Article No. 84

How would you rate this article?

Awful

Poor

Average

Good

Perfect

Create a Notebook

You currently don't have any notebooks.

If you want to save this article into a notebook, you can create one now.

Flag/Report
Article No. 84

Instructions:

What is the issue?

Let us know in a brief paragraph or sentence why you are flagging this article. Here are some other issues we, at CodeHealth, would like to know about:

  • Spam or misleading content
  • Violation of privacy
  • Unnecessary sexual content
  • Unnecessary violent content
  • Hateful or abusive content
  • Promotion of harmful acts

Report:

Flagged articles and users are reviewed withing 24 - 48 hours upon submission of a report. The report will be reviewed to determine whether the article and it's author (user) has violated CodeHealth's Terms of Use and/or Community Guidelines. Accounts are penalized for violations to either the Terms of Use and/or Community Guidlines. Repeated violations can lead to account termination.

Thank you for your report, it will be reviewed within the next 24-48 hrs.
Likelihood Ratios
ARCHIVE
RATE
REPORT/FLAG

Likelihood Ratios

(0)

Introduction

The likelihood ratio (LR) is the probability of finding an event or positive diagnostic test in a patient with the disease, divided by the probability of the same finding in a patient without the disease. It is a way of measuring the value of positive or negative diagnostic test.

The likelihood ratio is a way to determine how a positive or negative test result will affect the likelihood someone has a disease. Conversely, it also answers the question, "If a patient has disease Y, how likely is this patient to have a positive X test result compared to patients who do not have the disease.

Unlike negative and positive predictive values, likelihood ratios are not affected by prevelance.

Positive Likelihood Ratio (LR+)

The LR+ the the probability of a patient who has the disease testing positive divided by the probability of a patient who does not have the disease testing positive.

LR+ = sensitivity / (1 - specificity)

Negative Likelihood Ratio (LR-)

The LR- the the probability of a patient who has the disease testing negative divided by the probability of a patient who does not have the disease testing negative.

LR- = (1 - sensitivity) / specificity

Interpreting LRs

LR = 1, means that there is no change in the likelihood of the disease

LR >1, means that there is an increase in the likelihood of having disease

LR <1, means that there is a decrease in the likelihood of having disease

Practically speaking, in clinical practice you have a pretest probability of a patient having a disease based on things such as their history of present illness or physicial exam. You obtain further testing such as blood tests or imaging which has a likelihood ratio (either positive or negative) that will influence your pretest probability of that patient have a particular disease to a post-test probability. This factor that changes your pre-test to post-test probability is the likelihood ratio which can be plotted on the nomogram below.

 

About the Author

Mark Suguitan
Emergency Physician with interests in education, critical care and toxicology
Add Section
  • New Section
    Creates a new section with
    subtitle and text

  • Text Block
    Insert text without a subtitle

  • Image
    Insert an inline image

Add Margin
  • Text Block
    Insert text without a subtitle

  • Image
    Insert an inline image

Upgrade Your Account
Case Set Name