Over the last several decades, numerous preventive and therapeutic interventions have been introduced with the aim of helping people live longer and improve their overall quality of life. Such programs typically fall into four broad categories (see Compas et al., 1998): (1) interventions designed to decrease health risk behaviors, such as alcohol and substance abuse or smoking, or to increase health-promoting behaviors, such as exercise and following a healthy diet; (2) interventions aimed at facilitating effective coping with chronic or life-threatening diseases and conditions, including cancer, HIV/AIDS, asthma, diabetes, arthritis, and stroke; (3) interventions to help manage specific symptoms or problems, such as chronic headaches, back pain, and abdominal pain; and (4) interventions addressing psychopathologies such as bulimia nervosa, anorexia nervosa, and body dysmorphic disorder.
The first two intervention categories emphasize behavioral and social aspects of illness and will be the principal focus of this section. The majority of research on interventions, moreover, has focused mostly on modifying health-risk behaviors. In this report we describe additional ways that behavioral and social factors can be mobilized to improve health, prevent illness, and enhance quality of life. As research identifies new ways that social and behavioral factors affect health, it creates new targets for intervention research and practice.
Preventive and therapeutic interventions can be directed at different levels. At the individual level, interventions activate internal, or psychological resources possessed by the individual and focus on teaching indi-
vidual skills or strive to change individual behavior. At the social level, interventions attempt to bring to bear the broader resources deriving from contacts with the individual's family, friends, or social network or to change behavior patterns in family groups. At the organizational level, interventions are implemented in specific settings or units such as work sites or schools and target group change. Interventions at the population level are actions targeted at entire communities, towns, or states. In practice, intervention programs can be directed at one level at a time or can cross levels (Sorenson et al., 1998). Indeed at any one time, for example, public health efforts could promote better diets at the community level, schools could offer heart healthy lunches instead of the current fast food options, and individuals could be advised by their primary care providers to reduce fat in their diets.
Behavioral and social interventions can target either prevention or treatment, although these goals frequently coalesce in practice. For example, altering diet or increasing exercise can prevent the initial onset of cardiovascular problems, improve recovery, and prevent reoccurrence of cardiovascular problems. In general, interventions aimed at altering health risk behaviors have both preventive and treatment effects.
Successful intervention programs function on multiple levels (Sorensen et al., 1998). The benefits of targeting individuals at high risk due to their previous or current behavior, such as heavy cigarette smokers, or to a genetic susceptibility such as that of cholesterolemia, is apparent. However, clinical models that intervene with only high-risk individuals miss the potential for preventing disease by addressing other underlying causes contributing to elevated risk. When underlying causes of illness, such as low socioeconomic status are widely distributed in segments of the population, small changes at the population level are likely to have significant effects on overall population-level health. Indeed, many of the new social risk factors, including poverty and social isolation, are better addressed at the family, organizational, or population level than at the level of the individual. Similarly, when risk, such as widespread physical inactivity and overweight, is widely distributed, small changes at the population level to encourage activity (Chesney et al., 2001) are likely to yield greater improvements in the population-attributable risk than larger changes among a smaller number of high-risk individuals (Velicer et al., 1999).
The success of many health-related activities depends on the decisionmaking competence of the individuals involved. In their day-to-day lives people need to make good choices about diet and exercise, about the safety of their homes and vehicles, about the management of alcohol and anger, and about how to monitor their health status. When problems arise, they must decide when and how to present themselves to health care professionals as well as which treatments to follow. As practitioners, health care
professionals must present options in clear and effective ways and protect patients from unwarranted pressures. Health care professionals must also be responsive to cultural diversity in how to work effectively with individuals in the implementation of intervention programs (Kleinman, 1981, 1989).
EXPERIENCE WITH INTERVENTIONS
The following illustrates a range of individual, family, organizational, and population-level interventions that have been undertaken and is not intended to be a comprehensive review. The intent is first to describe an area, smoking cessation, where extensive work has been done to show the scope of social and behavioral interventions possible. Then, attention will be given to newer areas of intervention that show promise. Social and behavioral efforts to prevent smoking initiation or promote cessation have been studied for several decades (Warner, 2000). Extensive reviews document the success of these programs at the individual, school, and community or population levels.
Smoking prevention and cessation research illustrates a domain in which multilevel approaches (Warner, 2000) have been particularly effective. Extensive programs for children and teens have tried to prevent youth initiation of tobacco use through school education and counteradvertising campaigns in the media (U.S. DHHS, 1994). A meta-analysis of school interventions aimed at reducing smoking found that effect sizes were largest for interventions that focus on social reinforcement for the target behavior, moderate for those with either a developmental orientation or a focus on changing social norms as a way to influence the target behavior, and small for interventions with a health education focus (Bruvold, 1993). The effect of price increases (taxes) has also been shown to discourage youth smoking (see subsequent section, “Organization- and Population-Level Interventions ”).
