A variety of lessons can be drawn from the early implementation of the SafeFutures initiative. Some lessons are common to any complex demonstration; others are encountered less frequently and may result from SafeFutures’ emphasis on collaboration combined with implementation of specific component programming. The findings identified here are grouped into three categories: lessons related to funded demonstration projects, lessons related to community-based collaboratives, and lessons related to service provision. Classifying a given point is not always easy because some findings undoubtedly apply to more than one category.

• An iterative, flexible approach is needed on the part of funders and demonstration sites to implement complex, multifaceted initiatives. SafeFutures uses cross-site cluster conferences for this purpose, in addition to the ongoing training and technical assistance on key program elements.

  ross-site conferences also provide opportunities for sites to exchange information and to learn from each other’s experiences.

  Cross-site conferences have facilitated an ongoing dialog that has clarified and shaped the viewpoints of both local administrators/program staff and Federal agency representatives regarding how SafeFutures should be implemented with respect to systems reforms and specific programmatic components. Cross-site conferences also provide opportunities for sites to exchange information and to learn from each other’s experiences—facilitating cross-fertilization of successful or innovative programs or techniques employed to address challenges or issues. Electronic communication also can serve this purpose.

  OJJDP modified some policies in response to difficulties encountered by sites. For example, in year 3, OJJDP modified its expectations related to mentoring programs to allow sites flexibility regarding the costs of making and supporting each “match.” OJJDP originally required sites to meet the cost estimate of $1,000 per match (which was based on a study of costs associated with Big Brothers Big Sisters mentoring programs). After several sites reported difficulty matching this cost, OJJDP reviewed the study and costs associated with other mentoring programs. OJJDP determined that this amount might not be an appropriate “benchmark” for the kinds of mentoring programs operated by SafeFutures sites (e.g., focusing on high-risk youth or those already in the juvenile justice system) and that the requirement could be relaxed where sites could justify their inability to meet it.

• Communities need access to ongoing training and technical assistance to help implement components with highly structured requirements—such as mentoring or the gang-free schools and communities component with its required implementation of the Spergel model of gang intervention—and to adapt generic models to the local context. Assistance also is needed to reconcile differing interpretations by local and Federal actors with respect to the theoretical principles underlying SafeFutures and their practical use (e.g., what constitutes adequate procedures to accomplish community risk and resource assessments). Training and technical assistance resources also have been deployed to assist sites with implementation of evolving concepts and approaches.

  Flexibility in providing additional training and technical assistance regarding areas with which sites are experiencing difficulty also is desirable. For example, OJJDP enabled sites to obtain additional technical assistance regarding systems change and the Spergel model, aspects of SafeFutures that were perceived as particularly challenging. Further, to facilitate collaborative, multidimensional initiatives such as SafeFutures, training and technical assistance supported by OJJDP has gone beyond traditional “categorical” approaches focused on a single program to address topics such as systems improvement and comprehensive strategic planning.

  Continuity of technical assistance providers is important—particularly with those the community views as a “good fit” for them and with whom the community has established a satisfactory working relationship.

• Small service providers generally need more and different types of technical assistance and training than large, well-established organizations. Small, community-based organizations are often less familiar with some aspects of program implementation such as accountability, recordkeeping and record reporting, program evaluation, and other requirements commonly associated with demonstration programs. Staff of such organizations often are particularly concerned about recordkeeping and tracking requirements associated with monitoring clients to assess the outcomes of a demonstration program. This is due partly to concerns about maintaining the confidentiality of sensitive information. In addition, service delivery staff are primarily focused on working with youth/families and may believe that recordkeeping and monitoring take time away from their service provision role. Local program administrators frequently need to spend more time and effort with such organizations to ensure that they understand and are able to comply with various requirements of an initiative.

  emonstration sites need flexibility to exercise cultural sensitivity and competence in program implementation.

• Demonstration sites need flexibility to exercise cultural sensitivity and competence in program implementation. Although several programs in each SafeFutures community were specifically tailored to the cultural context of a targeted population, staff and service providers encountered difficulties adapting some components, such as mentoring, to ensure their relevance to the cultural context. Two sites experienced problems because the concept of one-to-one mentoring is not part of the culture of the respective target populations in those sites.

• Communities and program staff appear to be reluctant to impose eligibility criteria to ensure that they serve youth at highest risk or in greatest need. Many staff seem to regard all youth residing in target areas as being at risk, without seeing a need to make efforts to identify those at greater risk. Some programs rely heavily on self-referral or parental referral for client recruitment. This is logical from the perspective of ensuring that clients are willing to participate in program activities, but it may result in a failure to serve youth at highest risk because such individuals might be less likely to self-select for participation. By the end of year 3, several sites began using tools and processes to identify and focus on higher risk youth.

