Academic journal article Child Welfare

Predictors of Family Preservation Outcomes and Child Welfare Success in Colorado

Academic journal article Child Welfare

Predictors of Family Preservation Outcomes and Child Welfare Success in Colorado

Article excerpt

A sample of 4,589 Colorado child welfare cases that closed between October 2007 and September 2009 was analyzed. All cases involved child abuse or neglect. Permanency and follow-up outcomes were scored using a methodology developed for the study. Scores were used to create an ordinal measure of success for the case. A cumulative logit statistical model examined the relationship between the newly-developed ordinal success measure and six predictor variables: number of caregivers, risk of abuse, poverty, risk of neglect, age of primary caregiver, and substance abuse issues. Case profiles are provided. Limitations and implications for practice are discussed.

This study emerged from a long-held interest by county child welfare agencies in Colorado to define and understand child welfare success more holistically Participating counties use principles of family preservation in the provision of services. A family preservation services model is built upon the conviction that most children can be protected and treated within their homes when parents receive services and support and when parents become empowered to change their lives. A child welfare success scoring methodology was developed for this study to measure permanency and subsequent involvement outcomes. It was developed using the principles of keeping children at home or reunifying children with parents, whenever it can be done safely.


Typically, predictive modeling in child welfare seeks to identify demographic, case, and family characteristics that are predictive of the likelihood of future maltreatment (e.g., Dubowitz, Kim, Black, Weisbart, Semiatin, & Magder, 2011). The study adds to the evidence base by using predictive modelling to examine child welfare success as defined by permanency and follow-up outcomes. The study was designed to address the following research questions:

1. How should a successful child welfare case be defined?

2. What are the predictors of a successful child welfare case?

The knowledge derived from this study will assist county and state agencies in advocating for more effective child welfare policy and in enhancing practice to improve effectiveness in attaining positive child welfare outcomes.


The following review highlights prior findings related to key variables used in the predictive model: number of caregivers, age of primary caregiver, caregiver substance abuse, poverty, and risk of abuse and neglect. The number of caregivers in a family has received empirical support as a predictor of permanency. The literature indicates that children from two-parent households tend to reunify faster than do children from other family structures (Courtney, 1994; Shaw, 2010; Wells 8c Guo, 1999). Studies have found that single-parent families experience lower rates of reunification than do other family structures (McDonald, Poertner, 8cJennings, 2007; Rockhill, Green, 8c Furrer, 2007; Wells 8c Guo, 1999). Furthermore, children living in single-parent homes appear to be at greater risk of foster care reentry (Shaw, 2006; Wulczyn, 2004) and maltreatment recurrence (Fuller, 2005), while being less likely to exit foster care (Glisson, Bailey, 8c Post, 2000). In contrast to number of caregivers, research on the relationship between caregiver age and child welfare outcomes is rather scarce. However, children with younger caregivers are less likely to reunify than children with older caregivers (Davis, Landsverk, 8cNewton, 1997; Fraser,Walton, Lewis, Pécora, 8c Walton, 1996). Also, Zuravin, Benedict, and Somerfield (1993) found that for every one year increase in maternal age, the likelihood of maltreatment decreased by four percent.

Parental substance abuse has been identified as one of the strongest predictors of child welfare outcomes. Children with substance-abusing parents experience lower rates of reunification (McDonald et al., 2007; Rosenberg 8c Robinson, 2004; Shaw, 2010). Parental substance abuse is also associated with an increased risk of reentry (Brook 8c McDonald, 2009; English, Marshall, Brummel, 8c Orme, 1999; Frame, Berrick, 8c Brodowski, 2000; Miller, Fisher, Fetrow, 8c Jordan, 2006; Terling, 1999). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.