Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Science Science is nothing but trained and organized common sense. âThomas H. Huxley The keynote address for the session on science was delivered by Dr. Dean Ornish. Ornish surveyed the key fields related to integrative medi- cine, providing examples of the effectiveness of integrative interventions in improving patientsâ health. The panel discussion was moderated by Dr. Bruce McEwen of the Rockefeller University, with panelists offering their perspectives and priorities surrounding the development and ad- vancement of the evidence base for integrative medicine. Topics reviewed focused on the complex interplay of biology, be- havior, psychosocial factors, and environment shaping health and dis- ease. Often, as Ornish pointed out, these interactions can produce synergistic resultsâfor good or ill. This very complexity calls for a sys- tems approach in health care and in health sciences research that could evaluate multiple variables interacting in dynamic ways. As several pan- elists, including Dr. Lawrence Green, noted, this is a significant short- coming of randomized trial methodology, which tests one variable at a time and is not designed to evaluate multifaceted preventive approaches, such as the lifestyle interventions described by Ornish. New, more ap- propriate assessment methods are under development. They range from improved effectiveness trials at the community level to studies of im- mune system biomarkers at the molecular level, to an array of study methods being used at the National Institutes of Health (NIH) National Center for Complementary and Alternative Medicine (NCCAM), de- scribed by Dr. Josephine Briggs. Advances in genomic sciences are increasingly illuminating the con- tribution of genes to health and disease, explained Dr. Richard Lifton. New, more effective treatments may result that can be tailored to a per- sonâs genetic profileâa possibility that will greatly facilitate personal- 77
78 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC ized medicine and person-centered care. However, as several panelists pointed out, âwe are more than our genes,â and it is the epigenetic influ- encesâthe interactions of genes with other factorsâthat shape health and illness. Dr. Mitchell Gaynor reported on numerous studies indicating the influence of diet and other environmental factors on the expression of genes and, consequently, their effects on health. Psychosocial factors including stress, loneliness, and depression, all mediated by the brain, were described by McEwen and Dr. Esther Stern- berg as strong contributors to health and disease. As the brain responds to stress, hormones are released that can, for example, interfere with the immune response and metabolic processes and damage the cardiovascu- lar system. Something as simple as having a support group, a wide social network, or a nurturing belief system helps people manage stress and recover from illness, Ornish and Sternberg said. People with high levels of stress can be found throughout society. Dr. Nancy Adler noted that people in lower socioeconomic strata are particularly vulnerable to the effects of stress, which is reflected in their lower health status and pre- mature aging at the cellular level. Panelists said that lifestyle choices are important not only because unhealthy choices contribute to many of the leading causes of mortality, but also because healthy choices hold the potential to outperform com- monly prescribed drugs, increase brain function, and affect the expres- sion of genes. As individuals around the world begin to adopt an âAmericanâ lifestyle, especially our diet, the challenges of preventive medicine are becoming global, Ornish said. SCIENCE KEYNOTE ADDRESS Dean Ornish, Preventive Medicine Research Institute and University of California, San Francisco The good physician treats the disease; the great physician treats the patient who has the disease. âSir William Osler Ornish is a scientist and clinician who has spent much of the last three decades conducting clinical studies in what is now called integra- tive medicine. He described integrative medicine, or âprospective medi-
SCIENCE 79 cine,â as predictive, preventive, personalized, and participatory (Sny- derman and Williams, 2003; Weston and Hood, 2004). While it provides the best conventional care, its principal focus is on the preventive main- tenance of health by attention to all components of lifestyle, including diet, exercise, stress management, and emotional well-being. Lifestyle choices offer many opportunities to improve health, as these decisions are made multiple times throughout every day. Integrative medicine may also be perceived as âfunctional medicine,â which focuses on the under- lying pathways of chronic disease, such as inflammation, genetics, and metabolism (Bland, 2008; Hyman, 2007). Integrative medicine takes a systems approach to improving patient health, and analyses of its effec- tiveness also must examine systemic, synergistic effects. Integrative medicine contrasts with that of conventional medicine, where, Ornish said, clinicians spend a great deal of time mopping up the floor around an overflowing sink instead of just turning off the faucet. Evidence in integrative medicine accumulates, not through studies involving one independent variable and one dependent variable, but rather through studies of the effects of multiple factors working together in systematic ways. Similar to the components of a light bulb, the whole is greater than the sum of the parts. Ornish gave an example of the synergy of integrative medicine that is demonstrated in a study of curcuminâa component of the spice tur- meric, an ingredient in Indian curries. Curcumin, he noted, is known to have anti-inflammatory and antitumor properties, suppressing the onco- gene MDM2; further, as a free radical scavenger, it inhibits oxidative DNA damage and may help prevent Alzheimerâs disease. Researchers have hypothesized that Indiaâs substantially lower incidence of Alz- heimerâs disease might be associated with curcumin ingestion because it suppresses inflammation that, in turn, leads to amyloid deposits in the brain that lead to Alzheimerâs disease. Initial studies of curcumin alone failed to show any beneficial effect. However, he reported, when it is combined with black pepper, as it often is in local cuisine and in Ay- urvedic medicine, there was about a 20-fold increase in curcuminâs bioavailability, demonstrating a synergistic effect of the spices (Shoba et al., 1998). Similarly, other studies show that a diet rich in fruit and vegetables protects against heart disease and some forms of cancer, although vita- mins alone usually do not. For example, foods containing beta-carotene and other carotinoids have various health benefits, most likely owing to their antioxidant effect. Yet, in some circumstances, beta-carotene die-
80 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC tary supplements appear to have a pro-oxidant effect and actually in- crease the risk of lung cancer among smokers. Smoking causes inflam- mation in the lungs, a condition that increases free radicals that can damage DNA. In a recent study, beta-carotene supplements were shown to further increase free radicals in a smokerâs airways (van Helden et al., 2009). This increase occurs because supplemental beta-carotene inhibits an enzyme called myeloperoxidase and increases the formation of hy- droxyl free radicals, leading to increased inflammation and oxidative stress. Thus, dietary beta-carotene inhibits inflammation whereas sup- plements can increase it; studying beta-carotene in isolation (supple- ments) can miss the benefit of beta-carotene in foods and result in misleading information. The third example Ornish gave of synergy in integrative medicine research was a 1989 study in which a group of women was treated with chemotherapy, radiation, and surgery for metastatic breast cancer. One study cohort, randomly assigned to a support group that met 1 hour a week, exhibited a 5-year survival rate twice as high as that of control subjects who did not participate in a support group (Spiegel et al., 1989). âIf a new drug had been shown to do that, it would be malpractice not to prescribe it,â Ornish said. However, more recent studies failed to repli- cate this result. The possible explanation for this result came in 2007, in research re- porting that there was no overall survival in support group participants, but there was a survival advantage for women with estrogen receptor (ER)-negative breast cancer (Spiegel et al., 2007). Apparently, hormonal therapy improved so much after 1989 in ER-positive breast cancer (but not ER-negative breast cancer) that it washed out the contribution of support groups for those who were ER-positive, while the benefit of sup- port groups still occurs among ER-negative women. Among ER-negative women, those who participated in the support group still had survival rates 25 percent greater than those who did not (Spiegel et al., 2007). Without looking at the larger contextâthe systemâthis effect would have been missed. Integrative Medicine and Pathways of Disease Awareness, in Ornishâs view, is the first step in healing. One prop- erty of science is its power to raise awareness. Science can help us un- derstand that what we do, how we feel, and what happens to us are all
SCIENCE 81 interconnected. Clinical research in various aspects of integrative medi- cine have produced dramatic evidence of the effects of lifestyle on health status and suggested the importance of taking a systems approach and considering factors other than those traditionally held responsible for causing not only chronic disorders but also infections diseases. Health and disease are much more multidimensional than once thoughtâfor example, not everyone who is infected with a virus gets sick. Psychosocial factors significantly affect this varying susceptibility. For example, HIV-positive patients who were depressed were twice as likely to develop AIDS, and to die of it, than were HIV-infected patients without depression (Burack et al., 1993; Mayne et al., 1996). Depression prompted a significantly more rapid health decline, measured by cell death and reduced lymphocyte counts. Similarly, volunteers infected with a rhinovirus were less likely to develop head colds if they had ex- tensive social supports (Cohen et al., 1997). Subjects with at least six interpersonal relationshipsâmanifested in phone calls or visitsâwere only one-fourth as likely to develop cold symptoms. Reactions to stress are a well-recognized factor in immune function and can be either health protecting or health damaging. Stress generally suppresses immune function, while relaxation and meditation enhance it. Thus, there can be synergy between meditation and allopathic interven- tions. For example, psoriasis patientsâ skin cleared much faster (60 per- cent after 50 days) if they received standard photochemotherapy and listened to mindfulness-based stress management tapes; lesions of pa- tients who did not use the tapes were only 20 percent clear after 50 days (Bernhard et al., 1988). Integrative Medicine and Lifestyle Change Many people mistakenly believe that only a new drug or a new and expensive technology can succeed against disease. Ornish and his team use high-tech, state-of-the-art measurement techniques to show the bene- fits of low-tech, low-cost, and, in some cases, ancient interventions and forms of disease prevention. Comprehensive lifestyle changesâ including changing what we eat, how we respond to stress, moderate ex- ercise, and greater love and intimacyâcan yield remarkable improve- ments in health. Equally as important, these improvements are experienced within a short timeframe, which helps individuals feel better and facilitates the sustainability of their lifestyle modifications.
