Early Childhood Interventions: Theories of Change, Empirical Findings, and Research Priorities

INTERVENTIONS, PART I

Jack P. Shonkoff

Brandeis University

Dr. Shonkoff introduced his presentation as “a view from 30,000 feet”—an overview of the underlying science and reflections on the challenges facing the field. He began by outlining four characteristics of the current landscape of behavioral and social intervention:

  • Public skepticism. “We face widespread questioning about whether we really know how to change behavior and influence developmental trajectories.”

  • Expanding yet incomplete science. “The rich and growing knowledge base that guides the design and implementation of behavioral and social interventions is conceptually strong but empirically uneven.”

  • Demonstrated efficacy but inconsistent performance. Model programs provide credible evidence that we have the capacity to intervene effectively, but successful demonstration projects typically “have different characteristics from the full range of interventions that are actually delivered when promising programs are brought to scale.”



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 16
Early Childhood Interventions: Theories of Change, Empirical Findings, and Research Priorities INTERVENTIONS, PART I Jack P. Shonkoff Brandeis University Dr. Shonkoff introduced his presentation as “a view from 30,000 feet”—an overview of the underlying science and reflections on the challenges facing the field. He began by outlining four characteristics of the current landscape of behavioral and social intervention: Public skepticism. “We face widespread questioning about whether we really know how to change behavior and influence developmental trajectories.” Expanding yet incomplete science. “The rich and growing knowledge base that guides the design and implementation of behavioral and social interventions is conceptually strong but empirically uneven.” Demonstrated efficacy but inconsistent performance. Model programs provide credible evidence that we have the capacity to intervene effectively, but successful demonstration projects typically “have different characteristics from the full range of interventions that are actually delivered when promising programs are brought to scale.”

OCR for page 16
Complexity of successful service delivery. “Interventions that work are rarely simple, inexpensive, or easy to implement.” Given the diversity of interventions that have been proposed and implemented, Dr. Shonkoff indicated that his remarks would focus on early childhood intervention as a prototype—to serve “as a heuristic model for thinking more broadly about how we might approach behavioral and social interventions across different ages and venues.” Early childhood intervention, he said, is a useful model because it rests on a sound theoretical framework, builds on a strong experimental base, and provides promising foundations for a life-span strategy because of its prevention orientation. Effective interventions in the early childhood years have a number of distinguishing features, Dr. Shonkoff said. The first is the importance of an individualized approach linked to specific objectives. In contrast, programs that are built on a one-size-fits-all model and guided by broad generic goals are relatively ineffective. A second feature of successful programs is the high quality of their implementation. Central to this success is a well-designed intervention strategy, appropriate staff training, and careful monitoring of service delivery over time. A third feature of effective interventions in the early childhood period is the quality of the relationships that are built between the people who provide the service and those who receive it. The positive “effects of relationships on relationships” may be at the heart of what makes early childhood interventions work, Dr. Shonkoff said. “That is to say, the provider-parent relationship influences the parent-child relationship, which, in turn, can result in positive outcomes for both the child and the parents.” A fourth feature—that early childhood intervention be family centered, community based, and coordinated—“is embedded in a strong theoretical framework,” Dr. Shonkoff said, “but has not been sufficiently validated empirically.” For example, the widespread belief that programs are more effective when delivered through parents rather than focused directly on children may or may not be true in all circumstances, as we do not have sufficient experimental data on this dimension of service delivery. This “is an important issue because high levels of parent involvement in early childhood programs are more difficult to achieve than they were when fewer mothers were in the workforce.” The last feature of effective interventions relates to the critical dimensions of program timing, intensity, and duration. Here again, “the field is

OCR for page 16
replete with strong opinions,” Dr. Shonkoff said, but “the empirical knowledge base is thin.” We do have persuasive indications about the benefits of earlier initiation and longer duration of services—for example, demonstration programs for children in poverty that have been the most effective started either prenatally or in early infancy and extended up through school. “But answers to questions about cutoff points for ‘early’ versus ‘late,’ and hard data on intensity and frequency of specific service components, await further study.” Dr. Shonkoff then proceeded to discuss some of the “persistent challenges” that remain to be addressed by the early childhood intervention field. These include: Expanding access and participation. “Many young children who have the greatest need for services often don’t get them,” he said, “either because the programs don’t reach out effectively into communities with the most vulnerable populations or because families choose not to participate.” Ensuring greater quality control, particularly when bringing successful models to scale. “We have a problematic track record,” Dr. Shonkoff said, “of taking interventions that have been demonstrated to be effective in model settings, and then trying to do them ‘on the cheap’ by serving larger numbers of children with fewer staff who are trained less well and compensated more poorly.” Defining and achieving “cultural competence.” One of the important contexts in which young children develop is the culture of the family and of the community in which they live. Consequently, the call for early childhood intervention services that are culturally competent has become a growing political mandate. Dr. Shonkoff noted, however, that “we have very little hard knowledge” about this compelling and complex issue. “How we define cultural competence, how we teach it, and how we embed it in all of our intervention programs is an emerging area of scientific inquiry.” Identifying and responding to the special needs of distinctive subgroups. Most traditional models of early childhood intervention are not well designed to address significant family problems that can have major adverse impacts on child well-being. “Family violence, substance abuse, and parental mental illness, particularly maternal depression, are three common examples,” Dr. Shonkoff said. The challenge is to reconcile the core competencies that must be available within all early childhood programs with the specialized professional expertise that may be required to address a wide variety of serious family needs.

