National Academies Press: OpenBook
« Previous: Introduction
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

have “lives.” A partnership between these two groups offers the best chance to bridge this divide.

Further, it is clear that children should be a major focus of intervention efforts. Many of the risk factors observed in adults can be detected in childhood, such as high blood pressure, overweight, and poor respiratory function. The evidence is that interventions in early life can change the trajectory of these risk factors. A compelling case can be made for focusing attention on ensuring that children get a strong start by providing, for example, appropriate early education, immunization, injury prevention, nutrition, and opportunities for physical activity.

Opportunities for public health intervention do not end at childhood; evidence indicates that intervention efforts with adolescents, adults, and older adults can be successful if these efforts address major developmental tasks at each stage and address major sources of health risk at multiple levels. Adolescents, for example, are heavily influenced by their peers' conduct and attitudes, yet peer group norms can be influenced to improve health behaviors and attitudes, as long as these efforts are supported by consistent and complementary efforts in schools and communities.

The committee's review of all of these materials strongly suggests that interventions need to:

  1. focus on generic social and behavioral determinants of disease, injury, and disability;

  2. use multiple approaches (e.g., education, social support, laws, incentives, behavior change programs) and address multiple levels of influence simultaneously (i.e., individuals, families, communities, nations);

  3. take account of the special needs of target groups (i.e., based on age, gender, race, ethnicity, social class);

  4. take the “long view” of health outcomes, as changes often take many years to become established; and

  5. involve a variety of sectors in our society that have not traditionally been associated with health promotion efforts, including law, business, education, social services, and the media.

FINDINGS AND RECOMMENDATIONS

Overarching Recommendations

The committee finds that two fundamental issues must be addressed in developing interventions based on social and behavioral research. These include the need to address generic social and behavioral determinants of health, and the need to intervene at multiple sources of health influences (see Box 1).

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

BOX 1. Overarching Recommendations

Recommendation 1: Social and behavioral factors have a broad and profound impact on health across a wide range of conditions and disabilities. A better balance is needed between the clinical approach to disease, presently the dominant public health model for most risk factors, and research and intervention efforts that address generic social and behavioral determinants of disease, injury, and disability.

Recommendation 2: Rather than focusing interventions on a single or limited number of health determinants, interventions on social and behavioral factors should link multiple levels of influence (i.e., individual, interpersonal, institutional, community, and policy levels).

Addressing Generic Determinants of Health

As noted above, risk behaviors account for approximately half of preventable deaths in the United States. In addition, health care expenditures on chronic and preventable diseases and injury account for nearly 70% of all medical care spending (Fries et al., 1993). These data illustrate the tremendous toll, in both economic and human costs, of poor health behaviors and suggest strongly that initiatives to develop and implement effective interventions to reduce risk behaviors are an important component of efforts to improve the nation's health and to lower health care costs. Further, as illustrated above, risk behaviors are closely linked to social and economic conditions, such as economic inequality, social norms, and other forces (Kaplan, Everson, and Lynch, Paper Contribution A). Interventions are therefore needed to address these generic factors, and to broadly reduce risks across large segments of the population.

As yet, however, this need has not been met. The Robert Wood Johnson Foundation recently conducted an assessment of progress in population-based health promotion in six areas of behavioral risk (tobacco use, sedentary lifestyle, alcohol abuse, drug abuse, unhealthy diet, and sexual risk behavior), and concluded that more progress has been made in developing and implementing individually oriented interventions than in developing and implementing more environmentally focused interventions, even though the latter are necessary for greatest population impact (Orleans et al., 1999). Broad intervention strategies that address generic social and behavioral determinants of health must therefore receive greater attention.

Recommendation 1: Social and behavioral factors have a broad and profound impact on health across a wide range of conditions and disabilities. A better balance is needed between the clinical approach to disease, presently the dominant public health model for most risk factors, and research and intervention efforts that ad-

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

dress generic social and behavioral determinants of disease, injury, and disability.

Intervention at Multiple Levels

As the preceding discussion illustrates, there is substantial heterogeneity in health outcomes by social factors such as place, race and ethnicity, gender, and socioeconomic position. These findings indicate that it is not realistic to think that one might identify a single set of explanatory determinants, and thereby identify a single focus for interventions. Rather, as is pointed out across the reports in this volume, a multilevel approach to interventions is needed.

The framework for this approach is not new. A social ecological model provides a structure for intervening at multiple levels of influence, including the individual, interpersonal, institutional, community, and policy levels, as articulated in the paper by Emmons (see Paper Contribution F). This model assumes that health outcomes are the result of a dynamic interplay between multiple factors such as individuals' physiological processes; behavioral patterns and personality structures; physical, social, and economic contexts; and larger social processes that broadly influence health risks. However, interventions targeting health behavior change have a tradition of focusing first on individual factors, such as increasing knowledge, motivation, or skills related to health behavior change.

On the other hand, a multilevel approach is not meant to emphasize population- or societal-level interventions to the exclusion of individual-level approaches; rather, there is a need for integration of interventions across levels. By linking interventions, consistent intervention messages, support, and follow-up can be provided over time. Support for a multilevel approach is provided in this volume by Perry, Emmons, and Warner (see Paper Contributions E, F, and K). Perry, for example, observes that the effects of school-based smoking prevention programs are sustained when changes in the larger community are also present and when there is reinforcement of the intervention over time. Maintenance of health behavior change appears to be limited when behaviors promoted by school-based programs are inconsistent with larger community norms, so that healthy behavior change is not supported and reinforced after the program is completed.

Another excellent example of the promise offered by multilevel interventions is tobacco control, reviewed in this volume by Warner (see Paper Contribution K). The case of tobacco control is a major success story for public health. The prevalence of smoking in the United States dropped to 25% in 1997, down from about 45% in 1963, the year prior to the first publication of the Surgeon General's report on smoking and health * (U.S. Department of Health, Education,

*A major exception to this trend has occurred among adolescents, whose use of tobacco increased steadily through the first half of the 1990s before declining slightly in the latter half of the decade. In 1999, over one-third of high school students and one in eight middle school students reported using some form of tobacco in the past month (Centers for Disease Control and Prevention, 2000).

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

and Welfare, 1964). Extensive efforts to intervene exclusively at the individual level, using behavioral and pharmacological interventions, have yielded only marginal success. However, Warner, Gostin, and House and Williams (see Paper Contributions K, J, and B) all conclude that major progress in reducing the prevalence of smoking has resulted from population-wide interventions, including economic disincentives (e.g., increases in the cigarette excise tax), and laws and regulations (e.g., clean indoor air laws, minimum age of purchase laws, and regulation of advertising). These interventions are most effective at the population level when applied concurrently with individual-level interventions. The committee finds the evidence compelling that coordinated, multilevel interventions offer the greatest promise to address most public health needs.

Recommendation 2: Rather than focusing interventions on a single or limited number of health determinants, interventions on social and behavioral factors should link multiple levels of influence (i.e., individual, interpersonal, institutional, community, and policy levels).

