Use of Mind-Body Medicine and Improved Self-Rated Health: Results from a National Survey

Transcription

1 ARTICLE Use of Mind-Body Medicine and Improved Self-Rated Health: Results from a National Survey Long T. Nguyen PhD MPH a, Roger B Davis ScD b, Ted J Kaptchuk c, Russell S Phillips MD d a Research Fellow b Associate Professor of Medicine c Associate Professor of Medicine d Professor of Medicine Division of General Medicine and Primary Care, Beth Israel Medical Center, Harvard Medical School. Boston, Massachusetts, USA Abstract Background: Among the 20% of US adults who use mind-body (MB) medicine, half also use other types of complementary and alternative medicine (CAM). Our prior work showed that CAM use is associated with better selfreported health. However, little is known about the association between the use of MB therapies and perceptions of health. Method: Objective: To determine the association between the use of mind-body therapies and self-rated health. Design: We analyzed cross-sectional data from the 2007 US National Health Interview Survey (n=23,393). We partitioned the CAM-user respondents into 3 mutually exclusive groups: 1. MB-only, 2. Both MB and Non-MB-CAM (MB-Plus), and 3. Non-MB-CAM only. We used multivariable logistic regression to model the likelihoods of respondents reporting Excellent health and reporting Better health than the prior year. We controlled for socio-demographic, clinical and behavioral confounders and evaluated the models. Results: 8% of U.S. adults used MB-Only, 18% used non-mb-cam, and 10% used both. Compared to users of Non-MB- CAM, users of MB-Only (adjusted-or=1.13,95%ci=[0.96,1.33]) and MB-Plus (adjusted-or=1.09,95%ci=[0.94,1.26]) were not significantly more likely to report Excellent health. However, compared to users of Non-MB-CAM, users of MB- Plus were more likely (adjusted-or=1.48,95%ci=[1.28,1.71]) to report Better health over the prior year while users of MB alone were not. Conclusion: Respondents who used MB in addition to other CAM therapies were significantly more likely to report improvement in self-rated health over the prior year. Large prospective trials are needed to establish whether MB therapies, used as an adjunctive treatment with other CAM, will lead to improvement in health. Keywords Adjunctive therapy, complementary and alternative medicine, health improvement, mind-body medicine, NHIS survey, selfrated health status Correspondence Address Dr. Long T Nguyen, PhD, MPH, Division for Research and Education in Complementary and Integrative Medical Therapies, Harvard Medical School, 401 Park Drive, Suite 22-A West, Boston, Massachusetts, 02215, USA. Long@DrNguyen.org. Accepted for publication: 13 April 2011 Introduction The interaction of the mind and the body and the impact of the mind on healing is recognized in many healing traditions including Chinese and Ayurvedic medicine in the East and also in the Hippocratic writings (ca. 400BC) in the West [1,2]. Biomedicine has only recently acknowledged the deleterious effect of mental stress on health and made significant discoveries on the interaction between the mind and the body [3-5]. Mind-Body therapies comprise a significant component of what is called Complementary and Alternative Medicine (CAM), a collection of therapies and diagnostic disciplines that are not taught in conventional medicine. The demarcation between CAM and 1 The International Journal of Person Centered Medicine

2 Nguyen, Davis, Kaptchuk and Phillips Mind-body medicine and self-rated health conventional medicine, however, is in flux and changes through time and between countries. Despite the absence of conclusive evidence of a causal link between CAM therapies and health, CAM is used extensively to complement, or as an alternative to, conventional treatments [6] and has been reported to be associated with improved self-rated health [7]. CAM has diverse origins. Most CAM such as Ayurveda, acupuncture, yoga, taichi, and qigong, originated in the East, while others such as homeopathy, naturopathy, osteopathy, chiropractic and autogenic relaxation and hypnosis originated from the West. All CAM therapies, however, share a fundamental belief that the body can heal itself and healing often involves restoring the balance in the body, mind, and spirit [2, 8]. CAM therapies are grouped into five broad categories [6]: (i) alternative medical systems [1, 8]; (ii) energy healing [8]; (iii) manipulative and body-based therapies [8]; (iv) biologically-based therapies [8] and (v) mind-body therapies [8]. The CAM categories of alternative medical systems, energy healing, manipulative and body-based therapies and biologically-based therapies involve hands- in on treatments given by a healer, or substances found nature such as herbs, foods, and vitamins, taken by patients to effect healing. Mind-body therapies (MB), on the other hand, rely mainly on the user s focus of attention, without the use of an external agent such as a healer or a substance, to evoke a certain psychophysiological state of