With regard to adult smoking cessation, individual-level programs combining behavioral strategies for quitting with such pharmacological agents as nicotine patches continue to be at the forefront of research and practice. A challenge to these interventions has been maintenance of the positive changes that have been observed following initial training programs (Compas et al., 1998). Smoking cessation is an area that has not only experienced this challenge but where studies to identify effective strategies have been conducted. For example, one study examined alternative maintenance strategies in 744 adults from a health maintenance organization smoking cessation program (Stevens and Hollis, 1989). Participants who
achieved smoking cessation (79 percent) were randomly assigned to relapse prevention skills training, group discussion, or no further treatment. Group discussion and no-further-treatment conditions were equivalent in effectiveness (34 percent and 33 percent abstinence after one year), while the skills-training group was significantly superior (41 percent). Further efforts are needed to determine maximally effective strategies to maintain behavior change.
Formal cessation programs commonly succeed in helping 15-25 percent of participants to quit (U.S. DHHS, 1996; Warner, 2000), a figure dramatically higher than all other tobacco control interventions. Nonetheless, it is the case that relatively few smokers participate in these programs, and the vast majority of smokers who quit do so without the aid of a formal program.
Like smoking cessation, there is a portfolio of interventions designed to change other risk behaviors, including diet, physical inactivity, and alcohol and substance use. Many of these interventions are important both to prevent illness in the healthy and to prevent recurrence or delay illness progression in those who are managing chronic illness, such as coronary heart disease. This brings us to the second category of behavioral interventions, involving coping with chronic illness. This is a particularly important area, given the growing segment of the population that lives with chronic illness. Psychological interventions have shown considerable promise in the management of cancer. These interventions have been shown to help individuals manage the side effects of chemotherapy, and there is also evidence that psychosocial interventions can increase disease-free intervals and length of survival for cancer patients (Compas et al., 1998). Moreover, short-term psychiatric group intervention was associated with long-term changes in the natural killer cell (NK) system in a group of patients with newly diagnosed melanoma and good prognoses (Fawzy et al., 1990). At six months, 100 percent of the intervention group showed increases in CD 16 NK cells, 74 percent showed increases in CD 56 NK cells, and 94 percent showed increases in Leu-7 large granular lymphocytes. These changes indicate a consistent increase in the number of NK cells, seemingly in response to the intervention, suggesting that the NK cells' system might be responsive to psychological or behavioral influences. It remains to be determined whether these perturbations in cell immunity have downstream health consequences.
Some evidence supports the effectiveness of social interventions for prolonging survival of cancer patients. One study assessing the effect of group therapy on patients with metastatic breast cancer, for example, found that those participating in weekly group therapy for a year not only experienced reduced anxiety, depression, and pain but survived significantly longer than did controls—by an average of nearly 18 months, measured at a 10year follow-up (Spiegel et al., 1989). However, other studies have not
found positive results. Another study assessing the effect of several types of “supportive” group therapy on breast cancer patients observed no measurable psychological benefit of participation in the group programs compared to routine control care and found no significant difference in survival time (Gellert et al., 1993). It is unlikely that an intervention failing to provide psychological benefit would yield survival advantage.
Effective psychosocial treatments aimed at pain management not only recognize the importance of biological factors but also emphasize the influences that psychological factors (e.g., anxiety, depression, perceived control) and social factors (e.g., family and work environments) can have on the experience of pain (Compas et al., 1998). Several approaches (e.g., cognitive behavioral therapy, operant behavioral therapy, and biofeedback training) have proven efficacious for managing rheumatic diseases, chronic pain syndrome and low back pain, migraine headaches, and irritable bowel syndrome. Considerable evidence supports the effectiveness of cognitive behavioral therapy for reducing bulimia nervosa, showing roughly 80 percent reduction in binge-purge episodes and 50 to 60 percent of patients achieving complete remission (Craighead and Agras, 1991).