Promising Intervention Strategies
Defining “Promising Interventions”

The literature on behavioral and social interventions that purport to reduce injury and disease and promote health is large and growing. Perhaps the most significant contribution of behavioral and social sciences to health research is the development of strong theoretical models for interventions. Not all such research, however, has been guided by sound theory. In some cases, empirical findings suggest that interventions may be effective, but without sound theoretical underpinnings, the mechanisms through which interventions operate remain unclear. The committee defines promising interventions as those that have both a theoretical basis for efficacy and empirical evidence supporting at least some parts of the theoretical model. Promising interventions can be identified by studies showing beneficial effects on select populations, accompanied by a theoretical model for supporting expansion of the intervention to other populations. Alternatively, similar promise can be demonstrated by studies that reveal only modest effects of an intervention on wider application, but where a strong theoretical basis is present for expecting improved efficacy by changing the intervention in some way. The committee notes, however, that “promising” is not the same as “proven” —proven interventions would be those already known to be efficacious on wide-scale application. The promising strategies identified by the committee are listed in Box 2.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

BOX 2. Recommendations for Intervention Strategies

Recommendation 3: Intervention strategies must attend to aspects of the social context that may hinder or promote efforts at behavioral change and health risk reduction. Modifying the social capital of communities and neighborhoods offers promise to enhance social contexts. These interventions should therefore be developed, implemented, and evaluated.

Recommendation 4: Research consistently indicates that life circumstances associated with low socioeconomic status, gender, race, and ethnic disadvantage and other high-risk circumstances are strongly associated with a wide range of poor health outcomes. Policies that take account of and enhance positive health outcomes among subgroups at high risk of poor health outcomes need to be designed and implemented.

Recommendation 5: Improvements in reproductive outcomes will require addressing social and behavioral influences on the health of women well prior to conception, in addition to the perinatal and postnatal periods.

Recommendation 6: High-quality, center-based early education programs should be more widely implemented. Future interventions directed at infants and young children should focus on strengthening other processes affecting child outcomes such as the home environment, school and neighborhood influences, and physical health and growth.

Recommendation 7: Multi-level interventions should address aspects of adolescents' social environment as they affect health outcomes, including peer norms, role models, performance expectations, social and neighborhood supports, and ties to community institutions.

Recommendation 8: Community and work site interventions and evaluations are needed to promote behavioral change, prevent injuries, reduce exposure to occupational risks, and increase healthy environments. Such interventions should include multiple points of leverage, such as individual-level attributes, social supports and social norms, family and neighborhood factors, and environmental and social policies.

Recommendation 9: Interventions to improve the health of older adults should focus on the social, environmental, and behavioral conditions that minimize disability and promote continuing independence and productive activity. Interventions that enhance the social support and self-efficacy of older adults are particularly promising. Community senior centers and community-assisted housing are two examples of interventions that are especially promising for the frail elderly.

Interventions Directed at the Social Context

The preceding discussion of multilevel interventions offers examples of the importance of attending to the social context in which people live and work in the development of health interventions. The social context is comprised of socioeconomic conditions and factors in one 's physical, social, and cultural environment that influence access to health information, social support, social networks, social norms, and cultural beliefs and attitudes regarding health. Because

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

these factors directly and indirectly influence health and health risks, interventions must address the social context as a critical element of intervention design. For example, Syme's (1978) study of people with hypertension illustrates that addressing underlying social and economic conditions appears to enhance the management of hypertension and improve the effectiveness of antihypertensive therapy. In a randomized controlled study, blood pressure was most likely to be controlled among patients receiving visits from community health workers who not only provided information about hypertension, but also discussed family difficulties, financial strain, employment opportunities, and, when appropriate, provided support and assistance.

Beyond adapting to differences in social contexts, interventions must also directly influence attributes of the social context that promote health. Sampson and Morenoff and Wallack (see Paper Contributions I and H) discuss one attribute of social contexts—social capital—that holds promise for improving population health. Social capital is defined as a characteristic of communities stemming from the structure of social relationships that facilitates the achievement of individuals ' shared goals. This includes the quality of social networks, as well as the by-products of these networks—such as shared norms and mutual trust—that facilitate cooperation. Social capital is therefore an attribute of social structures (e.g., organizations, communities, neighborhoods, extended families) rather than individuals.

The importance of social capital is best illustrated by research that demonstrates an association between neighborhood cohesiveness, stability, and measures of mutual “trust” and a range of public health outcomes, including the health and mental health status of community residents and levels of crime and violence. Communities characterized by high poverty, low residential stability, and a high percentage of single-parent households often possess low levels of social capital, and are linked to poor individual health outcomes such as coronary risk factors and heart disease mortality, low birthweight, smoking, and other negative outcomes, even when individual attributes and behaviors are taken into account. Analyses of the Alameda County Health Study reveal, for example, that self-reported fair/poor health was 70% higher for residents of concentrated poverty areas than for residents of nonpoverty areas, independent of age, sex, income, education, smoking status, body mass index, and alcohol consumption (Sampson and Morenoff, Paper Contribution I). In addition, Kawachi (1999) has found that measures of distrust correlate strongly with age-adjusted mortality at the state level, even while controlling for individual risk factors.

Further, experimental and quasi-experimental studies lend strength to the hypothesis that improvements in community socioeconomic environments lead to better individual health and behavioral outcomes. Sampson and Morenoff (see Paper Contribution I), for example, cite a study of housing project residents in Boston indicating that children of mothers randomly assigned to an experimental group that received housing subsidies allowing them to move into low-poverty neighborhoods had a significantly lower prevalence of injuries, asthma attacks, and personal victimization than children of mothers who received no special

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

housing assistance or conventional housing assistance. Because the experimental design of this study controlled for individual-level risk factors, the study lends support to the view that changes in community socioeconomic conditions can lead to improvements in individual health.

Questions remain as to why these associations persist. Sampson and Morenoff suggest that benefits stem from social processes involved in collective aspects of neighborhood life, such as information exchange, reciprocal assistance, voluntary associations, and parental support. Perhaps more importantly, the collective efficacy of communities —that is, the trust shared among residents and willingness to intervene in support of public order—reflects the potential of communities to mobilize to improve community conditions. Sampson and his colleagues have found a strong association between collective efficacy and rates of neighborhood violence, controlling for concentrations of poverty, residential stability, and other individual-level characteristics (Sampson and Morenoff, Paper Contribution I).

Wallack (see Paper Contribution H) suggests that a “lever” to enhance social capital is presented by media and marketing strategies. He notes that traditional behaviorally oriented media campaigns, while useful, have been of limited effectiveness in creating and sustaining significant behavior change. Media campaigns have the potential, he argues, to address more fundamental aspects of the social context in which health behaviors occur, most significantly by enhancing involvement in civic life and encouraging social justice, participation, and social change. While experimental evidence for this approach is scant, Wallack discusses findings from three areas of research—civic journalism, media advocacy, and photo voice—that encourage community mobilization to advance public policies that promote health. Civic journalism involves efforts by the media to engage the community in civic life by providing information and other support to increase community debate and public participation in problem-solving. Media advocacy is commonly used in combination with community organizing to advocate and draw attention to the need for improvements in public policies. Photo voice is a tool used by community members to enhance awareness of social and economic conditions in their communities that are detrimental to health. Rigorous evaluation and refinement of these strategies is very much needed.