relaxation or some specific outcome to affect healing. MB focuses on the interactions among the brain, mind, behavior and body and ways in whichh emotional, social and mental factors can affect health [8, 9]. MB have been reported to be effective in treating cardiovascular diseases and pain [9], improving mood, quality of life and coping, as well as reducing the side-effects of cancer treatment. MB can be used as effective adjuncts to conventional medicine [9]. MB is relatively low-risk, as part of his or her own schedule. MB is ncreasingly popular and, with 1 out of 5 adults using MB, low cost and can be practiced by the patient it is among the most commonly used CAM in the United States [6]. However, four of ten CAM users used more than one category of CA [10]. Self-ratings of health are among the most frequently assessed measurement in health research [11]. Poor self- greater use of resources [11,12] and subsequent mortality, independent of objective health status [11]. Excellent self- rated health is associated with more functional limitations, rated health and health improvement have been associated with CAM use [7]. Little is known about how the use of MB therapies, Non-MB CAM therapies and MB and other CAM therapies together affects self-rated health and its change over time. In this context, we evaluated the relation between the use of these subcategories of CAM therapies and self-ratings of health and improvement of health among the respondents to a national survey. 2 Method Data Source We used data from the 2007 National Health Interview Survey (NHIS). The NHIS is a computer-assisted, face-to- representative estimates. It was conducted in English and Spanish by the National Center for Health Statistics, in the face annual survey designed to provide national households of the non-institutionalized, civilian population of the United States [13]. The survey asked information on socio-demographic characteristics, health status, insurance status and health care access and utilization for each family member. One adult and one child were then randomly selected for further questions about common medical conditions and health care utilization. In 2007, the selected adult and child were also asked about their past-12-month categories (Figure 1) [6,8,13]. The mind-body category included deep breathing exercises, meditation, yoga/taichi/qigong, progressive relaxation, guided imagery, hypnosis, and biofeedback. The biologically- based category included non-vitamin, non-mineral, natural products; diet-based therapies; and chelation therapy. The manipulative-and-body-based category included chiropractic or osteopathic manipulation, massage, and movement therapies. The alternative medical system category included homeopathic treatment, acupuncture, traditional healers, naturopathy, and ayurveda. The energy healing therapies included reiki and therapeutic touch [6,8]. Following common practice in CAM research, we use of 36 CAM therapies in 5 broad included all 5 CAM categories, but excluded use of prayer, vitamins, and minerals from our analysis [14]. Figure 1: Prevalence of CAM categories Collected data Interviews weree completed in 29,266 households with 75,764 persons. From these households, 23,393 adults responded to the CAM survey (final response rate = 67.8%). The International Journal of Person Centered Medicine Volume 1 Issue 2 pp x-xx

3 We focused our analysis on the type of CAM therapies that the respondents reportedly used and their answers to the demographic, clinical, behavioral, and health status questions. Demographic data included age, gender, race/ethnicity, birth region, marital status, income, education, residence region, health insurance and usual source of care. Clinical data consisted of conditions such as asthma, emphysema, heart attack, stroke, ulcer, liver condition, arthritis, diabetes, weak/failing kidneys, cancer, functional and cognitive impairments and mental health. Behavioral data included body mass index, amount and frequency of alcohol and cigarette use and type and frequency of physical activity. To obtain data on CAM use, respondents were asked a series of questions such as: During the past 12 months, did you see a practitioner for (specific therapy)? For disability, respondents were asked: By yourself, and without using any special equipment, how difficult is it for you to do/perform (activities)? ; What condition or health problem causes you to have difficulty with (these activities)?. For health conditions, respondents were asked: Have you ever been told by a doctor or other health professional that you had (specific condition)? ; and During the past 12 months have you had (specific condition)? [8,13,15]. Our outcomes of interest include a global assessment of health status and whether the respondent s health status had improved over the prior year. The specific questions used in NHIS to obtain the outcome information were: Would you say your health in general is excellent, very good, good, fair, or poor? and Compared with 12 months