<?xml version="1.0" encoding="utf-8"?><!DOCTYPE article  PUBLIC '-//OASIS//DTD DocBook XML V4.4//EN'  'http://www.docbook.org/xml/4.4/docbookx.dtd'><article><articleinfo><title>FAQ/likrat</title><revhistory><revision><revnumber>5</revnumber><date>2015-02-02 14:41:41</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>4</revnumber><date>2015-02-02 14:41:22</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>3</revnumber><date>2015-02-02 14:40:32</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>2</revnumber><date>2015-02-02 14:39:34</date><authorinitials>PeterWatson</authorinitials></revision><revision><revnumber>1</revnumber><date>2015-02-02 14:35:59</date><authorinitials>PeterWatson</authorinitials></revision></revhistory></articleinfo><section><title>Positive and negative likelihood ratios</title><para>Slide ten of these <ulink url="https://lsr-wiki-02.mrc-cbu.cam.ac.uk/statswiki/FAQ/likrat/statswiki/FAQ/likrat?action=AttachFile&amp;do=get&amp;target=posneglik.pdf">pdf slides</ulink> indicates rules of thumb for assessing positive and negative likelihood ratios to evaluate them to assess discrimination of the model. </para><para>where </para><para>Positive likelihood ratio = Sensitivity / (1 - Specificity) </para><para>Negative likelihood ratio = (1 - Sensitivity) / Specificity </para><para>The slides also briefly mention Hierarchical ROC curves which are useful for plotting curves combining within and between study variances from studies in meta-analyses. </para><para><emphasis role="underline">Reference</emphasis> </para><para>Jaeschke R, Guyatt, GH and Saclett DL (1994) Users guides to the medical literature. III. How to use an article abotu a diagnostic test. B. What are the results and will they help me in caring for my patients? The Evidence-Based medicine Working Group. <emphasis>Journal of the American medical Association</emphasis>, <emphasis role="strong">271(9)</emphasis>, 703-707. </para></section></article>