Twenty years ago, the Annie E. Casey Foundation launched an initiative aimed at tackling a persistent problem in the juvenile justice system: Too many young people who came into contact with the system were being confined unnecessarily in secure detention. Building on research showing that such confinement leads to significantly worse outcomes for youths, the foundation set out to help local agencies implement alternatives to detention.

Today the Juvenile Detention Alternatives Initiative (JDAI) is being implemented in more than 200 counties in 39 states and the District of Columbia. One in four US youths lives in a participating community. The use of secure confinement in these communities has dropped 43 percent, and there has been no decrease in public safety. Although JDAI has not been implemented in every community in the country, the approach, tools, and lessons learned have been shared broadly, and other juvenile justice leaders are taking action on their own.

JDAI is one example of a philanthropic initiative that has dramatically scaled up its impact over time. The Edna McConnell Clark Foundation’s pioneering work, the opportunities provided by the US government’s Social Innovation Fund, and the hard work of legions of social entrepreneurs mean that many proven programs now are reaching more people who need them. It’s an impressive, energizing story. But we still have a long way to go before we can say we’ve solved the problem of how to scale up human services programs effectively.

From Program Replication to True Scale

Replicating proven programs with fidelity is both critical and tough to do right, but we should be careful not to confuse program expansion with achieving population-level scale. Achieving that scale means attaining a meaningful, measurable result for a specific population. Supporting a particular evidence-based program or model to expand reach, grow in size, and capture a greater percentage of market share may be a necessary part of the path toward a large scale. But we shouldn’t declare victory until, for instance, all children in Baltimore enter school ready to learn, or all youths between the ages of 18 and 24 in Maine are connected to school or work, or all children in the United States read proficiently by the end of third grade.

To reach these types of ambitious goals, the road to scale inevitably will run through public systems. And decades of experience tell us that a bad system will trump a good program—every time, all the time. Whether programs focus on youths involved in the juvenile justice system, students in public schools, families in the child welfare system, or young mothers receiving public health services, even the greatest programs cannot succeed in a lasting way if they depend on dysfunctional systems. Programs can sink or swim, depending on how systems handle issues from intake, eligibility, and case planning to the selection and compensation of private providers. Similarly, policy decisions that determine program priorities, budget allocations, or staffing levels can accelerate or impede progress toward greater scale.

The JDAI Story

Replicating proven programs with fidelity is critical and tough to do right, but we should not confuse program expansion with achieving population-level scale.

In developing the JDAI strategy, the Casey Foundation noted the many ways that the juvenile justice system could trump any programmatic intervention we could mount in communities, so we decided to start with changing the system itself. JDAI works directly with the local agencies responsible for juvenile detention. We help them in such critical areas as adopting screening tools and processes for objectively assessing risk and making admission decisions; implementing case processing reforms to minimize unnecessary delays; collecting and using data to track the young people’s progress; and developing effective non-secure alternatives in the community. A common thread in the JDAI approach is strong collaborations among the important actors in the system, including the courts, probation officers, prosecutors, defenders, and community groups.

The results speak for themselves. In addition to reductions in confinement in the targeted communities, we’re beginning to see the national needle moving, too. In 1997, almost 28,000 young people were in detention. In 2011, only 19,000 were in detention. Recent federal data indicate that both the number and rate of young people confined have decreased by more than 40 percent nationwide since the mid-1990s, when this issue first became a priority for the Casey Foundation and others, notably the John D. and Catherine T. MacArthur Foundation. Although many factors have influenced this trend, this is genuinely a population-level shift that would have been hard to imagine just 20 years ago, when youths were being described as “super-predators” and mass incarceration was at its peak.

With JDAI’s strategy of working directly with the public system to reform core functions, inappropriate detention is prevented for every youth who comes into contact with the system in that community. In other words, we move the needle—achieve scale—for an important outcome in a targeted population.

The strategy of directly engaging the public system also helps ensure that JDAI implementation won’t be sloughed off in the next change of leadership or budget crisis, events that can derail even the strongest programs. The goal of JDAI is to advance fundamental system reforms so they become deeply rooted and cannot easily be removed or reversed. In addition, the program changes how existing resources are used rather than requiring new dollars, so there are no savings to be had by downsizing or dropping JDAI from the budget. In fact, by preventing unnecessary secure detention, communities have been able to close facilities or avoid building new ones, and the resources saved can be reinvested to expand alternatives to detention. Ending JDAI could actually cost jurisdictions money.

Following the Evidence

Of course, not every outcome we want to change is best achieved by working directly with a public system in this way. But few programs are not at risk of being trumped by bad systems.

In addition, just as public systems can affect what programs are able to achieve, so too can effective programs influence public systems. Innovative approaches can function as proof points and help build evidence for new ways of serving a public system’s clients. As we glean more information from evaluations of proven programs, we can see common principles of effective service delivery that can guide changes in public systems in areas from case processing to procurement criteria for contracted services. Given that these systems serve whole populations, it is incumbent on all of us who care about those populations to help public systems follow the evidence.

Twenty years is a long time to stick with a single program, especially for a foundation. At the Casey Foundation, we have stuck with JDAI for a variety of reasons, including its evidence of effectiveness and the fact that we’ve found cost-effective ways to help an accelerating number of communities adopt the program. But for any foundation that is focused on tough, pivotal problems, the most important reason to stay the course with a program like JDAI is that it is moving the needle for whole populations in community after community, state after state. That’s scale, and it’s something we can’t achieve through program replication alone.