Prevention Science

, Volume 9, Issue 3, pp 166–177

Implementation Science: Promoting Science-Based Approaches to Prevent Teen Pregnancy


DOI: 10.1007/s11121-008-0094-9

Cite this article as:
Philliber, S. & Nolte, K. Prev Sci (2008) 9: 166. doi:10.1007/s11121-008-0094-9


This paper reports the results of a project funded by the Centers for Disease Control and Prevention intended to promote the use of science-based approaches to teen pregnancy prevention. As with other efforts to promote diffusion of innovations, adoption of these successful programs faced a number of barriers including lack of knowledge of programs that work, lack of funding for training and materials, devaluing science-based approaches, complexity of successful programs, politics, funding streams and compatibility with particular community characteristics. Nevertheless, five state and three national teen pregnancy organizations provided intensive technical assistance, produced materials, and provided training to encourage use of programs that work. Local barriers to their work included the fact that teen pregnancy rates were already dropping, instability of funding to pay for such programs, turnover of agency staff, the need for intensive follow-up to promote adoption, the internal organization of the initiative, and the fragility of local teen pregnancy prevention coalitions. Still, in each of five states, there was increased adoption of science-based approaches to prevent teen pregnancy.


Science-based Teen pregnancy Diffusion of innovation 

In the past two decades, there has been a rapid accumulation of evidence about effective programs to reduce teen pregnancy. Perhaps one of the most telling indicators of this growth in effective interventions is that in his first comprehensive summary of strategies to reduce teen pregnancy, Kirby (1997) reviewed some 50 programs in a document cautiously titled No Easy Answers. Each rigorously evaluated, 29 or 58% of these programs had a positive impact on some prevention-related behavior such as use of contraception, fewer sexual partners, delay in onset of sexual intercourse, or on pregnancies themselves. Only 4 years later, Kirby (2001) published the more hopeful Emerging Answers and reviewed 74 rigorously evaluated programs, 44 or 59% of which had a behavioral impact. By 2005, a review of 83 sex and HIV education programs in developed and developing countries found 54 or 65% of them to have positive impacts on a teen pregnancy-related behavior (Kirby et al. 2005). Over time then, there have been gains in developing and testing effective programs to reduce teen pregnancy. However, the literature on adoption of intervention suggests that the actual adoption of these programs will take some time. Local communities are often still implementing ineffective and/or unevaluated programs, and using programs developed locally, usually using parts of various curricula and strategies.

In the field of addiction treatment, McGovern et al. (2004), lamented:

The research to practice gap is a major concern in addiction treatment delivery (2004, p. 305).

Similarly, Elliott and Mihalic (2004) predicted:

The new frontier in ...intervention research involves building a scientific knowledge base for the replication and dissemination of those programs that have been demonstrated to be effective (p. 4).

Thus the teen pregnancy prevention field has discovered what others have before: It is not enough to have strategies that work. It is also necessary to get program adopters to use what works and use it with fidelity.

In 2002 the Centers for Disease Control and Prevention (CDC) launched a 3-year initiative called “Coalition Capacity Building to Prevent Teen Pregnancy.” The purpose of the initiative was to increase the capacity of local organizations to use a science-based approach (SBA) to prevent teen pregnancy and promote adolescent reproductive health including abstinence, and prevention of sexually transmitted infections (STIs) and HIV. This paper reports on the activities and lessons of the first phase of this program, after a brief review of relevant literature on adoption of innovations and the relevance of this literature to teen pregnancy prevention.

Specifically, the paper: 1) briefly describes the program that CDC funded; 2) describes both local and contextual barriers to adoption of science-based approaches to teen pregnancy prevention, identified over the course of project; 3) describes how some of these barriers were overcome; 4) provides examples of increased adoption of SBAs; and 5) makes recommendations for future work.

The CDC Program

The “Coalition Capacity Building to Prevent Teen Pregnancy” program was designed to increase the capacity of local organizations to use a science-based approach (SBA) to prevent teen pregnancy and promote adolescent reproductive health, including abstinence, and prevention of STIs and HIV.

