What are some examples of safety and risk assessments?

At its core, the responsibility of child protection agencies is to ensure that children are safe. To fulfill this mandate, the Child Abuse Prevention and Treatment Act requires states to have procedures for the immediate screening of child abuse and neglect reports, as well as assessment of a child’s safety and risk. Further, the Indian and Child Welfare Act requires that active efforts, such as conducting a comprehensive assessment, be made to preserve American Indian/Alaska Native families. Applying the principles of the Indian Child Welfare Act more broadly also would benefit all families. Although life-changing decisions — such as the one to separate children from their families — are based on safety assessments, evidence indicates that decisions about child safety are made inconsistently, sometimes even within a given jurisdiction. Given the importance of making unbiased decisions — especially decision made within a milieu of limited information, time pressure, and intense family emotion — reliable safety and risk assessment tools and processes must be in place.

These structured tools and related processes can provide a framework for effectively assessing children’s current safety and risk of future harm, as well as inform service plans to strengthen and support families at all stages of involvement with the child welfare system, from initial contact with hotlines to beyond case closure.

This information packet is intended to aid in the exploration and selection of safety and risk assessment tools and decision-making models, and includes specific examples.

Definitions of safety and risk

While often discussed together, safety and risk are not interchangeable terms. Safety refers to immediate or impending danger or serious harm. It is assessed categorically — that is, a child either is safe or not safe. Some states use a third category: safe with an agreement or safety plan in place (sometimes called conditionally safe). Risk refers to the likelihood of future maltreatment. It is assessed on a continuum from low to high risk.¹

Safety assessments

A safety assessment is the systematic collection of information about threatening family conditions and current, significant, and clearly observable threats to the safety of a child — as well as family and other resources that can reduce or eliminate those threats — so that a caseworker can determine the degree to which a child is likely to suffer maltreatment in the immediate future. Safety assessments are completed first to determine whether the child can stay safe in the current situation.

Language and policy regarding safety assessments vary by state, but they generally include six domains:

1. Extent of maltreatment
2. Nature and history of maltreatment
3. Adult functioning
4. Parenting practices
5. Disciplinary practices
6. Child functioning

When training caseworkers to conduct safety assessments, staff must be taught not only how to complete the assessment, but also how to recognize personal and systemic biases that may be at play, especially when considering whether and how to incorporate information and findings from reports of prior maltreatment.² Engaging parents in ways that are respectful and honor cultural differences is also a critical skill, since understanding the role that a family’s culture might play in parenting and disciplinary practices is not always captured in standardized assessments, but is an important factor to effectively assess child safety.

A valid safety assessment also must take into account family, extended family, and community strengths and resources that may be available to counteract safety threats. Considering both threats to the child and available strengths and resources, caseworkers must then determine to the best of their ability if the child is safe right now.

Risk assessments

A risk assessment is the structured collection and analysis of information to determine the degree to which key factors that increase the likelihood of future harm to a child are present in a family situation. Risk assessments are conducted to gauge the risk of future maltreatment, determine whether services should be provided, and if so, inform what services would be appropriate. Risk assessments generally include both a neglect index and an abuse index.

The neglect index typically assesses the number of children in the home and their needs, whether the family has safe and stable housing, and the caregiver’s physical and mental health, current level of substance use, and ability to provide care.

The abuse scale typically assesses:

1. The caregiver’s number of prior reports of maltreatment
2. The caregiver’s prior involvement with child protective services
3. The caregiver’s current level of substance use
4. Any prior or current incidents of domestic violence
5. Significant problems with caregiving skills
6. The child’s special needs or history of delinquency

This information, however, needs to be considered in light of what we know about potential biases in the data, as Black families and families living in poverty are more likely to be reported to child protective services and therefore may have a history of greater involvement with the system.

There are two primary approaches to risk assessment in child welfare: consensus-based models, which often combine items from numerous instruments and rely on clinical judgment; and actuarial models, which use statistics to identify and assess factors that predict future maltreatment. Given the disproportionate number of American Indian/Alaska Native families and Black families historically and currently impacted by the child welfare system, there is some debate whether the reliance on historical data to inform future risk could perpetuate further harm to these families. Therefore, in both approaches, clinical judgment and a critical lens are necessary to try to mitigate potential biases.

Examples of safety and risk assessment tools and practice models

Some of the more frequently used safety and risk assessment tools are presented below.

The Safety Assessment and Family Evaluation (SAFE) model, developed by ACTION for Child Protection, is considered to be the first comprehensive safety decision-making model. It is a consensus-based safety assessment and decision-making support tool for the assessment of danger threats, child vulnerability, and caregiver protective capacities. Used at all decision-making points from referral to case closure, the SAFE model helps determine whether a child is safe, conditionally safe, or unsafe. It emphasizes family engagement and partnership, and assesses family strengths and protective capacities.

