[Eleven years of core data set in intensive care medicine. Severity of disease and workload are increasing]

Abstract

Background: In the year 2000 a working group of the German Interdisciplinary Association for Intensive Care Medicine (DIVI) defined a core data set on quality assurance for the first time. In the following years the participating intensive care units sent data to the registry on a voluntary basis and received an annual report on benchmarking data. Alterations in the quality in the field of intensive care medicine have so far only been published to a very low extent.

Aim: This study analyzed the core date set of the DIVI between 2000 and 2010 in respect to changes in disease severity using the simplified acute physiology score (SAPS II), the sequential organ failure assessment (SOFA), the need for therapeutic interventions with the therapeutic intervention scoring system (TISS 28) and intensive care unit (ICU) mortality.

Material and methods: Inclusion criteria were participation in the registry for at least 4 years, SAPS II, SOFA, TISS28 scores available and data on ICU discharge. A standardized mortality rate (SMR) was calculated for each year.

Results: The mean SAPS II score including 94,398 patients increased by 0.23 points/year with a standard error (SE) of 0.02 to 26.9 ± 12 points (p < 0.001). Similarly, the SOFA score on admission to the ICU increased by 0.14 points/year (SE 0.04) to 3.4 ± 2.7 points (p < 0.001), the proportion of patients with a two organ failure doubled to 7.1 % and the number of patients dependent on ventilation increased by 13.6 % to 59.8 %. The mean time on ventilation increased by 0.17 ventilator days/year (SE 0.01, p < 0.001) to 3.1 ± 7.5 days/patient. The mean number of therapeutic interventions increased by 8.7 % to 26.3 ± 8.3 TISS 28 points/day. The mean length of stay on the ICU (4.3 ± 8 days) and the age of the patients (63.2 ± 17.0 years) remained unchanged. The readmission rate showed no significant changes between the years 2004 and 2010. The readmission rate to the ICU within 48 h after primary discharge was 3.1 % with a 95 % confidence interval (CI) of 3.0-3.3 in contrast to 1.5 % (95 % CI 1.4-1.6) for readmission to the ICU after 48 h. The length of stay in hospital before admission to the ICU decreased for patients with scheduled surgery (6.3 ± 9.7 days vs. 4.2 ± 6.9 days), increased slightly for patients with medically indicated admission to the ICU (2.4 ± 8.2 days 3.1 ± 8.6 days) and remained unchanged for patients with unscheduled admission to the ICU after surgery (4.1 ± 8.6 days). The SMR decreased between 2000 and 2004 from 0.97 to 0.72 and increased again thereafter to 0.99 (ICU mortality 8.5 %).

Conclusion: The severity of disease on admission to the ICU, the proportion of patients on ventilation and the workload of therapeutic interventions increased between 2000 and 2010 in German ICUs but the length of stay of patients in the ICU remained unchanged. The SMR decreased until 2005 and increased thereafter to return to the initial values. The overall ICU mortality was low compared to international data.