Open Access Research article

Identifying patients with medically unexplained physical symptoms in electronic medical records in primary care: a validation study

Madelon den Boeft*, Johannes C van der Wouden, Trudie R Rydell-Lexmond, Niek J de Wit, Henriëtte E van der Horst and Mattijs E Numans

BMC Family Practice 2014, 15:109  doi:10.1186/1471-2296-15-109

PubMed Commons is an experimental system of commenting on PubMed abstracts, introduced in October 2013. Comments are displayed on the abstract page, but during the initial closed pilot, only registered users can read or post comments. Any researcher who is listed as an author of an article indexed by PubMed is entitled to participate in the pilot. If you would like to participate and need an invitation, please email, giving the PubMed ID of an article on which you are an author. For more information, see the PubMed Commons FAQ.

Comment to 'Better results when defining frequent attenders proportionally and using Problem diagnoses?'

Madelon den Boeft   (2014-09-17 10:33)  EMGO Institute for Health and Care Research, VU University Medical Center email

Dear Dr. Smits,

Thank you for your interest in our article.

With this reply I hope to clarify the issues that  you raised.

Regarding the issue that a sex and/or age specific selection method would have been more sophisticated, we agree without hesitation.  Using sex and age to refine our selection method, may lead to a higher prevalence of patients at risk for MUPS identified. However, this differentiation was not part of the procedure in this stage of development of the selection method we tested and compared, as we were not involved in this stage.

Regarding your second point, the choice between Episode codes and Problem status codes is to a large extent dependent on coding routines of general practitioners and characteristics of the EMR program they use. In our experience, as we have used the dataset before, the Problem status option is underused, so relying on this category of MUPS diagnoses would mean that a lot of eligible patients would be missed.  

Furthermore, we agree that an age/sex specific selection method as mentioned under the first point would also have consequences for the proportion of patients with chronic disease to be identified. We did address this issue in our Discussion section .

Regarding your last raised point: we included all contacts and not only GP contacts in our analyses.

Thank you for your valuable suggestions.

With kind regards,

Also on behalf of the co-authors,

Madelon den Boeft, GP and PhD fellow

Competing interests

There are no competing interests


Better results when defining frequent attenders proportionally and using Problem diagnoses?

Frans Smits   (2014-07-01 13:01)  Academic Medical Centre, university of Amsterdam, the Netherlands email

Dear Madelon and co-authors,

With great interest I read your article " Identifying patients with medically unexplained physical symptoms in electronic medical records in primary care: a validation study".

Because I am interested in this topic, I have a few comments and questions:
- You use frequent attenders (FAs) as "fishing pond" for MUPS. Why do you define FAs using a fixed number of consultations per year for all ages and both sexes? The disadvantage of this method is that you select predominantly the elderly and women. It doesn’t surprise me that you mention this later on in your article. To select patients who visit the GP exceptionally often, you better select the FAs for each age and sex group. We found in proportionally defined FAs a higher prevalence of MUPS than in your research.
- Is there any special reason why you use Episodes? In episodes with MUPS, the GP has not yet finally decided that the diagnosis is unexplained. If the doctor gives a so-called Problem status to a MUPS diagnosis, this usually implicates that the diagnosis is unexplained.
- You exclude patients with chronic complaints. Why? Patients with chronic conditions (e.g. diabetes) may also have MUPS? Our study showed that FAs often have both physical and psychological diagnoses (including somatoform). If you had defined FAs proportionally, it would not have been necessary to exclude patients with chronic problems.
- Because practice staff has mostly contacts for chronic conditions, I wonder whether you counted all contacts or only contacts with the GP.

I am curious whether your results would be different (better?) when using proportional defined FAs and problem diagnoses!

 With kind regards,

 Frans Smits, GP


Competing interests

non declared


Post a comment