  Communities have varied in the type and number of youth on whom their initiatives, or specific program components, focus. Some sites have chosen to spread resources around to reach as many youth in the community as possible, while others have concentrated resources to provide more intensive services to a smaller number of higher risk youth. The former approach is more commonly associated with prevention programs, while the latter is more common to intervention programming.

  eplication of programs that worked in other communities does not guarantee similarly positive results in a new setting.

• Replication of programs that worked in other communities (or under different circumstances within the local community) does not guarantee similarly positive results in a new setting. It may be difficult to isolate and completely duplicate the features specifically responsible for success. The original program’s success may have been due to charismatic leadership that is not easily transplanted. Other program factors also may not translate well in a new environment, and adapting a program may dissipate and undermine its effectiveness.

• It is desirable to address program sustainability well in advance of anticipated termination of Federal support. Some SafeFutures programs—such as Volunteers in Probation (mentoring) in Contra Costa County—took steps to establish selected programs as nonprofit entities. This enables these programs to begin fundraising and building a base of support before SafeFutures Federal funding ends.

  Communities often seek to sustain a demonstration program by applying for other grant funds for its continuation. This approach may seem more expedient and feasible than securing “hard money” from budgets of local governments or service providers. However, it often results in modifying key features of an existing program to make it fit the objectives of the next funder. It also may result in piecemeal sustainability, in which some elements of the initiative’s continuum of services receive ongoing support, while others fade away.

• Implicit in the SafeFutures approach is an emphasis on bringing about more comprehensive and more holistic treatment of youth and families. The road to systems reform can be seen as a continuum with gradations and permutations. Bringing together actors from different institutional contexts who logically need to interact, but have not previously done so, can be viewed as an early indication of systems reform. At the other end of the continuum is wholesale systems change, including changes in policies and practices of institutions brought about collaboratively/jointly to accomplish mutually agreed-upon reforms. Systems change can also occur within a single institution, not only across several institutions. Some reform efforts might be implemented in a single location or may be introduced on a limited scale with the intent of expanding them systemwide if they appear successful. The SafeFutures sites provide evidence of systems reform at various stages along the continuum.

• It takes a considerable amount of time for communities to develop viable collaborations. These are complex mechanisms that involve organizations with different institutional climates and levels of autonomy, flexibility, and power; individuals with differing levels of experience and expertise; and diverse cultural contexts that give rise to different ways of defining issues and solutions. Organizations need to develop the trust necessary to agree on how to work together and to decide what to do. Once established, collaborative relationships need to be nurtured and maintained over time. Consequently, collaboration is not easy and takes much time and effort. Organizations must work to overcome histories that include turf issues, longstanding isolation, dissension and mistrust among key parties, and real shortages of resources.

• In collaborative ventures, the differing perspectives of staff from different systems need to be recognized and respected if partnerships are to succeed. For example, staff from juvenile justice system agencies and social service agencies may work together to deliver a particular service or program, as in Contra Costa County’s Summit Center. In doing so, they bring different perspectives on the appropriate way to address particular types of youth behavior, which may lead to conflict among staff. Partners need to “learn each other’s language” and develop an understanding of the values and norms of their respective fields in order to work as a team. Cross-training helps promote team development and extends staff abilities to deliver holistic services.

  uccessful collaborations need individuals in positions of authority to exert their leadership to secure resources and support.

• Successful collaborations need individuals in positions of authority to exert their leadership to secure resources and support. Given the complexity and cross-cutting nature of comprehensive community initiatives, higher level forces—often external to the collaboration or even the local community (e.g., State-level officials)—can exert positive or negative influence on an initiative. High-level decisionmakers in the public arena and key local or regional private-sector actors should be informed about the critical nature of an initiative, to gain their support at the outset. (For funded demonstrations, it may be useful for the funders to do this.) Such individuals would not be expected to be active members of the local collaborative (in the sense of hands-on involvement in committees or direct service delivery) but might be asked to use their clout in support of the initiative and to mitigate obstacles that threaten to undermine the achievement of key objectives.

• Key community leaders face multiple demands for participation in various collaborative and strategic planning efforts. Many initiatives—of Federal or local origin—request (or require) collaboration and/or strategic planning. Some communities have multiple initiatives of this type in operation at the same time. As a result, the same key leaders are called upon to participate in multiple periodic planning, needs assessment, or similar events for the various initiatives. Because of the multiple demands on their time, it becomes more difficult for staff of a particular initiative to secure participation of these key leaders. Increasingly, as funders mandate comprehensive interagency strategies, local communities will need to develop mechanisms to “collaborate the collaboratives.”

  urnover among elected officials and administrators of key partner agencies can have a negative impact on collaborative efforts.

