Introduction: The assessment of children at risk of abuse and neglect is a critical societal function performed by child protection workers in situations of acute stress and conflict. Despite efforts to improve the reliability of risk assessments through standardized measures, available tools continue to rely on subjective judgment. The goal of this study was to assess the stress responses of child protection workers and their assessments of risk in high conflict situations. Methods: Ninety-six child protection workers participated in 2 simulated scenarios, 1 non-confrontational and 1 confrontational. In each scenario, participants conducted a 15-minute interview with a mother played by a specially trained actor. Following the interview, the workers completed 2 risk assessment measures used in the field at the time of the study. Anxiety was measured by the State-Trait Anxiety Inventory at baseline and immediately following the completion of each interview. Physiological stress as measured by salivary cortisol was obtained at baseline as well as 20 and 30 minutes after the start of each interview. Results: Participants demonstrated significant stress responses during the 1st scenario, regardless of whether the interview was confrontational or not. During the second scenario, the participants did not exhibit significant cortisol responses, however the confrontational interview elicited greater subjective anxiety than the non-confrontational scenario. In the first scenario, in which the workers demonstrated greater stress responses, risk assessment scores were higher on one risk assessment tool for the confrontational scenario than for the non-confrontational scenario. Conclusion: The results suggest that stress responses in child protection workers appear to be influenced by the novelty of a situation and by a parent's demeanor during interviews. Some forms of risk assessment tools appear to be more strongly associated than other with the workers' subjective and physiological stress responses. This merits further research to determine which aspects of risk assessment tools are susceptible to the emotional elements of intake interviews. (Contains 2 tables.)