82 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC Early in his career, Ornish conducted several studies to evaluate the effects of lifestyle on heart disease. In the first study, 10 heart disease patients who undertook comprehensive lifestyle changes for only 1 month experienced a 90 percent reduction in angina as well as improved blood flow (myocardial perfusion) to their hearts, the first study showing that coronary heart disease may be reversed by changing lifestyle (Ornish et al., 1979). In the second study, a randomized controlled trial, a similar set of patients experienced a 91 percent reduction in angina and a signifi- cant increase in the heartâs ability to pump blood, while the control group experienced a decrease (Ornish et al., 1983). In later randomized trials, patients who made comprehensive lifestyle changes showed a significant reduction in coronary artery blockages (atherosclerosis) as measured by quantitative coronary arteriography after 1 year and an even greater re- versal of heart disease after 5 years, whereas the randomized control group showed a worsening of coronary atherosclerosis after 1 year (Or- nish et al., 1990) and greater worsening after 5 years (Ornish et al., 1998), as shown in Figure 4-1. Cardiac PET scans revealed that 99 per- cent of patients who made these lifestyle changes were able to stop or reverse the progression of their heart disease (Gould et al., 1995). In a subsequent study, comprehensive lifestyle changes appeared to significantly slow, stop, or reverse the progression of early-stage prostate cancer, the first time that an integrative medicine intervention was shown to affect the progression of any form of cancer in a randomized con- trolled trial (Ornish et al., 2005). In both the cardiac and prostate cancer studies, the more people changed their lifestyle, the more benefit they received in a doseâresponse effect. If borne out, this finding is empower- ing, as the degree of improvement may be more a function of adherence than age or disease severity. Lifestyle changes can even outperform drugs in the secondary pre- vention of disease. In diabetes, lifestyle modifications functioned better than metformin in preventing the adverse effects of diabetes, including damage to the eyes, nerves, and kidneys (Knowler et al., 2002). Oneâs own body, Ornish suggested, may be able to regulate blood sugar levels more consistently than drugs, in these conditions. The idea that taking a pill is easy but changing lifestyle is difficult, if not impossible, is not supported by the evidence, said Ornish. Studies show that two-thirds of patients prescribed statin drugs are not taking them only 4 months later because these drugs do not make people feel
SCIENCE 83 60 â 55 51.9 50 Diameter stenosis, % Control 45 â½ 41.3 42.3 40 40.7 38.5 37.3 35 Treatment 30 Baseline 1 year 5 years FIGURE 4-1 Mean percentage diameter stenosis in treatment and control groups at baseline, 1 year, and 5 years. NOTES: Mean percentage diameter stenosis in treatment and control groups at baseline, 1 year, and 5 years. Error bars represent SEM; asterisk, P = .02 by between-group 2-tailed test; dagger, P = .001 by between-group 2-tailed test. SOURCE: Ornish et al., 1998. Reproduced from JAMA 280(23):2001-2007 (De- cember 16, 1998), with permission. Copyright 1998 American Medical Associa- tion. All rights reserved. better (Ornish, 2009). In contrast, Ornish said that comprehensive life- style changes are dynamic, and often cause people to feel so much better, so quickly, that it reframes the reason for changing lifestyle from risk factor reduction (which is boring) or fear of dying (which is too frighten- ing) to joy of living. Ornish noted that moderate changes in lifestyle create only moderate benefits. For example, people who are diagnosed with hypercholes- terolemia are often asked to make moderate changes in diet, but these relatively small changes cause negligible reductions in LDL cholesterol. Conversely, those who made more intensive changes in diet and lifestyle
84 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC experienced a 40 percent average reduction in LDL cholesterol, compa- rable to what can be achieved with statin drugs without the costs ($18 billion for Lipitor last year) and potential side effects (e.g., muscle and liver damage) (Ornish et al., 1998). Effects of Lifestyle Changes on Brain Function and Gene Expression Ornish provided several examples of how lifestyle changes can im- prove brain function. Evidence exists that healthy diets, effective man- agement of stress, moderate exercise, and being in loving relationships can produce growth in neurons, through the newly appreciated phenome- non of neuroplasticity. Consuming such foods as chocolate, blueberries, and tea, as well as alcohol in moderate amounts, can cause neurogenesis, whereas saturated fats, sugar, nicotine, and excessive consumption of alcohol can speed up the death of brain cells. Interestingly, he said, can- nabinoids may increase neurogenesis, but opiates and cocaine decrease it. Depression and chronic feelings of stress reduce hippocampal volume and therefore impair memory (Campbell et al., 2004; Conrad, 2006; Sheline et al., 2003). In contrast, increased exercise has been associated with an increase in hippocampal volume within only a few months, as demonstrated in Figure 4-2 (Erickson and Kramer, 2009). Increases in brain size in the hippocampus and frontal cortex have been also associ- ated with cognitive therapy, stress management, and psychotherapy (de Lange et al., 2008). Changes in lifestyle affect gene expressionâas nurture, in some cases, trumps nature. The Gene Expression Modulation by Intervention with Nutrition and Lifestyle (GEMINAL) studies found beneficial effects in 501 genes within 3 months after comprehensive lifestyle changes, in- cluding meditation. Changes in gene expression included the down- regulating of genes that promote heart disease or cancer and up- regulation of tumor-suppressing genes (Dusek et al., 2008; Ornish et al., 2008a). Ornish reiterated that genes are not destiny, and that genomics can be an important platform in studying the effects of integrative medi- cine. Lifestyle change also has been shown to affect telomere length, which plays a role in aging and longevity. The effect of stress on te- lomere length was demonstrated in a study of women who felt highly stressed, as long-time caregivers of children with autism or birth defects.
