The prevalence of and risk factors for healthcare-associated infections in Slovenia: results of the second national survey

A cross-sectional study (one-day prevalence survey) was conducted in all Slovenian acute-care hospitals during three weeks in October 2011. The SNHPS protocol had been adapted from the ECDC point prevalence survey protocol (EU PPS Protocol) (2) to ensure comparability of our results with other European member states participating in the EU PPS, and also with the results of the first Slovenian national one-day survey of HAIs in acute-care hospitals conducted in 2001. We included patients from all acute-care wards, including one acute psychiatric ward and neonatal ICUs. We excluded patients from long-term care wards and accident and emergency departments. Patients were included, if they were admitted to the ward before or at 8:00 am and not yet discharged from the ward at the time of the survey. Neonates on maternity and paediatric wards, if born before or at 8:00 am on the day of the survey, were also included. We excluded patients undergoing same-day treatment or surgery, those seen at outpatient departments and outpatient dialysis patients. Standard information was collected for all eligible patients by trained teams led by SNHPS coordinators for data collection in hospitals. It included age, sex, admission date, date of data collection (survey date), and ward specialty. As proposed by McCabe, patients were classified according to the severity of their underlying condition into three categories: non-fatal disease (expected survival >5 years), ultimately fatal disease (expected survival 1-4 years), and rapidly fatal disease (expected death within 1 year) (3). Exposures to indwelling devices (central vascular catheter, peripheral vascular catheter, urinary catheter and intubation) on the day of the survey, surgical procedures within 30 days prior to the survey, and insertion of implants within 12 months prior to the survey were recorded. All types of HAIs were identified by reviewing all medical records available at the time of the survey and through consultations with physicians and nurses. We used European standard surveillance definitions for different types of HAIs (2). The prevalence of HAIs was defined as the number of patients with at least one HAI present on the day of the survey (signs and/or symptoms) or still receiving treatment for HAIs on the day of the survey (previously present signs and/or symptoms). In contrast to the ECDC EU PPS Protocol, we have collected the data so that we were able to distinguish between patients with HAIs present on the day of the survey and those still receiving treatment for previous HAIs on the day of the survey. The onset of the signs and/or symptoms suggestive of a HAI was on day 3 of the current admission or later (day 1 was the day of admission). Infections present at admission or occurring on day 1 or day 2, or still being treated on day 1 or day 2, were also counted as HAIs, if fulfilling the HAIs surveillance definitions and additional criteria. These were: (a) If the respective patient has been discharged from acute-care hospital less than two days before re-admission; (b) when the patient had surgical site infection and surgery within 30 days before the survey (deep or organ/space surgical site infection and implant within a year before the survey); (c) in case of a Clostridium difficile infection, when the patient had been discharged from acute-care hospital less than 28 days before the current admission; and (d) if an invasive device was placed on day 1 or day 2 and that resulted in a HAI.

Completed data collection forms were checked for possible errors, missing information and inconsistencies. SNHPS hospital coordinators were approached for clarifications and to obtain missing information. Data were double-entered using Epi Info (Epi Info, version 3.5.4, CDC, Atlanta, GA, USA). Code range and filter checks were built in. Discrepancies due to entry mistakes were checked against the information on data collection forms and corrected.

Descriptive analyses were performed using the statistical software Stata (Version 11.2, StataCorp, College Station, Texas, USA). Characteristics of patients and their exposure to invasive procedures, including surgery, were described. The proportion of patients with at least one HAI (overall prevalence) and the prevalence of different types of HAIs were computed (overall and according to ward specialties). Univariate and multivariate analyses of association of having any HAI acquired during current hospitalization with selected risk factors studied were performed. Univariate analyses were first performed using the classical method for analyses of 2×k contingency tables and then repeated using logistic regression. Maximum likelihood estimates of odds ratios (ORs) together with 95% CIs and results of likelihood ratio tests for significance were computed. Risk factors that were found to be significantly associated with any HAI (p<0.05) were fitted into a series of multivariate models adding one at a time. They were kept in the multivariate model, if they remained significantly associated with any HAI after the adjustment for other risk factors in the model, except for age (borderline significance, p=0.05), which was identified a priori as explanatory variable of interest or confounder. Maximum likelihood estimates of aORs together with 95% CIs and results of likelihood ratio tests for significance were computed for all risk factors remaining in the final model.

All 21 Slovenian acute-care hospitals participated. 5628 patients were surveyed. Over half of them (56.8%) were hospitalized in three hospitals with over 650 beds, and only 10.0% in nine hospitals with fewer than 200 beds. The number of patients, overall and according to specialties of wards for each participating hospital, is shown in Table 1.

