Rich JonesAssociate Professor of Psychiatry and Human Behavior, Associate Professor of Neurology
I am an epidemiologist and methodologist. I lead a Quantitative Science Program (QSP) at the Brown University Warren Alpert Medical School, serving primarily the Department of Psychiatry and Human Behavior, the Department of Neurology, and the Norman Prince Neuropsychiatric Institute. My main research areas concerns the cognitive reserve hypothesis and measurement and methodology in the clinical neurosciences. My main office is at Butler Hospital (click here for annotated map and directions).
I am an investigator in the area of mental health and aging, with a specific focus on the epidemiology of cognitive aging, cognitive reserve, and environmental influences on cognitive function. Cognitive reserve is a hypothesis proposing that personal experiences and characteristics (e.g., educational attainment) provide a buffer against the clinical effects of underlying disease. I am probing a possible causal role for educational attainment and other markers of brain reserve.
A particular area of my research and expertise lies in the use of modern psychometric methods (e.g., item response theory and structural equation modeling) to better understand and quantify constructs in health research settings. I have applied these methodologies to the study of dementia, delirium, and depression, and numerous other and diverse areas. I am also a committed translational researcher. I try to use my expertise in psychometrics and epidemiology to improve measurement devices for clinical and research use. Examples include development of new mental status screening instruments for older adults ( Fong et al., 2011
; Yang et al., 2013
) and neuropsychological assessment batteries and summary measures ( Jones et al., 2010
). I am committed to help improve the rigor and sophistication of research by (a) promoting advanced methodology in aging research by co-founding the Gerontological Association of America Special Interest Group on Measurement, Statistics, and Research Design
, (b) teaching a latent variable methods workshop
, (c) participating in a R13-funded Conference on Advanced Psychometrics in Cognitive Aging
which has as a principal mission of training junior faculty (K-awardees), and (d) developing and freely distributing software
to facilitate the use of complex statistical software and integrate such model estimation into users' general purpose statistical software.
My research centers substantively on the study of social and environmental correlates of successful cognitive aging. Special areas of focus include cognitive reserve, delirium, and the social and environmental correlates of cognitive aging. I am currently investigating the cognitive reserve hypothesis the idea that personal experiences and characteristics (e.g., educational attainment) provide a buffer against the clinical effects of underlying neuropathological processes. Cognitive reserve is a theory that is used to explain the discrepancy between level of neuropathology and observed level of cognitive impairment. When the burden of neuropathology is significant but cognitive impairment relatively preserved, this discrepancy is explained as the action of an enriched neural substrate and/or cognitive resources that facilitate plasticity and/or compensation. One of the main sources of evidence for the cognitive reserve hypothesis is the very high correlation between educational attainment and mental status and neuropsychological test scores. My early work focused on alternative (to cognitive reserve) explanations for this relationship. I hypothesized that the relationship may be spurious and reflect test item bias due to familiarity with testing situations, and approached the question with an item response theory and differential item functioning approach. What I found was that only a small fraction of this difference is due to measurement bias ( Jones and Gallo, 2002
). Now, I am probing a possible causal role for educational attainment and other markers of brain reserve. I have proposed that typical epidemiologic approaches to measuring reserve are conceptually flawed in a recent review ( Jones et al., 2011
). In this review I lay out an approach to measurement and experimentation (controlled and natural experiments) that can be used to validate the theory of cognitive reserve. I am currently addressing this framework in a research project attached to a program project: The Role of Reserve in Delirium (Professor Sharon Inouye, PI).
I am very interested in the detection of measurement bias (differential item functioning, DIF) using item response theory and related methods. My first R01 concerned "basic psychometric science" analyses of the relative accuracy of different algorithms to detect DIF using Monte Carlo methods. I continue with this line of research, expanding in to questions regarding how measurement bias can be detected in the context of adaptive assessment (i.e., computerized adaptive assessment, CAT). We are also working and published (possibly the first) application of a bifactor measurement model for the detection of differential item functioning ( Yang et al., 2009
). I participate in the PROMIS (Patient Reported Outcomes Measurement Information System) NIH Roadmap initiative as a consultant on multiple projects.
In addition to measurement, I am also interested in longitudinal data analysis (and in particular the overlap of longitudinal data analysis and cognitive aging). Recently we published an application of profile mixture modeling in cognitive aging ( Hayden et al., 2011
). This study characterizes the heterogeneity of trajectories of cognitive aging, and dispels the myth that cognitive decline in old age is normal or inevitable. I have also published in the area of latent class or profile mixture models, characterizing different cognitive phenotypes in delirium ( Yang et al., 2009
R01-AG044518 (Jones & Inouye, MPI)
06/01/14 – 03/31/19
NIH/National Institute on Aging Development and Validation of a Delirium Severity Toolkit
The goal is to develop a Delirium Severity Toolkit, a dynamic set of new measures developed with expert clinical judgment using modern psychometric theory to capture the severity of delirium in various phenomenological presentations.
P01-AG031720 (Inouye, PI)
04/15/2010 – 03/31/2016
NIH/National Institute on Aging Interdisciplinary Study of Delirium and its Long Term Outcomes
This Program Project seeks to elucidate novel risk factors and to examine the contribution of delirium to long term cognitive and functional decline. Dr. Jones is leader of Project 4 (The role of reserve in delirium) and co-leader of the Data Management and Analysis Core.
Role: Project Leader
R01-AG051170 (Jones, PI)
9/8/2015 - 6/30/2020
NIH/National Institute on Aging Psychometric Integrative Technology for Cognitive Health Research
The goal of this research is to harmonize cognitive assessments collected in the Health and Retirement Study and 15 international sister studies.
NIMH Traineeship in Psychiatric Epidemiology, Johns Hopkins University, Pre-doctoral T32 award, 1995-1998
Exceptional Reviewer, Medical Care, 2007, 2008, 2010
Excellence in Tutoring, Harvard Medical School, Clinical Epidemiology AC701 First year medical students. 2009
Top Reviewer Award, Journal of the American Geriatrics Society, 2011
Abrantes, Ana Associate Professor of Psychiatry and Human Behavior (Research)
Armey, Michael Assistant Professor of Psychiatry and Human Behavior (Research)
Battle, Cynthia Associate Professor of Psychiatry and Human Behavior (Research)
Bond, Dale Associate Professor of Psychiatry and Human Behavior (Research)
Case, Brady Assistant Professor of Psychiatry and Human Behavior (Research)
Daiello, Lori Assistant Professor of Neurology (Research), Assistant Professor of Health Services, Policy and Practice (Research)
Dickstein, Daniel Associate Professor of Psychiatry and Human Behavior, Associate Professor of Pediatrics, Associate Professor of Diagnostic Imaging
Additionally, I lead a summer workshop that provides an introduction to measurement research (e.g., factor analysis, item response theory) and longitudinal data analysis (e.g., latent growth curve models, growth mixture models; info at lvmworkshop.org ).