2022 Marshall Joffe Methods Award Winner
Stephen R. Cole works to build robust, accurate, and impactful knowledge at the intersection of epidemiology and statistics. He is interested in study designs and analyses that accurately estimate parameters of central interest to population-health scientists. These study designs include randomized experiments and observational studies. In particular, he is interested in infectious diseases, primarily HIV, and birth outcomes.
2021 Marshall Joffe Methods Award Winner
“I am a ‘case of crossover,’ oscillating between academe and government, trying to combine the active skepticism of Tom and Iain Chalmers with the skeptical activism of Dr. Bernard Lown, co-founder of International Physicians for Prevention of Nuclear War. While I was an Assistant Professor at Harvard School of Public Health, helping Dr. Lown with a peace project (1986-88), I met another IPPNW activist, cardiologist James Muller, who invited me to help him design what became the Myocardial Infarction Onset Study. Read more
In 1991, I migrated to promoting trials of policy and programs in British Columbia PharmaCare, inspired by Tom Chalmers’s 40 years of randomizing. That led to a long collaboration with Sebastian Schneeweiss including my continuation as an Adjunct Professor at Harvard until 2016. Drug policy studies led me to meet pharmacoepidemiologists in Odense, Denmark, who were planning case-crossover studies, notably Jesper Hallas with whom I continue to work, including clarifying parallels between case-crossover and symmetry analysis. On leave from PharmaCare, 2002-2006, as Professor at University of Victoria’s School of Health Information Science (thanks to a career award from BC’s Michael Smith Foundation for Health Research), I helped create a BC node in Murray’s Stroke Onset Study. In 2007, SER invited me to run for president, which gave me a podium in 2009 to urge epidemiologists to be leaders against climate change.
Murray’s application of case-crossover methods to study healthcare worker injuries gave me confidence to compete for a Professorship entitled BC Chair in Patient Safety, endowed by the Ministry of Health in the Dept of Anesthesiology, Pharmacology and Therapeutics at UBC. From 2009-2012, my Professorship encompassed my government position as Director of Research and Evidence Development in PharmaCare. In those years, with David Madigan’s FDA-sponsored Working Group on Case Only Designs (2011), I worked out many similarities and several key differences between case-crossover and self-controlled case series designs for drug safety monitoring. In 2012, my initiative in promoting policy-relevant research triggered my involuntary oscillation back to full-time academe (7 staff were fired for taking initiative that new bosses didn’t understand) until the Ministry apologized and rehired me as Research Scholar in Residence (2018-2020) to develop rapid-evaluation with Designed Delays (control groups).
In the interim, with Andy Carson-Stevens, a patient safety activist in Wales, I developed a plan to apply case-crossover methods to routine adverse event reporting in hospitals but our progress has been slow. In 2016, I joined Suzanne Cadarette’s Self-controlled Crossover Observational PharmacoEpidemiology (SCOPE) Group in producing guidance (‘Control Yourself’, in press) and tools (forthcoming) for case-crossover, self-controlled case series and symmetry analyses. Through Will Dixon’s ‘Cloudy with a Chance of Pain’ Study in Manchester UK (2017-19), I have become excited about emerging opportunities for self-controlled analyses of data from smart phones. I see these as platforms for potential n-of-1 studies in diverse domains. Meanwhile I focus on improving quality of quality improvement methods: surgical quality outcome reports and pragmatic trials of audit-and-feedback to family physicians using ‘prescribing portraits’ made by Colin Dormuth’s Pharmacoepidemiology Group in the Therapeutics Initiative at UBC.”
2020 Marshall Joffe Methods Award Winner
Johns Hopkins Bloomberg School of Public Health
Elizabeth A. Stuart is a professor of mental health, biostatistics, and health policy and management in the Johns Hopkins Bloomberg School of Public Health, where she is also Associate Dean for Education. Her research involves causal inference and missing data methods and their application to mental health, substance use, education, and public health more broadly. She has done influential work in propensity score methods for non-experimental studies, and in methods for assessing and enhancing the external validity (generalizability) of randomized trials to target populations. She received her AB from Smith College and her Ph.D. in statistics from Harvard University. After working for two years as a researcher at Mathematica Policy Research, she joined the Johns Hopkins faculty in 2006. Read more
She is a Fellow of the American Statistical Association, chair of the SERV (Mental health services research) NIH study section, and recent recipient of the Gertrude M. Cox Award from the Washington Statistical Society and RTI International. She is also heavily involved with the Bloomberg American Health Initiative, for which she leads the evidence workgroup.