Medically prescribed and supervised physical activity forms the keystone of cardiac rehabilitation, and regular exercise by patients with coronary artery disease is associated with reductions in mortality from all cardiovascular causes except sudden death (Naughton, 1992). The addition of psychosocial treatments to standard cardiac rehabilitation regimens also reduces morbidity, psychological distress, and some biological risk factors (Linden et al., 1996). For example, interventions combining psychosocial strategies such as stress management and behavioral risk factor reduction, including exercise, have been shown to reduce morbidity and favorably affect the index of myocardial ischemia in cardiac patients (Blumenthal et al., 1997). Despite such results, however, as few as 10 percent of all eligible patients who could potentially benefit from cardiac rehabilitation services actually participate in formal rehabilitation programs due to cost, inconvenience, or lack of motivation (Wenger et al., 1995). Such findings underscore the future importance of clinical management of disease that effectively integrates current medical regimens with knowledge of optimal motivational strategies for patients.
Family and Network Interventions
Chronically ill patients, especially with life-threatening diseases like cancer or AIDS, must contend with a series of stressful life events. Evidence suggests that availability of or increase in social support during times of stress may lessen or prevent mood disturbances, thereby improving chances of recovery and survival from cancer (Koopman et al., 1998). Overall
social isolation is associated with a greater than twofold elevation in the relative risk of all-cause mortality, comparable to that associated with cigarette smoking or elevated serum cholesterol (House et al., 1988; see also Chapter 5). Although many studies suggest that the presence of social networks reduces mortality risk, most studies are limited by relying on onetime assessments of social support to predict disease outcome years later or use of inadequate proxy measures of social support (Spiegel and Kato, 1996). Studies are needed that systematically evaluate the benefits of intervening at the family or social network level. The studies of group support interventions suggest that such an approach would be worthwhile.
One program providing coping effectiveness training for men living with HIV includes an explicit social intervention component (Chesney et al., 1996). Individuals identify their support networks and characterize the persons in their networks into one of two groups of support providers: those who primarily provide problem-focused support, such as advice, and those who primarily provide emotion-focused support, such as listening and understanding. Although measures of the effectiveness of separate components in this program are not available, exit interviews with program participants indicate that they find the social intervention component to be especially helpful.
Families have been targeted in programs aimed at changing health risk behaviors of children. The involvement of at least one parent in programs addressing childhood obesity typically increases the effectiveness compared to controls where no parents are involved (Epstein et al., 1994). Evidence from one study comparing standard behavioral treatment with social support strategies suggests that involvement of friends as well as family members increases the effectiveness of intervention substantially (Wing and Jeffery, 1999). Of those recruited alone, 76 percent completed treatment and 24 percent maintained their weight loss in full from months 4 to 10, whereas of those recruited with friends 95 percent completed treatment and 66 percent maintained their full weight loss.
In general, however, families and proximal social networks, which are central to human health (see Chapter 5), have received comparatively little attention in the intervention realm. How spouses, parents, and children can contribute to each other's effective health practices (e.g., diet and exercise, adherence to treatment regimens, avoidance of harmful behaviors) is an important target for future inquiry.
Organization- and Population-Level Interventions
The classic examples of successful community interventions are the North Karelia Project (Pietinen et al., 1996; Puska and Koskella, 1985), the Stanford Three Community Study (Altman et al., 1987; Farquhar et al.,
1977), and the Stanford Five-City Project (Farquhar et al., 1990; see also Sorensen et al., 1998). All three targeted change and risk factors for coronary heart disease, including high blood pressure, elevated blood cholesterol levels, cigarette smoking, and obesity. The level of change observed in the North Karelia and Stanford Three Community Study, however, generally has not been replicated in subsequent community-based intervention trials. It should be noted that the level of motivation was high in North Karelia because of its identification as having the highest heart attack risk worldwide. Although community interventions have demonstrated that they can change health behaviors, the effectiveness and interactions among separate components of such programs are not well understood.
Work sites are now considered key channels for delivery of interventions designed to change behavior to prevent diseases and promote health among adult populations. Key targets for these workplace-based interventions include smoking cessation, improvement in diet, and physical activity (Sorensen et al., 1998). Other current lines of workplace research having potential to inform work site interventions are studies of stress and health risks in repetitive work and supervisory monitoring work (Lundberg and Johansson, 1999) as well as stress responses to low-status jobs and their links to musculoskeletal disorders (Lundberg, 1999). It is also important to note that, while programs in institutions have proven effective where studied as part of a specially funded effort, general adoption of successful programs into work site employee programs or school curricula has not frequently occurred.