ago, would you say your health is better, worse, or about the same?. These questions were included previously in the MOS-SF 36, a validated and internationally used instrument [16]. Analysis Primary Independent Variable Type of CAM Used We partitioned the respondents who reported using CAM therapies in the previous twelve months into three mutually exclusive groups based on the type of CAM therapy used. The Mind-Body group (MB-Only) consisted of respondents who practiced only MB. The Non-Mind-Body CAM group (Non-MB CAM) consisted of those who practiced any of the CAM types other than MB (excluding prayer). The mixed group (MB-Plus) consisted of those who practiced at least one MB and at least one other CAM therapy. Covariates - Comorbidity Index and Other Correlates The Charlson Comorbidity Index (CCI) is a measure that has been used in health services research to predict mortality and resource use based on the patient s clinical conditions [17]. To characterize the clinical condition of the respondents, we used the modified CCI (mcci) that was used and described in previous NHIS studies. To assess mental health conditions within the last 30 days, as in our previous study [7], we used the validated Kessler-6 score (K6), which ranged from 0 to 24, based on six mental health questions. For both CCI and K6, higher scores indicate more comorbidities. To characterize health habits, we included data on body mass index, smoking status, alcohol intake and physical activity level. For physical activity assessment, we used previously validated criteria to categorize respondents as having high, medium, or low activity level [7]. Logistic Regression Modeling We developed two multivariable logistic regression models of the binary dependent variables Excellent health and Better health than in the prior year to assess the association of types of therapies used by respondents with these outcomes. We included, as independent variables, the primary factor of interest, classified as MB-only, Non MB CAM, and MB Plus use and potential confounders, including the aforementioned socio-demographic, clinical, and behavioral variables. In a sensitivity analysis, we compared those who reported excellent, very good, or good health to all others. We also explored models that had MB Only, Non-MB CAM and MB-Plus use as dependent variables and health improvement among the independent variables. We report the summary results of these analyses. We developed our models incrementally. We used socio-demographic covariates previously reported as significantly correlated with use of CAM therapies as the first set of explanatory variables and then added clinical and behavioral variables to see how these health-related individual characteristic would affect the model. We retained the following socio-demographic factors in the model (age, sex, education, race/ethnic, birth region, and residence region) as well as covariates with p-values We assessed covariates for collinearity and eliminated those with tolerance computed index > 30. We report the Wald p-values, adjusted-odd ratios (adjusted- OR) and 95% confidence intervals of the covariates in each model using the Taylor linearization method to estimate variances. To characterize the discrimination, we report the c-statistic for each model. We evaluate the calibration of the models using the Hosmer-Lemeshow goodness-of-fit test and report their p-values. To accommodate for the complex survey design, we used SAS-callable SUDAAN 3 The International Journal of Person Centered Medicine

4 Table 1: Characteristics of CAM Therapy Groups (% of each group total) Estimated Percent of US Adult Population MB Only Therapies n= % Other CAM Therapies n=4149, 17.74% MB Plus Other CAM Therapies n=2442, 10.44% Health Status p = Excellent Very good Good Fair Poor Health Compared to 12 Months Ago p< Better Sex p< Female Age (years) Mean (SE) p< (0.47) (0.34) (0.40) Ethnicity p< White Black /African American Others Region of Birth p< United States Central & South America Elsewhere Marital Status p< Married Single/Divorced/Widow Never Married Region of Residence p< Northeast Midwest South West Education Level p< <HS Graduate HS Graduate Some College College Graduate Health Insurance p = Private Public Uninsured Usual Place of Care p = Yes NHIS-Charlson Comorbidity Index (CCI) p= or More Told Have Hypertension p = Yes Kessler-6 Mental Health Index p< or More (Emotional Distress) Body Weight Status (BMI) Mean (SE) p = (0.18) (0.13) (0.14) Activity Level p = Low Medium High Alcohol Status p< Abstainer Former/Current, Light Current, Moderate Current, Heavy The International Journal of Person Centered Medicine