SCIENCE 85 4500 4450 4400 Volume (mm3) 4350 4300 Control group 4250 Exercise group 4200 4150 Time 1 Time 2 Time 3 FIGURE 4-2 Hippocampus volume change. SOURCE: Reproduced from British Journal of Sports Medicine, K.I. Erick- son and A. F. Kramer, 43(1):22-24, 2009 with permission from BMJ Pub- lishing Group Ltd. These women exhibited lower levels of telomerase, the enzyme that re- pairs and lengthens damaged telomeres (Epel et al., 2004). The womenâs perception of stress was highly correlated with shorter telomeres. In con- trast, comprehensive lifestyle changes led to almost 30 percent increases in telomerase, and thus telomere length, within 3 months, as shown in Figure 4-3 (Ornish et al., 2008b). Ornish commented that this was the first study showing that any intervention can increase telomerase, another example that integrative medicine interventions are not only as good as pharmaceutical interventions but often better, much less expensive, and that the only side effects are good ones.
86 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC 2.4 p=0.031 (two-tailed) 2.3 Telomerase activity (natural logarithm) 2.2 2.1 2.0 1.9 1.8 1.7 1.6 1.5 Baseline 3 months FIGURE 4-3 Increase in telomerase activity from baseline to 3 months. SOURCE: Reprinted from Lancet Oncology, Vol. 9, Ornish, D., J. Lin, J. Daubenmier, G. Weidner, E. Epel, C. Kemp, M. J. Magbanua, R. Marlin, L. Yglecias, P. R. Carroll, and E. H. Blackburn, Increased telomerase activity and comprehensive lifestyle changes: A pilot study, Pages 1048-1057, Copyright 2008, with permission from Elsevier. A globalization of chronic diseases such as heart disease, diabetes, and cancer is occurring, as other countries are starting to eat like us, live like us, and die like usâa trend that is almost completely preventable, said Ornish. Through lifestyle change, preventive medicine can reap benefits on a global scale. Ornish also pointed out that the choices we make in our personal lives affect our planet as well. For example, some suspect that more global warming is caused by livestock consumption (due to methane production) than from all forms of transportation com- bined. Limitations of Conventional Medicine and Traditional Research Diet and lifestyle changes that have already proved effective could prevent 95 percent of heart disease (Yusuf et al., 2004), Ornish said. However, a series of randomized controlled trials has shown that the usual course of treatment that consists of angioplasty and bypass surgery
SCIENCE 87 and costs the nation more than $100 billion annuallyâdoes not prevent heart attacks or prolong life for the great majority of patients who receive them (Boden et al., 2007). An evidence-based approach to health care is necessary, he said, and should be applied across the board to conven- tional and integrative interventions alike. Instead, the dominant approach tilts toward reimbursement-based medicine rather than evidence-based medicine. Conventional methodologies available for gathering evidence on in- tegrative medicine are often limited. Randomized controlled trials (RCTs) may work well for controlling bias in drug trials and are consid- ered the gold standard for evaluating new treatments. However, lifestyle interventions and systems approaches often introduce uncontrollable sources of bias, making RCTs a less suitable evaluation method for com- plex lifestyle interventions. The randomization of participants can be challenging in lifestyle interventions. All participants must go through a series of baseline tests and agree to commit to make lifestyle changes if they are assigned to the experimental group. However, participants who are then randomized to the control group may become disappointed or even angry about the assignment, because they believe the intervention may have been beneficial to them, and they often drop out of the study as a result. Unlike a drug study, in which access to a new drug can be lim- ited to experimental group participants, researchers cannot prevent the control group from making positive lifestyle changes on their own, which will confound the studyâs results. The study of integrative medicine cannot easily be limited to the study of only one independent variable and one dependent variable, noted Ornish. For example, if the intervention consists of exercise, addi- tional independent variables may be involved, such as increased social interaction with others who are exercising with the participant. Such in- teractions provide encouragement and support, and reinforce the individ- ualâs sense of meaning, self-efficacy, and purpose. People who begin to exercise regularly are also likely to improve their diets. These additional variables can complicate the interpretation of study results. New research methods are being devised to better assess integrative medicine interventions. One alternative, developed by Marvin Zelen, consists of a randomized invitational design. In this approach, partici- pants are assigned to a study group before being interviewed, so that the intervention is not described in detail to those assigned to the control group, and they do not have to commit to participate in the intervention before randomization. Members of the control group are asked only to
88 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC agree to be tested; subsequently, they are less likely to drop out and tend not to be upset about the assignment. Crossover designs, in which the control group does not receive the intervention initially (and can serve as a nonintervention control group) but later receives the intervention, also can improve study design. Sham interventions are another approach that has been used in acupuncture studies. In this approach, members of the control group receive needle placements but not in the radial locations used in acupuncture (Haake et al., 2007). Innovative experimental de- signs and systems approaches such as these can lead to more persuasive evidence. Transforming Human Experience To Ornish, the most prevalent epidemic among Americans is not heart disease, cancer, or obesity, but rather loneliness and depression. Antidepressants are one of the nationâs most frequently prescribed drugs. Study after study has shown that people who feel lonely and depressed are many times more likely to get sick and to die prematurely than peo- ple who have a sense of love, connection, and community, said Ornish. This effect is, in part, because lonely and depressed individuals are more likely to smoke, overeat, drink too much, and work too hard, but it also involves mechanisms that are not completely understood. For example, within 6 months after a heart attack, patients who were depressed were six times more likely to die than patients who were notâa finding that is independent of traditional risk factors, such as blood pressure and choles- terol levels, as shown in Figure 4-4 (Frasure-Smith et al., 1993). Animal studies show similar results: rabbits that were talked to, petted, played with, and essentially loved had 60 percent less plaque in their arteries than a group that was neglected (Nerem et al., 1980), even though these rabbits were genetically comparable and on the same diet. In a final observation, Ornish described that sustainable changes in lifestyle are based on joy, fun, pleasure, and freedom, not austerity and deprivation. The language of behavioral change, unfortunately, often has a moralistic qualityâor what he termed a fascist quality: âI cheated on my dietâ and âI ate bad food, so Iâm a bad person.â Willpower and pa- tient compliance are based on restricting and manipulating behavior, which is not sustainable. The mechanisms that respond to comprehensive lifestyle changes are much more dynamic than had once been realized,
SCIENCE 89 30 25 % Mortality 20 15 Depressed (n=35) 10 Nondepressed 5 (n=187) 0 1 2 3 4 5 6 Months post-MI FIGURE 4-4 Cumulative mortality for depressed and nondepressed patients. NOTE: MI = myocardial infarction SOURCE: Frasure-Smith et al., 1993. Re- produced from JAMA 270(15):1819-1825 (October 20, 1993), with permission. Copyright 1998 American Medical Association. All rights reserved. and, as described earlier, they allow patients to experience significant improvements in how they feel in a relatively short time. As Ornish de- scribed, âJoy of living is a more sustainable and powerful motivator than fear of dying. How well we live is more motivating than how long we live.â Meditation, for example, was not used by spiritual teachers to un- clog arteries or improve genes, but rather to help quiet minds and bodies in order to experience an inner sense of peace, joy, and well-being. Ornish concluded by sharing an anecdote from a friend and teacher with whom he had studied, Swami Satchidananda. When asked the dif- ference between wellness and illness, Satchidananda replied by writing the words on a blackboard and circling the I in illness and the we in wellness. Healing occurs when we move from the loneliness, isolation, and depression of the I toward the sense of support, connection, and community of the we.
