Generation of LRs

AG Alice Garrett
CL Chey Loveday
LK Laura King
SB Samantha Butler
RR Rachel Robinson
CH Carrie Horton
AY Amal Yussuf
SC Subin Choi
BT Beth Torr
MD Miranda Durkie
GB George J. Burghel
JD James Drummond
IB Ian Berry
AW Andrew Wallace
AC Alison Callaway
DE Diana Eccles
MT Marc Tischkowitz
KT Katrina Tatton-Brown
KS Katie Snape
TM Terri McVeigh
LI Louise Izatt
EW Emma R. Woodward
NB Nelly Burnichon
AG Anne-Paule Gimenez-Roqueplo
FM Francesco Mazzarotto
NW Nicola Whiffin
JW James Ware
HH Helen Hanson
TP Tina Pesaran
HL Holly LaDuca
AB Alexandre Buffet
EM Eamonn R. Maher
CT Clare Turnbull
request Request a Protocol
ask Ask a question
Favorite

We generated positive LRs and confidence intervals based on the rate of the entity under study in positives (true positive rate) compared with the rate of entity under study in negatives (false positive rate), (a/a + c)/(b/b + d), where a = true positive, b = false positive, c = false negative, and d = true negative.22 We generated a negative LR based on the rate of absence of the entity under study in negatives (true negative rate) compared with the rate of absence of the entity under study in positives (false negative rate), (d/b + d)/(c/a + c). When 1 or more cells contained 0 counts, we universally applied to those analyses a Haldane correction (adding 0.5 to each cell): this correction dampens a signal of association toward the null and thus is inherently conservative.

Do you have any questions about this protocol?

Post your question to gather feedback from the community. We will also invite the authors of this article to respond.

post Post a Question
0 Q&A