(WITHDRAWN) The Use of Algorithmic Clinical Decision Support Systems in Social Work: A Systematic Review and Recommendations

Background: Research on decision-making in social work with children and families illustrates the influences on how workers make decisions, which may be demographic, personal or organizational, leading to the conclusion that decision-making is a highly subjective and potentially biased process, based more on experience, judgment, attitudes, and predispositions rather than on clearly articulated procedures or models (Schuerman & Vogel, 1986). As a result, states are searching for practice processes and tools that can improve the decision-making capabilities of social workers to ensure positive outcomes for those being served. While one approach to this improvement in decision-making is to sharpen the critical thinking skills of practitioners, in recent years efforts have been made to address this problem using clinical decision support systems (Savaya, Monnickendam, & Waysman, 2001). While clinical decision support systems (CDSS) have been demonstrated in other fields to be an important tool for evidence-based practice, and the few extant studies about social workers’ use of these systems also indicate they have the potential to change the way services are delivered. Therefore, this study seeks to enhance our understanding of the diffusion and utilization of algorithmic CDSS in social work practice by examining the facilitators and barriers to CDSS adoption among social workers using Rogers’ (1995) diffusion of innovation theory. This study provides an overview and critical appraisal of the current use of algorithmic CDSS in social work and aims to make social workers aware of these developments and engage in debate about CDSS implementation and development processes.

Methods: A systematic review was conducted, analyzing fourteen articles reporting empirical evidence about applications of CDSS in social work practice. The studies were performed in various areas of social work including child and family (including clinical settings), youth services (including youth probation and homeless youth), rehabilitation, mental health, and in-home support services. All studies included in this review utilized algorithmically driven CDSS that embody principles of expert systems.

Results: This analysis revealed that CDSS remain an emerging phenomenon in social work practice. The introduction of CDSS into the social work profession represents a seismic shift away from traditional service delivery and decision-making methods, as social workers usually rely on personal experience, professional judgment, and consultation with colleagues for referrals. With greater demands on social service agencies for accountability and demonstrated efficiency, more meaningful use of CDSS is likely to be in large public social service organizations in which the delivery of service is the responsibility of staff members with relatively little professional training.

Conclusion: In agencies such as child welfare, extremely important decisions must be made by people who are often ill-equipped to make them. Good supervision and in-service training programs help, however, CDSS may provide an additional practical approach by making knowledge readily available to all frontline workers. Therefore, CDSS can support significant improvements in evidence-based practice (EBP) concepts and allows for more rapid dissemination in the field.