Recommendation 3: Intervention strategies must attend to aspects of the social context that may hinder or promote efforts at behavioral change and health risk reduction. Modifying the social capital of communities and neighborhoods offers promise to enhance social contexts. These interventions should therefore be developed, implemented, and evaluated.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Understanding and Reducing Socioeconomic, Racial/Ethnic, and Gender Disparities in Health

The preceding discussion consistently notes that important health disparities exist among population groups. In particular, socioeconomic disparities in health are large, persistent, and increasing (House and Williams, Paper Contribution B). Racial and ethnic disparities are generally consistent, in that African Americans, Hispanics, and Native Americans face higher rates of age-adjusted morbidity and mortality from a number of diseases associated with behavioral and social factors. A number of important exceptions to this pattern exist; many foreign-born Hispanics and individuals of African descent enjoy better birth outcomes than American-born Hispanics and African Americans (House and Williams, Paper Contribution B). The implications of these exceptions are discussed below (see “Research Needs”). In addition, while women in the United States live 7 years longer on average than men, these additional life-years are characterized by greater levels of impairment in functional status and morbidity (Rogers et al., 1989).

As House and Williams illustrate (see Paper Contribution B), understanding these disparities requires further research and intervention into psychosocial and physiologic pathways that translate disparities into poor health, and into the broader social, cultural, economic, and political processes that determine the nature and extent of disparities. Socioeconomic status, for example, is commonly thought to exert its effects on health through discrete risk factors such as individuals' coping skills, health knowledge, health behaviors, and access to health-enhancing resources. House and Williams present compelling evidence, however, that disadvantaged socioeconomic status exerts its effects by shaping the experience of and exposure to virtually all known behavioral, psychosocial, and environmental risk factors for health. Thus, the effects of socioeconomic status on health are multiply determined, exerting its effects across all of the developmental stages, social contexts, and community attributes discussed above.

Further, House and Williams note that while racial and ethnic disparities in health are to a great extent the result of socioeconomic disparities among these groups, minority racial and ethnic status are associated with adverse health outcomes beyond those explainable by socioeconomic differences. This disparity is best illustrated by data indicating that for most causes of death and disability, ethnic minorities suffer from poorer outcomes relative to whites, even at equivalent education and income levels. House and Williams note that these disparities may result from racism and discrimination, including racism inherent in the health care system, and the ways in which racism and discrimination restrict socioeconomic and housing opportunities and elevate stress among their victims. The authors note further that because of the pervasive and insidious fashion in which low socioeconomic status, racism, and discrimination shape virtually all risks for poor health and poor health outcomes, interventions to improve access to medical care and reduce behavioral or psychosocial risks have only limited potential to reduce health disparities among these groups. Rather, they argue,

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

broader efforts to reduce socioeconomic and racial/ethnic disadvantage remain the single most powerful tool for alleviating health disparities. House and Williams cite evidence that socioeconomic change and policies that improve the status and life circumstances of socioeconomically and racially/ethnically disadvantaged populations have the most significant effects on reducing health disparities. For example, they cite research indicating that policies adopted at the height of the civil rights movement successfully removed barriers to occupational and educational integration. As a result, the socioeconomic status of blacks relative to whites improved, and concurrently, between 1968 and 1978 blacks experienced a larger decline in mortality rates (on both a percentage and an absolute basis) than whites.

Although basic biological differences by sex clearly influence health outcomes, potent health consequences emerge from the social construction of gender. Gender differences in health spring from many aspects of the social context, including differences in economic, political, and social power (Sorensen, 2000). Research is needed to better understand factors that contribute to gender inequalities in health, including stress related to gender roles, the types of jobs men and women hold, the amounts of money they are paid, and the patterning of household responsibilities. For example, men earn more than women, despite women's higher educational levels relative to men in similar occupations. While gender roles may be associated with poor health effects for both men and women, women continue to carry primary responsibility for household duties, in addition to work-related demands. Women 's socioeconomic disadvantages and their irregular career trajectories, generally related to family responsibilities, limit the availability of pensions and savings in later life. Given that employed women may be balancing job roles and family responsibilities, it is not surprising that most studies find that employed women experience more stress and distress relative to employed men (Williams and Umberson, 2000). And as noted by Korenbrot and Moss (see paper contributions), social constructions of gender roles shape the views and experiences women have of sex, contraception, and pregnancy, as well as their health behaviors and interactions with the health care system. These structural factors may also have long-term and often unforeseen consequences. For example, Kaplan (see paper contributions) points out that economic policies that alter work force participation by women may depress and delay reproduction, thereby increasing the risk of breast cancer.

It is critical to address socioeconomic, racial/ethnic, and gender disadvantages, not only to improve the health of individuals in disadvantaged groups, but also because the greatest opportunity for improvement of the nation's health as a whole lies in improving the health of these groups.

Recommendation 4: Research consistently indicates that life circumstances associated with low socioeconomic status, gender, race and ethnic disadvantage, and other high-risk circumstances are strongly associated with a wide range of poor health outcomes. Policies that take account of and enhance positive health outcomes

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

among subgroups at high risk of poor health outcomes need to be designed and implemented.

Opportunities for Intervention Across the Life Span

Preconception, Prenatal, and Postnatal Influences on Health. An implicit yet often unstated assumption in public health is that early intervention programs offer promise for improving health, with the potential benefit of reducing long-term costs to individuals, communities, and the larger society. Perhaps the most significant of such efforts are interventions to improve the health of very young children, starting at birth. Poor birth outcomes (e.g., low birthweight) are related to a range of poor cognitive and physical health indicators as infants grow and develop (Korenbrot and Moss, see Paper Contribution C). Substantial public health resources have therefore been devoted to improving birth outcomes, ranging from improved prenatal care, to health education and outreach, nutrition and psychosocial services, and other efforts. Some successes have been achieved; infant mortality, for example, has declined dramatically, due in large part to improvements in medical technologies that improve survival among preterm and low-birthweight infants.

Korenbrot and Moss (see Paper Contribution C), however, note that little progress has been made in improving other birth outcomes, despite major policy and program efforts. Improvements in the use of and access to prenatal care generally have not resulted in reductions in low birthweight and other poor birth outcomes, although prenatal and perinatal medical interventions may reduce morbidity substantially in those born with complications. In particular, racial/ethnic and socioeconomic disparities have not been reduced, suggesting strongly that overall rates of poor birth outcomes are less related to reductions in risk behaviors (e.g., smoking, adolescent childbearing) or improvements in prenatal health care in individuals, and are more likely the direct consequence of disparities between socioeconomic and racial/ethnic groups.

Factors that mediate this relationship are poorly understood. For example, African-American women are more likely to have low-birthweight infants than white women at all levels of educational attainment, suggesting that factors other than socioeconomic status influence pregnancy outcomes (Kaplan, Everson, and Lynch, Paper Contribution A). Another highly suggestive piece of evidence is that in mixed-race couples, the risk of adverse pregnancy outcomes is increased when the woman is African American but not the man (Collins and David, 1993). Other evidence suggests that early influences on women's health may play an important role. This evidence includes the fact that women who were born at lower birthweights are more likely to have low-birthweight infants (Korenbrot and Moss, Paper Contribution C). Thus, interventions to enhance reproductive outcomes must begin well prior to conception, and must attend to social and economic conditions that confer and moderate risk. This broader focus on women 's health would also reinforce the importance of implementing interventions during pregnancy to enhance the health of the woman, as well as

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

that of the fetus. Finally, to ensure that women are able to undertake their roles as parents and to prepare for subsequent children, increased attention is needed to postnatal health and health habits of new mothers.

Recommendation 5: Improvements in reproductive outcomes will require addressing social and behavioral influences on the health of women well prior to conception, in addition to the perinatal and postnatal periods.

