“Our responsibility is to create foresight about the changing nature of risk before anything goes wrong.” – David Woods
David Woods, one of the pioneers of safety science as applied today in industries such as nuclear power, airlines and health care, offers a critical insight into what continues to be missing from risk and safety analyses in child protective services (CPS). The insight is found in the phrase “changing nature of risk.”
Risk can be both a prospective judgment about the probability of a future event and an evolving dynamic aspect of present conditions. The dynamic aspect is what’s often missed in abuse and neglect fatalities of children who were left in the care of their parents or reunified after a stint in foster care. Both situations are linked to Woods’ “changing nature of risk.”
For the most part, the development of risk assessment models for application in CPS has focused on classifying families as to the likelihood of repeat maltreatment in general and repeat maltreatment specifically resulting in serious harm or death following a current maltreatment report.
But that way of thinking has essentially front-loaded risk assessment to the very beginning of cases. It has left CPS largely unprepared to gauge the important changes to risk that occur on a daily basis in the lives of many families.
In the early days of risk assessment, little differentiation was made between recurrence of maltreatment in general and recurrence with severe consequences. This began to change in the mid-1980s as Wayne Holder and his colleagues sought to differentiate between general risk of recurrence and safety.
While this distinction had profound implications, both risk and safety assessments continued to focus on a point-in-time classification of risk and safety levels rather than an analysis of the mercurial nature of risk and safety as Woods describes it.
Proponents of an actuarial approach claimed superior performance of their models relative to consensus-based models when classifying families as to the likelihood of maltreatment recurrence. More recently, predictive analytics, often referred to as “big data,” has emerged to assert itself as having even greater classificatory power due to its use of a broader array of data when estimating the level of risk.
Even so, predictive analytics continues the practice of producing a single risk estimation that does not adapt itself to daily fluctuations in the levels of threats of harm, protective capacities and child vulnerability inside the family.
Both actuarial and predictive analytics risk approaches do have value in supporting decisions about which families require an intervention to prevent future maltreatment. However, once families are triaged into an intervention category, point-in-time risk and safety approaches lose a considerable amount of their utility due to their inability to recalculate risk or safety levels based on current and evolving family conditions.
In family preservation and in-home case management scenarios, as well as following reunification, caregivers have full access to a vulnerable child. Risk and safety can fluctuate hourly based on caregiver stress and a multitude of other factors. The challenge CPS agencies face under these circumstances is recognizing when the safety threshold is about to be crossed and acting before anything goes wrong.
This begs two questions. First, how much surveillance is needed? Second, what should be monitored so as to gain foresight about the changing nature of risk?
The safety threshold is crossed when the complexity of a situation surpasses the adaptive capacity of the person responsible for safety. In families, this can occur when the adaptive capacity of caregivers is exceeded. It can also occur when the complexity of the family situation exceeds the adaptive capacity of the caseworker or CPS agency’s response.
Most CPS agencies tend to have minimal contact requirements, calling for once-a-month contact with the caregiver. Brief monthly contacts are hardly sufficient to produce the foresight needed to recognize that conditions are deteriorating relative to a child’s safety.
Caseworkers are given tools to use for initial assessments, but they are given little meaningful guidance as to what to monitor over time, and how to monitor it to gain foresight about safety once the case is opened. That is how a child may be considered as safe right up until the point of a fatal or near fatal incident.
Tracking safety over time requires frequent monitoring for changes in the intensity of multiple variables. A few of the many examples of significant changes to patrol for:
- The caregiver may have experienced a substance abuse relapse.
- A child’s behavior becomes more challenging, leading parents to resort to harsh physical punishment.
- Caregivers demonstrate escalating mental health symptoms, such as deepening depression thereby weakening protective capacities.
- Signs that a caregiver’s empathy for the child is decreasing, that the caregiver is being overwhelmed by mounting demands, or lacks the skills to manage new challenges presented by the child’s developmental needs.
These examples are by no means all-inclusive, but are meant to illustrate the fluid nature of variables influencing risk and safety when a child is in the full care and physical custody of a caregiver. A point in time approach does not capture the trend lines.
Accurately triaging families for interventions is certainly a critical part of the CPS response. But once this is done, risk and safety analyses take on a very different challenge. That challenge involves creating foresight about the changing nature of risk before anything goes wrong.
This aspect of risk and safety assessment remains comparatively embryonic. Its development couldn’t be more important if CPS agencies are to avoid high profile fatalities and serious harm to children in open CPS cases. A second and different approach to risk and safety assessment is needed once the case is opened.
Tom Morton most recently served as the child protection practice specialist for the Commission to Eliminate Child Abuse and Neglect Fatalities. Prior to that he served as director of the Clark County, Nevada Department of Family Services for five years and as founder, president and CEO of the Child Welfare Institute for 22 years.