• Turnover among elected officials and administrators of key partner agencies can have a negative impact on collaborative efforts. The presence of new leaders may introduce different visions and strategies that dramatically diverge from those previously endorsed.

  Turnover in leadership positions may undercut the pursuit of a coherent policy, undermining the credibility of the effort and staff morale. At the least, in such cases, time must be spent rebuilding relationships with the new administration and reestablishing an understanding of the initiative and its objectives. Turnover in elected legislative and judicial positions often leads to subsequent changes in department heads and other influential appointed positions. As a result, program directors may have to cope with multiple cases of turnover in relatively short time periods. Turnover can be particularly disruptive if there are prolonged periods where there is an “acting” administrator, who may have little incentive to maintain ties to the collaborative or little power to bring to bear on its behalf. In one site, for example, the departure of the police chief created a leadership void in the gang task force, which became virtually inactive for a period of months when that position was held by acting chiefs.

• Implementation of services and activities in multiple components took longer than either the local communities or the funders originally anticipated. Sites that had well-developed preexisting strategies were able to implement program services or activities more quickly, to the extent that SafeFutures categoric requirements permitted the incorporation of the preexisting local plans. Some sites primarily provided services through subcontracts or other agreements with agencies that already operated similar programs (e.g., afterschool or mentoring programs). This usually facilitated relatively early implementation of programming, but the downside may be the continuation of “business as usual” rather than the careful consideration of whether reforms are necessary. In cases where staff had to be hired and new programs established, services to youth were invariably delayed by startup activities.

• The SafeFutures sites introduced a variety of accountability mechanisms over the course of the initiative, in part to ensure that multiple service providers were fulfilling their obligations. For example, St. Louis instituted a monthly reporting system for case-managed youth at the start of the initiative. This management information system (MIS) is used to record information on service needs and referrals. The MIS has been used to make decisions about retaining partner agencies and to encourage agencies to make more referrals. Boston created positions for contract monitors among its administrative staff to provide better oversight of subcontractors. SafeFutures administrators hold mandatory monthly grants management meetings with contractors to convey information about administrative and reporting requirements and other contracting issues. Sanctions are imposed when contractors are not in compliance with their contracts. In Seattle, the city government recently introduced outcome-based budgeting, which affects all city agencies and subcontractors, including those working with SafeFutures. As a result of this initiative, contractors are required to set quarterly outcome goals and face funding reductions if they fail to meet them.

• Programs operated in partnership with other agencies can be discontinued for reasons unrelated to the initiative—even in cases where the program is considered successful. In one site, an afterschool/family strengthening program that was being replicated because SafeFutures partners believed it was successfully reaching and retaining high-risk youth was discontinued because of a decision by the Board of Directors of the agency that operated the program. This decision was related to financial difficulties of that organization, not to the effectiveness of the program. In such cases, the local initiative may have little or no ability to influence the decision of the agency responsible for the program, unless it is able to provide full funding to enable continued operation.

or most communities, implementing innovative programming had both beneficial and detrimental effects.

• Communities are willing and able to implement programming that is innovative in the local area. To some extent, each of the demonstration sites engaged in risk taking by implementing at least some services or activities that no one had ever tried or that were new to their locale. For most communities, implementing innovative programming had both beneficial and detrimental effects. New approaches were enticing in that they offered the opportunity to fill a previously unmet need or gap in service; however, there was no formulaic approach to success that could be followed. In most cases, service providers experienced learning curves and had to find creative ways to redress unanticipated difficulties, many of which were logistical. For instance, in one site, staff learned that they could not combine programming for at-risk girls and girls who were juvenile offenders since the latter intimidated the former and undermined the purpose of the group.

• Staff turnover in leadership and other key positions can seriously hinder program implementation and stability. Several SafeFutures sites experienced turnover in the position of project director during the first 2 years of the initiative—and some sites had more than one turnover in the same position within the first 3 years. Turnover affects the continuity of program implementation and requires site staff to reestablish linkages and, in some cases, restart programs that lapsed during periods of staff change. Loss of institutional memory also occurs in such cases, especially if the program does not have written guidance and other documentation detailing specific project operations to facilitate such transitions.

• Filling positions—especially positions involving specialized skills—can be a problem, particularly in rural areas that have a limited professional workforce. The relative isolation and low wages common to rural areas make it difficult to recruit individuals with relevant expertise to fill certain positions. Programs that arrange to provide training to make the existing workforce more qualified may then face the possibility of losing those staff after their enhanced skills make them more desirable to other agencies.

  ecruiting mentors and other volunteers is particularly challenging in low-income areas.