Infancy and Early Childhood. Fuligni and Brooks-Gunn (see Paper Contribution D) discuss social and behavioral influences on the healthy development of young children, noting that various domains that can be considered to define school readiness, including cognitive, emotional, and physical health outcomes, are closely correlated and should be considered simultaneously in intervention efforts. As with birth outcomes, socioeconomic status remains a key predictor of school readiness. Low socioeconomic status is associated with low cognitive test scores, behavioral problems, and a range of physical health indicators including growth stunting, low rates of childhood immunizations, high blood-lead levels, and other health outcomes. Fuligni and Brooks-Gunn explore possible pathways by which socioeconomic indicators, such as parental education and income, operate to affect young children's development, and suggest that family processes (including parental emotional and physical health, parental childrearing styles, and the quality of educational experiences in the home), the quality of child care (e.g., as reflected in home-based versus center-based care), and neighborhood influences (e.g., racial and economic segregation, inadequate social support, hazardous and unsafe living conditions, and limited access to resources such as high-quality schools) underlie these disparities.

These multiple sources of influence on young children's development suggest a number of targets for intervention, including those directed at children and their caregivers, family-level interventions, social supports, and broader social and economic policies that influence neighborhood characteristics. Research data on the effectiveness of these interventions, however, are mixed, and much of the research has focused on the cognitive and social-emotional development needed for school readiness. Comprehensive, high-quality, center-based early education for infants and children has been demonstrated to improve a range of child educational outcomes. These effects may extend to reduce socioeconomic disadvantage. The implementation and evaluation of future programs should focus on planned variations in elements and intensity of such approaches to identify the most appropriate and least costly interventions for specific communities.

Although the efficacy of center-based programs is well established, home visitation programs, larger-scale school-based programs, and family support programs have not consistently produced positive outcomes. Other interventions, such as comprehensive community initiatives, attempt to address neighborhood characteristics that influence child outcomes by promoting antipoverty programs and enhancing community-building strategies. Few of these interven-

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

tions, however, have focused directly on improving child developmental outcomes, and little evidence supports the effectiveness of these approaches. However, the strong theoretical and empirical evidence linking family, neighborhood, and child care processes to child outcomes would suggest that these approaches can be strengthened to be more consistently effective.

Recommendation 6: High-quality, center-based early education programs should be more widely implemented. Future interventions directed at infants and young children should focus on strengthening other processes affecting child outcomes such as the home environment, school and neighborhood influences, and physical health and growth.

Adolescence. Adolescence, as noted by Perry (see Paper Contribution E), is a time of significant developmental transition influenced by dynamic interactions between social forces and important biological changes. These forces interact to influence health outcomes for adolescents, but perhaps most significantly, they set the stage for later adult educational, vocational, and health outcomes.

Perry notes that while adolescence is generally the period of greatest health for most individuals—having outgrown the risks for childhood infectious diseases but not having reached the age of risk for chronic diseases more common to adults—it is also a period of risk for poor health. Because young adolescents typically have not fully developed abstract cognitive abilities, cannot judge the implications of their behavior for future consequences, and are particularly vulnerable to peer influences, they are susceptible to influences in the social environment that encourage risk behaviors such as smoking, violence, alcohol and drug use, and risky sexual behavior. For this reason, health messages directed to adolescents that focus on long-term consequences of behavior or that fail to provide concrete examples of health risks often fail to achieve desired results.

Because many health risks to adolescents are socially determined, research indicates that characteristics of the social environment can also serve to confer protective or health-enhancing effects. As Perry notes, the way in which the environment “fits” the needs of adolescents is a critical determinant of adolescent health behavior. Social contexts that provide positive social norms, offer appropriate role models, set appropriate expectations, provide useful information, and provide social and neighborhood supports are strongly associated with positive health outcomes for adolescents. Perry and her colleagues, for example, have found that community norms and role models accounted for nearly half of the variance in eighth grade children's alcohol use. Further, these attributes may mediate the relationship between poverty and adolescent health outcomes. Other health influences include media exposure (e.g., high exposure to violent television content is associated with subsequent aggressive behavior among adolescents), “family capital ” (e.g., parents' own educational attainment, parent-child activities and time spent together), and connectedness to school.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

These findings illustrate that intervention efforts must focus on adolescents' social environment, and address multiple sources of influence in these environments. In addition, these efforts are most effective when applied early (e.g., in primary school years) and sustained through adolescence. Perry presents data demonstrating that longitudinal, multicomponent interventions, such as the Seattle School Development Project, can reduce adolescent involvement in risk behaviors (e.g., violence, risky sexual behavior, heavy alcohol use). This intervention, which consisted of annual in-service teacher training, social competence training for students, and parent training classes, demonstrated long-term positive effects at age 18 for students who were exposed to the intervention in first through sixth grades.

Recommendation 7: Multilevel interventions should address aspects of adolescents' social environment as they affect health outcomes, including peer norms, role models, performance expectations, social and neighborhood supports, and ties to community institutions.

Adults. As noted earlier, behavioral and environmental factors contribute significantly to most major causes of adult physical and mental disorders and mortality. These factors are largely modifiable, and include those health risks due primarily to individual behavior (e.g., physical inactivity, smoking, diet, alcohol and drug use), to environmental or organizational-level exposures (e.g., occupational exposures and work practices, motor vehicle and other product design), and to social class (see discussion below).

As noted by Emmons (see Paper Contribution F), intervention strategies have typically focused on individuals, often addressing the needs of highly motivated volunteers. Individually based interventions may be effective for individual participants, particularly those at high risk, but may also miss important segments of the population who are not interested in participating in intensive programs. The potential population-wide impact of programs targeting individuals is further limited because these efforts often fail to consider the social and environmental attributes that support poor individual health behaviors. The population effects of interventions to improve the health of adults can be enhanced when interventions address multiple levels of influence, including individual factors (e.g., health knowledge and attitudes, motivation for behavior change), interpersonal processes (e.g., social support, social norms), institutional or organizational factors (e.g., workplace policies), community factors (e.g., neighborhood characteristics associated with poor health outcomes), and public policy (e.g., regulations and laws directed at reducing health risks). In addition, it is important that interventions be linked across multiple locations, such as work sites or communities, in order to provide consistent messages, support, and follow-up in support of healthful changes.

Again, the example of tobacco control provides evidence for the efficacy of a multilevel approach (Warner, Paper Contribution K). Individual-level be-

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

haviorally or pharmaceutically based approaches are most effective when linked with interventions that address social contextual factors that support or discourage tobacco use. These include community-and work site-based tobacco reduction programs that influence social norms and other social contingencies, and broader societal interventions, such as laws and regulations that limit access to tobacco products. Social supports and social networks can also serve to reinforce prevention and smoking cessation messages.

Recommendation 8: Community and work site interventions and evaluations are needed to promote behavioral change, prevent injuries, reduce exposure to occupational risks, and increase healthy environments. Such interventions should include multiple points of leverage, such as individual-level attributes, social supports and social norms, family and neighborhood factors, and environmental and social policies.

Older Adults. Social and behavioral research increasingly demonstrates that the effectiveness of interventions does not decline with older adults; rather, social and behavioral interventions offer promise to reduce functional disability and mortality among older adults. As Maddox illustrates (Paper Contribution G), while risk for chronic diseases and physical and cognitive impairment increases with age, behavioral and social factors predict substantial differences in morbidity and mortality among older adults. Further, behavioral and social interventions can modify trajectories of functional disability and the experience of aging in ways that reduce the high costs of health and social services and promote “healthy aging.”

