What's the point of indexing?

One of the questions we get to hear more often relates to plain text search versus indexing. In our first series of "HCM SPELLED OUT", our product experts therefore explain why good indexing is so important.

The core idea of JiveX Healthcare Content Management (HCM) is bringing together all medical data into one system. Consolidation of this kind makes searching for medical information far easier. Moreover, it also ensures that all available data on a case or a patient is taken into consideration - which improves the quality of treatment.

Keywords as a consequence of consolidation

In order to fully exploit these advantages in practice, however, the data cannot be gathered together and stored any old way. Rather, you need to ensure that users can rely on genuinely receiving all medical information. It sounds banal, but in actual fact it is not a matter of course - and it is a key quality characteristic of HCM.

This is because it enables digital classification of the medical data - a fundamental requirements for structured storage. Why is classification so important? Firstly, because it corresponds to the learned systems for paper files. It is only due to classification that data can be located at all in the paper file, because it specifies the necessary structure, within which the search is to be conducted.

JiveX HCM transfers these systems into the digital world - and even improves them. That's because, unlike the paper file, in which each document/each element of information can only be assigned once, digital information can be assigned to several keywords.

Classification leaves no room for doubt

There is another, more important argument in favor of classification: The demand for completeness of information in the search result. One alternative to classification, which at first sight is attractive for healthcare facilities, is unregulated keywording. That means that the search is conducted - as with the big Internet search engines - on the basis of keywords such as "CT", "lab" or "finding". A keyword search like this is equally easy for users and IT experts. For the user, because he does not need to work through a defined structure. For the IT expert, because he does not need to create such a defined structure.

At the same time, however, keyword search in the medical context is highly risky. After all, who guarantees that all relevant data is also displayed in relation to a keyword? What happens of a "CT examination" is stored under "CT" and (in the German case, two further variant spellings), "Computertomografie" and "Computertomographie"?  The standard use of the same terms sounds like a good solution for this problem, in theory, but in practice users actually tend not to be that disciplined.

So if healthcare facilities use unregulated keywording for the data search, there is always a residual doubt for the user as to the completeness of the information being displayed for his for the medical assessment.
Classification of the data is therefore absolutely necessary in order that documents can be reliably found with a defined search or a filter.  Incidentally,  classification also has a part to play when medical data is exchanged with other facilities. The IHE profiles intended for the cross-facility data exchange makes explicit provision for such classification.