Intended for healthcare professionals

Rapid response to:

Clinical Review

Diagnosis of ovarian cancer

BMJ 2015; 351 doi: (Published 01 September 2015) Cite this as: BMJ 2015;351:h4443

Rapid Response:

The modified NICE Flow Chart for the Diagnosis of Ovarian Cancer in Primary Care shown in the Figure lacks logical organisation and contains multiple errors.

This article is interesting and illuminating but the NICE Flow Chart [1] in the Figure is poorly organised and contains multiple errors making it difficult to understand and use, and the authors' changes add no value to it.

The flaws in the Flow Chart are as follows :-

1. The horizontal choices emanating from the “GP Consultation” Box are not mutually exclusive and encourage incorrect data flow from left to right instead of from top to bottom ( Figure 6 [2] ). The Flow Chart does not follow the order of the traditional diagnostic process namely History, Examination and Investigations followed by Diagnosis/Differential Diagnosis. Moving from left to right, the subject of the first Box is “Examination Findings”, the next two Boxes, “History Items”, and the last, “Unrelated Symptoms”. This last Box adds no value to the Flow Chart and should be omitted. A similar error can be found in a Flow Chart for the diagnosis of syncope [3].

2. The two “Symptoms” Decision Boxes should be combined into one Box as their outputs converge ( q.v. Figure 1 ).

3 .The two “Symptoms” Decision Boxes have no “Yes” and “No” choices
( i.e. Does the patient have any of these symptoms? ) so how can one select a valid response to the Decision Box containing the question immediately following them " Is ovarian cancer suspected?" which does have these choices?

4. The Decision and Process Boxes in the Flow Chart are of the same shape ( rectangular ) rather than decisions being placed in diamond-shaped boxes and processes in rectangular boxes.

5. The Decision Box based on the result of the “ultrasound of the abdomen and pelvis” should have a further Decision Box attached to it if the result is abnormal but not ovarian cancer ( q.v. Figure 1 ).

6. The decision process in the Flow Chart is cyclical and this property is not evident in the original Chart.

Figure 1 shows our re-designed Flow Chart that corrects the deficiencies in the original one. The Yellow Process Boxes indicate the possible end points of the Flow Chart. We have not included all the symptoms in the first Decision Box of the Flow Chart for the sake of clarity.


[1] National Institute for Health and Care Excellence.
The recognition and initial management of ovarian cancer.
(Clinical guideline 122.) 2011 Algorithms xxiv ( accessed September 2015 )

[2] Colman A, Richards B.
Clinical Algorithms: purpose, content, rules, and benefits.
International Journal on Biomedicine and Healthcare. 2014 ( 2 ), 28 –40. ( accessed September 2015 )

[3] Ammirati F, Colivicchi F, Santini M.
Diagnosing syncope in clinical practice. Implementation of a simplified diagnostic algorithm in a multicentre prospective trial - the OESIL 2 study
(Osservatorio Epidemiologico della Sincope nel Lazio)
Eur Heart J. 2000 Jun;21(11):935-40

Competing interests: No competing interests

28 September 2015
Peter McGill
Logician and 3rd Year Undergraduate Student
Andrew Colman, Post-doctoral Clinical Computer Scientist and Physician, Department of Medicine, QE Hospital, Gateshead, NE9 6SX, UK
Department of Physics and Astronomy
University College, London, UK