The Structured Decision Making (SDM) Model, developed by Evident Change, is one of the most commonly used safety and risk assessments. It provides different tools to be used at major case decision points — from intake to reunification — to improve child welfare decision-making. SDM’s primary goal is to reduce the likelihood of subsequent maltreatment in families where maltreatment has occurred. The logic is that if families at high risk of subsequent maltreatment are identified and provided services to meet their needs, the risk is reduced. The suite of tools provided through SDM helps ensure that case plans accurately reflect the strengths and needs of families. Efforts to support equity in the usage of SDM have been informed by participation of a diversity of stakeholders who have been involved in ongoing coaching, testing, and data analyses.

Signs of Safety (SoS), often used in combination with SDM, is an assessment and planning model developed in Australia that includes a comprehensive risk assessment while also mapping protective factors. It provides an inquiry-based approach that assesses safety through four key questions:

1. What are we worried about? (Past harm, risk of future danger)
2. What’s working well? (Family strengths and safety)
3. What needs to happen? (Risk assessment, short term goals, long term goals)
4. Where are we on a scale of 0 to 10? (10 means a child can safely stay at home and the case can be closed; 0 means the child may need to be removed from the home.)

SoS supports decision-making throughout the life of a case, aims for children’s voices to be included, and emphasizes family partnership and community context.

Safety Organized Practice (SOP) is a collection of best practices to engage, assess, and support children and families. Centering families as experts, it aims to support decision-making throughout the life of a case, emphasizing cultural humility, trauma-informed practice, and the inclusion of children’s voices. The three principles of SOP are: (1) increased safety, permanency, and well-being; (2) good working relationships; and (3) improved critical thinking. Among the many components of SOP are SDM, Signs of Safety, safety mapping, cultural humility, motivational interviewing, and Review, Evaluate, and Direct teams.

Considerations

Child protection agencies need structured tools and processes for understanding children’s safety needs, particularly in regards to identifying appropriate placement settings and informing effective service planning and delivery to support families. Those tools and processes need to be effective in real time and over the course of a child’s time in the child welfare system.

It is important for agencies to put processes in place for reviewing, piloting, evaluating, and selecting screening or assessment tools, in order to explore and evaluate their utility for the population of children who may need support or intervention. Tools should be easy to use, produce consistent results across users (reliability), measure what is relevant (validity), and function consistently across diverse families. Regardless of the type of tool used, the policies and processes that accompany the tools are critical. A tool is only as good as the information put into it. It is particularly critical to consider the role that bias might play in the information-gathering process, based on both current and historical data. Evidence is clear that American Indian/Native Alaskan families, Black families, and families from lower socio-economic areas are at greater risk for investigation and family separation.^3,4 It is vital to always engage families compassionately, respectfully, and with cultural humility. It is also important to gather sufficient information from all relevant sources, continue collecting and updating information as needed throughout the life of a case, and to bring a critical lens to the information that is presented rather than taking information at face value. Relatedly, agency policies and practices need to be aligned so that tools result in effective decision-making and case planning. In some situations, it may be essential to consult with a specialist in medical care, mental health, substance use, or other issue affecting the family dynamic.

1 The content in this section was informed by: Child Welfare Information Gateway. (2022). The use of safety and risk assessments in child protection cases. U.S. Department of Health and Human Services, Administration for Children and Families, Children’s Bureau.

2 Miller, K.M., Cahn, K., Anderson-Nathe, B., Cause, A.G., & Bender, R. (2013). Individual and systemic/structural bias in child welfare decision making: Implications for children and families of color. Children and Youth Services Review, 35(9), 1634-1642. https://doi.org/10.1016/j.childyouth.2013.07.002

Ellis, K. (2019, December 17). Race and poverty bias in the child welfare system: Strategies for child welfare practitioners. American Bar Association. https://www.americanbar.org/groups/public_interest/child_law/resources/child_law_practiceonline/january—december-2019/race-and-poverty-bias-in-the-child-welfare-system—strategies-f/

3 Baron, E. J., Doyle, Jr., J. J., Emanuel, N., Hull, P. & Ryan, J. P. (2023). Discrimination in multi-phase systems: Evidence from child protection. National Bureau of Economic Research Working Paper 31490.

4 Pac, J., Collyer, S., Berger, L., O’Brien, K., Parker, E., Pecora, P., Rostad, W., Waldfogel, J., & Wimer, C. (2023). The effects of child poverty reductions on child protective services involvement. Social Service Review, 97(1), 43-91.