5 Table 1: continued Estimated Percent of US Adult Population MB Only Therapies n= % Other CAM Therapies n=4149, 17.74% MB Plus Other CAM Therapies n=2442, 10.44% Smoking Status p< Never Former Current, Sometimes Current, Everyday Visits To A Health Care Provider, Past 12 Months Mean (SE) p< (0.07) 2.94 (0.04) 3.25 (0.06) Times In Emergency Room, Past 12 Months Mean (SE) p = (0.02) 0.31 (0.01) 0.37 (0.02) Days Spent In Bed, Past 12 Months Mean (SE) p = (0.31) 5.97 (0.64) 6.11 (0.58) v10.0 (RTI) analytic software and SAS statistical software (SAS institute, Cary, NC). Results As characterized in Table 1, approximately 1 of 10 United States adults used only MB therapy (MB-Only), 2 of 10 used other, non-mb, CAM therapies in the prior year (Non-MB CAM). Half of those who used MB (52%, 1 in 10) also used other Non-MB CAM therapies (MB-Plus). About 30% of each group rated their health as Excellent. Compared to the prior year, 27.5% of the MB-Plus group rated their health as Better, while only 22.7% of the MB- Only and 20% of the Non-MB CAM groups did so. MB users were more likely than Non-MB users to be female. Those who used MB alone were less likely to be born in the US. More Non-MB CAM users were married; more MB Plus users lived in the West or were college graduates, or were privately insured. Even though there were no significant differences in clinical conditions as included in the mcci among the three groups, the respondents in the MB-Plus group made more office visits in the prior year than those in the Non-MB groups while those in the MB-Only group reported the least number of office visits. However, those in MB and MB-Plus groups reported more bed days in the prior year than those in Non- MB group. Compared to the Non-MB CAM users, MB users and MB Plus users had higher K6, suggesting more mental conditions. The average respondents in all 3 groups were overweight. Among the 3 groups, the MB Plus group had the least alcohol abstainers and highest former/current light drinkers. The MB group had the most everyday smokers. Table 2 shows the adjusted-odds ratios based on types of CAM used for the Excellent health model. Compared to the Non-MB CAM users, MB-Only users and MB-Plus users were not significantly different in reporting their health as Excellent. Adjusted-for covariates included in this model are listed in the table. Table 3 shows the adjusted-odds ratios for the Better health than the prior year model. Compared to the Non- MB-CAM users, the MB-Only users were not significantly Table 2: Health is Excellent vs. Independent Variables Independent Variables Odds Ratio Lower 95% Limit OR Upper 95% Limit OR MB-Only (p=0.1554) MB Plus (p=0.2348) Non-MB CAM (reference) Adjusted-for Covariates Significant (p < 0.04): Body- Mass Index, modified Charlson Comorbidity Index, Mental Health, Education, Hypertension, Cigarette Use, Age, Region of Residence, Alcohol use. Non Significant (p> 0.05): Health Insurance, Ethnicity, Sex, Region of Birth, Marital Status, Activity Level Table 3: Health is Better Than a Year Ago vs. Independent Variables Independent Variables Odds Ratio Lower 95% Limit OR Upper 95% Limit OR MB Only (p=0.1865) MB Plus (p<0.0001) Non-MB CAM (reference) Adjusted-for Covariates Significant (p < 0.05): Age, modified Charlson Comorbidity Index, Hypertension, Health status, Alcohol use,, Activity Level, Sex Non significant (p>0.05): Body-Mass Index, Region of Residence, Ethnicity, Marital Status, Mental Health,, Region of Birth, Cigarette Use, Health Insurance, Education 5 The International Journal of Person Centered Medicine

6 Nguyen, Davis, Kaptchuk and Phillips Mind-body medicine and self-rated health more likely to report improved health. However, MB-Plus users were significantly more likely than the Non-MB- CAM users to report their health as Better than the prior year. Adjusted-for covariates included in this model are listed in the table. For our sensitivity analysis, we repeated the modeling process using good, very good or excellent health as dependent variables. We found that the results were similar. There were no substantial differences in the likelihood of reporting good health or better among MB- Only (adjusted OR= 1.18, 95%CI=[0.90, 1.54], p=0.2297), MB Plus (adjusted OR= 1.12, 95%CI=[0.89, 1.42], p=0.3375), and Non-MB CAM users (as reference). Both models demonstrated good calibration based on the Hosmer-Lemeshow test (Wald p= for the Excellent health model and Wald p= for the Better health than prior year model). The c-statistics were for the excellent health model and for the better health model. Discussion We found that approximately 1 in 10 respondents used a mind-body therapy alone, 1 in 10 used both mind-body therapy and another type of CAM, and 2 in 10 used other types of CAM exclusive of mind-body therapy. We found no difference in health problems between MB users and Non-MB CAM users, as indicated by the clinical problems included in the modified Charlson Comorbidity Index (mcci). However, MB users in general have more mental health issues as indicated by higher Kessler Score. We found no difference in likelihood of reporting excellent health or good or better health among the users of the three groups. In the model for improved health over the prior year, we found no difference between users who used MB-Only and users who used Non MB CAM; however, we found users in MB-Plus group (OR=1.44, 95%CI=[1.24, 