Surveillance and control of Hospital Acquired Infections in the Netherlands: ten-year experience in an acute care hospital.

Summary
Quality of care in hospitals is impaired by adverse patient outcomes, such as hospital-acquired infections. These infections, also called nosocomial infections, occur in patients during their stay in hospital and should be prevented as much as possible. Insight in the type and magnitude of the problem is required before one is able to solve the problem. Data collections on hospital acquired infections are known from the eighteenth century onwards (Chapter 1). Specific interventions did reduce the risk on nosocomial infections. After the introduction of antisepsis, asepsis and antibiotics the interest in nosocomial infection control decreased until 1950s. At that time Staphylococcal infections began to plague hospitalized patients and renewed efforts were directed at prevention and control of nosocomial infections. It was understood that assessment of the frequency of hospital-acquired infections is a prerequisite for judging if infection control measures have been successful or need to be adjusted. The continuous collection, collation and analysis of data on hospital-acquired infections, and, subsequently, the planning, execution and evaluation of infection control policies, with dissemination to those who need to know, became known as “surveillance”. The “Study on the efficacy of infection surveillance and control programs in preventing nosocomial infections in US hospitals “(SENIC) showed that surveillance is an essential element for any infection control program to be effective.
An infection surveillance and control program was started in 1984 in the 270-bed Oudenrijn Hospital, Utrecht, the Netherlands. The ultimate goal was to reduce the risk on nosocomial infections. To estimate the magnitude of the problem a study on the incidence of all types of bacteraemia was started along with hospital-wide surveillance of all types of nosocomial infections (Chapter2). In 1984, 197 episodes of bacteraemia occurred in 147 patients admitted to two general hospitals. The sources of infection and the moments of occurrence were studied. The incidence of bacteraemia was 1.15 per 100 admissions in the larger university-affiliated hospital and 0.84 in the smaller non-teaching hospital. 48% of these bacteraemias were due to urogenital tract infections, virtually always caused by Escerichia coli and other Gram negative bacteria of the family of Enterobacteriaceae. In 20% of the bacteraemias the source of infection was an infected wound, a decubitus ulcer or an intravascular catheter. The predominant causative agents isolated in these cases were Staphylococcus aureus and Staphylococcus epidermidis. 29 of the 174 (17%) patients died, 23 (13%) of them as a consequence of sepsis. 68% of the bacteraemias could be classified as nosocomial. A continuous active surveillance system for hospital-acquired infections was felt to be of prime importance.
Consequently hospital-wide surveillance and control of hospital-acquired infections (HAI) in the Oudenrijn hospital was continued and evaluated after 10 years (Chapter 3). From 1984 onwards surveillance-based preventive actions were targeted at nosocomial urinary tract infections, postoperative wound infections and central venous catheter-related sepsis. From 1984-1993 a total number of 2,772 HAI were found among 56,410 admissions representing 611,310 patient days in 92 months of hospital-wide surveillance. The overall incidence in these 10 years was 4.9 (CI954.7-5.0) per 100 admissions and 4.5 (CI954.4-4.7) per 1,000 patient days. The incidence of HAI increased with the age of the patients. The lowest incidence was found in the age category 1-14 years: 1.1% (CI950.8-1.6) and the highest incidences were found to be 7.0% (CI955.4-8.8) in patients from 65-74 years and 10.7% (CI958.5-13.6) in patients above 75 years. Infections most frequently found were urinary tract infections (43%), followed by surgical wound infections (19%), lower respiratory infections (11%), cutaneous infections (11%) and bacteraemias (10%). Post discharge infections were not included unless the patient was readmitted because of the HAI. The incidence decreased from 7.6 per 100 admissions in 1984 to 3.6 per 100 admissions in 1993; a 53% decrease. The incidence per 1,000 patient days (days at risk) decreased from 6.1 per 1,000 patient days in 1984 to 3.7 in 1993, a 39% decrease. The decrease in incidence of urinary tract per 100 admissions was 68%, the decrease per 1,000 patient days was 60%.
The predominant type of infections differed among the services. The high incidence in urinary tract infections in gynaecology was due to catheter-related urinary tract infections in the first years of surveillance. The infection rates in gynaecology decreased over the years from 16.5 per 100 admissions (19.4 per 1,000 patient days) in 1984 to 2.2 per 100 admissions (2.7 per 1,000 pati