In 2002, CDC reviewed 24 competitive applications and awarded cooperative agreements to three national and five state coalitions working in the teen pregnancy prevention field.1 The program began with a relatively simple logic model or theory of change. With CDC funding, the national organizations would provide technical assistance to the state organizations to increase their capacity to promote science-based approaches that, in turn, would help local organizations adopt science-based programs and practices to prevent teen pregnancy. Both the national and state organizations were to create materials, publish and disseminate information to the field, develop and implement training, and provide technical assistance to state and local organizations to reach this goal.

A science-based approach (SBA) was defined as the process for developing a strategy to prevent teen pregnancy to help ensure that the programs implemented have a greater chance of succeeding. Such an approach was defined to include the following:
  • Using demographic, epidemiological and social science research to identify populations at risk of early pregnancy and/or sexually transmitted infections, and to identify the risk and protective factors for those populations;

  • Using health behavior or health education theory to guide the selection of risk and protective factors that will be addressed by the program, and to guide the selection of intervention activities;

  • Using a logic model to link risk and protective factors with program strategies and outcomes;

  • Selecting, adapting, if necessary, and implementing programs that are science-based;

  • Conducting process and outcome evaluation of the implemented program and modifying the approach based on results.

Research on Barriers to Adoption of Effective Approaches

CDC is one of the first federal agencies to address the research-to-practice gap in teen pregnancy by promoting a science-based approach. Numerous federal agencies have encouraged adoption of evidence-based prevention approaches, including the National Institute on Drug Abuse (NIDA), the U.S. Department of Education’s Safe and Drug Free Schools Program, and the U.S. Department of Justice’s Office of Juvenile Justice and Delinquency Prevention (OJJDP). In spite of these efforts, Ringwalt et al. (2002) claim that only 30% of the nation’s schools are implementing any kind of evidence-based drug prevention program. Similarly, Ennett et al. (2003) reported that only 14% of school substance use prevention programs used both effective content and effective delivery methods. Why would this be? Why wouldn’t those who try to prevent negative outcomes want to use science-based practices?

One of the major struggles in adopting science-based programs has been the tension between fidelity of replication and adaptation of programs. Once program users know about evidence-based programs (the first hurdle to be overcome in encouraging their use), there is a tendency to adopt part of a program or to edit the program in some way. In a study to monitor these behaviors, Dusenbury et al. (2005) found that all of the 11 teachers asked to implement a new program on life skills made adaptations to it.

The motivations for adapting programs are many and varied. They include the desire to make the new program more culturally appropriate, to make it shorter, to make it longer, to use only the portions that make the program facilitator comfortable, to accede to political pressures to delete or add material, and to make the program easier to implement.

While much of this research has examined replication or adoption of drug, alcohol, or violence prevention programs, faithful use of teen pregnancy prevention programs may face even greater challenges. Local community pressures stemming from arguments over abstinence-only versus more comprehensive sexuality education programs can lead to severe editing of programs that have been shown to be effective as they were. School systems implementing such programs often feel vulnerable to parent complaints about the messages of these curricula.

Elliott and Mihalic (2004) found that failure to faithfully replicate science-based programs is often detrimental to their original effectiveness. Reporting on a study designed to track the replication of “Blueprint Programs” in violence prevention, they write:

“The adaptations that did occur... were typically made by front line treatment staff or teachers... They were typically made without developer consultations; without any knowledge of the theoretical/conceptual rationale for the program; and they typically involved dropping entire components, levels of required training and dosage levels.” (p. 51).

These authors also argue that the assumption that programs must have separate curricula for each age, gender and racial/ethnic group is “unwarranted,” and that those who insist that such adaptations be made are often politically, rather than empirically, motivated. Elliott and Mihalic (2004) further suggest that negotiating with a community about adapting effective programs rather than emphasizing implementation with fidelity “could backfire and undermine public confidence in scientific claims that we have programs that work...” (p. 52).