OCR for page 16
Reducing fragmentation and strengthening the service infrastructure. “The world of early childhood programs is characterized by highly fragmented policies and service systems that have been developed independently to address the needs of children living in poverty, children with disabilities, children who have been abused or neglected, and children who need generic care and early education. Consequently, we have many children whose complex needs are addressed separately by multiple service streams, with limited integration across systems,” Dr. Shonkoff said. Though dealing with that fragmentation is essentially a political issue, the scientific community can help by articulating the unified knowledge base that cuts across the multiple service systems. Assessing costs and making choices among alternative investments. The broad-based and multifaceted system of early childhood intervention has not had a tradition of looking carefully at costs and benefits or measuring cost effectiveness. Dr. Shonkoff urged that such concerns be given considerably more attention, particularly when assessing the impacts of complex interventions with multiple components. Dr. Shonkoff concluded his presentation by calling for a “dramatic rethinking about the interactions among the science, the policy, and the practice” of behavioral and social interventions across the life span, using early-childhood intervention as a model. Three issues were highlighted. First, he said, is the need to reconcile traditional service strategies with the economic and social realities of contemporary family life. Second is the need to improve the availability, training, and compensation of service providers in the field. Finally, Dr. Shonkoff underscored “the need to change the highly politicized context in which intervention programs are evaluated, which results in a high-stakes environment that undermines honest attempts to improve quality.” He noted that evaluators and service providers often underplay evidence of ineffective services and overstate the extent to which programs do work. Alternatively, Dr. Shonkoff called for “a more constructive culture of intervention research that asks hard questions about what is working, disseminates evidence of effective services and promotes their implementation, shines an equally bright light on programs that are not working, and goes ‘back to the drawing board’ to generate new approaches to behaviors that are more resistant to change.”

OCR for page 16
INTERVENTIONS, PART II Margaret Chesney University of California, San Francisco Dr. Chesney began her presentation by noting some of the progress over the past few decades in improving individuals’ health behaviors. Adult smoking prevalence, for example, has decreased by about 40 percent since the Surgeon General’s report of 1964. And behavioral interventions in studies over the last 25 years have increased average weight loss by 75 percent and physical activity frequency by up to 25 percent. Still, she noted, much remains to be done. Over 24 percent of adults in the United States still smoke. A majority of adult Americans—some 60 percent—are now considered overweight, and 18 percent of the adult population is deemed obese. In addition, there is evidence of a new epidemic of obesity among youth. Meanwhile, despite all the attention to exercise, only 24 percent of the U.S. population regularly engages in light-to-moderate activity. So on balance, Dr. Chesney said, we know that “behavioral interventions can lead to improvements in health, but that these improvements need to be maintained over time and reach all ethnic, racial, social class, and gender groups. They also need to be extended to the population at large across our neighborhoods.” Dr. Chesney said she is optimistic that these challenges can be met, and the remainder of her talk largely addressed her four basic reasons: The first reason is that an important shift in the basic nature of interventions has been occurring and will likely last. We are shifting from a treatment model—which has addressed unhealthy behaviors the way medicine approaches infections, as pathogens responsive to short-term therapy or surgical intervention—to “a model that recognizes that behavior is controlled by complex social contingencies.” If we want to change peoples’ diets, their level of physical activity, or other lifestyle factors, Dr. Chesney said, “It is not like a bacterial infection, for which one could administer five sessions of health counseling like an antibiotic and expect that the unhealthy dietary habits or the physical inactivity would be ‘cured.’ Changing behavior is more like managing diabetes; it requires monitoring and care over time.” The second reason for optimism, she said, is “the increasing diversity