90 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC PANEL ON THE SCIENCE BASE FOR INTEGRATIVE MEDICINE Panel Introduction Bruce S. McEwen, The Rockefeller University McEwen began the science panel with a brief summary of the biological effects of stress and the important role of the brain in mediating them. The brain is the central organ of stress; it perceives and decides what is threatening, producing physiologic responses that lead to adaptation and behavior that promotes or damages health, as shown in Figure 4-5. Although the body has a formidable capacity to defend itself against stress, this capacity is compromised by allostatic loadâthe cumulative, damaging effects of the overuse or disregulation of stress response systems over timeâthat results from chronic stress and behav- ior associated with experiencing stress, such as overeating, poor sleep, and lack of physical activity. Environmental stressors Major life events Trauma, abuse (work, home, neighborhood) Perceived stress (threat, helplessness, vigilance) Behavioral Individual responses differences (fight or flight; (genes, development, experience) personal behaviorâdiet, smoking, drinking, exercise) Physiologic responses Allostasis Adaptation Allostatic load FIGURE 4-5 The stress response and development of allostatic load. NOTES: The perception of stress is influenced by oneâs experiences, genetics, and behavior. When the brain perceives an experience as stressful, physiologic and behavioral responses are initiated, leading to allostasis and adaptation. Over time, allostatic load can accumulate, and the overexposure to mediators of neu- ral, endocrine, and immune stress can have adverse effects on various organ systems, leading to disease. SOURCE: McEwen, 1998. Reprinted, with permission, from New England Journal of Medicine. Copyright 1998 by Massachusetts Medical Society.
SCIENCE 91 Stress mediators, such as adrenalin and cortisol, are a part of a com- plex nonlinear network in which mediators regulate each other. These mediation pathways include parasympathetic as well as sympathetic ac- tivity. Other players in the network are oxidative stress, which is essen- tial for life but when excessive has deleterious effects, and anti- inflammatory and inflammatory cytokines, which are highly important in the adaptation of the immune system. These important mediators and others concurrently affect brain and metabolic functions, as well as the cardiovascular and immune systems, sometimes producing comorbid- ities. In considering the science of integrative medicine, McEwen empha- sized the lasting effects of early-life adversity and stress. As the CDC Adverse Childhood Experiences studies and others have shown, low so- cioeconomic status can lead to problems in systemic pathophysiology as well as poor cognitive abilities and learning skills. Abuse, neglect, and chaos at home can result in a sense of helplessness, low self-esteem, dis- tress, and poor self-regulatory behaviors. All of these stressors can cause early-life obesity, blood pressure elevation, cardiovascular reactivity, lasting inflammation, mental health problems, and, ultimately, a shorter life span (Lantz et al., 1998; Mare, 1990; Pamuk et al., 1998; Pappas et al., 1993). Family stressors are found not only among people of low so- cioeconomic status, but span all socioeconomic strata and have many physical and mental health consequences for the children and adults (Repetti et al., 2002). McEwen noted that the panel was organized to begin broadly with social determinants of health and progress through to genetics, epigenet- ics, and research methods. Advancing scientific methods have provided rapidly expanding data on gene expression and patterns of inheritance. Yet, as several other speakers indicated, the social environment exerts a powerful influence on health. Social Determinants of Health Nancy E. Adler, University of California, San Francisco Dr. Nancy Adler offered the perspective that, even though many types of risk occur in all socioeconomic groups, the ability to make life- style changes and to engage in healthy lifestyles is often patterned by social status. She reminded the summit participants that the people most
92 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC in need of the integrative medical interventions being discussed are fre- quently the people least able to access them, due to social barriers. Socioeconomic status (SES) typically includes the triad of income, education, and occupation. The higher one is positioned on any of these three ladders, the better oneâs health status is likely to be. More than 3 in 10 adults living below the federal poverty level 1 rate their health as fair or poor, whereas less than 7 percent of adults in the highest income group rate their health fair or poor (Robert Wood Johnson Foundation, 2008). This effect is most prominent in the health status gap between people below the poverty level compared to those just above it, as indi- cated in Figure 4-6a. Every study of the relationship between health and income confirms that health status is directly related to income. Education shows similar results, as seen in Figure 4-6b. There is a four- to five-fold health difference in reported fair or poor health be- tween college graduates and those with less than a high school diploma, with improved health among people at each step of additional education (Robert Wood Johnson Foundation, 2008). Although income may be af- fected by reverse causationâpoor health can cause a drop in incomeâ education is relatively free of this effect. Children exhibit the same general pattern, when socioeconomic status is measured in terms of their parentsâ income and education, as shown in Figures 4-7a and 4-7b. Fortunately, children generally are much healthier than adults, so the percentages of children in poor health are smaller. One way to conceptualize the effect of socioeconomic status (SES) on health is that lower SES accelerates the aging process by increasing allostatic load, as discussed by McEwen. People who are disadvantaged experience the physical changes of aging earlier in life, including higher blood pressure, higher body mass index, greater abdominal fat deposi- tion, and shallower evening drop in cortisol, said Adler. The powerful relationship between health and SES is apparent even at the level of cellular aging, as reflected in telomere length. One illustra- tion of this involves the third component of SES, occupation. English 1 Each year, the Department of Health and Human Services publishes the Federal Pov- erty Income Guidelines in the Federal Register. These guidelines are often referred to as the âfederal poverty level,â and they provide the total annual income level at which fami- lies of various sizes are considered impoverished.
SCIENCE 93 40 30 Percent of adults, ages â¥ 25 years, with poor/fair health* 25.7 Percent of adults, ages â¥ 25 years, with poor/fair health* 35 30.9 24 30 25 18 21.2 20 13.9 12 11.0 15 14.0 10.1 10 5.8 6 6.6 5 0 0 Family income Educational attainment (Percent of federal poverty level) Less than high school <100% FPL High-school graduate 100â199% FPL Some college 200â299% FPL College graduate 300â399% FPL > 400% FPL â FIGURE 4-6a and 4-6b Relationship between income and education and reported adult health status. NOTE:*age-adjusted. SOURCE: Robert Wood Johnson Foundation, 2008. Reprinted, with permission. Copyright 2008 Robert Wood Johnson Foundation/Overcoming Obstacles to Health. Prepared by the Center on Social Disparities in Health at the University of California, San Francisco; and Norman Johnson, U.S. Bureau of the Census, using data from the National Longitudinal Mortality Study, 1988â1998. workers in manual occupations showed a decrease of 140 base pairs in telomeres when compared to same-age workers in nonmanual fields, as shown in Figure 4-8; this is comparable to 8 years of aging (Cherkas et al., 2006).