Maddox's review of successful psychosocial interventions for older adults indicates that these interventions can improve the management of chronic conditions (e.g., osteoporosis and diabetes), and can reduce excessive dependence on the part of frail nursing home residents. Common elements of these interventions are that they enhance social supports and social network integration for health and well-being, and enhance the self-efficacy of older individuals, which is an important predictor of successful intervention outcomes. For example, an intervention to reduce excessive behavioral dependence among older adults in nursing home settings revealed that both patients and caregivers could be trained to change behavioral patterns to promote a greater sense of autonomy and self-efficacy among patients. Moreover, Maddox presents theoretical evidence that an intervention focused on addressing the structural environment of frail older adults—assisted living housing —can increase social support resources and personal choice and control among residents. Assisted living housing allows residents to purchase housing and services appropriate to their assessed needs, but encourages shared responsibility among caretakers, family, and residents, and seeks to promote maximum autonomy and participation in decisions about the community.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

Recommendation 9: Interventions to improve the health of older adults should focus on the social, environmental, and behavioral conditions that minimize disability and promote continuing independence and productive activity. Interventions that enhance the social support and self-efficacy of older adults are particularly promising. Community senior centers and community-assisted housing are two examples of interventions that are especially promising for the frail elderly.

Research Needs

The committee's review of opportunities for intervention at specific developmental stages and “levers” for public health intervention yielded a great deal of information regarding promising intervention strategies. Significantly, however, this review also prompts several findings and recommendations regarding future research needs. These recommendations are summarized in Box 3.

Research on Physiologic Pathways

As noted above, there are multiple pathways through which psychosocial factors affect health outcomes; these influences occur indirectly, through behaviors such as alcohol and tobacco use, and directly, through alterations in the functioning of the cardiovascular, endocrine, and immune systems. However, relatively few of the interventions referenced above include physiological measures. Research that addresses biobehavioral mechanisms is therefore needed to understand pathways through which psychosocial factors affect disease processes (Baum, Paper Contribution L).

Evidence of the relationship between psychosocial and behavioral factors and health is large and growing. Research demonstrates that risk factors such as chronic psychological stress can lead to exacerbation of coronary artery atherosclerosis, likely via a mechanism involving excessive sympathetic nervous system activation. Furthermore, acute stress can serve as a trigger for myocardial ischemia, promote arrhythmogenesis, stimulate platelet function, and increase blood viscosity; among patients with underlying atherosclerosis, acute stress also enhances coronary vasoconstriction (Rozanski et al., 1999). The endocrine system serves as one important gateway across a spectrum of diseases because stressors can provoke the release of pituitary and adrenal hormones that have multiple effects, including alterations in cardiovascular and immune function. For example, social stressors can substantially elevate key stress hormones including the catecholamines and cortisol, and these hormones have multiple effects (e.g., chronic stimulation of cortisol and catecholamine secretion at lower levels has been linked to cardiovascular pathology).

Psychosocial factors such as depression can intensify a variety of health threats. Some of the strongest evidence comes from the cardiovascular realm, with convergent data from a number of well-controlled prospective studies: depressive

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

BOX 3. Recommendations for Research

Recommendation 10: Understanding psychosocial and biobehavioral mechanisms that influence health is critical to better understand and tailor intervention efforts. Research in this area should be encouraged.

Recommendation 11: A substantial new research effort is needed to understand the pathways through which behavioral and social factors affect pregnancy outcomes, and to address the issues of women's health across the life span and across generations.

Recommendation 12: Research on early childhood interventions needs to be extended to provide information useful to tailoring such interventions to ensure the least costly, most effective strategies for a variety of populations. Both the theory and the research must be expanded and integrated to include a full array of child outcomes (e.g., physical health as well as social, cognitive, and emotional outcomes) and the effect of interventions on these outcomes.

Recommendation 13: Research should identify sources of health strengths and resilience, as well as health risks, among individuals, families, and communities of low socioeconomic status and racial and ethnic minority groups.

Recommendation 14: Research is needed to identify pathways through which social contexts directly and indirectly affect disease pathogenesis and outcomes.

Recommendation 15: Legal and regulatory interventions represent a powerful tool for promoting public health. However, evaluations are needed to ensure that these interventions achieve their intended effects without imposing undue burdens on individuals or society.

Recommendation 16: Expansions of research methodologies are needed to address a broad range of research questions, to ensure that methods are appropriate to the research questions being asked, and to respond to the complexities of multilevel interventions. This includes the need to integrate both qualitative and quantitative research.

Recommendation 17: Cost-effectiveness analyses are necessary to assess the public health utility of interventions. Assessments are needed of the incremental effects of each component of multilevel, comprehensive interventions, and of the incremental effects of interventions over time. Such analyses should consider the broad influence and costs of interventions to target individuals, their families, and the broader social systems in which they operate.

Recommendation 18: Efforts to develop the next generation of prevention interventions must focus on building relationships with communities, and develop interventions that derive from the communities' assessments of their needs and priorities. Models should be developed that encourage members of the community and researchers to work together to design, train for, and conduct such programs.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

symptoms were associated with the development of ischemic disease, as well as poorer outcome among patients who had preexisting cardiovascular disease (Glassman and Shapiro, 1998). In addition to cardiovascular risks, depressive symptoms have been linked with higher morbidity and mortality in a number of other arenas, such as cancer risk in older persons (Penninx et al., 1998), all-cause mortality in medical inpatients (Herrmann et al., 1998), risk for osteoporosis (Michelson et al., 1996), and lessened physical, social, and role function, worse perceived current health, and greater bodily pain (Wells et al., 1989).

Evidence indicates that many social and behavioral interventions have biological consequences, rather than acting primarily indirectly through such channels as health behaviors. For example, stress management-oriented interventions clearly alter a number of aspects of cardiovascular, immune, and endocrine function; some smaller studies provide encouraging results for the alteration of disease progression as well, particularly in the cardiovascular realm. Social and behavioral interventions may enhance health because they decrease psychological stress, improve mental health, alter chronic life strains, improve health behaviors, and/or enhance social support or social connectedness; indeed, the most successful interventions have positive benefits across this spectrum, rather than within any single sphere.

Within the past two decades, ample evidence has accumulated that documents diverse psychosocial influences on morbidity and mortality. Interventions that modify psychosocial risks should also alter physiology, and a better understanding of these pathways will aid in the design of more effective interventions.

Recommendation 10: Understanding psychosocial and biobehavioral mechanisms that influence health is critical to better understand and tailor intervention efforts. Research in this area should be encouraged.

Research on Pregnancy

Abundant and continuing evidence exists for socioeconomic and racial/ethnic gradients in perinatal outcomes as measured by birthweight and gestational age, yet research is still a long way from illuminating how these gradients emerge (Kaplan, Everson, and Lynch; Korenbrot and Moss, Paper Contributions A and C). Moreover, evidence is accumulating that documents intergenerational risks (e.g., women who were themselves of low birthweight as infants are more likely to give birth to low-birthweight infants). Current knowledge of the complications of pregnancy, however, does not offer immediate opportunities to alter pregnancy outcomes; new approaches are needed. Korenbrot and Moss postulate that one potential mechanism increasing the risk of poor pregnancy outcomes might be stress-inducing characteristics of disadvantageous environments. Adaptation to such stress is affected by other factors such as the process of cultural adaptation and social support. While some evidence links stress and poor pregnancy outcomes, knowledge of the pathways by which this

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

effect occurs is very incomplete. For example, self-medication to deal with stress might include smoking or alcohol use, which in turn results in lower birthweights or congenital malformations. Research on pathways must also elucidate linkages between social and psychological variables and some of the well-established biological antecedents of poor pregnancy outcomes, such as bacterial vaginosis and pregnancy-induced hypertension. Moreover, elucidation of the pathways to poor pregnancy outcomes should include consideration of gender-based issues in reproduction and family roles (as noted earlier), and should address intergenerational risks.

