| Dr. Burt Holland is a Professor in the Department of Statistics at Temple University, Philadelphia, PA, where he served as Department Chair from1991 to 1996. Since receiving a Ph.D. in Statistics from North Carolina State University in 1970, he has become an experienced data analyst and has published extensively in both statistics and applied journals including the leading statistics journals Biometrics, Journal of the American Statistical Association, and Technometrics. Dr. Holland frequently serves as a referee for JAMA, American Journal of Public Health, Psychmetrika, the National Science Foundation, and many others. He is presently co-authoring a comprehensive intermediate graduate level statistical methods text. Dr. Holland has been a consultant to CBP® since 1993. |
AJCC Jan 2000 |
|
Spine to Publish CBP® Researchby
Burt Holland, PhD
In
December 1999, as principal investigator, Dr. Deed Harrison received a
letter from James Weinstein, DO, Editor-in-Chief of Spine, informing us
that one of our cervical projects had received final approval after
review at Spine. This project concerned reliability of different types
of radiographic analysis on the lateral cervical view. I was a co-author
of this paper with responsibility for the study design, data analysis,
incorporation of the statistical conclusions into the manuscript, and
addressing all editorial comments dealing with statistics.
Final acceptance of this article required my spending a great
deal of effort to convince Spine’s referees and editors that the
procedures conventionally used to calculate and evaluate the statistical
measures of interexaminer and intraexaminer reliability were not
applicable to our experiment, nor to many similar experiments previously
reported in the chiropractic and orthopedic literatures.
I was able to locate the correct measures for our experimental
setup in a manuscript under preparation by Kevin Keen, PhD, a
biostatistician at Case Western University. Dr. Keen was very kind to
share this research in progress with us. I then programmed his formulas
for the analysis of our data. The key distinction between the
conventional and correct analysis is that, in our experiment, the
experimental factors bore a “crossed” rather than “nested”
interrelationship. As a guest speaker at the upcoming CBP® Annual
Seminar in New Orleans in September 2000, I will further explain some of
the details in a non-technical fashion, and more generally, the role of
the statistician in clinical research.
As a CBP® team of researchers, we have accomplished and will
continue to produce pathbreaking basic and clinical chiropractic
research that is reported in some of the best Index Medicus journals.
CBP® research projects will appear/have appeared in Spine (2), Journal
of Spinal Disorders (2), Journal of Manipulative and Physiological
Therapeutics (28), Chiropractic Technique (6), Journal of Orthopaedic
Research (1), and Clinical Biomechanics (2). Publication of chiropractic
research in front-line orthopedics journals is a substantial feat, and I
am very proud of my participation in these accomplishments.
|