Estimating the annual burden of health care-associated infections in Finnish adult acute care hospitals

Am J Infect Control. 2009 Apr;37(3):227-30. doi: 10.1016/j.ajic.2008.07.004. Epub 2008 Dec 25.


Background: We estimated the burden of health care-associated infections (HAIs) occurring in Finnish adult acute care hospitals using national hospitalization data and estimates of HAI based on a recent national prevalence survey.

Methods: A total of 7531 non-HAI patients and 703 HAI patients (8.5%) identified in the prevalence survey were included in the study. Using the patients' national identity numbers and the prevalence survey date, we obtained data on hospitalizations, including discharge diagnoses from the National Hospital Discharge Registry (NHDR), and the dates and causes of death from the National Population Information System. We converted the prevalence of HAI into incidence using the Rhame-Sudderth formula, assessed the 28-day case fatality of the HAI patients, and then extrapolated the annual estimates of HAI burden from the total number of hospitalizations in adult acute care hospitals in 2005 (n = 804,456). We also assessed the sensitivity of the NHDR diagnoses in identifying HAIs.

Results: The estimated incidence of HAIs was 5.7% (95% confidence interval = 5.0% to 6.5%), and the 28-day case fatality was 9.8%. Thus, >8500 hospitalizations per million population annually would result in at least 1 HAI and approximately 270 HAI-associated deaths within 28 days. The sensitivity of the NHDR diagnoses was 34% (range by infection type, 0% to 67%).

Conclusion: Our disease burden estimates can be used in health care planning and resource allocation. The NHDR was not a reliable source for case finding of HAIs.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Cross Infection / epidemiology*
  • Cross Infection / mortality
  • Finland / epidemiology
  • Hospitals
  • Humans
  • Incidence
  • Prevalence