in general, in the subset of patients with a high educational level. Probably a high education level allows easier access to the media, internet and information about medicine. An aspect that has emerged in our dataset is the higher rate of CAM use in patients receiving their treatment in highly-specialized cancer centers rather than in peripheral hospitals/clinics. In the univariate analysis we also found that patients with anxiety disorders were more likely to use CAM though this association was not confirmed by the multivariate analysis; however, this lack of statistical significance may be due to the low number of patients suffering from anxiety present in our study (only 15.4%, 72/368). Our data are also consistent with the available literature regarding the absence of a relationship between CAM use and lack of trust in conventional medicine, other authors, in fact, have already observed that CAM are often used by patients who are fully satisfied with conventional health care and look for adjuvants rather than for substitutes to traditional practices [29,30]. As observed in previous studies [31][32][33][34] supplements and herbal medications are the main CAM used by cancer patients and they are more likely to be used as adjuvants, to reduce AC side effects, rather than as direct anticancer treatments. In contrast with prior observations, 85% of our interviewed subjects said that they had informed their oncologist about their CAM use, though such data should be carefully interpreted due to the administration of our questionnaire by physicians.
The costs sustained to buy CAM were between 100 and 300 euros per patient, per month for most patients (62.9%) and were paid for by the patients themselves in 98.6% of cases. Such finding is consistent with the estimated cost per year of supplements in Italy which accounts up to 2.5 billion of euros [35]. The most CAM used was aloe (75% of patients). In 67.7% of the cases, the CAM prescription was a self-prescription and only in 11.4% of the cases had a doctor prescribed the CAM. This aspect is very dangerous considering the risk of toxicities and potential interaction between CAM and standard/conventional medicine; regarding this aspect, 197 patients (86%) did not know about the risk of potential side effects. Finally, the high rate of satisfaction from CAM and the fact that approximately all the users (87%) recommended their use, indeed, deserves attention; maybe healthcare professionals in general and oncologists in particular should pay more attention to this aspect of the anamnesis at the time of the visit and check at each control, especially for patients on AC, if new conventional or unconventional drugs have been added to the patient's personal therapy. The routine anamnesis collection should be completed by explicit questions about the use of herbals, supplements, and any other preparations. In fact some CAM products can interfere with some AC through effects on the metabolic pathway and oncologists must be careful to avoid drugherbs interaction when they plan a treatment with herbal medications as supplement to AC ( Table 7). The use of CAM in oncologic patients is a growing issue that requires more attention by the scientific world and physicians. To know the causes that induce oncologic patients to inappropriately use CAM is of fundamental importance. To retrieve the distortion of the information sources and to improve the communication modalities are crucial steps to preserve the health of patients, trough the strengthening of the therapeutics alliance and of the adhesion to treatment with reliable efficacy. We firmly believe that to promote knowledge about CAM between patients and physicians with the aim of alerting them to potential toxicities is very important. The need to increase the evidence base of CAM therapies, using methodologies that are appropriate and sensitive to CAM cannot be overemphasized. This is also important from an economic point of view, as its use is a multibillion Euro business,  [36]. In the EU it is the second biggest growth industry [37]. Finally, appropriate legislation and regulation of CAM therapies in the EU is also necessary.

Patients setting
A cross-sectional descriptive survey design was used to collect data trough a questionnaire about CAM therapies. Nine hospitals with oncology and/or radiotherapy divisions were approached for possible collaboration in this study. Of the nine hospitals, five agreed to participate. A key person from each hospital was selected, based on interest in and/ or experience with CAM. Data collected from each hospital were returned to one of the investigators and then the data were coded for analysis. This survey was approved by the ethics committees and carried out between September 2014 and May 2015. Both metastatic and non-metastatic cancer patients were included from cancer hospitals, oncology units of general hospitals, day units, and radiotherapy units. All participating patients received information about the study. Patients were considered eligible if they met the following inclusion criteria, they were Italian-speaking, adult patients of either gender with a diagnosis of cancer; aware of their diagnosis; were receiving an anticancer treatment [antiblastic chemotherapy (AC), endocrine therapy, radiotherapy and surgery] at the time of the survey or had received an anticancer treatment no more than three years before participating in the survey; able to understand the questions; free from any condition that would make completing the questionnaire inappropriate or overburdening for the patients; and they were willing to participate in the study.

Procedures
The questionnaire was administered to the patients after they received information about the study, agreed to participate, and signed the consent form. Patients completed the questionnaire with the support of a physician. Participation was voluntary and did not interfere with medical treatment.

The questionnaire
The questionnaire used was based on the one developed by Molassiotis A et al. [26]. However, the questionnaire was modified for the purpose of the present study and some new items were added (for example: awareness about the disease and treatment received, trust in conventional medicine and oncological treatments, expectations from oncological treatments, awareness about CAM, knowledge about the differences between alternative and complementary medicine, source of knowledge about CAM and role of CAM in oncological treatment), and some others were modified to reflect Italian culture (for example: geographic area of origin) there were 42 items in total. These included demographic data (age, gender, occupation, and education), clinical data (site of primary cancer, standard treatments received previously and current standard treatment) and questions about CAM use, in particular: knowledge about what CAM is and whether it has ever been used. The questionnaire about CAM was completed only by the patients who declare that they have had, or had at the time of the survey, a personal experience with CAM and included eighteen questions about different aspects of such therapies concerning, e.g. efficacy, cost, side effects, satisfaction rating. On the basis of the available literature we decided to exclude from the analysis medical therapies routinely prescribed, such as iron, vitamin D, and calcium supplements, which are often used as support therapy.

Data analysis
In the descriptive analysis, the median, range, and relative frequencies were used. Analyses of frequencies and cross tables with χ 2 tests were done using IBM SPSS Statistics 20 [38]. In this study, we focused on those questions that were the same or similar in both group in order to compare patients and professionals. Only completely filled-out questionnaires were analyzed.