An interview with David L. Streiner: Truth teller of statistical concepts in medicine

There is too much pressure to publish. Lack of sufficient statistical power remains a significant methodological obstacle. LC: Tell us something about your background. How did you get started in statistics and what are you doing now? DS: I was born in the Bronx, New York and went to City College of New York, graduating in 1963 with a Bachelor of Arts and majoring in Psychology. I went on to study clinical psychology at Syracuse University, ultimately earning a Ph.D. in 1968. The political environment in the US at the time was troubled, and in 1968 we moved as a young family to Canada. I had the fortuitous opportunity to begin work at a nascent medical school in Hamilton, Ontario, and remained at McMaster University for 30 years, employed as professor for about 20 of those years (Department of Psychiatry & Behavioural Neurosciences, and Department of Clinical Epidemiology & Biostatistics), and then retired in 1998. Retirement was brief however, and I went on to join the faculty at the University of Toronto as Professor of Psychiatry and was the founding director of the Applied Research Unit at the Baycrest Centre for Geriatric Care where I was for about 10 years until retiring for the second time. Since then I am back at McMaster University (Professor Emeritus) one day a week and consult at the University of Toronto's Center for Addictions and Mental Health. Throughout my career I have been teaching, writing, and consulting; I have participated in many grant submissions and helped others with statistics and research methodology. I still do that even though I am “retired.” LC: How did you get started providing methodological advice to researchers? DS: I started doing that the after day after I arrived at McMaster. You need to understand that it was a brand new medical school. Because I had taken two graduate courses in statistics I was labelled as the “stats guru.” This began a multi-decade educational process to keep ahead of the questions and I have kept it up since. I started writing commentaries and tutorials for the Canadian Journal of Psychiatry in 1990, inspired by having done peer reviews of poorly-constructed submitted papers. Topics range from using meta-analysis in psychiatric research2 to path analysis.3 The editors of Community Oncology/Journal of Community and Supportive Oncology also invited me to contribute to a series on practical biostatistics, followed by requests from the editors of Chest and the Journal of Clinical Psychopharmacology. In my hands-on work in research I have worked with several disciplines including psychiatry, neurology, family medicine, and pediatrics. I was one of the founding editors of Evidence-Based Mental Health and worked on that journal for 10 years. LC: Of all the commentaries you have written, which are your favorites? DS: Well, the editors of Community Oncology/Journal of Community and Supportive Oncology gave me free rein to use any language as I see fit.4 Another article that comes to mind addresses a common practice (screening) that is done without thinking through the consequences or the issues involved.5 LC: What inspired you to write about P-hacking? DS: I am concerned about problems with lack of a priori hypotheses, multiplicity, and subsequent spurious findings. For example, brain imaging studies may involve only a handful of subjects and yet millions of voxels are being looked at for “findings.” Genome studies also are fraught with chance findings. The articles authored by John Ioannidis further describe some of the issues involved.6, 7 LC: Do you see any improvements in how research is being reported today compared to 10-20 years ago? DS: Sadly, no. There is too much pressure to publish. Lack of sufficient statistical power remains a significant methodological obstacle. I continue to see articles being submitted that have basic errors. LC: Sounds like we have quite a bit of work to be done! What did you think of the animation I sent you of a hapless researcher insisting his/her submission would be welcome because the “P-value is less than 0.05”? It was part of a prior editorial,8 and readers can find it at https://www.youtube.com/watch?v=KBALRk2hjMs. DS: Loved the animation. You must have been sitting in on some of my consultations with researchers! LC: What would readers not normally know about you? DS: About 40 years ago I got into wood working and I am currently the “Master Woodworker” at a train and trolley museum – these old trolley cars need constant repair. Sometimes I even drive them. LC: Thank you so much for your time in answering these questions. DS: It was delightful talking with you. No external funding or writing assistance was utilised in the production of this editorial. In the past 12 months, Leslie Citrome has served as a consultant to: Acadia, Alkermes, Allergan, Indivior, Intra-Cellular Therapeutics, Janssen, Lundbeck, Merck, Neurocrine, Noven, Otsuka, Pfizer, Shire, Sunovion, Takeda, Teva, Vanda. In the past 12 months, Leslie Citrome has served as a speaker for: Acadia, Alkermes, Allergan, Janssen, Lundbeck, Merck, Neurocrine, Otsuka, Pfizer, Shire, Sunovion, Takeda, Teva, Vanda. Other disclosures: stocks (small number of shares of common stock): Bristol-Myers Squibb, Eli Lilly, J & J, Merck, Pfizer purchased >10 years ago; royalties: Wiley (Editor-in-Chief, International Journal of Clinical Practice), UpToDate (reviewer), Springer Healthcare (book).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.