gms | German Medical Science

Kongress Medizin und Gesellschaft 2007

17. bis 21.09.2007, Augsburg

Regional differences in ICD-coding practices of hospital discharge diagnoses related to acute upper gastrointestinal (UGI) bleeding: analysis of the official hospitalization statistics of Germany for 2003

Meeting Abstract

  • Ingo Langner - Bremen Institute for Prevention Research and Social Medicine (BIPS), Bremen
  • Elke Scharnetzky - Bremen Institute for Prevention Research and Social Medicine (BIPS), Bremen
  • Iris Pigeot - Bremen Institute for Prevention Research and Social Medicine (BIPS), Bremen
  • Wolfgang Ahrens - Bremen Institute for Prevention Research and Social Medicine (BIPS), Bremen
  • Edeltraut Garbe - Bremen Institute for Prevention Research and Social Medicine (BIPS), Bremen

Kongress Medizin und Gesellschaft 2007. Augsburg, 17.-21.09.2007. Düsseldorf: German Medical Science GMS Publishing House; 2007. Doc07gmds232

Die elektronische Version dieses Artikels ist vollständig und ist verfügbar unter:

Veröffentlicht: 6. September 2007

© 2007 Langner et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen ( Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.



Introduction: For upper gastrointestinal bleeding the ICD-10 coding system allows both the coding of unspecific symptoms (e.g. haematemesis) and the more specific coding of the underlying disease (e.g. oesophageal variceal bleeding). Regional differences in coding could result in regional variations in disease incidence when analyzing data from hospital statistics. We investigated regional differences, especially East-West differences in coding practices with respect to diagnoses of UGI bleeding.

Methods: We investigated two groups of diagnoses: oesophageal bleeding (unspecific code K92.0 vs. specific codes I85.0, K22.6, K22.8) and gastrointestinal bleeding (unspecific code K92.2 vs. 16 specific codes). We used data from the 2003 German Official Hospital Statistics and the Population and Area Statistics (source: Federal Statistical Office). East-West differences for specific and unspecific coding and the combination of both for each diagnosis group were analyzed with separate Negative Binomial Regression Models, respectively. The models adjusted for age, sex, and population density.

Results: In total, 121,537 hospitalizations with a discharge diagnosis related to UGI bleeding were identified. Specific diagnoses of oesophageal bleeding were coded more often in East Germany (RR=1.23, 95% CI 1.15-1.30) than in West Germany whereas it was the reverse for haematemesis (RR=0.61, 95% CI 0.55-0.68). In the combined analysis of specific and unspecific oesophageal bleeding, this East-West difference was no longer present: RR=1.05 (95% CI 0.99-1.11). A similar pattern was seen for gastrointestinal bleeding: RR=1.29 (95% CI 1.23-1.34) for specific diagnoses, RR=0.87 (95% CI 0.84-0.91) for gastrointestinal haemorrhage with unspecified location (K92.2), and RR=1.07 (95%CI 1.04-1.11) in the combined analysis of these diagnoses.

Discussion: There appear to be different coding practices for hospital discharge diagnoses in East and West Germany related to UGI bleeding with more specific coding of the underlying disease in East Germany. Therefore regional different disease incidences should be interpreted with caution since this fact might be due to different regional coding practices.