Clinical correlation after testing

On many reports from radiology or genomic analysis you will see the words “Clinical correlation is recommended”.  These words encapsulate the wisdom that the test results add important new information that may dramatically change the differential diagnosis, and may prompt further examination and testing. 

In some situations, the clinical correlation involves re-examining studies already done to look for pertinent negative findings to assess diagnoses being considered.  Some specialties have formalized this type of clinical correlation by having radiology rounds or meetings to assess results of genomic testing.

Example to work through

Consider a 6-month-old boy with intellectual disability who has a CT scan showing brain calcifications (Registered users can click the images below to jump into the software; some videos may be helpful in navigating from there).

Intellectual disability and brain calcifications (click the image to jump into the software)

The differential diagnosis includes infections, ischemic encephalopathy and genetic disorders.  The Add Findings tab suggests various findings to check for the clinical correlation such as hepatomegaly. 

Useful findings (click the image to jump into the software)

Using the View Finding in the Differential Diagnosis screen one can see how hepatomegaly distinguishes among the diseases in the differential diagnosis.  Both experienced clinicians and the software take into account the frequency of finding by the patient’s age in considering the differential diagnosis.

Viewing hepatomegaly in the differential diagnosis (click the image to jump into the software)

Adding hepatomegaly to the phenotype moves the differential towards diseases with hepatomegaly, which is visually clear by the increased amount of black in the bars on the right side of the screen.

Adding hepatomegaly to the patient profile (click the image to jump into the software)

The error associated with not following up after an abnormal test result would be failing to get the maximal diagnostic value out of tests.

This page is part of a series on the Elements of Diagnosis.

Copyright © 2025 SimulConsult