Splitting combination codes in detailed longitudinal data

The Health-Hilmo and Avohilmo registers may use combination codes that consist of two ICD10 codes separated by a special character. Combination codes are a Finnish speciality, only seen in the Finnish ICD10 version, and not in the international WHO ICD10 version. When THL receives data for these registers they split the combination codes into separate fields. FinnGen will do the same if it receives any combination codes from the THL. The handling of the combination codes depends on the special character (*,#,& or +) that is between the codes:

Below is an example of how combination codes are split into CODE1, CODE2 and CODE3 columns in the detailed longitudinal data file:

The THL Hilmo Opas describes the handling of the combination codes as follows:

When Hilmo data arrives at THL, the combination codes are checked and the “reason code” is moved to its own field and the “symptom code” to its own, using the combination code punctuation (+, *, # and &). If there are no punctuation in the combination codes or the cause and symptom codes are not already in their own fields when they arrive at THL, the cause and symptom codes may be in the wrong fields in the registers. When compiling Hilmo data, the service provider must ensure that the patient information systems or the specifications and other measures related to the compilation of Hilmo data do not break the combination codes and thus leave them incomplete. Changes have been made to the patient information systems that will automatically decode codes into “cause” and “symptom” codes.

Last updated