Interventions can be classified according to increasing orders of coerciveness (e.g., degree to which they force behavior change). In the context of tobacco control the most coercive levels pertain to laws and regulations (Warner, 2000). These have contributed, along with other interventions, to considerable progress in tobacco control, with prevalence of smoking dropping from approximately 45 percent in 1963 (U.S. DHEW, 1964) to 25 percent in 1997 (CDC, 1999). With regard to early prevention, even modest increases in cigarette tax rates have been shown to reduce youth smoking (National Cancer Institute, 1993; Warner et al., 1995; Chaloupka and Warner, in press). Although the impact of rising prices is less evident for adults, smoking in adulthood is also responsive to price (Chaloupka and Warner, in press). The other principal regulatory intervention that affects smoking by adults is prohibition against smoking in public places. Ironically, indoor air laws were implemented for other reasons (i.e., to protect nonsmokers from the dangers of tobacco smoke), but they have along the way decreased smoking in the aggregate (Brownson et al., 1997) and for youth helped establish a nonsmoking social norm. These policy-level interventions warrant consideration in other public health problems, although lessons from tobacco control should proceed with caution (Warner, 2000).
Largely due to the human genome project (Pennisi, 2000) and advances in diagnostic testing, it is increasingly possible to identify people 's risk for certain diseases and conditions at earlier stages. For example, isolation of the BRCA1 gene, which affects susceptibility to breast and ovarian cancer, has led biotechnology companies to market genetic tests (see, e.g., Lerman et al., 1997). Early identification of vulnerability, however, can have mixed consequences. In addition to introducing the possibility of early treatment and possible cure, early diagnosis can discourage patients from seeking treatment. Cancer patients can face ostracism, dismissal, and hostility from others (Feldman, 1986), and awareness of this discrimination among patients living with chronic conditions such as HIV and cancer motivates many to choose not to disclose their status, even to health care providers (Chesney and Smith, 1999).
Past medical and psychosocial interventions have focused largely on the standard risk factors for chronic illness, such as inactivity, smoking, and unhealthy diet. These interventions, when studied and found to be effective, do not find their way into health plans for the individual. Psychosocial interventions for coping with chronic illnesses, such as cancer, or selfmanagement of such chronic conditions as arthritis (Lorig et al., 1998) are also effective in improving functioning and quality of life; however, these also are not supported through health plans. There is a need to translate research into practice and to study the best mechanisms by which to achieve this objective.
This report highlights new social risk factors, including social isolation and lower socioeconomic status (SES). Apart from the group interventions for cancer or other conditions discussed earlier, little research has been conducted on social-level interventions. The effective strategies for building support systems or networks have not been identified. Nor have there been studies to evaluate interventions designed to address the risk associated with lower SES. As detailed in Chapter 8, racial/ethnic and SES variation is particularly salient in the context of smoking. For example, smoking prevalence is particularly high among Native Americans and Alaskans compared to other ethnic minorities (CDC, 1999). Since the antismoking campaign began in 1964, the change in smoking prevalence has also varied dramatically by educational level, falling nearly two-thirds among college graduates but only by one-sixth among those without a high school diploma (CDC, 1999). “In short, smoking has moved from an ‘egalitarian' burden in the mid-1960s to a heavy weight today on those of low socioeconomic status” (Warner, 2000). This widening gap has received limited attention and clearly intersects with the problem of reducing social inequalities in health (see Chapter 7).
In addition to the needed emphasis on particular at-risk segments of society, there is also need for giving increased attention to interventions in families, organizations, and communities (see Sorensen et al., 1998). Given the growing awareness of the critical importance of social and contextual factors in health described in this report (see Chapter 3, Chapter 5, Chapter 6, and Chapter 7), consideration of a broader array of interventions is appropriate. The following elaborates this need.
Behavioral and psychosocial interventions historically were proposed as an alternative to biomedical approaches. Among the exceptions has been research on behavioral interventions to increase adherence to medical regimens. There is, however, a need to extend the work at the interface between behavioral and biomedical approaches. More sophisticated behavioral theories and strategies could, for example, be brought into play in studies of adherence, including the viewpoint that this is another health behavior that must be maintained over time. Similarly, biomedical approaches could be integrated with behavioral strategies to more effectively treat obesity, eating disorders, or cigarette smoking. Research investigating the efficacy of combined behavioral and pharmacological approaches often demonstrates a superiority for the combined approaches over either approach alone. Biobehavioral strategies could be extended to interventions directed at coping with chronic conditions and recovery from acute illness.
The behavioral interventions discussed in this chapter have been manual based but often delivered by professionals to individuals or groups, either separately or within work sites or schools. New technologies exist at various levels to offer innovative strategies at all levels. At the individual level, alpha-numeric pagers can prompt individuals to adhere to specific medication dosing schedules or to monitor blood glucose levels. Computers and telecommunications offer new opportunities for interactive decision making between patients and providers or for support among patient groups.
