Changes in hospital admissions across Europe: 1995-2003. Results from the EuroSIDA study

To describe changes in the proportions of patients admitted to hospital and the duration of admission during the month of March between 1995 and 2003 and to describe the factors related to admission for 9802 patients from EuroSIDA, a pan-European, observational cohort study.

Generalized estimating equations were used to determine changes over time in the proportion of patients admitted and the median duration of admission. Logistic regression was used to determine factors related to admission in March 1995, March 1998 and March 2001.

The proportion of patients admitted during March declined from 7.4% in 1995 to 2.6% in 2003. After adjustment, the estimated reduction in the proportion of patients admitted was 5.5% per year [95% confidence interval (CI) 2.5-8.5%; P=0.0004], a 26% reduction. The median duration of hospital admission declined by 58% from 12 days in 1995 [interquartile range (IQR) 5-19 days] to 5 days in 2003 (IQR 3-12 days), a significant decline of 0.7 days per year after adjustment (95% CI 0.5-0.9 days; P=0.031). Patients with a lower CD4 lymphocyte count, and with an AIDS diagnosis made within the 3 months prior to March, all had increased odds of admission during March 1995, 1998 or 2001. In March 2001, patients whose treatment regimen was changed as a consequence of toxicities had increased odds of admission [odds ratio (OR) 2.34; 95% CI 1.26-4.37; P=0.0074]. In addition, patients who were hepatitis C virus-positive during March 2001 (OR 1.66; 95% CI 1.02-2.68; P=0.041) had increased odds of admission.

There has been a considerable decline in both the proportion of patients admitted to hospital and the median duration of the stay. Patients with hepatitis C had increased odds of admission, but there was little evidence of an increase in admissions among patients taking highly active antiretroviral therapy (HAART) associated with serious adverse events, although longer follow up is required.