Others have argued that part of the challenge in getting widespread use of science-based programs is the simplistic view of those who create and evaluate such programs. Miller and Shinn (2005), for example, argue that there is often “a mismatch between what scientists design and what communities have the capacity to implement.” (p. 170) The program design phase may ignore values that are important in program choice in given communities. Miller and Shinn also argue that program designers and researchers may have a “pro-innovation bias,” believing that evidence-based programs must have benefits over indigenous programs that have not been studied (p. 174), a proposition not so easily accepted by those using these programs locally over some period of time. Even the very conditions under which these programs are tested; i.e., carefully controlled implementation, carefully drawn samples, ample funding, and with the best trained staff, may be antithetical to the “real-world” environments to which these programs are then disseminated (Glasgow et al. 2004). To overcome some of these obstacles, Miller et al. (2003) have proposed researcher/community partnerships to develop and test programs.

The Contribution of Diffusion of Innovation Research and Theory

From cornfields to high tech boardrooms, those interested in the diffusion of innovation have puzzled over the factors that lead to adoption of new technologies, gadgets, programs, and designs, and how to decrease the time between invention and adoption. The Diffusion of Innovation theory developed by Everett Rogers in 1962 refers to the spread and adoption of new ideas, techniques, behaviors or products throughout a population. In his classic text on diffusion of innovation, Rogers (2003) suggests there are four main elements of this process:
  1. 1.

    The innovation, or an idea or practice that is perceived as new by an individual or other unit.

This includes the perceived attributes of the innovation, its relative advantages over current ideas, its consistency with existing values of the adopter, how complex the innovation is and whether its advantages can be readily perceived. In teen pregnancy prevention work, many programs do not have a tradition of evaluating their programs and thus do not know how effective they have been in actually preventing early pregnancies. Some teen pregnancy prevention programs are complex to implement, and are thus harder to adopt.
  1. 2.

    Communication or the means by which messages get from one individual to another.

Channels of communication may include mass media, individual or interpersonal channels and other strategies. The speed and efficiency of learning about an innovation affect its rate of adoption. While many national organizations have published numerous documents about “what works” to prevent teen pregnancy, these communication strategies may not reach front-line youth workers who are selecting programs for local youth.
  1. 3.

    Time or the amount of time it takes to adopt the new idea or practice, including the decision process.

This includes time required to learn the new innovation or acquire it, as well as the time it takes to decide whether or not to adopt the innovation. In teen pregnancy prevention programs, it may require a great deal of time to be trained to use a new curriculum and a decision for a school board to adopt this curriculum may require multiple meetings, as well as some behind-the-scenes lobbying.
  1. 4.

    The social system into which the innovation is introduced.


These are the individuals, groups, and/or organizations interacting together to accomplish a common goal, such as adoption of an effective strategy for preventing teen pregnancy in the community. This system is likely to have opinion leaders, norms of its own, and a social structure that may be more or less cumbersome. In adopting teen pregnancy prevention innovations, some of the members of this group might be conservative, while others are more liberal—leading to conflicts about curriculum content or emphases. This framework has been applied to the studies of numerous and varied innovations.

In an extensive review of innovation diffusion in health service delivery, Greenhalgh et al. (2004) present a conceptual model on the determinants of such diffusion that includes some of the same factors cited by Rogers. They argue that many factors, including characteristics of the innovation itself (e.g., complexity, advantages) and the outer context (e.g., sociopolitical climate, environmental stability) have been found to influence adoption.

Pankratz et al. (2002), studied some of these principles and found three constructs related to schools’ adoption of a federal drug prevention policy:
  • advantage/compatibility or the degree to which an innovation is perceived as better than previous practice and is consistent with the existing values, past experiences, and needs of potential adopters;

  • complexity or the degree to which an innovation is perceived as difficult to understand and use; and

  • observability or the degree to which the results of the innovation are visible to others, so that small or non-noticeable changes are less likely to be adopted.

These specific factors are subsets of Rogers’ more general framework and advance our knowledge of potential barriers to the adoption of innovation.

In the field of teen pregnancy prevention, some of these same challenges and guidelines apply. The CDC program discussed here builds on this research literature.

Data Sources

To assemble the data for this paper, several strategies were used. Each of the eight funded sites provided reports to CDC every six months, detailing their progress, achievements, and challenges. Seven of the eight sites also provided evaluation reports from independent, local or national evaluators, including data on their events, their activities, or information from surveys. The same seven sites also supplied a cumulative 3-year report to CDC in 2005. All of these documents were reviewed to extract themes and specific data for this report. Themes were noted in each document and summarized as the analysis proceeded. Information from the needs assessments largely came from the evaluation reports provided by the evaluators attached to each project. These too were read for their themes and quantitative data.

