A growing chorus of voices is calling for a shift of resources away from responding to child abuse and neglect toward preventing its occurrence. Interest is coalescing around a newer idea that would combine universal reach with a response that is targeted based on a family’s risk, sometimes called targeted universalism. Several jurisdictions are already implementing initiatives based on this approach. Governments interested in adopting such a system need to resolve a number of questions concerning the system’s entry point, goals, lead agency, program content, and how to attract and retain the families that are most at risk. But the idea of targeted universalism is worth pursuing as it combines the advantages of both approaches.
The new focus on prevention should not be confused with the changes made by the Family First “Prevention” Services Act of 2018, which allows funds under Title IV-E of the Social Security Act to be be diverted from foster care to services to help keep children with their families. Despite its name, Family First funds can be used only for services to families in which abuse or neglect has already occurred. Such services are generally considered treatment, not prevention, although public health specialists refer to them as “tertiary prevention,” which mean preventing the recurrence of a problem. But this is not the meaning of prevention to the layperson, and the placement of “prevention” in the title of the act continues to cause confusion.
Preventing a problem, if possible, is certainly preferable to addressing it after it appears. As child maltreatment prevention expert Deborah Daro states in an issue of The Future of Children devoted to universal approaches to promoting healthy development, doctors don’t send away patients with precancerous cells and tell them to come back when they have Stage 4 cancer. Yet, that is exactly what we do in child welfare. As Daro points out, “our public response lacks an adequate early assessment when people become parents, and we often offer the appropriate level of assistance only after a parent fails to meet expectations or a child is harmed.” For this reason, many child welfare thought leaders like Daro are calling for a new emphasis on preventing maltreatment before it occurs.
The consensus on prevention still leaves the question of how much to invest in universal approaches (known as primary prevention) as opposed to “secondary prevention” approaches that target families who are deemed to be at risk. Secondary prevention has a lot going for it. Targeting a program to those who need it most can be justified on grounds of efficiency. Why spend money reaching people who do not need help? But investing only in secondary prevention has drawbacks, as described by Kenneth Dodge and Benjamin Goodman in the Future of Children issue referenced above. Even the lowest-risk groups have some risk of child maltreatment. And because they are much larger than the high-risk groups, they may account for most cases of maltreatment. Moreover, interventions with targeted groups rarely reach a high proportion of that population, and thus cannot have a detectable impact on the problem overall. Finally, targeted programs are often stigmatized and not politically popular, leading to lower funding–and less participation by targeted groups.
Dodge and Goodman point out that a debate over universal vs. targeted approaches played out when public schooling was first discussed in the United States. Some advocates argued that middle and upper class families could pay for their children to be educated, and that confining public education to the poor would save taxpayers money. Of course the proponents of universalism won out. Even though affluent families continue to be able to buy a more expensive education through higher property taxes and access to private school, one can only imagine the sorry state of our public education system if at been confined to the poor.
As Dodge and Goodman point out, not all universal programs must provide the same services for everyone, and they cite pediatric care as analogy. All children are seen for well-child visits, during which pediatricians screen them for conditions that might warrant services from specialists, and refer them accordingly. Therefore, they argue that “the best strategy may be to embed targeted interventions in a universal strategy that reaches the entire population while offering intensive interventions for targeted subgroups.” This approach, which they call “targeted universalism,” involves screening all families at a single point in time, such as the birth of a child, identifying the family’s risks and needs, and connecting them with community resources for addressing those risks and needs.
This is the approach that they embedded in the Family Connects Program, starting with a home visit fro a nurse. Nurse home visiting is a popular platform for a universal program, and there is a lot of precedent for a universal nurse home visiting program. Universal nurse home visiting is used by many European countries, where it is part of a comprehensive maternal and child health system, and is also used in other parts of the world. Perhaps the best developed application of this approach in the United States is Family Connects, which Dodge and his team at Duke University initiated in Durham, North Carolina in 2008 as Durham Connects. The model is now being implemented in over two dozen communities around the country, and Oregon is rolling it out statewide. Family Connects aims to reach every family giving birth in a community, assess the parents to determine their risks and needs, and refer them to appropriate services. The program rests on three “pillars:” home visiting, community alignment and data and monitoring, as described by Dodge and Goodman.
The first pillar is home visiting by a trained public health nurse who visits the family in the hospital to welcome the baby and offer a free home visit when the baby is about three weeks old. During the home visit, the nurse uses a structured clinical interview to assess risk in 12 key domains that predict adverse outcomes among children. At the end of the interview, the nurse works with the family to develop a plan of action which may include follow-up visits, phone calls, or contact with external agencies. With parental consent, this plan is shared with the baby’s pediatrician and the mother’s primary care provider. Four weeks later, a program staff member calls the family to check on their progress and determine if the referrals were successful. If the family has not succeeded in making the connection, the program either helps the family try again or makes another referral.