94 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC 5 5 Percent of children, ages â¤ 17 years, with poor /fair health* Percent of children, ages â¤ 17 years, with poor /fair health* 4.4 4.3 4 4 3 3 2.4 2.4 2 2 1.7 1.4 1.0 1 1 0.7 0.6 0 0 Family income Educational attainment (Percent of federal poverty level) Less than high school <100% FPL High-school graduate 100â199% FPL Some college 200â299% FPL College graduate 300â399% FPL > 400% FPL â FIGURE 4-7a and 4-7b Relationship between income and education and reported child health status. NOTE: *age-adjusted. SOURCE: Robert Wood Johnson Foundation, 2008. Reprinted with permission. Copyright 2008 Robert Wood Johnson Foundation/Overcoming Obstacles to Health. Prepared by the Center on Social Disparities in Health at the University of California, San Francisco. Source: National Health Interview Survey, 2001â 2005.
SCIENCE 95 7.2 Nonmanual 7.15 Manual Telomere length (kb) 7.1 # * 7.05 7 6.95 6.9 Age-adjusted Age, BMI, smoking and exercise adjusted *P = 0.01 #P = 0.04 FIGURE 4-8 Mean telomere length and standard error by manual vs. non- manual social class groupings. SOURCE: Cherkas et al., 2006. Reprinted, with permission, from Aging Cell, 2006. Copyright 2006 The Authors. SES shapes almost every aspect of peopleâs lives. It affects where they work, where they live, the social norms that govern them, where their children are educated, and their physical and social environments. In turn, these environments affect more proximal causes of disease and health, such as health behavior, access to health care, and exposure to toxins and pathogens, as well as to social threats that generate stress re- sponses. The current model of health focuses on the individual, involving ef- forts to promote behavior change, reduce stress, and increase patient em- powerment. However, to improve health, it may be more effective to attempt broader strategies, suggested Adler. For example, Kaiser Perma- nente in the San Francisco Bay area responded to low-income individu- alsâ lack of access to fresh fruits and vegetables by sponsoring farmersâ markets at its health facilities. Environmental, social, economic, and educational policy and inter- ventions can affect peopleâs health-related behavior and their heath status; these can occur at the national and/or local level. The state of the science now adequately documents the effects of social factors in health.
96 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC The next challenge, Adler said, is to determine how best to use a com- prehensive set of tools, ranging from policy change to clinical interven- tions, to make a difference in improving health. MindâBody Medicine Esther M. Sternberg, National Institute of Mental Health Biologic pathways mediated by the brain provide the central connec- tion between the external environmental factors described by Adler and health status. Sternberg described these pathways, particularly those re- lated to the immune response to stress. A stressor triggers a cascade of neuroendocrine activity. The hypothalamus and the pituitary gland pro- duce the stress hormones CRH and ACTH, and, in response, the adrenal glands release cortisolâthe bodyâs most potent anti-inflammatory hor- mone. Cortisol also is vital in regulating the immune system by provid- ing an opportunity for it to return to homeostasis after a stress event occurs. Concurrently, the adrenergic sympathetic nervous system re- leases norepinephrine, and the adrenal medulla produces adrenalin. To- gether, these nerve chemicals and hormones, regulate the immune response at the cellular and molecular levels. Health requires effective communication between the neuroendo- crine and immune systems. An imbalance between these systems can cause disease. Too much cortisol, for example, can increase susceptibil- ity to infections. People with excess cortisol may have more frequent or severe viral infections, less benefit from vaccines, prolonged wound healing, increased vascularization of solid tumors, and increased chro- mosomal aging (Armaiz-Pena et al., 2009; Cohen et al., 2007; Kiecolt- Glaser et al., 1995). By contrast, people with a cortisol deficiency will have a diminished anti-inflammatory effect and increased susceptibility to such autoimmune inflammatory diseases as inflammatory arthritis. Research indicates that the development of cardiovascular disease, diabe- tes, and depression can be influenced by cortisol-related inflammation. The road to healing also is modulated by the brain. Health-promoting activities, such as meditation, yoga, tai chi, and exercise, have biologic effects on the neuroendocrine systems. When people engage in these types of activities, the vagus nerve functions as a brake on the sympa- thetic nervous system, thereby increasing the power of certain compo- nents of the immune system. Also, such activities prompt release of
SCIENCE 97 powerful neuroendocrine system hormones (specifically, endorphins and dopamine). There is now broad acceptance that stress and allostatic load, or cu- mulative effects of stress, can foster disease. However, less is known about how health and healing are affected by a personâs beliefsâfeelings of purpose, peace, relaxationâwhich are enhanced by meditation, yoga, mild to moderate exercise, and several other activities. The biology of belief is difficult to study in humans and impossible in animals. How- ever, elegantly designed studies are beginning to shed light on brain mechanisms underlying such interventions. These include activation of the brainâs positive emotional centers, with release of dopamine and en- dogenous opiates, and activation of the vagus nerve and parasympathetic nervous system, with resulting down-shifting of the stress response. Sternberg conceded that she had been a skeptic of the relationship between belief and healing until she experienced inflammatory arthri- tisâan illness related to chronic stressâafter caring for her terminally ill mother. She began to be convinced of the relationship while on holiday in Crete, when a routine of swimming and meditation at the ruins of the temple of Aesclepius, the Greek god of medicine and healing, provided a beginning for her healing process. Sternberg reiterated Ornishâs point about the importance of allowing the body to heal itself, which often re- quires lifestyle change. Researchers need complex measures to understand how, biologically, a whole host of molecules work together in response to stress, relaxation, and other factors Sternberg referred to as belief. Ideally, these measures should provide a molecular signature of health or disease for a specific person, around which tailored advice can be developed. Researchers at the NIH have developed a way to obtain reliable measures of several of the immune biomarkers of stress through having people wear a sweat patch for 24 hours. Analysis of the sweat avoids having to draw blood, a process that, naturally, creates the stress that researchers are trying to measure. These biomarkers are evaluated through sophisticated method- ologies that use immunoaffinity chromatography and various types of sophisticated mass spectrometry analyses. In one such study, sweat patch analysis showed that asymptomatic women with a history of depression had elevated levels of proinflamma- tory cytokines and NPY, an adrenalin-like nerve chemical (Cizza et al., 2008). These women also exhibited a reduction in vasoactive intestinal peptide, a marker of the parasympathetic vagus nervous system. This signature indicated that their stress response had shifted away from
98 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC vagally mediated relaxation toward a sympathetic adrenalin-mediated response. The significance of this finding is its suggestion that, although these women were asymptomatic, biological remnants of their depression remained. Such advanced research techniques will enable development of a full picture of the brainâs hormone-production activities that both contribute to disease and promote wellness. Genomic and Predictive Medicine Richard P. Lifton, Yale University School of Medicine Dr. Richard Lifton emphasized the role of basic science in under- standing, preventing, and treating disease. Examples of integrative ap- proaches to understanding and eliminating disease can be found in the development of prevention and treatment strategies for infectious dis- eases. These diseases profoundly affected the course of public health his- tory and the lives of millions. Even our presidents have not remained untouched: Franklin Roosevelt contracted poliomyelitis, and Abraham Lincolnâs son, Willie, died in the White House of typhoid fever. Malaria was once endemic in the Foggy Bottom area of Washington, DC, where the summit was held. Many of these diseases were eradicated through integrative approaches that began with basic science, including the de- velopment of vaccines for prevention, targeted therapies for specific in- fections, and improvements in public health and sanitation. The current status of genomic and predictive medicine reflects 150 years of progress in the basic sciences, starting with Gregor Mendelâs 1865 recognition of genetic factors and Thomas Hunt Morganâs 1910 discovery that genes lie on chromosomes. Other milestones include the 1944 identification of DNA as the genetic material, the 1953 unveiling of the structure of DNA, the 1960 characterization of the genetic code, and the entire sequencing of the human genome in 2001. This last achieve- ment allows the identification of all genes and common variations, and the stage is now set for understanding the inherited and acquired varia- tions in genes and gene expression that contribute to the development of disease. Basic science and genetics have been necessary for an improved understanding of the pathogenesis of cardiovascular disease, chronic myelogenous leukemia, and HIV, among others. This greater understand- ing has, in turn, led to new treatments. Highly effective statins were
