Nilton Yhuri Carreazo Pariasca wrote:
> We are giving form to Biostatistics syllabus for Medicine students, and we
> would like include Critical Appraisal in this course. Does anyone have some
> experience or example of this?
I wrote a book about critical appraisal from a statistical perspective,
and much of the material I used for this book came from my old website,
* www.childrensmercy.org/stats
I also teach a series of short courses based on the book. Here's a brief
outline:
Stats 32: This discusses how a control group is selected for a research
study. It got a bit long, so I split it into three pieces.
Stats 32a: This covers randomization, how it is done, why it helps, and
when it cannot be used.
Stats 32b: This covers observational studies, defining cohort,
case-control, cross-sectional, and historical control studies. It also
discusses the strengths and weaknesses of these studies.
Stats 32c: This covers matching, stratification, and statistical
adjustments used when there is covariate imbalance between the
treatement and control groups.
Stats 33: This covers exclusions, refusals, dropouts, and failure to
comply, and the problems that these cause.
Stats 34: This covers clinical importance, blinding, surrogate outcomes.
Stats 35: This covers meta-analysis (systematic overviews) and describes
publication bias, heterogeneity and other issues.
Stats 36: This covers the criteria for causality first elucidated by Sir
Austin Bradford Hill (strength of association, dose-response, plausible
mechanism, specificity) as well as newer issues like conflict of
interest and fraud.
These courses are described on my training page:
* www.childrensmercy.org/stats/training.asp
though I had to remove the actual handouts out of respect for the
copyright holder for my book.
There is a general correspondence between the chapters of my book and
the course numbers but it is slightly out of order (Chapter 1=32, 2=33,
3=34, 4=36, 5=35).
Each of these classes takes about three hours to teach (except for 32
which is why I split it), so that's a fair amount of your syllabus. I
would argue though, that critical appraisal is more important to your
students than some of the computational issues stressed in most
statistics courses.
There are a number of good textbooks on critical appraisal besides mine.
I hesitate to single out any particular book, but I do like
* Studying a Study and Testing a Test: How to Read the Health Science
Literature Third Edition. Riegelman RK and Hirsch RP (1996) Boston:
Little, Brown and Company. ISBN: 0-316-74521-9.
You could also draw your material from a book about Evidence Based
Medicine, though EBM is more than just critical appraisal. Dan Meyer's
book is one of my favorites.
* www.childrensmercy.org/stats/weblog2006/FavoriteBooks10.asp
Steve Simon, Standard Disclaimer
I'm starting a new career as an independent consultant,
so please note the change in email address and web site.
My new website, www.pmean.com, is up and running!
|