James Hanley, SSC Award for Impact of Applied and Collaborative Work 2016
The award recognizes outstanding contributions by members of the SSC in collaborative research and applied work, the importance of which derives primarily from its relatively recent impact on a subject area outside of the statistical sciences, on an area of application or on an organization.
The 2016 recipient of the Statistical Society of Canada Award for Impact of Applied and Collaborative Work is Jim Hanley, Professor, Department of Epidemiology, Biostatistics and Occupational Health, McGill University; Associate Member, Department of Mathematics and Statistics, McGill University; and Senior Scientist, Division of Clinical Epidemiology, Royal Victoria Hospital.
Born in Ireland, Jim completed his undergraduate and Masters degree at the National University of Ireland in Cork before coming to Canada to complete his PhD at the University of Waterloo. Jim and his wife Ann Marie met at the University of Waterloo in 1972. They spent seven years in Buffalo and Boston where Jim worked at State University of New York at Buffalo and the Harvard School of Public Health, respectively. The first two of their boys were born in Framingham, and the other two in Montreal after Jim joined McGill University’s Department of Epidemiology and Biostatistics in 1980. Their sons remember the 1985-86 year spent at the WHO in Geneva and especially the 1992-93 year in Ethiopia. Jim and Ann Marie continue to travel. They recently welcomed a granddaughter, who joins their four grandsons.
Jim’s excellence as a statistical scientist working in epidemiology and human health is evident in his 90 peer-reviewed research papers and many invited editorials and other contributions to the statistical and medical literature. This work reflects Jim’s deep understanding of the substantive areas in which he works, and his passion for translating statistical knowledge to the collaborating scientists by using simple language and clear, intuitive explanations.
In 2013 Jim was commissioned by the World AntiDoping Agency (WADA) to validate test limits for drug testing in sports. Specifically, he has been working with scientists in Canada and Europe to set and validate limits for a biological measure (human growth hormone) that can be used by WADA to detect illegal usage of this substance. This work has been, quite literally, “game changing” and is now the legally upheld standard for the detection of cheating athletes, implemented for the first time in the Sochi Winter Olympics of 2014. As part of this work, Jim gave expert testimony at the Court of Arbitration for Sport in Switzerland, and assisted WADA in winning their case against an athlete who had appealed a drug ban.
Perhaps the greatest of Jim’s contributions to the statistical sciences in medicine stems from his very important and highly-cited papers with Barbara McNeil (Hanley and McNeil, Radiology 1982 and 1983). These papers were ground-breaking in for radiology and other areas of diagnostic testing and are cited 12,667 and 5,102 times respectively (Google Scholar) – very often outside of the statistical literature as highlighted in Dr. McNeil’s letter of support. It is fair to say that these papers revolutionized research in radiology and more broadly in diagnostic testing, and despite their relative age, are still both current and landmarks in the field. These papers opened up several avenues for methods development and for epidemiologic research. Jim followed these up with a long series of papers on these topics, and continues today to work on methods in the area. He has also collaborated on a number of important studies on diagnostics and screening. Alone, this line of work would represent a significant lifetime contribution to epidemiology and clinical research, and clearly shows Jim’s impact on medical research.
A third example of Jim’s collaborative impact is his work on screening. Jim recognized that clinical trials of screening tests in cancer and other diseases were designed in such a way as to underestimate the true effects of screening. Jim led the effort to develop statistical methods that used the trial data in a rigorous way to demonstrate that screening tests could have significantly larger effects than those detectable in trials, and applied this work in collaboration with clinical scientists in prostate cancer, breast cancer and other diseases. This work has been ongoing for several years, but has had direct and recent impact on screening recommendations and research and has potential to have important effects on public health.
Jim has also contributed to collaborative and applied research through his work as an Associate Editor for several journals including Biometrics, Statistics in Medicine, the Canadian Medical Association Journal, and Medical Decision Making, and as a consultant or in collaboration with a variety of health research groups including the Conseil d’évaluation des technologies de la santé du Québec, the Massachusetts Dental Survey and the New England Regional Burn Program. He has also served on grant and award review panels such as the FRQS Chercheur Boursier awards and the NHRDP Personnel Awards.
Jim has an avid interest in the history of statistics and data, including Student’s t-test and Guinness, Galton’s development of the early ideas of regression (going so far as to send his son to photograph the pages of Galton’s original record books from the archives at University College London), and the metaphysical and mathematical origins of the hazard function. Students and colleagues alike enjoy learning about Jim’s historical discoveries, and appreciate the breadth of knowledge and wealth of ideas and experience, which Jim is always cheerfully ready to share.
The citation for the award reads:
“To Jim Hanley, for ground-breaking research in methods for diagnostic testing, influential contributions to assessment of disease screening programs, and development of drug-testing standards in international sports; as well as for fostering research excellence through collaboration, mentorship, and devotion to training in epidemiological methods.”