gms | German Medical Science

49. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds)
19. Jahrestagung der Schweizerischen Gesellschaft für Medizinische Informatik (SGMI)
Jahrestagung 2004 des Arbeitskreises Medizinische Informatik (ÖAKMI)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie
Schweizerische Gesellschaft für Medizinische Informatik (SGMI)

26. bis 30.09.2004, Innsbruck/Tirol

Critical evaluation and interpretation of routine hospital data: incidence and geographical variation of cataract and cataract surgery in Austria

Meeting Abstract (gmds2004)

Search Medline for

  • corresponding author presenting/speaker Verena Barbieri - Department of biostatistics and Documentation, Innsbruck, Österreich
  • Eduard Schmid - Department of Ophthalmology and Optometry, Innsbruck, Österreich
  • Karl- Peter Pfeiffer - Department of Biostatistics and Documentation, Innsbruck, Österreich

Kooperative Versorgung - Vernetzte Forschung - Ubiquitäre Information. 49. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds), 19. Jahrestagung der Schweizerischen Gesellschaft für Medizinische Informatik (SGMI) und Jahrestagung 2004 des Arbeitskreises Medizinische Informatik (ÖAKMI) der Österreichischen Computer Gesellschaft (OCG) und der Österreichischen Gesellschaft für Biomedizinische Technik (ÖGBMT). Innsbruck, 26.-30.09.2004. Düsseldorf, Köln: German Medical Science; 2004. Doc04gmds360

The electronic version of this article is the complete one and can be found online at:

Published: September 14, 2004

© 2004 Barbieri et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.




Cataract is highly prevalent in elderly persons and cataract surgery is one of the most common interventions in Austria and worldwide. Geographical variability of cataract and cataract surgery can be examined using the Minimum Basic Data Set (MBDS) of Austrian hospitals of the years 2001 and 2002. Since every public and private hospital admission is documented in the MBDS, these data contain information on every patient such as age, sex, postal code, main- and additional diagnoses and procedures. Using this dataset the epidemiological analysis of routine data is critically discussed. Not only has the documentation bias to be taken into account, but also the application bias. This bias is arising since hospital documentation first of all is done because of financing.


Since patients' data are documented anonymously, the only way to refer persons to a region is via the postal code. Hospitalisation rates (HR) and standardized hospitalisation rates (SHR) were calculated on the district level and age/sex frequencies and patterns were assessed. The average length of stay was calculated. Different prognostic factors, aggregated at the district level, were considered to explain regional variability. Further variability between districts was taken into account, as adjacent districts influence each other. A model that considers the application bias as well as prognostic factors and regional autocorrelation on aggregated data was designed using the Bayes' approach. Incidence rates were compared to rates from two prospective population based cohort studies in USA and Australia [1].


In 2001 there were a total of 59597 public hospital admissions with cataract as main or additional diagnosis (ICD-10: H25, H26.1-H26.3, H26.5-H26.9, H28.0, H28.1, Q12.0).76.7 % of those underwent cataract surgery. 58.857 public and 6.480 private admissions with these diagnoses were documented in 2002. While in the public hospitals 78.7 % of the patients underwent surgery, 92.9 % were operated in the private clinics. The surgery rate men : women was 1:1.7 in public and 1:1.9 in private hospitals. After adjustment for age and gender regional variability was examined. Geographical patterns could be assessed, depending on health care supply and the number of specialists. The average length of stay with cataract surgery is depending strongly on the hospitals. In 2002 private clinics' average length of stay was 2.97 days. Public hospitals discharged patients on average after 4.21 days in 2002, and after 4.42 days in 2001.


The advantage of the use of routine data in epidemiology is the complete documentation of the hospitalisation data, since hospitals are constraint to document because of financing. Admission and surgery habits of the hospitals can be assessed and imbalance in health care supply can be shown. But there are several disadvantages too: The exact number of hospital admissions can be calculated, but, since admissions are documented anonymously, the exact number of persons having a diagnosis can't be assessed. Because there is no unique patient identifier the calculated incidence rate could be slightly higher. Comparison with the results of other studies shows similar tendencies, but the rates of cataract diagnoses and surgery are higher in Austria (e.g. 5-year incidence of cataract surgery in USA 6.3, in Australia 5.7 and in Austria 8.1 in people aged 50 years and older).While the documentation bias is negligible, the application bias is quite a problem, since different hospitals show very different admission and surgery habits. The influence of prognostic factors can only be assessed at the district level. This aggregation involves loss of information. Since routine data reflect reality but may be biased [2] by the application, they are very helpful in the documentation of the strength and weaknesses of the Austrian health care system. Further routine data can be very helpful in showing trends and regional patterns, but when patterns vary, the validity of the data in relation to its intended use must be carefully considered [3].


Panchapakesan J, Mitchell P, Tumuluri K, Rochtchina E, Foran S, Cumming RG. Five year incidence of cataract surgery: the Blue Mountains Eye Study. Br J Ophthalmol. 2003 Feb;87(2):168-72.
Lewsey JD, Leyland AH, Murray GD, Boddy FA. Using routine data to complement and enhance the results of randomised controlled trials. Health Technol Assess. 2000;4(22):1-55.
Hansell A, Hollowell J, McNiece R, Nichols T, Strachan D. Validity and interpretation of mortality, health service and survey data on COPD and asthma in England. Eur Respir J. 2003 Feb;21(2):279-86.