SCIENCE 99 developed following the mapping of the cholesterol biosynthetic path- ways, identification of a mutation in the LDL cholesterol receptor, and the recognition of a causal relationship between increased cholesterol levels and heart disease. Recent advances in the treatment of chronic myelogenous leukemia build on a long history of basic science: in 1963, researchers identified a chromosomal anomaly related to the disease; in the 1970s, they found a specific chromosomal rearrangement; and in the 1980s, the specific gene mutation. The pharmaceutical industry and a passionate physician then developed and directed the clinical trials that established the capacity of a single drug to induce remission for most patients. Similarly, basic sci- ence led to the identification of HIV-1, the virus that causes AIDS. Diag- nostic tests were developed, and a host of new drugs, including reverse transcriptase inhibitors, protease inhibitors, and integrase inhibitors, have transformed this rapidly fatal disease into a manageable chronic condi- tion. The sequencing of the human genome provides tremendous ability to characterize genes and identify their roles in human disease. A variety of study designs can be used to harness the knowledge resident in the ge- nome, including the single-gene mutations that produce large effects and suggest therapeutic targets; acquired somatic mutations in cancer; and comprehensive measurement of gene expression in diseased and healthy tissues, in order to identify underlying disease pathways. Additionally, scientists increasingly can resequence large portions of the genome, which will soon enable identification of important mutations. Lifton illustrated the impact of these genomic techniques with exam- ples of newly understood pathogenesis of certain diseases. In Alz- heimerâs disease, for example, rare mutations produce common lipoprotein variations that are major contributors to disease progression. Discoveries of inherited and acquired mutations of BRCA1 and BRCA2 genes have fundamentally changed the understanding of breast cancer. Within just the last year, mutations in the IDH1 gene have been identi- fied that appear to be essential to the formation of glioblastoma multi- forme in the brains of young patients. These advances allude to the tremendous promise of efforts to define the biology that connects genes to disease. Such discoveries also can help ameliorate racial and ethnic dispari- ties in disease prevalence and health outcomes, said Lifton. Examples include newly acquired knowledge about the role of common variants of MYH9 in causing kidney failure for African Americans with hyperten-
100 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC sion or HIV, and the role of common variants of PLPLA3 in increasing the risk of nonalcohol-related fatty liver disease in Hispanic Americans. The path from gene identification to therapy is difficult and unpre- dictable. To further advance the science base for integrative medicine, Lifton perceived a need for at least five resources: (1) a robust scientific enterprise, where investigators with subject-matter expertise proceed from gene discovery to finding plausible points for therapeutic interven- tion; (2) far broader interactions between academia, where the scientific strength in biology resides, and the pharmaceutical industry, where talent in medicinal chemistry prevails; (3) enterprising ways to manage con- flicts of interest; (4) well-designed clinical trials to define best practices; and (5) improved delivery systems to ensure that the public ultimately benefits from this substantial investment in science. Lifton offered two complementary visions of the future. In one, pa- tients are treated with drugs targeted to individual abnormalities, based on their genomic data. That will likely be the case for disorders in which differences among patients are large, such as in certain cancers. In the other vision, disease pathways and key nodes along those pathways will be identified, allowing for the development of population-based interven- tions. Examples of the latter approach could be used to address both be- havioral and environmental factors, such as reducing salt intake in order to control blood pressure and reducing cholesterol levels in order to pre- vent heart disease. Lifton said that realizing both visions require a fun- damental, science-based understanding of the causes of disease. Environmental Epigenetics Mitchell L. Gaynor, Weill-Cornell Medical Center and Gaynor Integrative Oncology Gaynor introduced the discussion of environmental epigenetics by noting that one in three Americans, at some point in his or her life, is go- ing to hear the words, âYou have cancer.â The question at hand is to what extent oneâs chances of acquiring cancer or other diseases are the result of oneâs genetic endowment or other, more controllable factors? The answer is a combination of the twoâwhile the genes one is born with cannot be controlled, how those genes are expressed can be. Gaynor illustrated this point by describing nutritional genomic and toxicoge- nomic studies that use mouse models. In these studies, toxicogenomics examines how various toxins can increase tumor-promoter genes and decrease tumor-suppressor genes. Conversely, nutritional genomics uses
SCIENCE 101 nutrients to reduce cancer risk by increasing expression of tumor- suppressing genes. Agouti mice are typically born with a pale yellow coat and an ele- vated risk of developing diabetes, obesity, and eventually, cancer. How- ever, pregnant female agouti mice that are fed a methyl-rich mixture of nutrients including folic acid, B12, choline, and betaine (found in beets) produced âpseudoagoutiâ offspring. These mice are born with brown coats and no elevated risk of obesity, diabetes, or cancer (Waterland and Jirtle, 2003). Researchers have found that the motherâs diet during preg- nancy changes gene expression in the offspring, a positive change that is transmitted to subsequent generations. Researchers have obtained the same overall result after administering a different nutrient, the soy extract genistein (Dolinoy et al., 2006). While the earlier diet provided additional methyl groups enabling the DNA methylation that caused the change in gene expression, the gen- istein affected histone formation, which also facilitated the methylation process. Endocrine disruptors exist in plastic products, cosmetics, and pesti- cides and are stored in the body for decades. When pregnant mice were exposed to either of two endocrine disruptors, vinclozolin (a fungicide used on grapes and other fruits) and methoxychlor (a pesticide now used instead of DDT), 90 percent of their male offspring had a 70 percent de- crease in sperm cells. This defect in the epigenome was inherited through four generations (Reik et al., 2001). It may be no coincidence that 11 percent of Americans have problems with fertility, Gaynor said. Bisphenol A (BPA) is another possible example. Agouti mice that were exposed to BPA and then fed either genistein or a diet rich in methyl groups produced pseudoagouti offspring. The nutritional supple- ment protected these mice from the BPA-induced heritable changes (Dolinoy et al., 2007). The incidence of breast cancer in American women has tripled, from one in 22 in 1960, to one in eight today. A geneticist at the University of Washington has found that women born with the BRCA1 or BRCA2 mu- tation before 1940 had a 24 percent risk of breast cancer, whereas such women born after 1940, as endocrine disruptors have become more prevalent in the environment, have a 67 percent risk. Simultaneously, the risk of ovarian cancer doubled for BRCA1 carriers, and increased 23 percent for BRCA2 carriers (King et al., 2003). These large increases in cancer risk suggest a significant geneâenvironment interaction, said Gaynor.
