David Robertson receives Young Biometrician’s Award 2015

The British and Irish Region of the International Biometric Society, jointly with the Fisher Memorial Trust, award a prize every two years for young biometricians (no more than 5 years since completing full-time education), who are members of the British and Irish Region of the International Biometric Society. The award will recognise the research of one paper published, or accepted for publication, in a refereed journal. This award comprises a diploma and a prize of £1000.

David RobertsonThe MRC Biostatistics Unit is delighted to share the news that David Robertson, a Unit PhD student, has been awarded the 2015 Young Biometrician's Prize, jointly run by the British & Irish Region of the International Biometric Society and the Fisher Memorial Trust.

The Prize is awarded on the basis of his contribution to the paper “Correcting for bias in the selection and validation of informative diagnostic tests” written with Drs Prevost and Bowden and published in Statistics in Medicine in 2015, and on the recommendation of the panel of three judges. They commented:

The paper provides an accessible and valuable approach to an important biostatistical problem. The development shows an impressive grasp of the practicalities of the issues involved, as well as a mastery of highly appropriate methodological techniques which, although well-established, are seldom seen nowadays.

David is expected to receive his prize at the Region's Annual General Meeting, which this year will be held at Rothamsted on October 8th. (Information provided by British and Irish Region of the International Biometric Society and David Robertson)

Congratulations David!

Paper summary

“Correcting for bias in the selection and validation of informative diagnostic tests”
When developing a new diagnostic test for a disease, there are often multiple candidates to choose from, and it is unclear if any will offer an improvement in performance compared to current technology.  A two-stage trial design can be used to select a promising test (if one exists) in stage one for definitive validation in stage two.  However, the first stage selection rules can lead to biased estimates of the true properties of the chosen diagnostic test.

We derived the most efficient unbiased estimator for a diagnostic tool's sensitivity in a two-stage design with arbitrary stage one selection rules.  We then applied our estimation strategy to data from a recent family history screening tool study, and were able to identify and successfully adjust for bias in the tool’s estimated sensitivity to detect those at high risk of breast cancer.