In addition, coordinators at each site were sent a brief survey to get their final perceptions of their biggest accomplishments and challenges during the 3-year project. Finally, program directors were called to obtain additional numbers, information, or clarification. This report represents a synthesis of this information.

Local Barriers to the Use of Science-based Approaches

While the generic barriers to adoption of new programs and strategies have been identified in the research and theoretical work described above, this project began by researching the specific local barriers that might hamper the adoption of effective teen pregnancy prevention strategies. CDC asked their state organization grantees to select five to ten local organizations for whom they would provide targeted intense technical assistance over the three years of the project. Each grantee designed its own needs assessment tool to determine the knowledge of and barriers to the adoption of science-based approaches and programs, understand the organization’s value and use of an SBA, and determine its capacity to select, implement, and evaluate an SBA. Selected organizations consisted of local teen pregnancy prevention organizations, school boards, and state health departments. The grantees also assessed through surveys, meetings, or focus groups, what strategies for overcoming these barriers would be most well received by these groups, including publications, video conferences, annual institutes or conferences, tailored technical assistance and local workshops.

Together these eight organizations surveyed hundreds of youth workers, agency heads, funders, decision makers, and state administrators, using a variety of sampling techniques, including their national membership lists, lists of funders known to support teen pregnancy prevention projects, lists of local organizations known to state coalitions, government lists of those directing teen pregnancy-related efforts at the state level and other sources. None of the samples was random but all were thoughtful convenience samples.

Perhaps the most remarkable result of these needs assessments, across different locations and different constituencies, was the congruence of their findings. Taken together, these surveys and interviews found barriers in each of Roger’s categories of factors affecting adoption of innovation:

Characteristics of the Innovation

Lack of Funds for Training and Materials

Most of the teen pregnancy prevention programs that “work” require training or materials to be successfully implemented. The most frequent barrier reported was lack of funds to buy materials or acquire such training.

Communication about the Innovation

Lack of Knowledge of Such Approaches

Many of those surveyed did not know what programs had been found to be effective in preventing teen pregnancy. They were not familiar with the characteristics of effective programs and did not know how to find or access such programs.

Time Required to Adopt the Innovation


Workers expressed the need for programs that were easily implemented, given their limited resources. Even if a program works, some agencies did not have the capacity to implement it if it required many staff, staff with special skills, or a great deal of organization.

Characteristics of the Social System into Which the Innovation is Introduced

Devaluing Science-based Approaches

Some of those surveyed did not see the reason to give preference to well-evaluated programs. They were working in systems that did not require evaluation or systems that valued certain messages over proven effectiveness in choosing programs. In this group too were some who were skeptical about the research showing effectiveness of programs, questioning the political motivations of those developing programs and questioning the quality of the research.


Many of those surveyed felt they lived in communities that would not be open to programs that included information about sexuality-related issues and contraceptive use.

Funding Streams

Some of those surveyed said that their communities had received funds that had to be used for specific programs or types of programs. Funds were reported in short supply for other programs, regardless of their demonstrated effectiveness.

Compatibility with Community Characteristics

Many of those surveyed did not believe that the programs found to be effective would necessarily work in their communities. Their concerns were most often about cultural appropriateness but some mentioned whether programs from the inner-city would work in rural areas, or whether programs that worked in the North would work in the South.

Grantee Implementation Strategies

Given the findings of these needs assessments, the eight funded organizations responded with multi-level strategies. Information was disseminated through national and state conferences, reports, scientific briefs, newsletters, fact sheets, websites, listservs, and other means, to increase awareness and knowledge. Structured learning time during trainings, roundtables, and networking sessions served to increase knowledge and skills. Intensive technical assistance provided through regular phone calls, site visits, and one-on-one consultations with national experts sought to increase motivation to use SBA. The volume of research briefs, longer publications, fact sheets and other publications was substantial. Among them, these eight organizations produced 93 research briefs or fact sheets and 18 longer reports. There were more than 100 presentations, roundtables or other trainings offered. Information could be easily accessed through the internet. The National Campaign alone distributed 550,000 products to the field of teen pregnancy prevention through its website and printed distribution including 71,000 downloads of PowerPoint presentations.

