THE IMPACT OF DATA-DRIVEN DECISION SUPPORT SYSTEMS ON GOVERNANCE AND POLICY IMPLEMENTATION IN U.S. INSTITUTIONS

Abstract

This systematic review examines how data driven decision support systems shape the day-to-day realization of public policy in U.S. institutions by synthesizing evidence on their effectiveness, efficiency, compliance and risk management, transparency and accountability, and equity. Guided by a preregistered protocol and PRISMA procedures, we screened multidisciplinary databases and grey sources, applied calibrated two stage eligibility checks, and extracted implementation context, study design, and outcome metrics into a structured coding frame. In total, 115 studies were reviewed and included in the final analytic corpus. Across sectors such as health and human services, justice and public safety, education, benefits administration, and inspections and transportation, we find that decision support systems consistently improve proximal outcomes when analytic signals are coupled to executable protocols inside operational workflows, with frequent gains in guideline adherence, cycle time, backlog reduction, and error rates. Programs that embed documentation, rationale capture, and monitoring show stronger compliance and quicker remediation, while equity performance is most durable where pre-launch subgroup audits and scheduled post launch checks are routine. Moderator analyses highlight that leadership sponsorship, formal data governance, provenance controls, human in the loop checkpoints, role specific training, and scheduled model monitoring are reliable predictors of sustained benefits, especially when tools are fully integrated into systems of record rather than used as standalone dashboards. We conclude that value from decision support is contingent on socio technical completeness that connects trustworthy data pipelines to intelligible models, workflow embedded actions, and accountable governance.