An Exploration into the Use of Differential Response in Ontario: Is There a Gap between Vision and Reality?

Background and Purpose: Child Welfare in Ontario has a dual mandate: to protect children from maltreatment and to protect children from environments that negatively impact their well-being and future development. In 2007 a Differential Response (DR) model was introduced to address the changing profile of families, it allowed workers to provide different response to families based on their presenting needs and type of maltreatment. Implementation of policies does not always reflect the intended outcomes. This study explored what the practice of DR looks like in Ontario 10 years after it was introduced. The purpose of this study was to determine if the type of maltreatment and level of harm to a child can predict the type of investigative approach used by the workers. 

Methods: The Ontario Incidence Study of Reported Child Abuse and Neglect (OIS) is a provincial study examining the incidence of reported child abuse and neglect in Ontario. The OIS-2013 data was used to explore the case factors influencing a worker’s clinical decision about the type of investigation to use with families. Trocmé et al. (2014) divided investigated reports of maltreatment into two categories; urgent protection investigations and other investigations, and assessments representing chronic need. This study utilized Chi Square Interaction Detection (CHAID) analysis was used to see if cases requiring urgent protection investigations utilize the traditional response, and if those categorized as chronic need are more likely to receive customized responses.

Results: Customized investigations were used in 70% of all urgent protection investigations, and vary from a low of 49% for severe physical harm to a high of 73% and 74% for neglect and physical abuse in children under four, respectively. 83% of all chronic need investigations were conducted using a customized response, and vary from a low of 72% for physical abuse in children over four to a high of 90% for exposure to IPV. Only police involvement and previous case openings displayed statistical differences in the types of investigations used. Overall, 81% (n= 4,191) of cases utilized a customized investigation approach, and traditional investigations were used in 19% (n=995) of cases. The most significant predictor of customized investigations was exposure to intimate partner violence, χ2 (1, n= 5,186) = 58.18, (p<.001).

Conclusions and Implications: According to most DR models, traditional investigations are used for high risk cases, including all reports of sexual abuse, physical abuse, or serious neglect, especially when criminal charges may be laid (Merkel-Holguin et al, 2006; Trocmé et al., 2003). This does not appear to be the way DR has been implemented in Ontario. The findings of this study suggest that DR has not assisted in reducing the number of families subjected to a child welfare investigation, nor that the model has assisted the system in more clearly identifying cases that require an urgent and thorough forensic investigation. Given that the system remains overburdened, and over 80% of the families investigated do not fit the criteria for urgent protection investigations, more research is needed into the implementation of Ontario’s differential response model.