The national organizations worked with the state coalitions to strengthen both internal operations such as strategic planning, leadership, fund-raising, board development and website development, and external capacities including services and products to local organizations.

Advocates for Youth developed a package of technical assistance and training tailored to each organization based on their needs assessment results. This package of technical assistance included aid on website development, board leadership, training-of-trainers, local council development and fund-raising. To support the state organizations’ work, Advocates created publications on science-based practices, integration of HIV and STI prevention with pregnancy prevention, state-based fact sheets, annotated bibliographies, and an extensive series of training modules.

Healthy Teen Network, together with project partner ETR Associates, provided capacity building technical assistance, training, presentation tools and other resources to increase the use of SBA. They provided training to state and local organizations on logic model development, risk and protective factors, characteristics of effective programs, integrating HIV, STIs and teen pregnancy prevention, and evaluation, among other topics.

The National Campaign developed and broadly disseminated easy-to-read research-based materials on teen pregnancy prevention and related topics. Science Says is one such effort. The series of 20 four- to six-page briefs summarized key survey data, research studies, and related information on adolescent reproductive health. The series was available free of charge on the Campaign’s website and was disseminated to organizations that work with youth on the national, state and local levels, and to parents and policymakers. The National Campaign encouraged those working directly on teen pregnancy prevention and those with broader agendas, to incorporate science-based approaches into their work. PowerPoint presentations were created for almost all full-length research reports and Science Says research briefs. States and communities downloaded these presentations from the National Campaign’s website. States and communities were able to adapt them for their own use in presentations to school boards, health departments, parents, and others.

The state organizations developed a multi-dimensional approach to providing intensive technical assistance to communities or local organizations. This technical assistance began with a needs assessment. The state organizations then implemented a series of trainings focused on the specific needs identified. Repeated communication, feedback, troubleshooting through phone calls, site visits, and roundtables throughout the life of the project sought to increase the capacity of local organizations to develop logic models, select a science-based program that fit the community values and was effective for their target population, or help them strengthen their current program to include the characteristics of effective programs.

The South Carolina Campaign to Prevent Teen Pregnancy visited 40 community adolescent pregnancy prevention programs semi-annually to provide on-site technical assistance. They organized one-on-one meetings between the adolescent pregnancy prevention programs and national experts to review the local organizations’ logic models and evaluation plans to assess feasibility and provide assistance to strengthen the work plans. It was this intensive, individualized approach to technical assistance that strengthened the local capacity to overcome barriers and be able to select, implement, and evaluate a science-based approach.

Thus, each of the barriers to adoption of science-based approaches was addressed in some way and by at least one of the eight agencies funded by CDC. All of them addressed the lack of knowledge about SBA and helped organizations find programs that matched both their resources and community preferences and characteristics. Some paid for training and materials so that organizations could afford to adopt a science-based approach. Still others tried to lessen system barriers by providing training about evaluation and its value, by convening community and funder groups to address the political and funding stream barriers and then facilitating selection of an appropriate approach.

Challenges to Increasing Capacity

In their semi-annual reports to CDC, these eight organizations cited a variety of contextual or organizational challenges that made it difficult to overcome barriers to adopting science-based approaches. These too, can be classified into Roger’s categories of factors affecting adoption of innovation and included:

Communication Issues

Turnover at All Levels

Turnover among local agency staff, state grantees, and local coalition partners made it difficult to build lasting capacity. Technical assistance focused on only a few key staff members could be lost, at least to the organization targeted for help.

Organizational Relationships

There were logistical challenges to the work as well—factors about how the program was organized that created issues in the beginning. One of these was the choice to fund three national organizations. These organizations are all pre-eminent in the field of teen pregnancy prevention but were not used to working together to coordinate their work. They did not all have the same expertise and each was tasked to provide training and technical assistance to the state and local organizations. Sometimes the state organizations felt they were inundated with “help.” Early in the program mechanisms had to be created to orchestrate how assistance was offered and coordinated.