OCR for page 16
of interventions and approaches that are being tested and implemented. In particular, interventions are increasingly being developed to respond to the needs of different community groups. Interventions are targeting higher-risk populations, tailored to individuals or to groups, and designed in ways that are more culturally appropriate.” The third and perhaps most compelling reason for optimism, discussed and embedded in the six National Research Council and Institute of Medicine (IOM) reports, is the importance of the social context in which the behavior occurs. This understanding is reflected, Dr. Chesney said, in the social ecology model developed by Daniel Stokols of the University of California, Irvine. An individual’s behavior, rather than being seen as an isolated event and the responsibility of the individual alone, is considered to be influenced by or the result of a number of factors: intrapersonal factors (including motivation, skills, attitudes); interpersonal factors (social networks, norms, the influence of one’s neighborhood); the institutions and organizations in which the person works or goes to school; and the public policies that broadly influence his or her life. Thus, as Tracy Orleans of the Robert Wood Johnson Foundation put it, “we need to expand the targets of successful interventions beyond the individual to the powerful social contexts in which they live.” Interventions may be focused at multiple levels to achieve change, including what John B. McKinlay of the New England Research Institutes has called the “downstream” level (individualized approaches and interventions), the “midstream” level (interventions at homes, work sites, schools, and churches), and the “upstream” level (efforts to change social policies through media and legislation that reward health). The impressive achievements in tobacco control, for example, may be attributed to the simultaneous efforts at each of these levels. Over the past decades, “downstream” interventions consisting of individual counseling and group smoking cessation programs have improved quit rates. At the same time, “midstream” interventions worked to prevent smoking initiation and to encourage smoking cessation with school-based, work site, and community programs. Tobacco control efforts also illustrate the impact of “upstream” interventions, with policy-level approaches such as the Food and Drug Administration (FDA) regulations designed to reduce the availability and impact of tobacco marketing aimed at youth. Tobacco control has moved even farther upstream. It became apparent that the reliance of farmers in the southeastern United States on tobacco

OCR for page 16
crops created pressure to maintain a tobacco market. Dr. Chesney described efforts to help tobacco farmers transition to new enterprises, new commodities, and new crops as “very upstream” interventions. In these efforts, it is especially critical to work not only with the farmers themselves but with their neighborhoods, communities, and churches—the social contexts in which they live—which for centuries have supported and have been supported by the tobacco crop. Dr. Chesney’s fourth and most important reason for optimism, she said, is that such multilevel interventions, which address risk behaviors in the social context that supports them, are beginning to show effects. She cited the Treatwell 5-A-Day Study, carried out by Glorian Sorenson and her colleagues at the Dana-Farber Cancer Institute, as an example. Community health centers in the Northeast were randomly assigned to one of three treatment arms: some community health centers had only a work-site intervention; others had a work-site-plus-family intervention; and a third group of centers, which had only a minimal intervention, served as the control. In the first group, people in participating community health centers received the Treatwell 5-A-Day series of 10 interventions aimed at improving diet in general and increasing their intake of fruits and vegetables in particular, and they were exposed to annual campaigns in nutrition education. In addition, health center staff actively worked with them to make changes in their work site that would increase the availability of healthy foods in snack rooms, vending machines, and throughout the work setting. “They were directly impacting the social environment in which people lived to change the way that they ate,” Dr. Chesney said. Community health centers assigned to a work-site-plus-family intervention did all of the above but also went one step farther. They provided a five-session “Fit in 5” program in which families could learn at home, along with newsletters and other follow-up incentives, events, and materials designed to motivate the family to change its dietary habits. At the end of this 19-month program, the groups that received the work site and family interventions showed significantly greater increases in fruit and vegetable intake than the control group (which received only a modest amount of dietary information). Most impressively, the increase in the work-site-plus-family intervention group was almost three times that of the group exposed to work site interventions alone (see Figure A). The focus of this talk, Dr. Chesney said, was the contribution of behavioral interventions to health promotion and disease prevention. But she

OCR for page 16
FIGURE A Treatwell 5-A-Day study in fruit and vegetable intake. also wanted to briefly mention psychosocial interventions, such as those aiming to reduce stress, depression, depressed mood, and social isolation. “Here we do have growing evidence that psychosocial interventions, which target coping skills and provide social support, can contribute to treatment, particularly in chronic disease management,” she said. “These interventions, which are typically individual- and group-based, also need now to move from the downstream level to midstream and upstream . . . so that they reach more diverse groups and populations across our entire nation.” We need a stronger science base there as well, she added. Dr. Chesney concluded by noting that the time has come “for us to design, to test, and to implement behavioral and social interventions to improve health across the life span, beginning with the very young, and including the growing numbers of the oldest old, and to extend these efforts to the diverse groups that populate our communities.” This will require an ambitious but attainable partnership of public health officials, researchers, and community members. Her hope, she said, is that “when I come here in 2010 and we talk about . . . the objectives of Healthy People 2010, we can say that we have actually hit 100 percent” of those objectives.