Similar programs need to be developed at the organizational and population levels. These programs would not need to treat all persons in the population the same way. Technology and self-selection would permit individuals or groups to obtain information tailored to their needs through community-level channels. Research indicates that such tailoring significantly improves the chances that recipients will thoughtfully consider the information and move toward self-assessment and intention to change (Kreuter et al., 1999).
People often have limited time, patience, and cognitive capacity for learning about health-related decisions. As a result, the content of health communications must be selected with great care. Risk, policy, and decision analysts have developed approaches to that selection process for communicating with professional audiences. These must be adapted to the
needs of laypersons, recognizing the diversity of the personal circumstances, values, and knowledge levels that they may have (Fischhoff, 1999; Lipkus and Hollands, 1999; Schwartz et al., 1999; Vernon, 1999).
Collectively, these trends underscore the importance of the multilevel approach to health interventions. Here again the experience in smoking prevention and cessation is instructive. Extensive review of programs in this area (Warner, 2000) as well as their translation to clinical practice (U.S. Public Health Service Report, 2000) underscores the importance of grounding intervention programs in solid science and of approaching tobacco control via multipronged strategies (e.g., educational programs, media advertising, counseling, social support, pharmacotherapies for nicotine dependence, cigarette taxes, prohibition against smoking in public places). These various levels almost certainly reinforce each other, although further research is needed to evaluate the effectiveness of multiple intervention channels running simultaneously.
FUTURE RESEARCH NEEDS AND DIRECTIONS
Behavioral and social interventions have traditionally emphasized acute change. The implicit assumption has been that the individual, group, community, or population had some adverse approach to health that could be corrected with a brief program, often lasting 8 to 10 weeks. Thus, for example, the overweight people in the community would self-assess their eating habits and participate in that community program for a period of weeks, change their eating behavior, and sustain that change for life. Employees at the work site would start exercising and sustain that behavior. Patients coping with HIV/AIDS or cancer would learn new coping skills and apply them as their progressive illnesses presented new challenges over the years. There are some health behaviors that fit this acute intervention model better than others. Most health behaviors and certainly approaches to managing chronic conditions require a different model. There is an urgent need for new models of behavior change that address sustaining effort in the face of forever-changing personal, social, and environmental circumstances. The failure of our current models to address the dynamic of time and circumstance may explain why some models, such as the transtheoretical model of behavior change or “stages of change,” while having considerable intuitive appeal, have proven not to be widely applicable. In recent years research on smoking (Herzog et al., 1999) and diet (Jeffery et al., 1999) found no association between stage and health outcomes achieved with intervention. (See, however, “Behavioral Factors” in Chapter 2 for more supportive evidence.)
In the early 1980s the National Heart, Lung, and Blood Institute and the National Cancer Institute suggested a sequence of research phases for
development of programs effective in modifying behaviors (NHLBI, 1983). These phases range from hypothesis development (Phase I) and methods development (Phase II) through controlled intervention trials (Phase III) to studies in defined populations (Phase IV) and demonstration research (Phase V). The need to improve our understanding of mechanisms at the physiological, psychological, social, and population levels documented in this report, however, suggests that funding should be available for such integrative research. Pilot studies and small-group research must precede largerscale studies. It is also important to note that some research problems and certain intervention approaches are optimally evaluated by randomized controlled clinical trials. Other problems and interventions, particularly those focusing at the organization or community level, may require new designs. Additional methods for demonstrating feasibility, sustainability, and cost effectiveness must be developed for population-level interventions.
There are four common features of population-level interventions that make traditional randomized control trials difficult, if not impossible:
The program contains multiple interventions acting simultaneously.
Participants are self-selected.
It is unknown whether the decision criteria used by volunteers are the same or different from those who do not participate.
A control group from the target population cannot be assembled.
Regarding the first feature, high-dimensional factorial experiments are required to assess the impact of each component intervention acting alone or in combination with other interventions. When the impact of a package of interventions is the primary concern, randomization in complex factorial designs is wasteful in time (taking many years to assess the effects of subsets of the full package) and resources.
With respect to the second feature, random assignment of persons to treatment and control groups can eliminate from consideration one or more central phenomena that require study. For example, in evaluating methadone maintenance programs, the target population is the set of chronic heroin users in a given community. Part of what one wants to understand is the characteristics of those who self-select to come into the program in the first place.
The third feature is related to the second in that it is useful to study the decision-making processes of volunteers versus nonvolunteers. Feature four is obviously outside the domain of a conventional randomized clinical trial. The investigator must make comparisons between the responses of voluntary participants and the known natural history of a given disease assessed in other studies.
