Emergency. 2016; 4 (3): 114-115

EV I D E N C E BA S E ME D I C I N E

Evolution of Fagan’s Nomogram; a Commentary
Abdelrahman Ibrahim Abushouk1∗

1. Faculty of Medicine, Ain Shams University, Cairo, Egypt.

Received: february 2016; Accepted: February 2016

Cite this article as: Ibrahim Abushouk A.Evolution of Fagan’s Nomogram; a Commentary. Emergency. 2016; NN(I):114-115.

Dear Editor

I read with interest your paper entitled “Pre and post-test

probabilities and Fagan’s nomogram” (1). I would like to add

a note concerning an update on Fagan’s Nomogram. Gener-

ally, the basic idea of most nomograms is having the scales

of 3 variables in a manner that if you draw a straight line be-

tween 2 values, the 3rd value is found where the line inter-

sects the 3rd scale (2). They were initially developed in the

1980s by Maurice D’Ocange. Nomograms remained popu-

lar in medical practice until the invention of pocket calcula-

tors and computers. Their use increased again with the in-

troduction of evidence-based medicine in clinical practice.

In a letter to the New England Journal of Medicine in July

1975, Dr Terry Fagan displayed a test characterization tool

that went on to carry his name as the Fagan’s nomogram (3).

This nomogram is a simple application of the Bayes’ theo-

rem, which establishes a rule to calculate the post-test prob-

ability of a disease. However, Fagan’s nomogram had a set

of drawbacks that limited its use in clinical practice. These

drawbacks included: 1- The original Bayes’ theorem is de-

signed to deal with odds ratios, not probabilities, so alge-

braic conversion is needed to calculate probability. 2- Most

diagnostic tests are characterized in terms of sensitivity &

specificity in the literature, which need special equations to

be converted into likelihood ratios (4). Noticing these diffi-

culties, in 2011, a group of researchers published a modern

version of the nomogram that they named “Bayes’ theorem

nomogram”. The new nomogram targeted the former prob-

lems using: A- Parallel lines for probability and odds on each

side of the nomogram figure. B- The inner lip along the en-

tire circle contains values for sensitivity and specificity that

can be connected to calculate the likelihood ratio of a cer-

tain diagnostic test (5). As illustrated in figure (A), a pretest

probability of 18% and a likelihood ratio (LR+) of 2.8 for a di-

agnostic test would give a posttest probability of 38%. Fur-

ther advantages of the modern nomogram include: - In Rare

∗Corresponding Author: Abdelrahman Ibrahim Abushouk; Ain Shams Uni-
versity, Cairo, Egypt. E-mail: abdelrahman.abushouk@med.asu.edu.eg .

Figure 1: The modern Bayes’ theorem nomogram with an example

of probability calculation as shown by the green line.

disorders, having a very low pretest probability implies per-

forming a diagnostic test with a fairly high LR (+). In Fa-

gan’s nomogram, the high values of LR are compressed in a

tight portion over that scale (4), while in this model; a more

spaced representation of high LR is feasible. - Replacing the

linear form with a circular one works better for complex di-

agnostic protocols where addition of multiple arrows for dif-

ferent diagnostic tests may be required (5). Considering the

mentioned superiorities of the Bayes’ theorem nomogram

over the conventional Fagan’s nomogram, it is highly recom-

mended for clinicians to use it in conducting diagnostic pro-

tocols and formulating therapeutic plans.

References

1. Safari S, Baratloo A, Elfil M, Negida A. Part 4: Pre and Post

Test Probabilities and Fagan’s Nomogram. EMERGENCY-

An Academic Emergency Medicine Journal. 2016;4.

2. Thimbleby H, Williams D, editors. Using nomograms to

reduce harm from clinical calculations. Healthcare Infor-

This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0).
Downloaded from: www.jemerg.com



115 Emergency. 2016; 4 (3): 114-115

matics (ICHI), 2013 IEEE International Conference on;

2013: IEEE.

3. Fagan TJ. Letter: nomogram for Bayes theorem. The New

England journal of medicine. 1975;293(5):257-.

4. Miettinen OS, Caro JJ. Foundations of medical diagnosis:

what actually are the parameters involved in Bayes’ theo-

rem? Statistics in medicine. 1994;13(3):201-9.

5. Marasco J, Doerfler R, Roschier L. Doc, what are my

chances. UMAP Journal. 2011;32:279-98.

This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0).
Downloaded from: www.jemerg.com


	References