The second pillar, called community alignment, is a compilation of community resources available to families at birth, including targeted home visiting programs, early care and education, and mental health. This directory is available in electronic for for nurses to use on their visits. The third pillar is an electronic data system that documents each family’s assessment, referrals, and connections with community agencies. These records, scrubbed of identifying information, are aggregated to provide information about each agency and on a community level to identify gaps between needs and services. To contain costs, Family Connects is limited to seven contacts (phone calls and visits) over the course of 12 weeks. Dodge and Goodman report that the cost of Family Connects ranges from $500 to $700 per family.
Family Connects has been evaluated with two randomized controlled trials (RCT’s) in Durham, NC and one field study in four rural counties in the state. Eighty percent of the intervention families in the first trial scheduled a visit and 86 percent completed it, for a total “completion rate” of 69 percent. In the second trial, the percentages were 77 percent scheduled and 84 percent completed for a total of 64 percent. The first and second RCT’s found 49 and 52 percent of families respectively to have moderate needs, 46 and 42 percent to have serious needs requiring referral to a community resource, and one percent with a crisis needing immediate intervention. Of the families referred to a community agency, 79 percent and 83 percent reported they had followed through and made the connection. In the first RCT, researchers found that intervention infants had 39 percent fewer referrals to Child Protective Services (CPS) than did the control infants by the age of 60 months, controlling for demographic risk factors. CPS data are not yet available from the second trial.
Nurse home visiting is not the only possible platform on which to base a targeted universal program, Other options for locating a universal service include pediatric practices. There are two different models based in pediatrician’s offices that have shown promise for preventing child maltreatment–SEEK and Healthy Steps. SEEK trains pediatric primary care providers (PCP’s) to use a questionnaire to assess for a specific set of risk factors. The PCP initially addresses identified risk factors and refers the parent to community resources, ideally with the help of a behavioral health professional. Healthy Steps, as described by Valado and coauthors in The Future of Children, functions as a targeted universal model with three tiers. All families receive screenings and access to a child development support line. Second-tier families receive short-term consultations, along with referrals, additional guidance and resources. Families classified in the highest-risk tier receive “a series of team-based well child visits incorporating a Healthy Steps specialist.”
Models based in pediatric practices have had some promising results. A study testing SEEK with a high-risk sample of patients from a pediatric primary clinic in Baltimore found a “striking” 31 percent reduction in CPS reports. The other SEEK study focused on a low-risk population and there were not enough cases of maltreatment to find impacts on abuse and neglect; however, the study found a lower rate of physical punishment and psychological maltreatment reported by participating mothers. A multisite evaluation of Healthy Steps, as described by Velado et al, showed similar effects, such as a 33 percent reduction in the use of severe physical discipline in the intervention group vs the comparison group when the child was 30-33 months of age. The philanthropic partnership Blue Meridian Partners has chosen Healthy Steps as one of five models to receive large grants to help bring them to a national scale as a potential solution to poverty and lack of economic mobility.
Neither of these primary care-based models has been used universally throughout a jurisdiction. Moreover, pediatric care based models have less reach than models based on the birth hospital. Almost all babies are born in hospitals, but fewer infants attend their regularly scheduled well-baby visits. But according to the National Survey of Children’s Health for 2019, only an estimated 89 percent of children aged 0 to five had experienced one or more preventive care visit in the past year. While hopefully the percentage is greater for infants, it is probably less than the nearly 100% who are born in hospitals.
Many questions must be answered in developing a “targeted universal” child maltreatment prevention system.
What should the entry point be, and should there be more than one? Having all families enter the program through the same portal (be it the GYN practice, birth hospital, or pediatrician’s office) would avoid overlap and inefficiency. Choosing the system that meets parents earliest–the OB-GYN office–would allow programs to make a difference at a crucial time but would also miss the children who get little or no prenatal care. The founders of Family Connects chose to use the birthing hospital because it covers the most families, even though they are missing the chance to address problems that begin prenatally. Combining two or more portals may increase a program’s reach and the opportunity to coordinate and extend services. In Guildford County, NC, Healthy Steps is being integrated with Family Connects. Michael Wald, in a forthcoming article in the Handbook of Child Maltreatment, proposes a prevention system that starts with OB-GYN’s and WIC programs in the prenatal stage and continues with universal services at birth through pediatricians, home visitors and family resource centers. All of these entry points in turn would refer families to targeted services. Using more than one portal requires linkages and procedures for hand-off or collaboration, adding complexity to the system, but increases potential coverage.
What should the goals of the system be? A basic question is whether the system would be framed as a child maltreatment prevention system or something broader. It is hard to separate the goal of preventing maltreatment from that of promoting healthy child development, and indeed most of the programs discussed above have broader goals. The mission of Family Connects is “to increase child well-being by bridging the gap between parent needs and community resources.” Healthy Steps has the goal of “promoting the health, well-being and school readiness of babies and toddlers.” Clearly it is hard to separate the goals of child maltreatment prevention and the promotion of child well-being and healthy development.