Another challenge was that there was no clear hierarchy in knowledge or experience between these national and state grantees. Even though the national organizations served a broader constituency, the state organizations sometimes had more experience in working with local agencies. The idea that all of the help would flow “down” did not always sit well with the state organizations. In fact, during the course of the project experience and expertise flowed in multiple directions.

Time Required to Provide Assistance

The Need for Individualized Follow-up

One training is not enough to affect change in a community or organization. After training about science-based programs or approaches, state organizations provided additional assistance with knowledge and technology transfer, funding, recruitment, help to prevent attrition among young people, and resistance to the continuing temptation of program implementers to edit or alter these programs. Such assistance was resource- and time-intensive even though state organizations reported that it did lead to increased adoption of science-based approaches.

Characteristics of the Social System or Context

Decreasing Teen Pregnancy Rates

For the decade prior to this project, teen pregnancy rates had been dropping. In some communities this had apparently given rise to some complacency about this issue, even though the nation still has the highest teen pregnancy rate of any developed country.

Instability of Funding for Teen Pregnancy Prevention Programs

While a general lack of funding for teen pregnancy prevention has already been noted above as a major barrier to use of science-based approaches to prevent teen pregnancy, in two grantee states, state funding for teen pregnancy prevention programs was cut drastically just as this initiative began. This meant that the local organizations and agencies that depended on this money had to reduce or eliminate their teen pregnancy programs, greatly decreasing the demand for training and assistance from their state organizations. In another state some agencies were unexpectedly so poor that they could not even find the transportation funds to attend free trainings offered by their state organization through this program.

The Fragility of Local Coalitions

State organizations that worked with local coalitions found them to be vulnerable, under-funded, largely volunteer and thus, fragile. In North Carolina, 3 of 10 of these local coalitions collapsed completely during the life of the project; in Massachusetts only 2 of the 17 local coalitions remained. This raises questions about whether these groups are the best entry point to effect change in use of science-based practices, rather than strong single agencies or even systems such as state or local health departments, or departments of education that already have networks in which they can influence change. These barriers to helping organizations as well as the local barriers to adoption science-based approaches to teen pregnancy prevention are summarized in Fig. 1, again using Roger’s conceptual framework.
Fig. 1

Local barriers to use of science-based approaches and solutions used by national and state organizations

Strategies to Overcome Barriers

As discussed above, the project succeeded in identifying specific barriers to use of science-based approaches to prevent teen pregnancy and barriers experienced by the CDC-funded organizations in lessening these barriers, all consistent with Roger’s categorization of factors affecting adoption of innovation. It also succeeded in developing training and technical assistance tools to help local and state organizations overcome these barriers. There is ample evidence of increases in awareness, capacity and use of science-based approaches among those reached by these grantees.

Characteristics of the Innovation

Teen pregnancy programs that are effective require materials and training. Since many local organizations said they could not afford the training or materials from science-based programs, the Arizona coalition used part of its funding to buy the Reducing the Risk curriculum and paid for local organizations to be trained in its use. The South Carolina Campaign sent copies of materials produced by the National Campaign to 80 providers and they became certified trainers for the Teen Outreach Program (a nationally evaluated science-based program shown to reduce teen pregnancy). This made local training in this program more accessible. The North Carolina coalition provided scholarships for training and bought curricula for local organizations.

Communication Barriers

To overcome lack of knowledge about effective programs, each state provided trainings on SBAs, logic models, evaluation, and characteristics of effective programs. Annual meetings also provided forums to highlight science-based approaches and provide networking opportunities. Advocates for Youth helped organizations improve their web site design, library and online information services to promote and disseminate information on science-based approaches. Improvements to the state organization websites included easier navigation, improved accessibility, easily printed content, accessible curriculum reviews, and new pages such as model science-based programs and current statistics. Healthy Teen Network, and the coalitions from Minnesota, North Carolina, South Carolina, and Massachusetts also have data showing improvements in knowledge among those they reached.