• Recruiting mentors and other volunteers is particularly challenging in low-income areas. Transportation and poverty issues also affect programs’ abilities to attract and work with volunteers. In rural areas and large counties, the need to travel long distances works against attracting individuals to serve as mentors or work as volunteers. Distance also makes it difficult to meet with volunteers as a group, either to provide training or to hold periodic meetings to support volunteer efforts. Difficulties in recruiting and retaining mentors in rural settings also make each match more expensive. In low-income areas, it is apparently more difficult to find adults who have the ability or motivation to donate their time. Many of these adults work more than one job, are preoccupied with the need to find a job, and/or are struggling to raise their own families with limited resources. In addition, some adults have personal or legal histories that make it inadvisable to use them as mentors (or as volunteers in other roles).

  In some cases, new mentoring programs established for SafeFutures by small, community-based organizations had difficulty recruiting mentors because they were competing for volunteers with large, well-known mentoring programs (such as BBBS) with more resources and experience in recruiting. Corporations and other organizations welcome recruiting efforts by such entities, but are less willing to provide access to their staff to unfamiliar organizations.

  Recruiting mentors is also more challenging for programs that serve youth already involved in the juvenile justice system or youth who may be perceived as high risk by potential mentors.

• “Hidden” resource requirements can pose challenges to program implementation. Several programs encountered unanticipated costs associated with transportation and food (discussed separately below). One program designed to reenroll clients in school found that once they helped youth return to school, these students lacked appropriate clothing to attend classes. The program was unprepared to meet that need or youth’s other needs related to school participation (e.g., activity cards, sports attire/gear, band equipment). At least one youth could not officially reregister in school until settling $180 in outstanding school library fines (the program helped negotiate a payment plan). Similarly, programs geared to youth employment unexpectedly found that clients had no way to get to their jobs or lacked money for lunches (one program arranged for free lunch program deliveries).

• The ability to offer food is an important element of many programs that serve a low-income population. Providing food and opportunities for group dining can play a part in enhancing group bonding, modeling prosocial behavior (e.g., etiquette, respect for others), and meeting the service needs of low-income youth/families or those in unstable home environments. Many programs provided healthy snacks during afterschool activities, both to attract participation and in an effort to address the nutritional needs of youth (who did not always receive adequate nutrition at home). The PREPP program in St. Louis provided a hot meal each afternoon, to which parents also were welcome, as an incentive to youth and to encourage parental involvement. In Contra Costa County, the weekly meeting of gang-involved youth includes a shared meal, since staff perceive that “breaking bread” together helps youth involved in different gangs to bond with each other and with program staff. Serving food also is viewed as an important drawing card for programs intended to attract parents or the community in general. Demonstration sites experienced some confusion or concern about the use of Federal SafeFutures funds to pay for food in various contexts. Some programs drew on other resources to underwrite food costs, while a few sought donations for this purpose.

  ransportation and location are critical considerations in programs serving low-income and at-risk youth.

• Transportation and location are critical considerations in programs serving low-income and at-risk youth. Providing easy access to programs (particularly afterschool programs) was a challenge in some sites. Transportation problems were commonly related to the relative lack of public transportation in low-income areas, the absence of a family vehicle, or the lack of a family member who could drive the youth to a program. In the rural SafeFutures sites, these problems were compounded by long travel distances between various locations. Some programs addressed the problem by arranging for free bus/subway tokens; others bought or made arrangements to share vans.

  Distance and accessibility concerns were further heightened in some cases by concerns about the safety of the neighborhood in which a program is located or whether it is located in an area perceived as the turf of a particular gang. The latter issues were key concerns in selecting the location of the SafeFutures Youth Center in Seattle, for example. Some programs made an effort to locate programs in areas that were considered gang neutral or sought to define their sites as such.

  Conversely, selecting an appropriate site for a program may pose challenges due to the “at-risk” characteristics of the clientele. One site that initiated school-based services for youth and families found that elementary school administrators were not comfortable with the older adolescents these programs were intended to attract and the schools did not welcome their presence. The site ultimately closed two of the programs based in elementary schools.

  t is difficult to get families of at-risk youth to support services for their children or to participate in family-focused services.