What should the lead agency be? A key question about universal prevention is which system should take the lead. Child welfare leaders like Jerry Milner, head of the Children’s Bureau under the Trump Administration, have expressed the desire to expand the role of child welfare to include primary prevention. But if the goal is the broader enhancement of child development, and if the main providers of universal services are health professionals rather than social workers, another agency like public health may be a more appropriate home. Moreover, the child welfare system is already overburdened and underfunded. The work of investigating existing abuse and neglect (which will never be totally eliminated), helping parents and children heal, and making sure children have a safe environment to thrive either temporarily or permanently, is work enough for this beleaguered system. It is interesting that many modern child welfare leaders are so discontented with their primary mission that they are crying out to take on prevention. Perhaps the answer lies in the current political climate, which disparages child protective services and foster care, rather than recognizing the crucial role these services play in protecting children from harm,
What should the targeted services be? Michael Wald raises the question of whether a limited set of evidence-based models should be supported as part of a prevention system or whether jurisdictions should be given free rein in program selection. There are arguments for both, but it may be more practical to allow local jurisdictions to choose their own programs, especially since most “evidence-based” programs have only modest effects. But there are many reasons to advocate that every family found to be high-risk receive early care and education (ECE) interventions such as Early Head Start and Educare, which reduce child maltreatment risks in so many different ways, as I argued in an earlier post. The proposed Child Care for Working Families Act should help make such quality programs more available around the country.
How can we engage the highest-risk families? This is perhaps the thorniest question of all. All of the existing and proposed programs discussed above are voluntary, and voluntary programs never succeed in involving all eligible families. Moreover, It is often hardest to enroll and retain the highest-risk families in parenting support programs. We have already seen that Durham Connects provided a home visit to only 69% and 64% of eligible families in its two RCT’s. And considering that only 79 percent and 83 percent accessed the recommended services, the actual completion rate goes down further. In a study of Durham Connects, the researchers found that parents with a higher risk based on demographic factors like age and income were more likely to agree to a home visit but less likely to follow through. They also found that infant health risk, as measured by low birth rate, birth complications, and medical diagnoses, was associated with lower levels of both initial engagement and follow-through. An HHS issue brief suggests many ways to improve a program’s performance in enrolling and engaging families in home visiting programs. Not listed is the idea of including a peer mentor in the program model, which is embedded in some newer initiatives like Hello Baby and the Detroit Prevention Project. But the fact that remains that a voluntary program will leave some children unprotected until they are actually harmed by abuse or neglect. Yet, it is clear that a mandatory child abuse prevention program will not be accepted in the U.S–just look at the debate over the COVID vaccine. We can hope that If a universal, voluntary prevention program is adopted, it will gradually gain in acceptance by high-risk and low-risk parents alike as a valuable benefit. In case that does not occur, we must record identifying information about the parents who do not accept the initial offer to participate and and follow up on future outcomes for their children in order to assess the efficacy of the program at preventing maltreatment among the highest-risk groups. Moreover, CPS must have access to this information when they receive a new report on a family.
Can the system be funded by cutting child welfare budgets? Congress has an unfortunate history of cutting funds to unpopular programs prematurely with the hope of achieving savings by a new approach, rather than waiting for a new approach to yield savings. One could mention mental health deinstitutionalization but a closer analogy might be the Family First Act’s moving money away from congregate care toward family preservation services, before states have the foster homes to replace these facilities, thereby simply shifting the cost of such facilities to states. We hope that establishing a robust system of prevention may well eventually result in a reduction of calls to Child Protective Services and in the need for foster care and in-home services. But we cannot reduce funding for traditional child welfare unless and until this effect has occurred, as Brett Drake argued in in a webinar recently conducted by the American Academy of Political and Social Science (AAPSS) in conjunction with their 2020 volume on child maltreatment. Even then it may make sense to retain current funding as the system is currently stretched thin.
What about anti-poverty programs? Adopting a system of targeted universalism to prevent child maltreatment does not directly address poverty, a prime risk factor for abuse and neglect, as discussed in my last post. Targeted universal programs will not change the level of welfare benefits or housing availability. However, it is important to remember that most poor families do not neglect their children, as University of Maryland’s Brenda Jones-Harden mentioned in the AAPSS webinar. Those families that are functioning well will find a way to get the resources they need to care for their children, and the supportive services provided through targeted universalism may help the others do the same. But at the same time, prevention advocates should fight to improve economic supports so that no child is deprived of what he or she needs to grow up into a healthy and functioning adult. Some of the new programs already passed or under discussion under the Biden Administration and the new Congress, such as the expanded child tax credit and the Child Care for Working Families Act, will help make targeted interventions more effective by addressing some of the poverty-related risk factors that cannot be addressed by targeted programs.
There is a growing consensus in the child welfare world that we must focus on prevention of abuse and neglect, and there is an increasing interest in using the approach of targeted universalism. The current historic expansions of the safety net will provide the perfect backdrop for such an initiative by addressing the economic risk factors for child abuse and neglect. With the motivation to build a better society in the wake of the pandemic, the time may be right to develop a universal, targeted system to prevent child maltreatment and allow every child the conditions for optimal development.