102 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC Women in societies that consume a great deal of soy, such as Japan, have one-seventh the incidence of fatal breast cancer as American women. Asian men living in Asian countries have one-thirtieth the inci- dence of fatal prostate cancers as American men. Gaynor suggested a close relationship between environmental public policy and personalized medicine, saying that often âThe environment outside us is the same as the environment inside us.â Intervention Evaluation and Outcomes Measures Lawrence W. Green, University of California, San Francisco Green indicated that he approaches integrative medicine from the opposite end of the biopsychosocial spectrum than the previous panelists. Namely, he approaches it from the standpoint of public health. He noted progress has been made in public health since the development of the disease prevention and health promotion initiative in the late 1970s under Surgeon General Julius B. Richmond, Dr. Michael McGinnis, and others. Drawing from this experience, he contended that, in order to achieve more evidence-based practice, we need more practice-based evidence. Patient-centered medicine, which is a hallmark of integrative care, challenges the supremacy of RCTs in evidence-based medicine, said Green. This challenge was acknowledged in 1999 in the preface to the second edition of a book by Archie L. Cochrane, whose previous contri- butions earned him the sobriquet of âfather of evidence-based medicine,â and it appropriately may be termed the âpost-Cochrane challengeâ (Cochrane, 1999). A Venn diagram illustrates this challenge, as shown in Figure 4-9. There is a large circle of information (labeled C) that is potentially useful to patients in their decision making. This sphere overlaps a smaller sphere (B) of information that is potentially evidence based. The overlap is small, consisting of information that is both potentially useful to pa- tients and potentially evidence based. Only a tiny sector of this overlap contains information already grounded in good evidence (labeled A), usually resulting from RCTs. Ten years after Cochrane offered this con- struct, researchers are finally recognizing the need to apply alternative designs and to develop new study designs to obtain a large body of evi- dence that will also be useful to patients, said Green. Difficult to under- take, and tethered to conditions that may be far removed from patientsâ true circumstances and desires, RCTs provide information of only lim- ited value or utility for prevention.
SCIENCE 103 Information of Area where there is currently good importance to patient evidence-based information which is of choice which is not even importance to patients in making choices potentially of evidence-based type, e.g., information about the process of delivery of health care C A B Information of importance to patient choice which is potentially of evidence-based type A. Information which is currently based on good evidence B. Information for which good-quality evidence-based information could be available in due course C. Information which is potentially of importance to patients in making health care choices FIGURE 4-9 Evidence-based medicine and integrative, patient-centered medicine. SOURCE: Hope, T. 1997. Evidence-based patient choice and the doctor-patient relationship. In But will it work, doctor? London: Kings Fund. Reprinted, with permission from Tony Hope, The Ethox Center. Research in integrative and person-centered medicine can shift the spotlight from the mediating variables that focus on the mechanisms of change to the moderating variables that focus on the characteristics of individual people and the context in which they live. Moderating vari- ables are reflected in the types of questions a clinician asks patients in order to learn about their identity, values, lifestyle, and life conditions. Green offered two strategic suggestions for meeting the post- Cochrane challenge. The first is to shape interventions in clinical re- search around moderating variables. Second, he emphasized the value of strategically blending theory and practical experience with evidence. An obsessive emphasis on best evidence has tended to crowd out good ob- servational and theoretical considerations, he said. Theory permits us to generalize evidence to other populations, settings, and circumstances. This is especially useful when, as is almost always the case, replication
104 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC of evidence may be too cumbersome, Green noted. Theory can help pro- vide solutions to problems, so long as investigators draw on theories eclectically and do not start with a theory and then look for problems on which to test it. In the aphorism of computer scientist Jan van de Snep- scheut, âTheory and practice are the same thing in theory, but they are not the same thing in practice.â The same could be said of best evidence and practice. Modalities in Complementary and Alternative Medicine Josephine P. Briggs, National Center for Complementary and Alternative Medicine Briggs expressed the National Center for Complementary and Alter- native Medicineâs (NCCAM) commitment to integrative health care. NCCAMâs mission is to build the evidence base for complementary and alternative medicine interventions. The NIH core principles that govern this mission include rigorous peer review, investigator-initiated science, and partnerships. NCCAMâs annual budget of approximately $122 mil- lion constitutes only about 4.1 percent of all NIH expenditures. However, she reported that the centerâs programs generate a significant share of interest and excitement. All of NCCAMâs large programs are operated through partnerships; partners include other NIH institutes and centers, other federal agencies, such as the Agency for Healthcare Research and Quality (AHRQ), and private-sector organizations. The programs occupy four spheres: basic science, translational re- search, efficacy studies, and effectiveness research, as shown in Figure 4-10. Basic science consumes roughly half of NCCAMâs resources. Cur- rent investigations in this area are evaluating the neuroscience of medita- tion, the biology of the placebo effect or expectancy effect, the neurobiological correlates of acupuncture, and neuroplasticity. An NCCAM-funded study in neuroplasticity, for example, demonstrated that a stroke patient imagining moving an arm affected by stroke increased movement and function in the affected arm. This may support the hy- pothesis that mental practice helps build neurocircuits for moving that arm (Page et al., 2007).
SCIENCE 105 Basic Translational Efficacy Effectiveness science research studies research FIGURE 4-10 NCCAMâs mission: Building the evidence base for integrative medicine. Other basic science research includes investigation of natural foods and other natural products; epigenetics, including effects of stress on gene expression; and the role of prebiotics and probiotics in health. Al- though the basic science portfolio is thriving, NCCAMâs limited funding means that these studies have the lowest grant application funding rates. Efficacy studies make up about 12 percent of NCCAMâs budget, in- cluding its signature RCTs of natural products and dietary supplements. For example, these investigations so far have found: no impact of Echi- nacea on the severity or frequency of head colds; positive substantial impacts of fish oil and omega-3 fatty acids on cardiovascular risk factors; and mostly negative results for ginkgo biloba, glucosamine and chondro- itin, and St.-Johnâs-wort. The results of these studies tend to affect con- sumer use of the products; for example, St.-Johnâs-wort, which was widely used in 2002, is no longer among the top 20 nonvitamin, nonmin- eral, natural products used by adults, as reported by the 2007 National Health Interview Survey. Sometimes these types of studies lead to addi- tional research. Further analysis of St.-Johnâs-wort revealed that the ac- tive ingredient hypericin may not be the only active ingredient, as was initially expected. It is now thought that hyperforin, another compound in St.-Johnâs-wort, may also be an important active ingredient. Improved methodologies will be needed to evaluate complex com- plementary and alternative interventions in real-world settings. Effec- tiveness research, which is an expanding area of activity for NCCAM, is particularly ripe for new study designs. For example, use of sham inter- ventions in acupuncture studies may describe certain effects but fails to show whether acupuncture is more effective than standard care. Despite some assertions to the contrary, rigorous scientific studies can substantiate practices used in complementary and alternative medi-