Staff turnover and thus loss of capacity was addressed by making sure that all materials produced about SBAs were readily available to incoming staff at each local organization. South Carolina went a step further and developed a quarterly “new provider” orientation training program and manual. The 2-day workshop was designed to introduce new providers in the field to a variety of topics related to planning effective teen pregnancy prevention programs and SBAs.

Evaluations completed by each state and national organization revealed a strong desire on the part of various groups to receive information on their own schedules and in convenient short formats. Websites of information created for this project experienced heavy use and follow-up data from those receiving publications revealed that shorter publications were more likely to be read, shared and used than were longer ones.

Social System Barriers

To overcome political controversy, the National Campaign summarized the content of science-based programs so that communities could choose those curricula most suitable for their values and target population. As part of the work in Massachusetts, town hall meetings were held to help a community come to consensus on how to approach teen pregnancy and select an effective program that met the community’s needs.

As noted above, the loss of state funding for teen pregnancy prevention programs in some of these states dealt a crippling blow to the work of some of these organizations. Since they had to turn to survival issues, agencies normally doing teen pregnancy prevention work could hardly be interested in refining the quality of their programs. Interestingly, in the two states hardest hit by these budget cuts, the policy work and coalition building work done under this CDC-funded program contributed to restoring, and even increasing, available funds. In Massachusetts, faced with having almost all state funding for teen pregnancy prevention cut in 2002, the Massachusetts Alliance successfully turned the crisis into an opportunity by providing leadership, information, and interventions that shifted all teen pregnancy prevention programming in the state toward a science-based approach. As a result of their work, state funding for teen pregnancy prevention has increased by $1 million dollars. Three science-based “Request for Responses” were developed in Massachusetts with state and local funding. The state went from no science-based teen pregnancy prevention programs being implemented across the state to now 19 science-based programs being implemented and reaching 1050 adolescents.

Figure 1 shows these strategies against the barriers they sought to address, again using Rogers’ framework.

Adoption of Science-based Approaches

Perhaps most importantly, there are examples in each funded state of increased adoption of science-based programs (see Fig. 2). Sometimes this was adoption of a specific curriculum shown to be effective in preventing teen pregnancy, usually across a set of programs or in an entire school district or county. Thousands more children were reached with programs and approaches that have been shown to be effective.
Fig. 2

Examples of increased adoption of science-based programs and approaches

The South Carolina Campaign to Prevent Teen Pregnancy reported that at the inception of the program none of the 40 community adolescent pregnancy prevention programs had a logic model. By the end of the three years 100% had a diagrammed logic model in which 88% clearly identified risk and protective factors for teen pregnancy, and 79% clearly identified activities designed to change associated determinants. At the end of the first year only seven programs used an SBA. By the end of the third year all programs reported using structured, planned curricula in their interventions and 94% stated they used at least one science-based curriculum during the year. Satisfaction surveys were replaced by 98% of the programs conducting process evaluations, and 86% of the programs reported measuring changes in risk and protective factors.

The Massachusetts Alliance on Teen Pregnancy Prevention provided intensive technical assistance to a community in Western Massachusetts. Working with a task force of local stakeholders, they designed a community-wide science-based strategy to prevent teen pregnancy. In an eight-month partnership, the Alliance and this community selected a research-based teen pregnancy prevention program that met community needs and gained community-wide support.

As part of this technical assistance, the Alliance trained the task force members in science-based approaches to prevent teen pregnancy. They compiled and distributed an orientation binder on science-based approaches and presented at task force meetings. The Alliance also kept the work moving by contacting members to insure that tasks were being completed, and helped prepare materials to present the chosen intervention to the larger community. Supportive materials provided by the Alliance included data on local teen birth rates, the logic model of the intervention, a press release, a letter to the editor by a local teen, an overview of the selected program, and information on science-based approaches (Massachusetts Alliance on Teen Pregnancy Prevention 2005).

Another result of this project is that teaching others to use these approaches led some of the state and national organizations to adopt such approaches themselves. For example, in one state that had never before created a logic model for its own work, there is now a logic model for the state coalition as a whole and for each of its projects.