• It is difficult to get families of at-risk youth to support services for their children or to participate in family-focused services. Despite the emphasis in OJJDP’s Comprehensive Strategy for Serious, Violent, and Chronic Juvenile Offenders on families as the primary providers of care for children, few sites have been able to meaningfully involve parents in the programs or services designed to support their children. For the most part, at-risk youth and juvenile offenders were being served independently of their families. There appear to be several reasons for the difficulty in engaging parents. Some parents may feel intimidated by the institutional settings or staff. Others may be used to resolving personal issues within family/kinship networks or skeptical about risking privacy to accept assistance from outsiders. Further, there are parents who fear that program participation might reveal their own dysfunctional behaviors, such as substance abuse, gang involvement, criminal behavior, or poor parenting practices. The transience of some low-income families also works against establishing relationships with them. Similarly, the severity of family problems (e.g., substance abuse, imprisonment, death of parent or sibling) makes service provision more challenging. In some cases, parents are not truly supportive of changes that they perceive as turning the child away from the family or displacing the values or authority of the parents. Some parents undermine program efforts, for example, by canceling meetings between the mentor and their child, not letting the child participate in specific program activities, or failing to reinforce program-supported behaviors at home.

  Logistical challenges also affect programs’ abilities to involve parents in services for themselves or in parent activities associated with youth programming. Limited access to transportation (public or private) in low-income areas (especially rural communities), work hours that may not conform to a “9 to 5” schedule, jobs that do not provide time off to participate in family activities, and absence of childcare to enable participation in activities are among the factors that make participation difficult for parents who wish to receive, or be involved in, services.

  Parents also may be reluctant to participate in classes or programs designed to improve parenting skills because they believe this indicates that they are not perceived as good parents. To overcome such feelings, some SafeFutures programs have made efforts to present such classes in ways that do not imply that participants lack parenting skills. In Boston, for example, case managers at health centers and public housing facilities reach out to parents with information on how the school and juvenile justice systems work, to engage parents who might resist activities labeled as parenting classes.

• The considerable stigma associated with mental health services affects providers’ ability to obtain clients or to serve those in need of such assistance. Reluctance to participate in mental health services—on the part of youth or on the part of parents (who are reluctant to receive such services or allow their child to receive them)—is fairly common, particularly in low-income or minority communities. In some cases, parents have opted to have their child spend time in a juvenile detention facility rather than participate in residential mental health treatment since the former is viewed as a common (and therefore acceptable) occurrence in their community and usually involves a shorter stay than the latter. Some youth/parents may simply be resistant to mental health or substance abuse treatment and may require the additional leverage of a court order (for juvenile offenders).

  rograms need to be developmentally appropriate in terms of both substance and setting.

• Programs need to be developmentally appropriate in terms of both substance and setting. Programs seemingly experienced more difficulty attracting and retaining older teens than they did middle or elementary school-age children. Many SafeFutures programs, particularly afterschool programs, experienced this problem. There are several possible reasons for difficulties attracting older teens.
  • They may be more deeply entrenched in peer groups that support higher risk activities.

  • Older teens appear to be reluctant to be associated with programs that are perceived as serving children and younger teens.

  • The types of activities commonly offered in afterschool programs, such as loosely organized sports, free play, and arts and crafts, may not appeal to older youth.

  • Older teens may have more options for their free time (e.g., afternoon sports practice in high schools).

  • Low-income youth’s need (or desire) for afterschool jobs apparently mitigates against their participation in programs.

  Programs that focused exclusively on older youth, such as Seattle’s SafeFutures Youth Center, did not appear to have as much difficulty attracting older teens as those that included a broader range of ages. Some programs, such as Imperial County’s Boys & Girls Club, scheduled activities for older youth at different times than those for younger ones.

  Staff attempting to attract older youth often are hard pressed to identify programming that will hold the interest of older adolescents, many of whom are involved in gangs or the justice system and not all of whom are committed to engaging in a prosocial lifestyle. In working with such youth, staff often seek to engage them long enough to have a chance to shape their behavior. Staff are often reluctant to enforce rules or requirements that might cause youth to leave a program since participation itself is often seen as an initial indication of movement toward prosocial behavior.

  t is more difficult—and takes longer—to see results of program efforts with youth who are beyond the at-risk stage.

• It is more difficult—and takes longer—to see results of program efforts with youth who are beyond the at-risk stage, such as those already deeply involved in the juvenile justice system, gangs, or substance abuse. Prevention programs often are able to see changes in client behavior and attitude in the short run. Intervention programs may see little or no real change after considerable lengths of time. The lack of measurable program outcomes in such cases makes it difficult for administrators and funders to determine whether intervention programs are working and should continue receiving support or whether modifications, or alternative programs, are needed.

Comprehensive Responses to Youth At Risk: Interim Findings From the SafeFutures Initiative

OJJDP Summary • November 2000