Taken together, these attributes mean that multiple criteria should be
used to evaluate the effectiveness of such interventions and that intervention outcomes should be compared with those of other programs or with outcomes absent of interventions (see Singer, 1986). Moreover, an administrative structure that is minimally intrusive to the patient may be of as much importance as the treatment itself. Randomized trials that included variations in administrative structure would be prohibitively costly. A more effective strategy is to combine performance-based ratings of program organization and implementation with outcome assessments (Singer, 1986). The complexity of such intervention programs suggests that what is deemed effective should be installed initially and that one or more components should be adjusted at regular intervals, based on outcome indices.
It should be recognized that intervention programs will seldom if ever be based on complete understanding of mechanisms at all levels described in this report. In this respect, intervention research is analogous to brain research, which often focuses on specific mechanisms in a particular group of cells or functional system without accounting for all interactions of that component with other parts of the brain or all processes within that component. Similarly, intervention research needs to selectively focus on specific mechanisms. Research designs should specify the mechanisms of interest and measures for assessing implementation of the designed intervention as well as health outcomes at appropriate levels.
The great successes of the past in convincing people to take better care of themselves have generally occurred when there was broad social consensus on appropriate behavior (even among those who were still acting otherwise). The persuasive approach, however, is less acceptable where such consensus is lacking. Individual circumstances may be sufficiently different that distinct courses of action may be indicated or there may be conflicting or culturally specific interpretations of the facts that might not be subject to reconsideration. The educational components of interventions must accommodate the full contextual setting. Overall, the need for longitudinal data, heterogeneous populations, and multiple intervention levels calls for integrative and linked studies to illuminate broad interventive targets, as described in this chapter. Development of an overall strategy and coordination of separate efforts will be of utmost importance. Across the entire spectrum of interventions, there is a great need to increase dissemination of preventive and treatment strategies that are found to be effective to the larger public.
The National Institutes of Health should support a new generation of intervention studies with the following emphases:
development of strategies for extending successful social and behavioral interventions to more heterogeneous populations, including those focused on prevention via early identification of persons at risk: the design of interventions that take into account the dynamic and more chronic aspects of health should be given priority;
expansion of implementation and dissemination activities so as to reduce the gap between research progress and practice;
development of an overall strategy for intervention research that integrates behavioral, psychosocial, and biomedical approaches and spans multiple levels, from the individual to the societal;
intervention research that capitalizes on new opportunities created by technological innovation should be given priority.
Altman DG , Flora JA , Fortmann SP , Farguhar JW. 1987 “The cost-effectiveness of three smoking cessation programs” American Journal of Public Health 77/2 : 162-165 .
Blumenthal JA , Jiang W , Babyak MA , Krantz DS , Frid DJ , Coleman RE , Waugh R , Hanson M , Appelbaum M , O'Connor C , Morris JJ. 1997 “Stress management and exercise training in cardiac patients with myocardial ischemia. Effects on prognosis and evaluation of mechanisms ” Archives of Internal Medicine 157/19 : 2213-2223 .
Brownson RC , Eriksen MP , David RM et al. 1997 “Environmental tobacco smoke: Health effects and policies to reduce exposures” Annual Review of Public Health 18 : 163-185 .
Bruvold WH. 1993 “A meta-analysis of adolescent smoking prevention programs: A case study in Massachusetts” American Journal of Public Health 83/16 : 872-880 .
CDC (Centers for Disease Control and Prevention) . 1999 “Cigarette smoking among adults–United States, 1997” Morbidity and Mortality Weekly Report 48 : 993-996 .
Chaloupka FJ , and Warner KE. In press “Economics of smoking” in Culyer AJ , Newhouse JP (eds.) Handbook of Health Economics ( Amsterdam : Elsevier ).
Chesney MA , Smith AW. 1999 “Critical delays in HIV testing and care: The potential role of stigma ” American Behavioral Scientist 42/7 : 1162-1174 .
Chesney MA , Folkman S , Chambers D. 1996 “Coping effectiveness training for men living with HIV: Preliminary findings” International Journal of STD and AIDS 7/suppl.2 : 75-82 .
Chesney MA , Thurston RC , Thomas KA. 2001 “Creating social and public health environments to sustain behavior change: Lessons from obesity research” in Schneiderman N , Gentry J , da Silva JM , Speers M , Tomes H (eds.) Integrating Behavioral and Social Sciences with Public Health ( Washington DC : American Psychological Association ) pp.31-50 .
Compas BE , Haaga DA , Keefe FJ , Leitenberg H , Williams DA. 1998 “Sampling of empirically supported psychological treatments from health psychology: Smoking, chronic pain, cancer, and bulimia nervosa” Journal of Consulting and Clinical Psychology 66/1 : 89-112 .