The mean age of patients was 54.8 years (range from 0 to 102, median 60.7 years), 34.1% were less than 50 years old, 50.9% were from 50 to 79 years old, and 15.0% were aged 80 years or more. There were fewer males (47.8%) than females. Of 5602 patients (99.5% of all surveyed) who were categorized according to McCabe index, 5.1% had rapidly fatal diseases and 16.4% had an ultimately fatal disease. The average length of hospital stay from the admission to the survey day was 11.4 days (median five days). The length of stay was 1-3 days for 33.4% of patients, 4-7 days for 33.4%, 8-14 days for 16.2% and ≥15 days for 17.0% of patients. 30.5% of patients had undergone surgery since admission, 31.6% during the month before the survey, and 9.5% of patients had an implant inserted in the year before the survey. On the day of the survey, 46.8% had a peripheral vascular catheter, 7.3% a central vascular catheter, 16.2% a urinary catheter, and 3.0% were intubated. Exposures to indwelling devices were most common in ICUs. The prevalence of exposures to different indwelling devices, overall and according to the ward specialty, is shown in Table 2.

On the day of the survey, 3.8% (95% CI: 3.3%-4.4%) of patients had at least one HAI, and additional 2.6% (95% CI: 2.1%-3.0%) were still in treatment because of at least one HAI (not present on the day of the survey), corresponding to the overall prevalence of 6.4% (95% CI: 5.7%-7.0%). The prevalence of HAIs ranged from 0.0% to 8.5% in different hospitals. The prevalence was higher among patients hospitalized in large hospitals with more than 650 beds (7.3%), and lower among those hospitalized in small hospitals with less than 200 beds (3.7%). Overall, the prevalence of urinary tract infections was the highest (1.4%), followed by pneumoniae (1.3%) and surgical site infections (1.2%). Excluding HAIs that were not present on the day of the survey, but for which patients were still being treated, the corresponding prevalence estimates were 0.7%, 0.8% and 0.8%. The proportion of patients with at least one HAI or still treated for at least one HAI was the highest in ICUs (35.7%), followed by surgical wards (6.4%), general medical (5.9%), paediatric (2.4%), and gynaecology and obstetrics wards (2.2%). Excluding HAIs that were no longer present on the day of the survey, but for which patients were still being treated, the corresponding prevalence estimates were 26.6%, 3.7%, 3.1%, 2.1%, and 1.3%. The numbers of patients with different types of HAIs and their respective prevalence, overall and according to different ward specialties, are shown in Table 3.

396 episodes of HAIs occurred in 358 patients (322 had one, 34 two and 2 had three episodes). 335 episodes (84.6%) started during current hospitalization, of which all were attributed to current hospitalization. 15.4% of all HAIs were present at admission, of which 59.0% were associated with a previous stay in the same hospital, and 23.0% were surgical site infections. The median duration of hospital stay until the onset of HAIs acquired during current hospitalization was 10 days (mean 15.6 days).

Among 396 episodes of HAIs, urinary tract infections were most common (19.4%), followed by pneumoniae (18.7%) and surgical site infections (16.7%). The mean number of HAI episodes per infected patient was 1.1. There were 7.0 episodes of HAIs per 100 patients. The corresponding ratio was the highest in ICUs (42.5/100), followed by surgery (7.1/100), general medicine (6.5/100), paediatrics (2.4/100), and gynaecology and obstetrics (2.2/100).

The prevalence of HAIs acquired during current hospitalization, according to patient characteristics, ward specialties, exposure to extrinsic risks and the length of hospital stay is shown in Table 4, along with the results of univariate analysis of association.

The results of multivariate analysis of association of different risk factors with any HAI acquired during current hospitalization, together with the relevant results of univariate analyses, are shown in Table 5. Patients aged 80 years or more had 2.2 times higher odds for any HAI than those less than 50 years old (adjusted OR (aOR) 2.2; 95% CI: 1.4–3.4). In comparison to patients without fatal diseases, those with rapidly fatal diseases had 2.1 times higher odds for any HAI (aOR 2.1; 95% CI: 1.4–3.3). The presence of at least one HAI was further independently associated with surgery during current hospitalization (aOR 1.9; 95% CI: 1.5-2.5), the presence of a central vascular catheter (aOR 4.0; 95% CI: 2.9–5.7), the presence of a peripheral vascular catheter (aOR 2.0; 95% CI: 1.5–2.6), the presence of intubation (aOR 2.3; 95% CI: 1.4–3.5), and the presence of a urinary catheter (aOR 1.8; 95% CI: 1.3–2.4).