A recent RAND report found that test-driving autonomous vehicles is not a feasible way to determine when they will be safe enough for consumer use. This resulted in a lot of questions.
Oct 25, 2016 The RAND Blog
Susan Paddock is a senior statistician at the RAND Corporation and a professor at the Pardee RAND Graduate School. Her research includes developing innovative statistical methods, with a focus on Bayesian methods, hierarchical (multilevel) modeling, longitudinal data analysis, and missing data techniques. Paddock is the principal investigator of a project sponsored by the National Institute on Alcohol Abuse and Alcoholism to develop methods for analyzing data arising from studies of group therapy–based interventions. She was the principal investigator of a project sponsored by the Agency for Healthcare Research and Quality (AHRQ) to improve the science of public reporting of healthcare provider performance. She is the co-principal investigator of a project to conduct analyses related to the Medicare Advantage Plan Ratings for Quality Bonus Payments. She previously led a study sponsored by AHRQ to investigate statistical methods for the analysis of longitudinal quality of care data for non-ignorable missing data. Her substantive research interests include health services research, substance abuse treatment, drug policy, mental health, quality of health care, and health care provider performance assessment. She was the project statistician for the external program evaluation of the quality of care provided by the Veterans Health Administration to patients with mental health conditions. Paddock has been involved with several evaluations of group cognitive behavioral therapy–based interventions for treating co-occurring depression in substance abuse treatment clients. She has previously been involved with research on developing, monitoring, and refining a prospective payment system for inpatient rehabilitation care for Medicare beneficiaries. Paddock has served on editorial boards for the Annals of Applied Statistics, Journal of the American Statistical Association, and Medical Care, and is currently serving on committees for the American Statistical Association and the Institute of Medicine. She received her Ph.D. in statistics from Duke University.
Paddock SM, "Statistical benchmarks for healthcare provider performance assessment: A comparison of standard approaches to a hierarchical Bayesian histogram-based method," Health Services Research, 49(3):1056-1073, 2014
Paddock SM, Savitsky TD, "Bayesian Hierarchical Semiparametric Modelling of Longitudinal Post-treatment Outcomes from Open-Enrollment Therapy Groups," Journal of the Royal Statistical Society, Series A, 176(3):795-808, 2013
Savitsky TD, Paddock SM, "Bayesian Non-Parametric Hierarchical Modeling for Multiple Membership Data in Grouped Attendance Interventions," Annals of Applied Statistics, 7(2):1074-1094, 2013
Paddock SM, Woodroffe A, Watkins KE, Sorbero M, Smith B, Mannle TE, Solomon J, Pincus HA, "The quality of mental health care for veterans of Operation Enduring Freedom and Operation Iraqi Freedom," Medical Care, 51(1):84-89, 2013
Paddock SM, Hunter SB, Watkins KE, McCaffrey DF, "Analysis of group therapy data under a rolling client admissions policy using conditionally autoregressive priors," Annals of Applied Statistics, 5(2A):605-627, 2011
Paddock SM, Louis TA, "Percentile-based Histogram Estimates for Performance Evaluation of Healthcare Providers," Journal of the Royal Statistical Society Series C (Applied Statistics), 60(4):575-589, 2011
Paddock SM, Ebener P, "Subjective prior distributions for modeling longitudinal continuous outcomes with non-ignorable dropout," Statistics in Medicine, 28:659-678, 2009
Paddock SM, "Bayesian variable selection for longitudinal substance abuse treatment data subject to informative censoring," Journal of the Royal Statistical Society, Series C (Applied Statistics), 56:293-311, 2007