1.67]) were more likely than users in Non- MB CAM to report that their health were better than in the prior year. While comorbidities were not significantly different among the three groups (based on the mcci), they were significantly associated with both outcomes, We found that MB-Plus respondents reported having higher Kessler 6 scores than the Non-MB CAM, but were also more likely to report that their health had improved and been better than in the prior year. This seeming contradiction, similar to our previous observation [7], may stem from the fact that the reported health status reflects the assessment made at the time of the interview while the answers to the mental health questions referred to what happened in the prior period. Therefore, these seemingly contradictory findings may in fact be plausible. This finding of improved health reported by those who used MB in addition to a Non-MB CAM seems to be similar to the adjunctive effect between MB therapies and conventional treatments previously reported [9]. One possible explanation is that MB practices, through changes in hormones, neurotransmitters/neuropeptides and cytokines [19], may potentiate the effects of other modalities and facilitate healing. Similar to our recent report [7], we used the novel approach of modeling health status using logistic regression. In previous studies conducted by others on the relation between CAM use and self-rated health, health status or change in health status was frequently treated as an independent variable to model CAM use (CAM use was used as the response variable). We, however, took the opposite approach and treated health status and change in health status as dependent variables in our logistic regression models. The main factor of interest in these regression models was the type of CAM : MB-only, Non- MB, or MB-Plus that the respondents used. We also included other covariates to characterize the demographic, socio-economic and clinical status of the respondents. In these models, we model the health status and health improvement of the respondents, with the type of CAM used as the independent variable and control for other respondent characteristics as potential confounders. In the model that predicts health improvement, we also control for health status as an additional potential confounder. Most previous studies on the relationship between health status and CAM use have examined CAM use as a group, instead of categorizing CAM use by subcategory or modality. As it is possible that MB therapies affect the body distinctively differently from other CAM modalities [19], we separated MB from non-mb modalities, creating 3 mutually exclusive groups of users: MB only users, Non- MB users, and MB-Plus users. Even though CAM effectiveness research, in general, has been inconclusive, MB studies have tended to have more positive results and shown evidence of benefits for blood pressure, preventing falls, alleviating low back pain and irritable bowel syndrome [7]. In addition, MB was also reported as an effective adjunct treatment to conventional medicine for several clinical conditions [9]. On the other hand, participation in healing rituals can confer subjective perceptions of benefit independent of any changes in pathophysiology or symptomatology [20]. Our study has limitations that are often associated with survey research. The measure of MB and other CAM use prevalence is affected by respondents recall bias and is limited to the therapies included in the survey. The absence of data on quantity, duration and timing of use limits our ability to discern the effect of repeated use from that of single use and to ascertain any causal impact of treatment and dose response effects on health status. The self-reporting nature of clinical data may not meet standard clinical definitions. The survey was administered only in English and Spanish and may have under-represented smaller and more recent immigrant populations who do not speak these languages. Finally, our decision to categorize health status as excellent compared to other responses may have affected our results. However, in a sensitivity analysis, we repeated the analysis grouping Excellent, 6 The International Journal of Person Centered Medicine

7 Very Good, and Good self-rated health together. Our conclusions were essentially unchanged. Although MB use is increasingly popular, research on its effectiveness, similar to CAM research in general, is still in its early stages. Methodological issues, such as insufficient statistical power, lack of randomization, inadequate controls and poor specificity of eligibility criteria and interventions, have been frequently reported and hindered the interpretation and generalizations of results [21-23]. Our findings, however, suggest that, on a population basis, MB therapies, used as an adjunctive to Non-MB CAM, may have implications for better health improvement over time than using Non-MB CAM alone. Large-scale randomized controlled studies are needed to determine if there is