Moaven Razavi
Dr. Moaven Razavi is a Senior Research Scientist at CISWH, where he focuses on estimating the economic impact of social work services on patient outcomes and healthcare costs. In his prior work he has demonstrated the value of care provided by nurse practitioners and physician assistants making the case that investment in training and promoting these practitioners improve the efficiency of the healthcare delivery system. Using insurance claims and electronic health records, he applies advanced statistical and machine learning methods to compare health outcomes and costs of the patients who receive social work interventions with those who do not. By emulating clinical trial designs within real-world data, Dr. Razavi aims to generate robust evidence on how social workers contribute to improved health and cost savings across various conditions, advancing the understanding of social work’s value in healthcare.
Dr. Razavi is a health policy expert with a multidisciplinary background in computer science and engineering, economics, healthcare policy, and health data science. From 2003 to 2005, he served as Director of the Bureau for Research at the Institute for Management and Planning Studies (IMPS) in Tehran, and consulted with the World Bank and WHO-EMRO on healthcare financing reforms in the Middle East and North Africa. He joined Brandeis University as a Ford Foundation Fellow in 2005 and later earned a PhD in Health Policy, contributing to research at the Schneider Institutes for Health Policy and Global Health.
His applied policy work includes serving as task lead for the State of Vermont’s inefficiency analysis in healthcare spending, co-investigator on Medicare health plan value methodologies, and key contributor to federal projects like the CMS Beacon Community Medicare data initiative and the Episode Grouper for Medicare. Across these efforts, Dr. Razavi has focused on risk adjustment, performance measurement, and improving care attribution methods, bringing a deep understanding of health system challenges and solutions.
An active contributor to the field of healthcare data science, Dr. Razavi leads research on applying machine learning and natural language processing (NLP) to large healthcare datasets. He serves on the Technical Advisory Committee for the Tufts CTSI Informatics Strategy, helping shape the Tufts Analytics Platform (TAP), which supports translational research through advanced ML and AI tools. His current work includes predictive modeling on post-surgical complications, HIV-related health-seeking behavior, and unsupervised machine learning to identify hidden patterns in health data and cluster analysis of hospitals for mapping into the two-dimensional cost-quality quadrant.
Dr. Razavi holds a Ph.D. in health policy, an M.S. in healthcare management and policy from Brandeis University, as well as an M.A. in economics from Institute for Management and Planning Studies in Tehran, Iran. He received a BSE in computer science and electronic engineering from Khajeh-Nasir Technical University in Tehran, Iran. His doctoral dissertation was evaluation of the impact of economic reforms on healthcare financing using a quasi-experimental design with a difference-in-differences construct applied to a multi-year data from household surveys of income and expenditures.
