Preliminary Evaluation of Structured Professional Judgment to Assess Removal in Child Protection Practice

Abstract

This study is a pilot validation of a newly devised evidence-based clinical instrument that assists professional judgment for decisions relating to child removal. The Child Protection Removal Assessment (ChiPRA) instrument adopts a structured professional judgement (SPJ) approach to decision-making and was developed from a literature review of studies identifying factors associated with severe child abuse. A study comparing the predictive validity of ChiPRA and an actuarial instrument was conducted using court file data from 298 child protection cases. A logistic function from all ChiPRA items, χ2(11) = 147.546, p < 0.000, correctly classified 86.3% of cases (Area Under the Curve [AUC] = 0.799, p < 0.000, 95% confidence interval: 0.738–0.859). The abuse subscale of the actuarial instrument yielded a modest but significant AUC (0.595, 95% CI: 0.530–0.660). Results indicate an increased reliance upon dynamic factors by magistrates when determining child removal. SPJ instruments warrant further research including prospective studies measuring reliability and validity studies.