Recommendation 11: A substantial new research effort is needed to understand the pathways through which behavioral and social factors affect pregnancy outcomes, and to address the issues of women's health across the life span and across generations.

Research on Interventions in Infancy and Early Childhood

A strong body of evidence supports the effects of family, neighborhood, and child care processes on child cognitive and social-emotional development, and has led to effective interventions (Fuligni and Brooks-Gunn, Paper Contribution D). However, as noted above, many of the most successful center-based interventions are single-site, relatively small programs conducted in populations homogeneous for socioeconomic status or race. Many questions remain about extending these approaches to more heterogeneous groups, taking the programs to a larger scale, and comparing curricula and the intensity of participation. For example, to what extent do such programs enhance child outcomes among children of middle- and upper-income families? In addition, more research is required to identify the successful elements of other modalities such as home-visiting or family support programs when center-based interventions may not be appropriate. Finally, we are just beginning to learn about strategies at the neighborhood level to enhance collective efficacy, and the effect of these strategies on child development must be assessed.

However, a larger issue looms. The most well established body of evidence deals with child cognitive and social-emotional development. As Fuligni and Brooks-Gunn note, these are but two of the components of complete child development. The evidence about socioeconomic and racial/ethnic gradients in physical health and growth, the processes leading to such gradients, and the impact of serious child health problems on child development is much less systematically developed, although certainly some literature exists. Information on child health outcomes needs to be expanded and integrated with that for cognitive and social-emotional development to provide a more complete picture of effects of early childhood interventions. We must also understand the effect of interventions proposed to enhance cognitive and social-emotional development on the physical health and growth of children and on their parents. For example, the cognitive, social, and emotional benefits of center-based early childhood

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

interventions may involve a trade-off between increased risk of infectious disease for the child and increased time lost from work for the parent.

Recommendation 12: Research on early childhood interventions needs to be extended to provide information useful to tailoring such interventions to ensure the least costly, most effective strategies for a variety of populations. Both the theory and the research must be expanded and integrated to include a full array of child outcomes (e.g., physical health as well as social, cognitive, and emotional outcomes) and the effect of interventions on these outcomes.

Research on Health Resources and Strengths of Low Socioeconomic and Ethnic Minority Groups

As noted in the discussion of racial and ethnic health disparities, in some cases ethnic minorities experience lower rates of illness and mortality than whites. For example, foreign-born mothers of African and Hispanic descent experience generally better birth outcomes than American-born black and Hispanic women. In addition, lower rates of some psychiatric problems are found among African Americans, despite their relatively greater risk for poorer health outcomes. This suggests that cultural attributes protect against negative social and economic conditions, and should be studied to identify new intervention methods. Further, despite evidence that societal-level interventions are needed to assist communities in solving many of the problems that label them as “disadvantaged,” there remains a subset of children and families that succeed in these environments. Efforts should be expended to find out what qualities exist in these children and families that can be developed in others who share the same risk status.

Recommendation 13: Research should identify sources of health strengths and resilience, as well as health risks, among individuals, families, and communities of low socioeconomic status and racial and ethnic minority groups.

Research on Social Contexts and Social Capital

Research is needed to identify ways in which social context influences health behaviors, environmental exposures, psychosocial responses, and health outcomes and mediates the influence of interventions. Fuligni and Brooks-Gunn provide an important example in their discussion of the need to specify processes through which neighborhoods influence children's outcomes. Neighborhoods may include a variety of resources, such as availability of and access to quality schools or day care; relationships, such as social support networks available to parents; and norms or collective efficacy, such as the presence of physi-

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

cal risk to residents. The effects of these are likely to be both direct and indirect. Research that helps to elucidate these causal pathways may provide other important points for interventions.

Recommendation 14: Research is needed to identify pathways through which social contexts directly and indirectly affect disease pathogenesis and outcomes.

Research on Legal and Regulatory Interventions

Governments frequently enact laws and regulations to promote public health by discouraging unhealthy behavior (e.g., indoor smoking ordinances, outlawing nonprescription use of mood-altering drugs), encouraging and mandating healthy behavior (e.g., childhood immunizations, mandatory use of safety belts), and encouraging and mandating companies to produce safe products and protect the environment. In addition to direct community appeals through public education, governments can influence individual choices through taxing and spending powers, and can penalize harmful activities through the imposition of civil and criminal penalties. Within the constitutional limits on free speech, governments regulate advertising and product labeling to ensure that the public receives accurate information about food, drugs, and other products. Finally, governments enact consumer product safety regulations, occupational health and safety laws, environmental ordinances, and building codes to reduce the likelihood of harm from hazardous products or environments (Gostin, Paper Contribution J).

Despite the magnitude and scope of government efforts to promote health, the effectiveness of legal and regulatory strategies is poorly understood, with a few exceptions such as motor vehicle safety. Few initiatives have been subjected to rigorous evaluation. In light of the cost of these interventions, and the legal and moral implications of using government power to influence individual behavior, the effects and public acceptance of interventions of this type should be studied.

The best evaluations of legal and regulatory interventions, however, provide compelling evidence that such strategies can have a profound impact on the incidence of injury and disease. For example, the decline in heart disease deaths that has occurred over the past 40 years can be attributed, at least in part, to significant reductions in the rate of cigarette smoking. This was due in no small part to the Surgeon General's report on smoking (U.S. Department of Health, Education, and Welfare, 1964) and subsequent government efforts to bring the health risks of smoking to the attention of the public (Gostin, Paper Contribution J). In addition, the rate of motor vehicle crash deaths per million miles driven has decreased dramatically since 1950. This is not due to greater skill on the part of American drivers, but rather to government-mandated improvements in the crashworthiness of automobiles, mandatory use of safety belts, and improvements in road design (Centers for Disease Control and Prevention, 1999). Further, laws that require parents to produce proof that their children have been vaccinated against measles, rubella, and diphtheria before they can be enrolled

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

in school have dramatically increased rates of childhood immunization (Briss et al., 2000). This has resulted, in turn, in a marked reduction in morbidity and mortality from these communicable diseases.

Clearly, the force of law can be a powerful tool for public health. However, the anticipated benefits of using legal and regulatory strategies to promote public health must be balanced against the costs, including the loss of personal freedom and the potential for unintended consequences. When the benefits of the intervention are great and widely appreciated (e.g., fluoridation of water) and the burden of the intervention is slight, legal and regulatory interventions generally enjoy widespread support. When an intervention is perceived as producing limited benefit, or is viewed as unnecessarily burdensome by a substantial segment of the population, it is likely to be much more controversial. In general, individuals are much more willing to support laws that protect them from the actions of others and from involuntary exposure than from risks voluntarily assumed.

Enactment of public health laws and regulations involves political calculus and trade-offs between various interest groups. Since society devotes substantial resources to regulation, policymakers need high-quality evaluation data to determine which strategies work and which do not. They also need accurate and objective data on the cost and potential consequences of various policies. Policy cannot be entirely divorced from the political environment in which it is made, but it is essential that those involved in the process of government have access to objective information.