Craighead LW , Agras WS. 1991 “Mechanisms of action in cognitive-behavioral and pharmacological interventions for obesity and bulimia nervosa” Journal of Consulting and Clinical Psychology 59/1 : 115-125 .
Epstein LH , Valoski A , Wing RR , McCurley J. 1994 “Ten-year outcomes of behavioral family-based treatment for childhood obesity” Health Psychology 13/5 : 373-383 .
Farquhar JW , Fortman SP , Flora JA , Taylor CB , Haskell WL , Williams PT , Maccoby N , Wood PD. 1990 “Effects of community-wide education on cardiovascular disease risk factors: The Stanford Five-City Project.” Journal of the American Medical Association 264/3 : 359-365 .
Farquhar JW , Maccoby N , Wood PD , Alexander JK , Breitrose H , Brown BW Jr , Haskell WL , McAlister AL , Meyer AJ , Nash JD , Stern MP. 1977 “Community education for cardiovascular health” Lancet 1(8023) : 1192-1195 .
Fawzy FI , Kemeny ME , Fawzy NW , Elashoff R , Morton D , Cousins N , Fahey JL. 1990 “A structured psychiatric intervention for cancer patients: II . Changes over time in immunological measures” Archives of General Psychiatry 47/8 : 729-735 .
Feldman FL. 1986 “Female cancer patients and caregivers: Experiences in the workplace ” Women and Health 11/3-4 : 137-153 .
Fischhoff B. 1999 “Why (cancer) risk communication can be hard” Journal of the NationalCancer Institute Monographs 25 : 1-7 .
Gellert GA , Maxwell RM , Siegel BS. 1993 “Survival of breast cancer patients receiving adjunctive psychosocial support therapy: A 10-year follow-up study” Journal of Clinical Oncology 11/1 : 66-69 .
Herzog TA , Abrams DB , Emmons KM , Linnan LA , Shadel WG. 1999 “Do processes of change predict smoking stage movements? A prospective analysis of the transtheoretical model” Health Psychology 18/4 : 369-375 .
House JS , Landis KR , Umberson D. 1988 “Social relationships and health” Science 241/4865 : 540-545 .
Jeffery RW , French SA , Rothman AJ. 1999 “Stage of change as a predictor of success in weight control in adult women” Health Psychology 18/5 : 543-546 .
Kleinman A. 1981 . Patients and Healers in the Context of Culture ( Berkeley : University of California Press ).
Kleinman A. 1989 The Illness Narratives: Suffering, Healing, and the Human Condition ( New York : Basic Books ).
Koopman C , Hermanson K , Diamond S , Angell S , Spiegel D. 1998 “Social support, life stress, pain and emotional adjustment to advanced breast cancer” Psycho-Oncology 7/2 : 101-111 .
Kreuter MW , Bull FC , Clark EM , Oswald DL. 1999 “Understanding how people process health information: A comparison of tailored and nontailored weight-loss materials” Health Psychology 18/5 : 487-494 .
Lerman C , Biesecker B , Benkendorf JL , Kerner J , Gomez-Caminero A , Hughes C , Reed MM. 1997 “Controlled trial of pretest education approaches to enhance informed decision- making for BRCA1 gene testing” Journal of the National Cancer Institute 89/2 : 148-157 .
Linden W , Stossel C , Maurice J. 1996 “Psychosocial interventions for patients with coronary artery disease: A meta-analysis” Archives of Internal Medicine156/7 : 745-752 .
Lipkus IM , Hollands JG. 1999 “The visual communication of risk” Journal of the National Cancer Institute Monographs 25 : 149-162 .
Lorig K , Gonzalez VM , Laurent DD , Morgan L , Loris BA. 1998 “Arthritis self-management program variations: Three studies . Arthritis Care and Research 11/6 : 448-454 .
Lundberg U. 1999 “Stress responses in low-status jobs and their relationship to health risks: Musculoskeletal disorders . Annals of the New York Academy of Sciences.
Lundberg U , Johansson G. 1999 “Stress and health risks in repetitive work and supervisory monitoring work” in Backs RW and Boncsein W (eds.) Engineering Psychophysiology: Issues and Applications ( Mahwah NJ : Lawrence Erlbaum ) .
National Cancer Institute . 1993 The Impact of Cigarette Excise Taxes on Smoking AmongChildren and Adults: A Summary Report of a National Cancer Institute Expert Panel( Bethesda MD : Division of Cancer Prevention and Control, NCI ).