The state coalitions uniformly reported that intensive technical assistance, provided to local agencies and coalitions, was an effective strategy for producing adoption of science-based approaches. Data from the South Carolina Campaign show that providers who attended seven or more trainings conducted by the Campaign had greater confidence in creating logic models, implementing a curriculum with fidelity, and collecting and using evaluation data to improve programs. Some state organizations were concerned however, that such a strategy is not sustainable. One state coalition wrote in its final report:

“...although the more intensive TA efforts in Years 2 and 3 appeared to be more productive in advancing the use of science-based practices among community agencies, that level of intensity is not feasible to maintain with current staffing patterns.”


The same barriers to diffusion of innovation offered by previous theory and much research, although cloaked in teen-pregnancy specific language, have been encountered in this project. There are barriers that have to do with the characteristics of the innovation itself (costly materials and training are required for adoption), with communication (some programs still do not know what works to prevent teen pregnancy), with time required from both innovation adopters and from those promoting the innovation (program complexity and intensive technical assistance), and with the social system into which the innovation is offered (the political barriers, lack of funding, the perceived seriousness of the problem and, specific community characteristics).

This project developed multiple strategies for overcoming these barriers including materials, training, and intensive technical assistance. Across the funded sites, use of these strategies resulted in numerous examples of adopting science-based approaches to teen pregnancy prevention.

There are three particular areas on which more work is needed. One is that more materials and trainings are needed to focus on the specific barriers to the use of science-based approaches to teen pregnancy prevention. For example communities need to know how to frame teen pregnancy prevention to create political will, how to handle controversy effectively, and how to choose programs most suited to their populations. They need to understand which programs are least controversial and which are most easily implemented. These barriers are nearly ubiquitous and need to be anticipated.

Another area that received little attention in the first round of funding was partnering with funders and systems to help them advocate for the adoption of science-based strategies within their sphere of influence. Two sites were exceptions and in fact, contributed to a statewide adoption of science-based programming. Their work appears to demonstrate that a great deal can be accomplished by working with systems and funders, and adding advocacy to technical assistance and training

Still a third challenge is the ever-present tendency of program adopters to alter science-based curricula, to leave out parts of programs, or in other ways to fail to replicate the program with fidelity. Elliott and Mihalic (2004) have taken the position that “bargaining” with communities to let them modify a program is not a productive strategy and that program effectiveness would be better served by insisting on and making possible a faithful replication. Still, the tendency to alter programs is connected to their complexity, to political pressures, and to program resources. In teen pregnancy prevention, these forces are unlikely to disappear and more work is needed to either minimize these pressures or to create some sound principles by which such alteration could be done without serious damage.

This CDC-funded program to promote use of science-based approaches to prevent teen pregnancy is now beginning its second phase of work. The first three years, reported here, have been successful in identifying barriers to the adoption of such approaches, in developing training and TA strategies for overcoming these barriers, and in producing adoption of science-based approaches to prevent teen pregnancy. In the second phase of work, CDC has funded 16 national, regional and state organizations in hopes of increasing the coverage, scope and impact of this effort.

To guide the second phase of this work, a more complex conceptual model has been adopted. The initiative is clearer on its desired outcomes, its definitions, and its strategies. In the next five years, all of the regional, state and local organizations funded or reached will complete common needs assessments. While during the first phase, each grantee conducted unique evaluations, during this next funding period, common evaluation elements are being introduced to even more thoroughly measure the results of the whole initiative.


Advocates for Youth; Healthy Teen Network (then the National Organization for Adolescent Pregnancy, Parenting, and Prevention or NOAPPP); the National Campaign to Prevent Teen Pregnancy; the Arizona Coalition on Adolescent Pregnancy and Prevention, the Massachusetts Alliance on Teen Pregnancy Prevention; the South Carolina Campaign to Prevent Teen Pregnancy; the Minnesota Organization on Adolescent Pregnancy, Prevention and Parenting; and the Adolescent Pregnancy Prevention Coalition of North Carolina.


Copyright information

© Society for Prevention Research 2008

Authors and Affiliations

  1. 1.Philliber Research AssociatesAccordUSA
  2. 2.The Georgia Campaign for Adolescent Pregnancy PreventionAtlantaUSA

Personalised recommendations