• Together with the German ESSCA partners, the three Dutch partners and B. Knudsen from Copenhagen are the first to install and use the WinAlldat/ESSCA software.
• Funding for ESSCA will be sought for from manufacturers and industrial associations; additionally, A. Schnuch and W. Uter are preparing a EU proposal under the 5 FP, area ”Environment & Health”.
• In the VU department, the program runs under Windows 2000 and MS Office/Access 2000, respectively. While this is not actually supported by the program developers, it seems to be possible without stability problems.
• It is possible to work in a network with WinAlldat (as several German partners do), but it is recommended not to use more than five clients simultaneously.
• Personal data, including hospital or national patient ID-numbers, are not transferred with the set of data exported for ”central pooling”. The export files contain plain ASCII; they are automatically compressed to one zip-file.
• In the unlikely case of the whole database having to be transferred to the data centre / support in Göttingen (instead of applying a set of transactions locally to the database, e.g., for future extensions and amendments) the patient data will be encrypted locally, so that they are not readable elsewhere, and decrypted only on return of the updated database. R. Arnold will supply ESSCA partners with such a tool; e.g., via the Göttingen FTP-server. For additional security, the database would be treated on a PC not connected to the Internet in Göttingen.

Catalogues

• Occupations:
  The full list of occupations is considered too long to be practicable. However, if only a very concise version, including only the most common job titles is used, with the instruction to consult the full list if the actual job title of a patient cannot be found, the latter may not be done, resulting in too gross or mis-classification. As a compromise, to make the European standard ISCO-88COM applicable for ESSCA, W. Uter will present a carefully shortened list.
• Current diagnoses:
  All Dutch ESSCA partners require the recording of more diagnoses than the ones included in the ”minimal dataset” (ACD, ICD, endogenous eczema, ”Other” diagnosis [to be specified]). As a compromise combining this apparent need with ESSCA ”minimal dataset” requirements, W. Uter suggested to include 4 diagnoses (with identical pick lists, containing the three above mentioned diagnoses and any additional terms) together with an additional qualifier for each of these 4 diagnoses (i.e., ”exclusively”, ”predominantly”, ”contributing”).
• Substances:
  The names of the allergens can be edited (=changed, e.g., translated) via MS Access, as long as the full meaning is preserved and the key number is not altered. Occasional duplicates may occur despite utmost care when introducing new allergen names. If they have not yet been used in a test series, they may be invalidated by typing ”XXX...” in the ”name” data field. If they are used, the same can be done, but the substance-ID has to be changed into the ID belonging to the remaining term in the table ”tblTestsubst” via MS Access (this table links the allergens names with concentration etc.; the ”testsubstid” is quoted in the table ”tblBlocksubst” which is then linked to ”tblBlock”, containing the name of the series etc.)
  Test series containing special substances currently under investigation which should not be transferred to the data centre for pooling can be given the test method attribute ”epicutaneous in loco” (or, actually, any other – possibly newly created – attribute than ”epicutaneous”): only results with substances contained in test series marked ”epicutaneous” (ID=A) are selected for transfer. In future, an additional attribute of the test series will serve as a respective flag.
• Sites:
  Altogether six sites are considered necessary; as with additional diagnoses, the present possibility to include these as ”additional questions” is considered too clumsy (question: diagnosis-related, i.e., six sites for each diagnosis? To be clarified)

• Screen and other designs:
  Several improvements have been suggested concerning screen design, which can help to ease the use of the program:
• Moving the left ”final diagnosis” button upwards
• Using the free space for a header ”printouts”, referring to the 4 buttons at the bottom of the main dialogue
• Use the same terms on buttons and on the entries in the pull-down menu lists
• Narrower page margins for printout, especially test protocols, to enable longer test series to fit under the first-page header.
  
• Reports
  The reports presently included are partly rudimentary. They are derived from statistics included in the old IVDK software. As further development of this essential function (data entry would not be worthwhile if data cannot be analysed locally also) R. Arnold and W. Uter propose the creation of a second database, containing a transformed copy of the original database (which is optimised for minimum storage and maximum update flexibility) in a much more aggregated fashion, largely identical to the data structure of centrally pooled data. This would mean that the same queries can be used locally and centrally (=synergistic effect); additionally, queries developed locally by experienced MS Access users (as C. Scholten) can be distributed (e.g., via the data centre / FTP-server) to other ESSCA partners, because most of the statistics will be of interest for others, too. However, in this case, the purpose and logic of such a report must be carefully documented and explained, so that it can be fully understood by those applying it.
• This and that:
• Instead of travelling to the ”OK-button” with the mouse pointer, the ”Pos-1” key can be activated for ”save” and the ”End” key for ”cancel” (pull-down menu ”Extras” à ”Options” à ”Keyboard”)
• The paper ”history form” (a Winword 97 document which may be edited, e.g. translated locally) will be distributed together with installation files
• Data fields containing unstructured text should either be visible (e.g., ”unformatted text” in the main dialogue, as ”relevance description” is visible in the relevance dialogue and – at least partly – at the bottom window of the main dialogue) or some marking should be displayed showing that text has been entered (e.g., comments to the reading of a test series at a certain reading time).
• During the meeting of ESSCA after the ESCD meeting in May, 2000, already it was suggested (and will be implemented in the next program version) that the user should be able to pick single standardised test-substances and compile these to an individualised test series (as with materials brought in by the patient) without loosing the possibility to include these in analyses concerning these standardised allergens (as yet, patient materials from individual test series are stored and analysed differently). At the moment the only possibility is to choose a test series containing the desired allergen(s) and mark all other allergens with ”NT” for not tested.