Naughton J. 1992 “Exercise training for patients with coronary artery disease: Cardiac rehabilitation revisited” Sports Medicine 14/5 : 304-319 .
NHLBI (National Heart, Lung, and Blood Institute) . 1983 Guidelines for Demonstration and Educational Research Grants ( Washington DC : U.S. Government Printing Office ).
Pennisi E. 2000 “Human genome. Finally, the book of life and instructions for navigating it” Science 288/5475 : 2304-2307 .
Puska P , Koskella K. 1985 “Community-based strategies to fight smoking: Experiences from the North Karelia Project in Finland” in A Blum (ed.) The Cigarette Underworld( Mahwah NJ : Lyle Stuart Inc. ) pp.95-98 .
Pietinen P , Vartiainen E , Seppanen R , Arro A , Puska P. 1996 “Changes in diet in Finland from 1972 to 1992: Impact on coronary heart disease risk” Preventive Medicine25/3 : 243-250 .
Schwartz L , Woloshin S , Welch G. 1999 “Risk communication in clinical practice” Journalof the National Cancer Institute Monographs 25 : 124-133 .
Singer B. 1986 “Self-selection and performance-based ratings: A case study in program evaluation” in H Wainer (ed.) Drawing Inferences from Self-Selected Samples( New York : Springer-Verlag ) pp .29-49 .
Sorensen G , Emmons K , Hunt MK , Johnston D. 1998 “Implications of the results of community intervention trials” Annual Review of Public Health 19 : 379-416 .
Spiegel D , Kato PM. 1996 “Psychosocial influences on cancer incidence and progression—Review” Harvard Review of Psychiatry 4/1 : 10-26 .
Spiegel D , Bloom JR , Kraemer HC , Gottheil E. 1989 “Effect of psychosocial treatment on survival of patients with metastatic breast cancer” Lancet 2/8668 : 888-891 .
Stevens VJ , Hollis JF. 1989 “Preventing smoking relapse , using an individually tailored skills-training technique” Journal of Consulting and Clinical Psychology 57/13 : 420-424 .
U.S. DHEW (U.S. Department of Health, Education, and Welfare) . 1964 Smoking and Health: Report of the Advisory Committee to the Surgeon General of the Public Health Service( Washington DC : U.S. Department of Health, Education, and Welfare , Public Health Service , Center for Disease Control) PHS Publication No. 1103 .
U.S. DHHS (U.S. Department of Health and Human Services) . 1994 Preventing Tobacco Use Among Young People: A Report of the Surgeon General( Washington DC : U.S. Government Printing Office , U.S. Department of Health and Human Services , Public Health Service , Centers for Disease Control , National Center for Chronic Disease Prevention and Health Promotion , Office of Smoking and Health ).
U.S. DHHS (U.S. Department of Health and Human Services) . 1996 Smoking Cessation, Clinical Practice Guideline Number 18 Public Health Service , Agency for Health Care Policy and Research , Centers for Disease Control and Prevention . ( Rockville MD : AHCPR Publication No. 96-0692 ).
U.S. Public Health Service Report . 2000 “A clinical practice guideline for treating tobacco use and dependence ” Journal of the American Medical Association 283/24 : 3244-3254 .
Velicer WF , Prochaska JO , Fava JL , Laforge RG , Rossi JS. 1999 “Interactive versus noninteractive interventions and dose-response relationships for stage-matched smoking cessation programs in a managed care setting” Health Psychology 18/1 : 21-28 .
Vernon SW 1999 “Risk perception and risk communication for cancer screening behaviors: A review” Journal of the National Cancer Institute Monographs 25 : .
Warner KE. 2000 “The need for, and value of, a multi-level approach to disease prevention: The case of tobacco control” in Smedley BD and Syme SL (eds.) Promoting Health: Intervention Strategies from Social and Behavioral Research ( Washington, DC : National Academy Press , Committee on Capitalizing on Social Science and Behavioral Research to Improve the Public's Health , Division of Health Promotion and Disease Prevention , Institute of Medicine ) pp .326-352 .
Warner KE , Chaloupka FJ , Cook PJ et al. 1995 “Criteria for determining an optimal cigarette tax: The economist's perspective” Tobacco Control4 : 380-386 .
Wenger NK , Forelicher ES , Smith LK , and the Cardiac Rehabilitation Guideline Panel of the Agency for Health Care Policy and Research . 1995 Cardiac Rehabilitation( Rockville MD : U.S. Department of Health and Human Services ).
Wing RR , Jeffery RW. 1999 “Benefits of recruiting participants with friends and increasing social support for weight loss and maintenance” Journal of Consulting and Clinical Psychology 67/1 : .