Latent class analysis (LCA) has been used for determination of tests'
sensitivity and specificity in the absence of the gold standard and has been
pioneered by Dr. Stephen Walter of McMaster University (see his review
article on the topic in J Clin Epidem 1988;41:923-937).
hope this helps
ben d.
Benjamin Djulbegovic, MD,PhD
Associate Professor of Medicine
H. Lee Moffitt Cancer Center & Research Institute
at the University of South Florida
Division of Blood and Bone Marrow Transplant
12902 Magnolia Drive
Tampa, FL 33612
Editor: Evidence-based Oncology
http://www.harcourt-international.com/journals/ebon/
e-mail:[log in to unmask]
http://www.hsc.usf.edu/~bdjulbeg/
phone:(813)979-7202
fax:(813)979-3071
> -----Original Message-----
> From: Trisha Greenhalgh [SMTP:[log in to unmask]]
> Sent: Monday, April 03, 2000 9:17 AM
> To: evidence-based-health
> Cc: patkenny
> Subject: Validating diagnostic tests in the absence of an established
> gold standard
>
> Dear list members
>
> Pat Kenny has emailed me from Argentina an interesting question, and I
> think it would be helpful to share some dialogue in this list. The
> clinical area he's interested in is the diagnosis of gastro-esophageal
> reflux, but the general question is this: what do we do when we wish to
> develop a clnical study about a diagnostic test in an area where there is
> no gold standard? See his email below.
>
> I (and many others) have previously written that in the absence of a true
> gold standard one needs to establish and use the best available
> combination of existing tests and remain mindful of their potential
> limitations, but that sounds like a bit of a cop-out. It's probably time
> for interested parties in EBHC to explore this area and come up with some
> general guidance, if not guidelines, for people like Pat.
>
> My own feeling is that we should collect some specific examples of areas
> where people have addressed this problem, find out what they did, and try
> to draw generalisable lessons from their successes or failures, rather
> than start with hypotheticals which may not incorporate the practical
> problems of the real world.
>
> Any suggestions? Please copy Pat into your reply as he's not to my
> knowledge a member of this list. Thanks for your input.
>
> trish
>
> Trisha Greenhalgh
> Senior lecturer in primary care
> Department of Primary Care and Population Sciences
> Holborn Union Building
> Whittington Campus
> London N19 3UA
> Tel 0171 288 3246
> Fax 0171 281 8004
>
>
> -----Original Message-----
> From: Patricio Kenny [mailto:[log in to unmask]]
> Sent: 01 April 2000 23:50
> To: Trisha Greenhalgh
> Subject: Consultation from Argentina
>
>
> Dear Prof Greenhalgh:
> I´m a paediatric gastroenterologist from Buenos Aires, Argentina, who
> follows closely your papers and publications, specially those related to
> how to read/develop a medical paper.
> The purpose of this, is to ask you for some help in developing a clinical
> study about a diagnostic test in an area where there is no gold standard
> to be used for validation of the new test. In other words, what
> combination of criteria do we have to develop and justify against which
> the new test could be assessed.
> The specific area that we are dealing with is "gastroesophageal continuous
> overnight scintigraphy in children suspected to have aspiration by
> gastroesophageal reflux". The point is that there is no gold standard for
> "aspiration" by GER. So I ask for your expert counseling.
> Many thanks in advance, and best regards.
>
> Pat Kenny
> Paediatric Gastroenterologist
> Director
> Center for Paediatric Digestive Diseases
> Buenos Aires British Hospital
> Buenos Aires
> Argentina
> e-mail: [log in to unmask] <mailto:[log in to unmask]>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|