The establishment of a statewide surveillance program for hospital-acquired infections in large Victorian public hospitals: a report from the VICNISS Coordinating Centre.

BACKGROUND: A 1998 survey of acute Victorian public hospitals (VPH) revealed that surveillance of hospital-acquired infections (HAI) was underdeveloped, definitions and methodology varied considerably, and results disseminated inconsistently. The survey identified the need for an effective surveillance system for HAI. OBJECTIVE: To develop and support a standardized surveillance program for HAIs in large acute VPH and to provide risk-adjusted, procedure-specific, HAI rates. METHODS: In 2002, the independent Victorian Nosocomial Infection Surveillance System (VICNISS) Coordinating Centre (VCC) was established to develop and support the standardized surveillance program. A multidisciplinary team was recruited. A communication strategy, surveillance manual, user groups, and Web site were developed. Formal education sessions were provided to participating infection control nurse consultants (ICCs). Surveillance activities were based on the US Centers for Diseases Control and Prevention's National Nosocomial Infection Surveillance System (NNIS) surgical site infection and intensive care unit (ICU) components. NNIS methods were modified to suit local needs. Data collection was paper based or through existing hospital software. An advisory committee of key stakeholders met every second month. RESULTS: The surveillance program was rolled out over 12 months to all 28 large adult VPH. Data on over 20,000 surgical procedures performed at participating sites between November 11, 2002, and December 31, 2004, were submitted. Thirteen hospitals contributed to the ICU surveillance activities. Following aggregation and analysis by the VCC, hospital- and state-level results were posted on the Web page for hospitals to review. CONCLUSION: A standardized approach for surveillance of HAI was established in a short time frame in over 28 VPH. VICNISS is a tool that will continue to provide participating hospitals with a basis for continuous quality improvement.

Infections after coronary artery bypass graft surgery in Victorian hospitals--VICNISS Hospital Acquired Infection Surveillance.

OBJECTIVE: To establish a surveillance program reporting surgical site infection rates after coronary artery bypass graft surgery (CABGS) in Victorian public hospitals. METHODS: The VICNISS Coordinating Centre was established in 2002 to implement and co-ordinate a standardised surveillance system for hospital-acquired infections in acute care Victorian public hospitals. Using validated definitions and methodology from the Centers for Disease Control and Prevention's National Nosocomial Infection Surveillance (NNIS) program, data on risk-adjusted surgical site infection (SSI) rates were collected and submitted to the Coordinating Centre for collation and reporting. RESULTS: Six large Melbourne metropolitan hospitals contributed data for CABGS for the period 11 November 2002 to 30 June 2004, comprising a total of 3,482 patient records. Of 3,398 complete records, the aggregate SSI rates per 100 procedures for NNIS risk category 1 and 2 were 4.4 (95% Cl 3.7-5.3) and 6.0 (95% Cl 4.5-7.8) respectively. The deep sternal SSI rates were 0.6 (95% Cl 0.4-1.3) and 0.5 (95% Cl 0.5-2.4 for patients in risk category 1 and 2 respectively. The most common pathogen identified was Staphylococcus aureus. CONCLUSION: This early data from VICNISS demonstrates similar CABGS SSI rates to those reported by NNIS in the USA, but higher than reported by the German Nosocomial Infection Surveillance System. IMPLICATIONS: The adoption of a statewide, co-ordinated surveillance program using validated internationally accepted methodologies allows hospitals to benchmark their infection rates against aggregate local and international data and to examine infection prevention interventions.

Rotavirus detection and characterisation in outbreaks of gastroenteritis in aged-care facilities.

BACKGROUND: Although rotavirus is a major cause of gastroenteritis in children, its role in adult gastroenteritis and the sensitivity of different methods for its detection in specimens collected from adults are less well understood. OBJECTIVES: (1) To examine the frequency and seasonality of rotavirus-associated gastroenteritis outbreaks in aged-care facilities in Victoria, Australia. (2) To determine rotavirus type in these outbreaks. (3) To determine whether other enteropathogenic agents are present in specimens from these outbreaks. (4) To examine the sensitivity of different methods (electron microscopy (EM), reverse transcription-polymerase chain reaction (RT-PCR), enzyme immunoassay (EIA) and latex agglutination (LA)) for the detection of rotavirus in specimens from adults. STUDY DESIGN: Specimens from gastroenteritis outbreaks in aged-care facilities forwarded to this laboratory for the years 1997-2000 were tested for enteropathogenic agents by various methods. Epidemiological, clinical and seasonal data from the rotavirus-positive outbreaks were analysed. RESULTS: Rotavirus was detected by EM in 18 out of 29 individuals associated with seven out of 53 (13%) gastroenteritis outbreaks in aged-care facilities; norovirus was detected in 22 outbreaks (42%) and astrovirus in one outbreak (2%). No mixed viral infection was found in any outbreak. All rotaviruses were typed as Group A by RT-PCR. The rotaviruses in the seven outbreaks were G-typed as follows: G2 (three outbreaks), G4 (two outbreaks), G1 (one outbreak) and G9 (one outbreak). The rotavirus-associated outbreaks were concentrated in mid-winter to mid-spring. The relative sensitivities of the Group A rotavirus detection methods (for the 29 specimens tested) were EM (18), first-round RT-PCR (11), second-round PCR (19), EIA-visual (19), EIA-photometric (19) and LA (13). CONCLUSIONS: In Victoria, Australia, outbreaks of gastroenteritis associated with rotavirus are quite common in aged-care facilities. They involve Group A rotavirus and have a winter/spring seasonality. G-types G1, G2, G4 and G9 were all detected. EIA, second-round PCR and EM proved sensitive methods for rotavirus detection whereas first-round RT-PCR and LA did not.