a causal relationship between the use of mind-body medicine as an adjunct treatment to other non-mind-body CAM treatments and health improvement. Acknowledgement Funding/Support: Dr. Nguyen was supported by an Institutional National Research Service Award (T32AT00051) from National Institutes of Health (NIH). Prof. Ted Kaptchuk is supported by a Mid-Career Investigator Award from the National Center for Complementary and Alternative Medicine (NCCAM), NIH (K24 AT004095). Drs. Roger Davis and Russell Phillips are supported by a Mid-Career Investigator Award from the NCCAM, NIH (K24 AT000589). Role of the Sponsor: The funding organizations had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript. Disclaimer: The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of NCAAM or the NIH. References [1]. NCCAM. (2004). Whole Medical System: An Overview. NCCAM Backgrounder [2]. Kaptchuk, T.J. (2000). The web that has no weaver : understanding Chinese medicine. Rev. ed. Chicago, Ill: Contemporary Books [3]. Cannon, W. (1932). The Wisdom of the Body. New York, NY: Norton [4]. Ader, R., Cohen, N. & Felten, D. (1995). Psychoneuroimmunology: interactions between the nervous system and the immune system. Lancet 345 (8942), [5]. Kerr, C.E., Shaw, J.R., Wasserman, R.H., Chen, V.W., Kanojia, A., Bayer T. et al. (2008). Tactile acuity in experienced Tai Chi practitioners: evidence for use dependent plasticity as an effect of sensory-attentional training. Experimental Brain Research 188 (2), [6]. Barnes, P.M., Bloom, B. & Nahin, R.L. (2007). Complementary and alternative medicine use among adults and children: United States, National Health Statistics Report 10 (12), [7]. Nguyen, L.T., Davis, R.B., Kaptchuk, T.J. & Phillips, R.S. (2011). Use of Complementary and Alternative Medicine and Self-rated Health Status: Results from a National Survey. Journal of General Internal Medicine 26 (4), 399. [8]. NCCAM. What is CAM? (2007). NCCAM Backgrounder; Feb. [9]. Astin, J.A., Shapiro, S.L., Eisenberg, D.M. & Forys, K.L. (2003). Mind-body medicine: state of the science, implications for practice. Journal of the American Board of Family Practice 16 (2), [10]. Tindle, H.A., Davis, R.B., Phillips, R.S. & Eisenberg, D.M. (2005) Trends in use of complementary and alternative medicine by US adults: Alternative Therapies in Health and Medicine 11 (1), [11]. DeSalvo, K.B., Bloser, N., Reynolds, K., He, J. & Muntner, P. (2006). Mortality prediction with a single general self-rated health question. A meta-analysis. Journal of General Internal Medicine 21 (3), [12]. Fylkesnes, K. (1993). Determinants of health care utilization--visits and referrals. Scandinavian Journal of Social Medicine 21 (1), [13]. CDC-NCHS National Health Interview Survey (NHIS) Survey Description Document. June 2008 [cited 2010 Oct 4]; Available from: ftp.cdc.gov/pub/health_statistics/nchs/dataset_documentation /NHIS/2007/srvydesc.pdf [14]. Kaptchuk, T.J. & Eisenberg, D.M. (2001). Varieties of healing. 2: a taxonomy of unconventional healing practices. Annals of Internal Medicine 135 (3), [15]. CDC-NCHS National Health Interview Survey Questionnaire - Sample Adult & Adult CAM. [Electronic - pdf file] 2007 [cited 2010 Oct 4]; 2007 National Health Interview Survey]. Available from: ftp.cdc.gov/pub/health_statistics/nchs/survey_questionnaires/ NHIS/2007/English/qalthealt.pdf [16]. Ware, J.E. & Kosinski, M. (2001). Interpreting SF-36 summary health measures: a response. Quality of Life Research 10 (5), ; discussion [17]. Chaudhry, S., Jin, L. & Meltzer, D. (2005). Use of a selfreport-generated Charlson Comorbidity Index for predicting mortality. Medical Care 43 (6), [18]. Birdee, G.S., Wayne, P.M., Davis, R.B., Phillips, R.S. & Yeh, G.Y. (2009). T'ai chi and qigong for health: patterns of use in the United States. Journal of Alternative and Complementary Medicine 15 (9), [19]. Vitetta, L., Anton, B., Cortizo, F. & Sali, A. (2005) Mindbody medicine: stress and its impact on overall health and longevity. Annals of the New York Academy of Sciences 1057, [20]. Kaptchuk, T.J. (2002). The placebo effect in alternative medicine: can the performance of a healing ritual have clinical significance? Annals of Internal Medicine 136 (11), [21]. Rains, J.C. & Penzien, D.B. (2005). Behavioral research and the double-blind placebo-controlled methodology: challenges in applying the biomedical standard to behavioral headache research. Headache 45 (5), [22]. Rains, J.C., Penzien, D.B., McCrory, D.C. & Gray, R.N. (2005). Behavioral headache treatment: history, review of the empirical literature, and methodological critique. Headache 45 (Supplement 2), S The International Journal of Person Centered Medicine

8 Nguyen, Davis, Kaptchuk and Phillips Mind-body medicine and self-rated health [23]. Bloom, B.S., Retbi, A., Dahan, S. & Jonsson, E. (2000). Evaluation of randomized controlled trials on complementary and alternative medicine. International Journal of Technology Assessment in Health Care 16 (1), The International Journal of Person Centered Medicine