106 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC cine. To illustrate, in the management of chronic lower back pain, a peer- reviewed study supports the use of yoga (Sherman et al., 2005), a sys- tematic (or Cochrane-type) review supports the use of manipulative ex- ercise (Slade and Keating, 2007), and an authoritative clinical guideline includes recommendations to consider acupuncture, massage, spinal ma- nipulation, and yoga among other approaches for patients who do not improve with self-care options (Chou and Huffman, 2007). Now in its 10th year, NCCAM is embarking on a strategic planning process. Briggs solicited input from stakeholders and summit participants in this endeavor. Panel Discussion Members of the panel responded to questions from audience mem- bers in a discussion that was moderated by McEwen. Selected points of discussion follow. Socioeconomic Status, the Environment, and Research In response to a question on how research on integrative medicine can better involve lower-socioeconomic status (SES) populations, Adler observed that one attribute of low-SES groups is that they tend not to be well organized. Greater involvement of these patients may require going to where they are, through community-based research. Green agreed and noted that the NIH, Centers for Disease Control and Prevention (CDC), and the AHRQ are all supporting community-based participatory re- search. Green noted that one disadvantage of this approach is the need for an investigator to spend adequate time building trust in the commu- nity, but an advantage is that the community can help frame high-value research questions. An audience member took issue with the notion that minority populations are reluctant to participate in research projects; his organizationâs research has shown that the main factor in nonparticipa- tion is that people were simply not asked to do so. Several panelists responded to a related question regarding the lack of research on the effects of environmental risk factors and toxins on low-SES communities, where risks are often prevalent. Gaynor observed that African American women face an increased incidence of breast can- cer as a result of increased exposure to endocrine-disrupting chemicals,
SCIENCE 107 which are found in many hair care products and cosmetics. He called for studying the relationships between endocrine disruptors and not only cancer, but also diabetes and obesity. Adler added that this point ties in with community-based research, as members of focus groups composed of inner-city residents have expressed concern about the health effects of toxins, dumpsites, and other local environmental hazards. Lifton noted that measuring the environment can be difficult. However, advances in measuring environmental effects on the body through evaluation of gene expression can be useful in recognizing harmful environmental agents. Sternberg suggested that proteomics can also be useful in analyzing bio- logical changes that result from exposures. Balancing High-Tech Interventions with Lifestyle and Environmental Changes Panel members responded to the question of how to balance high- tech medical interventions and diagnostics with the broader impact of lifestyle and environmental health changes. Sternberg said that technol- ogy should be used as an advantage in advancing integrative medicine. She noted that high-tech methods can be used to assess the effects and health outcomes of lifestyle changes. Adler commented that skyrocketing health care costs create incentives to seek relatively inexpensive, low- tech interventions. Lifton mentioned the possibility of greater investment in disease prevention strategies, such as through changes in Medicare reimbursement. Shifting the Paradigm Panelists were asked to discuss how alternatives to RCTs could be developed. Green cited growing interest in this topic, as a result of the recent economic stimulus legislation, which contained support for com- parative effectiveness research. He mentioned that large clinical trials often produce weak results; in the early 1980s, for example, the Multiple Risk Factor Intervention Trials (âMR. FITâ), a multi-intervention RCT, sought to reduce coronary heart disease in high-risk men. When no statis- tical difference in coronary deaths was found between the experimental and control groups, researchers sought to undertake subgroup analyses. Biostatisticians objected to analyzing the subgroups because they were
108 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC not randomized, and subgroup analysis fell immediately into disrepute. Green suggested that subgroup analysis nevertheless offers the greatest potential to understand variations in moderating variables. Briggs agreed with Green, indicating that RCT methods, developed in order to determine the efficacy of drugs, may not be sufficient to ex- amine other types of interventions. Nor are they suited to assess the ef- fectiveness of interventions in the real world, given all of the forms of heterogeneity in the U.S. population. Green urged that the different dis- ciplines along the biopsychosocial spectrum resist the centrifugal force that pulls them into silos, and instead develop more transdisciplinary re- search approaches. Gaynor commented that, in matters of protecting public health, the threshold for public policy interventions should be the precautionary principle, rather than 100 percent proof of effectiveness. The last half century could have been better spent in a war on carcinogenesis than in the war on cancer, he suggested. For example, much time was lost in ac- cepting the tobacco industryâs persistent claim that cigarettes were not proven to cause cancer. Adler added that a market model is lacking for making disease prevention profitable, and a better alignment of incen- tives is needed. PRIORITY ASSESSMENT GROUP REPORT 2 Advancing the Science Base Lifton provided the report for the priority assessment group on ad- vancing the science base. This summary includes the priorities discussed and was presented by the assessment group to the plenary session for its discussion and consideration; these priorities do not represent a consen- sus or recommendations from the summit. The assessment group began with a discussion of one challenge that had resonated throughout the summitâone of definitions. The group noted that a difference exists between integrative medicine and an inte- grated system. The former involves the care of individual patients, while 2 See Chapter 1 for a description of the priority assessment groups. Participants of this assessment group included Bruce McEwen (moderator), Richard Lifton (rapporteur), Donald Abrams, Kenneth Brigham, Margaret Chesney, Gary Deng, Wayne Jonas, Lixing Lao, Patrick Mansky, Rustum Roy, and Alan Trachtenberg.
SCIENCE 109 the latter extends from basic science discovery all the way through the mechanisms of delivery and organization of care and to broad public health interventions. In building the evidence base for integrative medicine, a pivotal task is establishing causal vectors. Genetics and genomics, in particular, have the capacity to demonstrate causation; within the next 3 to 5 years, these disciplines may revolutionize the understanding of some of the funda- mental causes of health and wellness. In advancing this area of science, one participant noted the importance of recognizing biological variations among individuals when it comes to assessing effectiveness research on interventions. Certain areas of biomedicine are not yet accorded sufficient attention among researchers, despite their importance and prevalence in popula- tion health. Areas that the group identified as requiring additional re- search and understanding were fatigue, the link between beliefs and biology, and the science of achieving behavior change. Fatigue is very prevalent in chronically ill patients, but scientists have not yet developed an understanding of its biologic foundations, nor is there funding support for research in this area. One disease where fatigue has received consid- erable attention is breast cancer. Findings in the area of fatigue should also be extended to increase the understanding of the role it plays in other conditions and diseases. An increased understanding of the link between beliefs and biology, as described by Sternberg and others throughout the summit, and the ef- fects of belief on physiology and human health are fundamental areas that the group suggested warrant additional research. Progress is also needed in the development of evidence-based inter- ventions for effective behavior change. Behavior change is a key strategy for preventing a very large number of diseases, but strategies for accom- plishing it lack a great deal of empirical evidence. Additional research is required to improve the understanding of the various components of be- havior change, including the effects of motivation, group support, and education. During the discussion with the audience, participants offered additional insight into ways to advance the science of behavior change. One participant suggested research involving a qualitative perspectiveâ that is, asking individuals about their experiences and what meaning they attribute to behavioral changeâmay be a promising line of inquiry. An- other suggestion was to apply the stages-of-change model used in psy- chology, which adjusts interventions to the individualâs progress along the continuum from precontemplation to action.
110 INTEGRATIVE MEDICINE AND THE HEALTH OF THE PUBLIC To advance the science base of integrative medicine, including the priority areas identified for additional research, the assessment group determined that the most important set of key actors is research funders, including the NIH and private foundations. The role of funders in re- search is vital to providing incentives for shifts in research approaches. The group also noted that, while the biomedical research workforce has a strong basis, it requires more training in integrative human biology, be- havior, and physiology. This additional training would show researchers how their work can apply to individual patients. One audience member reiterated this point and noted the importance of exposing doctorate-level bioscience students to the physicians and patients who ultimately use and benefit from their research.