Recommendation 15: Legal and regulatory interventions represent a powerful tool for promoting public health. However, evaluations are needed to ensure that these interventions achieve their intended effects without imposing undue burdens on individuals or society.

Research Methodologies

Because multilevel intervention approaches are a relatively new area of inquiry, a range of research questions must be addressed. Research is needed that will contribute to our understanding of how best to create linkages between levels of influence, and how to sequence or coordinate interventions across levels. While there are empirical examples of successful multilevel interventions and theoretical reasons to expect that multilevel interventions will be more successful than single-level interventions, research is also needed on the incremental cost-effectiveness of intervening on additional levels in order to establish the most efficient intervention methods (see discussion below).

Interpretation of the results of population-wide interventions should be based on different criteria from those used for trials targeting high-risk individuals. As Emmons (Paper Contribution F) points out, community intervention trials are likely to observe smaller changes in individual health behaviors than are clinic-based interventions with motivated volunteers. Indeed, the work of Geoffrey Rose (1992) provides a clear rationale for the expectation that

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

small changes at the population level can lead to large effects on disease risk. Rose argues that when risks are widely distributed in the population, small changes in the behavior of many individuals in the population are likely to have a greater impact on risk reduction than large changes among a smaller group of high-risk individuals. Interpretation of the impact of community-based interventions must therefore be judged as a function of the intervention's efficacy in terms of producing community-level changes, as well as its reach, or penetration within the population.

Interpretation of the effects of population-wide interventions must also address problems inherent in establishing causality and time effects. Many studies of population-wide interventions are cross-sectional, and may therefore miss intervention effects that occur beyond the evaluation time frame, or may overestimate intervention effects by capitalizing on population changes not due to the intervention. As noted above, interpretation of findings must be guided by sound theory to anticipate the magnitude and timing of intervention effects. It is important to have explicit models of change and to predict in advance when results should be observed. Also, because multilevel interventions require assessments at multiple levels of influence and change—for individuals, organizations, and communities—methods for measuring change across time in these various settings are needed. In addition, it is important to identify methods for assessing factors mediating change at these multiple levels, in order to understand the pathways through which change occurs.

Finally, it is clear from the above discussion that it is necessary to study the social and cultural context in which behaviors occur. The qualitative research paradigm allows for such a detailed inquiry of the social context. In addition, it allows for the identification of various segments in a community, including hard-to-reach populations, and the impact of changes in the social context on individual behaviors. Qualitative data collection includes participant observation, open-ended questions, in-depth interviewing, and focus groups. Participant observation allows investigators to observe behaviors in the natural setting. By linking observational data to information from in-depth interviews, researchers can triangulate information on what people do versus what they say they do. In addition, the dialogue among participants in focus groups allows for a rapid assessment of themes. The triangulation of qualitative data sources complements the information provided by quantitative data.

Further, qualitative interviewing often allows greater access to information regarding sensitive behaviors (e.g., health risk behaviors or illegal activities), and allows for a better understanding of the meaning of specific behaviors. Qualitative research can therefore illuminate why individuals engage in risk behaviors, despite their knowledge of risks and an intent to change their behavior. Also implicit in qualitative research is the recognition that local circumstances vary from a street block to a neighborhood, metropolitan area, state, or country. Qualitative research is also a useful tool for conducting systematic and comprehensive needs assessments and program evaluations.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

Recommendation 16: Expansions of research methodologies are needed to address a broad range of research questions, to ensure that methods are appropriate to the research questions being asked, and to respond to the complexities of multilevel interventions. This includes the need to integrate both qualitative and quantitative research.

Cost-Effectiveness Analyses

In the preceding sections, the committee has highlighted a number of promising approaches to public health intervention that draw upon important directions in behavioral and social science research. These interventions incorporate and integrate a range of approaches, spanning the continuum between traditional downstream approaches (i.e., interventions directed toward individuals to address health behaviors, attitudes, and/or knowledge) and upstream approaches (i.e., interventions directed toward public policies to address social and economic structures that influence health outcomes). While the scientific evidence base for these approaches is growing, these intervention strategies would prove impracticable for broader public health application should they fail to deliver significant gains relative to their cost and ease of implementation.

The relatively modest resources directed toward public health efforts must be carefully allocated, often while weighing alternative intervention strategies, and prioritized to yield the greatest benefit to target populations. Cost-effectiveness analyses (CEA) are therefore needed to help public health officials, foundations, and community leaders to establish priorities and assess the utility of interventions. (CEA is an umbrella term that includes other methodologies such as cost-utility analysis.) Such analyses compare alternate interventions, often using measures such as potential years of life gained or quality-adjusted life years that include changes in the quality of health as well as in the length of life, to assess the kinds of interventions that should be done, for whom, and how often. CEA therefore yields a ratio that expresses the value of an intervention relative to its cost.

In the context of the longitudinal, multilevel, comprehensive intervention strategies that are advocated in this report, CEA is critical to help assess which components of such interventions are most useful, for how long they must be applied, and with which populations they are most cost-effective. The incre-mental benefits of each component of multilevel intervention strategies must therefore be assessed. Further, the costs and benefits of interventions should be assessed at levels beyond that of the target individual(s). As Russell (2000) illustrates, interventions deliver effects not only on the target individuals, but also on their families, friends, social networks, communities, and so on. Interventions not only require the time and resources of providers of health care and health interventions, but also use the time and resources of individuals, as well as the social systems within which they operate. Interventions that tackle multiple levels of influence and multiple risk factors must therefore make the best

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

use of resources at each level. Such a societal perspective of CEA considers the impacts of interventions on all persons and systems significantly affected by the intervention.

Recommendation 17: Cost-effectiveness analyses are necessary to assess the public health utility of interventions. Assessments are needed of the incremental effects of each component of multilevel, comprehensive interventions, and of the incremental effects of interventions over time. Such analyses should consider the broad influence and costs of interventions to target individuals, their families, and the broader social systems in which they operate.

Community as Partners

A consistent theme across the papers included in this volume is a call for interventions and research conducted in partnership with communities—that is, research efforts that are conducted by communities rather than on communities, and interventions that are not strictly message-driven but rather are guided by the voice of the community. Historically, community intervention trials have relied on community organizing principles as a means of involving communities in program planning and implementation (Sorensen et al., 1998). This model may fall short of the partnership needed to engage communities fully in health issues. Partnership with the community implies that the community has a voice in problem definition, data collection and the interpretation of results, and application of the results to address community concerns. Although further evidence is still needed to establish the effectiveness of this approach, the papers included in this volume note that community partnerships are likely to influence community priorities in the direction of health and foster social norms supportive of healthy outcomes. In addition, community-level participation and buy-in are likely to enhance the sustainability of interventions. Further, by providing an infrastructure that remains in a community after a research project ends, interventions may be more likely to be maintained beyond the funded period.

Recommendation 18: Efforts to develop the next generation of prevention interventions must focus on building relationships with communities, and develop interventions that derive from the communities' assessments of their needs and priorities. Models should be developed that encourage members of the community and researchers to work together to design, train for, and conduct such programs.

Funding

Full implementation of the above recommendations regarding intervention strategies and needed research requires significant changes in the manner in

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

BOX 4. Recommendations for Funding

Recommendation 19: Payers of health care should experiment with reimbursement structures to support programs that promote health and prevent disease.

Recommendation 20: Government and other funding agencies and universities should promote the development of interdisciplinary, collaborative research and training. Among the mechanisms that should be encouraged are interdisciplinary research centers, special program project awards, fellowships, and other postgraduate training programs.

