Department of Biostatistics Seminar/Workshop Series

Conflict-of-Interest Concerns in Biomedical Statistics

Peter B. Imrey, PhD

Quantitative Health Sciences, The Cleveland Clinic Foundation

Monday, February 4, 2:00-3:00pm, MRBIII Conference Room 3131

Intended Audience: Persons interested in applied statistics, statistical theory, epidemiology, health services research, clinical trials methodology, statistical computing, statistical graphics, R users or potential users, and ethicists

Statisticians take pride as, with increasing frequency, statistical design and analysis frame discourse and are pivotal to business, scientific, and public policy decisions. This talk considers scrutiny as a price of such influence, with medical research controversies as point-of-departure.

Conflict of interest in medical research is drawing increasing concern from the public, governments, and prominent journals. With growing dependence of public-sector research on private-sector support, suspicions have been heightened by i) highly-publicized though apparently infrequent instances of selective dissemination, ghost authorship, and attempted data suppression, and ii) more common observance of research designs subordinating science to marketing, and financial links between regulatory advisory panelists and clinical practice guidelines authors and manufacturers of products whose use they influence.

Data from clinical studies are assembled and interpreted under statistical supervision. Statistical information may be privileged, statistical judgments may be technically debatable, and statistical decisions can greatly affect a study’s impact, and biopharmaceutical stock prices. In a published survey, many biostatisticians claimed awareness of fraudulent medical research. Conflict-of-interest concerns may thus inevitably impinge on biomedical statisticians. So a prominent medical journal now mandates academic replication of statistical analyses, selectively for industry research, while other journals assert rights to review raw data generally, and primary clinical trial statisticians are increasingly blinded to treatments.

The statistical community’s response, if any, to conflict-of-interest concerns, requires both understanding the motivations for these concerns and facing questions of professional identity and practice. What differentiates conflict of interest in applied statistical work from a statistician’s responsibility to a scientific team? Do statisticians have special responsibility, above that of other scientists, to maintain a dispassionate scientific posture? What are professionally appropriate limits of statistical advocacy? Do we share and educate to a sufficiently specific professional ethic? More down-to-earth, i) what support do we provide colleagues under pressure, and ii) what are the practical limitations and ramifications for statisticians and institutions undertaking oversight reviews of data analyses? Some thoughts on these will be offered as starting points for discussion and introspection.

Presenter Information
Topic revision: r3 - 26 Apr 2013, JohnBock

This site is powered by FoswikiCopyright © 2013-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback