Back 3CGP          

April 2, 2015

  Marc Jamoulle

3CGP ; Story & description

Core Content Classification in General Practice Family medicine is a classification mixing the International Classification of Primary Care and the Q-Codes.

At the Department of General Practice, University of Amsterdam, in the 80’, the burden of reading medical journals specific to general practice was devoted to teachers and assistants in the department.  Articles relevant to symptoms, process and diagnoses (ie clinical items) were indexed by ICPC, the International classification of Primary Care and non-clinical subjects were by Q-Codes. Using the letter Q, available in ICPC, the late Professor Henk Lamberts had opened this new category. This was before the Internet, before Medline become available on CD Rom. To our knowledge this list Q-Codes has  never  been published but a copy taken on the tables of the library of the department in 1987 has been preserved. We see in Figure 1 the first list of Q-Codes proposed by Professor Lamberts, also linchpin of development of ICPC in the Wonca International Classification Committee, the Wonca working group in classification field.

Figure 1 Q-CODES Amsterdam (+/- 1987)

Dep of gen practice. Prof Lamberts

From Q-Codes to 3CGP

 

In 2007 the annual conference of Wonca Europe is held in Paris. A complicity with some French colleagues allowed to have access to abstracts of the conference before it takes place.

The reading of 998 abstracts has permitted to empirically identify major themes presented by participants from many countries and cultures. The structure of what might be called an authority list took shape slowly. The tool should be combined with the ICPC classification. One therefore used a hierarchical structure. Taking the Q-Codes of Lamberts as a basis, gradually the 8 areas plus a rag bag appeared to be necessary and sufficient to introduce the major concepts encountered in the texts analyzed.

This empirical method of choice based on the experience of the researcher is described in the Sage dictionary of social research method as 'Purposive sampling' {Jupp 206};

“A form of non-probability sampling in which decisions concerning the individuals to be included in the sample are taken by the researcher, based upon a variety of criteria which may include specialist knowledge of the research issue, or capacity and willingness to participate in the research. Some types of research design necessitate researchers taking a decision about the individual participants who would be most likely to contribute appropriate data, both in terms of relevance and depth.”

 

In our case the "individual participant" is the concept identified as significant by the researcher specialized in the discipline he observes.

 

Figure 2 The 8 knowledge domains in GP/FM and one rag-bag (QO Other)

Information about the themes identified can be considered as meta-information. The name chosen for the tool storage was first “meta-classification”. The need to show we were at the heart of the business of GP/FM has given rise to a change in naming and meta-classification becomes the Q-Codes, which, associated to ICPC-2 became Core Content Classification in General Practice / Family Medicine whose acronym is 3CGP.

Figure 3 Opening of Q-Codes domain (here QT4 Training)

Q-Codes has a hierarchical structure. The 8 domains denoted each par their first letter are open into categories, open itself into subcategories and if necessary in basic headings. It does not list all the areas of interest of family doctors but those encountered by indexing till now. They can be grouped conceptually according to the logic of the semantic triangle of Ogden and Richard {Richard and Ogden 1923 & 1989}

It is observed that this procedure about the knowledge exhibited by physicians in the field is a bottom-up approach as opposed to the knowledge accumulated in the textbooks we have mentioned above which can be seen as a knowledge Top- bottom. In our case, the experts are the general practitioners. They talk about their concerns. In textbooks, experts are interpreters of what  they think is reality. Both methods have their value and are probably complementary but to get an intimate knowledge of family medicine, it is interesting to collate what is expressed in the field.

« What are they talking about?»

This is the question that must be asked by readers of communications to congresses. Thus the project goes much further than keywords or descriptors. The project is to identify the categories of concepts in GP/FM. When themes and major categories are emerging, one must tackle the thorny issue of subcategories. There is the dilemma of any classifier, namely the impossible combination of total inclusivity and exclusivity. A section must contain all the topics concerned with the exclusion of those who do not apply. That's the question sometimes insoluble whether you put the watermelon with fruits or vegetables. It is easy to separate the nails & screws but where does ne may put screws that can be nailed?

Anyway we tried to resolve these issues with the greatest possible precision, without, in 2007, to have time to look up the definitions, inclusions and exclusions that would probably refine the tool and report use the most fluid. This task is at the forefront of work to be undertaken. Computer techniques have recently made ​​great strides, we try to incorporate this approach to knowledge in the movement of distributed data and the Semantic Web.

Figure 4 Conceptualization of the themes of communications (here Vocational training) following the semantic triangle

But the first step  is to continue the slow discovery of the universe of GP/FM.

After Wonca 2007 and a 6 years latency period, we have had the chance to assist to the 2013 Portuguese congress of family medicine in Covilha, then to analyze the French CNGE 2013 congress in Clermont, followed by the SwissFamilyDocs congress , Zurich, 2014, the Belgian congress of GP/FM in Brussels 2014 and last one the LILLE CNGE Congress 201

We have succeeded to identify less than 200 themes. ICPC-2 contains 760 rubrics. ICPC-2 and Q-Codes form a tool of less than 1000 rubrics which aims is to index abstracts of communications of GPs to congress, offering new insight in the work of GPs

See the current list of Q-Codes

MJ  online april 2, 2015