We read the interesting and informative Clinical Review on Infectious Mononucleosis ( IM ) by the authors.
However, the Diagnostic Algorithm for IM in Figure 1 contains one major error ( a ) and one minor error ( b ).
Error (a ): After the first clinical features box, the Algorithm immediately wanders off the point by diverging into three choices, two of which terminate in dead ends without the user ever having an opportunity to make a diagnosis of IM or not.
Looking at the three choices from left to right:-
The first choice ("Full blood count ...") is the only option of the three relevant to the Algorithm. The second choice ("High risk patients ..." ) should be introduced further down the Algorithm after the diagnosis, or non-diagnosis, of IM. The third choice ( "Mild symptoms ..." ) is not relevant to the Algorithm and should be omitted.
Error (b ). If a patient has a lymphocyte count of exactly 4 x 10^9/L, the algorithm will fail as there is no option for this value in the Algorithm.
Our re-designed initial part of the Algorithm, shown in Figure 1, is much easier to understand and simplifies the logic of lymphocyte count interpretation and the presence, or absence, of atypical lymphocytes in the blood film. The yellow-coloured Decision and Process boxes show where “pregnant patients” and those at “high risk of HIV” should be placed in the Algorithm.
We appreciate that Clinical Algorithms in the form of Flow Charts are complex Diagnostic-aids to design, and we have recently identified simple design-errors in twenty such Flow Charts that can seriously affect their interpretation [1].
Reference
[1] Colman A, Richards B.
Clinical Algorithms: purpose, content, rules, and benefits.
International Journal on Biomedicine and Healthcare. 2014 ( 2 ),
28 –40. http://www.ijbh.org/ijbh2014-2.pdf ( accessed April 2015 )
Competing interests:
No competing interests
27 April 2015
Andrew W Colman
Clinical Computer Scientist and Staff Physician
Bernard Richards, Emeritus Professor of Medical Informatics, University of Manchester
Rapid Response:
The Diagnostic Algorithm in Figure 1 is confusing
We read the interesting and informative Clinical Review on Infectious Mononucleosis ( IM ) by the authors.
However, the Diagnostic Algorithm for IM in Figure 1 contains one major error ( a ) and one minor error ( b ).
Error (a ): After the first clinical features box, the Algorithm immediately wanders off the point by diverging into three choices, two of which terminate in dead ends without the user ever having an opportunity to make a diagnosis of IM or not.
Looking at the three choices from left to right:-
The first choice ("Full blood count ...") is the only option of the three relevant to the Algorithm. The second choice ("High risk patients ..." ) should be introduced further down the Algorithm after the diagnosis, or non-diagnosis, of IM. The third choice ( "Mild symptoms ..." ) is not relevant to the Algorithm and should be omitted.
Error (b ). If a patient has a lymphocyte count of exactly 4 x 10^9/L, the algorithm will fail as there is no option for this value in the Algorithm.
Our re-designed initial part of the Algorithm, shown in Figure 1, is much easier to understand and simplifies the logic of lymphocyte count interpretation and the presence, or absence, of atypical lymphocytes in the blood film. The yellow-coloured Decision and Process boxes show where “pregnant patients” and those at “high risk of HIV” should be placed in the Algorithm.
We appreciate that Clinical Algorithms in the form of Flow Charts are complex Diagnostic-aids to design, and we have recently identified simple design-errors in twenty such Flow Charts that can seriously affect their interpretation [1].
Reference
[1] Colman A, Richards B.
Clinical Algorithms: purpose, content, rules, and benefits.
International Journal on Biomedicine and Healthcare. 2014 ( 2 ),
28 –40.
http://www.ijbh.org/ijbh2014-2.pdf ( accessed April 2015 )
Competing interests: No competing interests