Empirical research
Making Evidence Actionable: Interactive Dashboards, Bayes, and Health Care Innovation
Authors:
Anupa Bir ,
RTI International, US
About Anupa
Anupa Bir, ScD, MPH, is a health economist who has focused on improving the evaluation methods applied to research on the well-being of vulnerable populations. She has a particular interest in evaluating innovative interventions and their impacts on outcomes across traditional silos; for example, evaluating the impact of health care innovation on the workforce or family strengthening interventions on the financial well-being of families affected by incarceration. Her methodological interests relate to integrating data from various sources (qualitative, administrative, survey) and studies (meta-evaluation) to synthesize evidence for policy making. Dr. Bir pursues research that connects various social service systems and varied research disciplines to improve policy and practice to support vulnerable children and families. She currently leads several contracts to evaluate complex health and social policy interventions. These interventions include innovative interventions to improve access to quality health care, provide financial incentives for asset development, and improve communication and family strength during stressful circumstances like incarceration and re-entry.
Nikki Freeman,
University of North Carolina, US
Timothy Day
Centers for Medicare and Medicaid Innovation, US
Abstract
The results of many large-scale federal or multi-site evaluations are typically compiled into long reports which end up sitting on policymaker’s shelves. Moreover, the information policymakers need from these reports is often buried in the report, may not be remembered, understood, or readily accessible to the policymaker when it is needed. This is not a new challenge for evaluators, and advances in statistical methodology, while they have created greater opportunities for insight, may compound the challenge by creating multiple lenses through which evidence can be viewed. The descriptive evidence from traditional frequentist models, while familiar, are frequently misunderstood, while newer Bayesian methods provide evidence which is intuitive, but less familiar. These methods are complementary but presenting both increases the amount of evidence stakeholders and policymakers may find useful. In response to these challenges, we developed an interactive dashboard that synthesizes quantitative and qualitative data and allows users to access the evidence they want, when they want it, allowing each user a customized, and customizable view into the data collected for one large-scale federal evaluation. This offers the opportunity for policymakers to select the specifics that are most relevant to them at any moment, and also apply their own risk tolerance to the probabilities of various outcomes.
How to Cite:
Bir A, Freeman N, Chew R, Smith K, Derzon J, Day T. Making Evidence Actionable: Interactive Dashboards, Bayes, and Health Care Innovation. eGEMs (Generating Evidence & Methods to improve patient outcomes). 2019;7(1):40. DOI: http://doi.org/10.5334/egems.300