Recommendation 21: Greater attention should be paid to funding research on social determinants of health and on behavioral and social science intervention research addressing generic social determinants of disease.

which health services and research funding are allocated. The committee 's recommendations for funding are described below and summarized in Box 4.

Need for Health Care Funding to Place Greater Emphasis on Prevention

The United States expends more than a trillion dollars annually on health care. This vast expenditure provides considerable leverage for motivating health professionals and patients to pursue behaviors that result in preventing illness and reducing disability.

Payers for services, especially Medicare and Medicaid, have significant potential influence in encouraging widespread improvements in health behavior through the structure of their programs. For example, by benefit design decisions payers can encourage individuals to take advantage of preventive services that have been demonstrated to be cost-effective, such as immunizations and cancer screening services. Similarly, reimbursement systems can be designed so providers will be encouraged to give attention to ways of organizing preventive and care services that promote effective health behavior and contribute to the prevention of disability. One example of such action is the “healthy aging” project. For this project, the Health Care Financing Administration (HCFA) has commissioned both systematic reviews and demonstration projects to assess how changing the Medicare benefit structure can encourage the use of screening and preventive services and risk reduction behavior such as smoking cessation, increased physical activity, and chronic disease self-management.

HCFA has also developed programs to integrate acute medical care and long-term-care services to improve function among older persons. Such demonstrations include social health maintenance organizations (social HMOs) and PACE (Program for All-inclusive Care of the Elderly). PACE, for example, integrates funding under the Medicare and Medicaid programs to develop more comprehensive service systems that seek to maintain community residence and function among older adults with substantial functional limitations.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×

Evaluation studies of social HMOs and PACE are continuing, but it is clear that these large reimbursement programs provide opportunities to develop and test programs of many kinds that encourage preventive health behavior, reduce risks of adverse events such as falls and malnutrition, and promote the recognition of such disabling conditions.

Recommendation 19: Payers of health care should experiment with reimbursement structures to support programs that promote health and prevent disease.

Need for Interdisciplinary Collaboration

The committee was impressed by evidence that the most successful interventions were those aimed not only at the behavior of individuals but also at the families, neighborhoods, and communities in which people live. Interventions worked best when they were supported also by legislative, media, and marketing efforts. Warner (Paper Contribution K) shows that the most successful tobacco control programs are those that include a combination of elements, including higher prices on tobacco products, laws regarding indoor air quality and restricting the use of vending machines, the development of media messages, and the establishment of smoking cessation programs.

Given the complexity required for effective interventions, it becomes increasingly important that social and behavioral scientists work closely with scientists representing other disciplines. The list of such disciplines is broad and includes genetics and human biology, clinical medicine, urban geography, law, journalism, nutrition, and education. Unfortunately, current training programs tend to focus on one or two disciplines at a time. Effective interdisciplinary collaboration is difficult because of important differences among experts in each discipline in vocabulary, history, and perspective. These experts attend different professional society meetings, train in different programs, read different journals, and rarely work together. There are now examples of programs, however, that have succeeded in developing truly interdisciplinary collaboration. One example is provided by the MacArthur Foundation Research Networks, and another by the programs of the Canadian Institute for Advanced Research. Emphasis needs to be given to developing and supporting similar programs on a broader scale. This should be done by government bodies as well as by private foundations.

Recommendation 20: Government and other funding agencies and universities should promote the development of interdisciplinary, collaborative research and training. Among the mechanisms that should be encouraged are interdisciplinary research centers, special program project awards, fellowships, and other postgraduate training programs.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Refocusing of Research Efforts to Address Fundamental Determinants of Health

Most government agency and private foundation support for health research and interventions is organized in terms of particular diseases. The largest such funders, the National Institutes of Health (NIH) and the Agency for Health Care Research and Quality, are organized primarily by disease categories, although some institutes and centers of NIH support cross-cutting basic research that may inform interventions for several diseases. The predominant organization of research funding based on a clinical classification of disease may be of value for the treatment of sick individuals, but it is not as useful for studies of disease etiology and prevention. Because social and behavioral research demonstrates that several clinical entities have one or more risk factors in common, it is more useful to consider a new disease classification that unifies these different clinical entities based on the social and behavioral features they have in common. This is not a new idea. For many years, infectious disease epidemiologists grouped together different clinical diseases based on their similarities in modes of transmission (e.g., whether they are water-borne, air-borne, vector-borne, or food-borne). This way of classifying disease may not be of direct value in the treatment of sick people, but it certainly is of value in identifying those aspects of the environment to which interventions could be directed. A comparable set of social and behavioral categories for many of the diseases of concern today, for the most part, does not exist. The reasons for this are not clear. It may be that diseases such as coronary heart disease, cancer, AIDS, and arthritis are viewed as diseases of the “individual,” and not as diseases influenced by social and environmental factors. Whatever the reasons, it is difficult for social and behavioral scientists to obtain funding for research and programs focused on generic determinants.

There already exist several examples of more appropriate funding arrangements to fit this model. The Centers for Disease Control and Prevention funds prevention centers, but nevertheless encourages proposals that focus on the prevention of specific diseases and conditions. The National Institute of Occupational Safety and Health deals with diseases associated with work, but a specific disease focus is often called for here as well. The National Institute on Aging and National Institute of Child Health and Human Development perhaps come closest to what is needed because their mission allows for the simultaneous consideration of several related diseases associated with human development and aging. Similarly, the Substance Abuse and Mental Health Services Administration's initiatives to fund research on the integration of prevention and treatment of comorbid substance abuse and psychiatric conditions is another example of moving the field toward integrated research and public health practice efforts.

The committee's findings demonstrate the importance of prevention efforts that are based on the reality that many diseases have numerous risk factors in common and that intervention programs are most effective when they deal with this commonality.

Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 6
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 7
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 8
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 9
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 10
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 11
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 12
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 13
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 14
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 15
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 16
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 17
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 18
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 19
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 20
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 21
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 22
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 23
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 24
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 25
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 26
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 27
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 28
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 29
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 30
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 31
Suggested Citation:"Findings and Recommendations." Institute of Medicine. 2000. Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: The National Academies Press. doi: 10.17226/9939.
×
Page 32
Next: Conclusions »
Promoting Health: Intervention Strategies from Social and Behavioral Research Get This Book
×
Buy Paperback | $55.00 Buy Ebook | $43.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

At the dawn of the twenty-first century, Americans enjoyed better overall health than at any other time in the nation's history. Rapid advancements in medical technologies, breakthroughs in understanding the genetic underpinnings of health and ill health, improvements in the effectiveness and variety of pharmaceuticals, and other developments in biomedical research have helped develop cures for many illnesses and improve the lives of those with chronic diseases.

By itself, however, biomedical research cannot address the most significant challenges to improving public health. Approximately half of all causes of mortality in the United States are linked to social and behavioral factors such as smoking, diet, alcohol use, sedentary lifestyle, and accidents. Yet less than five percent of the money spent annually on U.S. health care is devoted to reducing the risks of these preventable conditions. Behavioral and social interventions offer great promise, but as yet their potential has been relatively poorly tapped. Promoting Health identifies those promising areas of social science and behavioral research that may address public health needs.

It includes 12 papers—commissioned from some of the nation's leading experts—that review these issues in detail, and serves to assess whether the knowledge base of social and behavioral interventions has been useful, or could be useful, in the development of broader public health interventions.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!