Residential aged-care facility (RACF) residents are at risk of developing healthcare-associated infections (HAIs) for many reasons, including advanced age, comorbidities, functional disabilities, immunosuppression, and use of invasive devices.Reference Kirk, Hall, Veitch and Becker 1 – Reference Stevenson, Moore, Colwell and Sleeper 4 In 1991, to standardize the identification of HAIs within RACFs, McGeer et alReference McGeer, Campbell, Emori, Hierholzer, Jackson and Nicolle 5 published surveillance definitions specific to these facilities.Reference McGeer, Campbell, Emori, Hierholzer, Jackson and Nicolle 5 These were revised in 2012,Reference Stone, Ashraf, Calder, Crnich, Crossley and Drinka 6 including new requirements for clinical and microbiologic criteria to fulfill the case definition for urinary tract infection (UTI).
To estimate the prevalence of HAIs and antibiotic use within long-term aged-care facilities, the European Center for Disease Prevention and Control conducted the first European point prevalence survey (PPS) in 2010. 7 In 2011, an equivalent survey was performed in rural Australian RACFs in the state of Victoria. This survey was repeated in 2012 and 2013, and revised McGeer definitions were applied in 2013.
In Australian RACFs, urinary tract, respiratory tract, skin, soft-tissue, and mucosal infections have been shown to comprise the majority of HAIs. Limited use of microbiology testing has also been demonstrated, possibly related to difficulty obtaining clinical specimens from confused or debilitated residents, the clinical significance of culture results being difficult to interpret, insufficient access to diagnostic laboratories, and/or costs.Reference Stuart, Wilson, Bellaard-Smith, Brown, Wright and Vandergraaf 8 The impact of using the revised McGeer definition for UTI upon HAI surveillance in Australia has not been evaluated.
The objectives of this 2014 PPS were (1) to estimate the prevalence of HAIs, (2) to review the utility of microbiology testing for HAIs, and (3) to evaluate the impact of a revised case-definition for UTI upon prevalence estimates in Victorian RACFs.
METHODS
Study Population
The Rural Infection Control Practice Group (RICPRAC) represents Victorian infection control practitioners (ICP) employed in rural public health services, including 146 RACFs operated by state government. All members were invited to participate in the current study. Although not actively recruited, participation by ICPs working in the metropolitan 28 RACFs operated by state government was also permitted. Eligible residents were those who resided full-time in the RACF for ≥48 hours and were present on the survey day. In Victoria, RACFs provide continuous supported care ranging from assistance with personal care and daily tasks to 24-hour nursing care.
Survey Tool and Data Collection
Using resident medical records, data were collected by ICPs on a single day between July 1 and August 31, 2014. Prior to data collection, ICPs were educated regarding uniform methodology, and a study coordinator was available to assist (via telephone) on the survey day. A facility survey was completed for each participating RACF, including the total number, age, and sex of residents. A resident survey was completed for residents identified by ICPs or RACF clinical staff and confirmed by ICPs as having an HAI on the survey date.
Definitions
Presumed and confirmed UTIs were classified according to 3 case definitions: (1) revised McGeer clinical and microbiologic criteria met, (2) revised McGeer clinical (not microbiologic) criteria met, and (3) revised McGeer microbiologic (not clinical) criteria met.
Bloodstream infections, gastrointestinal tract infections, respiratory tract infections (RTIs), skin, soft-tissue, or mucosal infections, as well as unexplained fever were defined according to the revised McGeer definitions.Reference Stone, Ashraf, Calder, Crnich, Crossley and Drinka 6 HAIs were reported if not already present at the time of admission to the RACF. The prevalence of HAIs was calculated by dividing the number of residents having ≥1 HAI by the total number of surveyed residents.
RESULTS
Participating Facilities
Overall, 3,741 residents in 112 RACFs (99 rural, 13 metropolitan) participated in the PPS, corresponding to 64.4% of all state-government–operated RACFs in Victoria. The median number of eligible residents at participating RACFs was 28.5 (range, 7–110), the median proportion of residents aged >85 years was 52.3% (range, 0–89.3%), and the median proportion of male residents was 35.1% (range, 3.6–79.3%).
Hospital-Associated Infections
Using the revised McGeer definition, 109 residents had at least 1 HAI, corresponding to a crude prevalence of 2.9% (95% confidence interval [CI] 2.4%–3.5%). In total, 111 HAIs were identified, with 2 residents having 2 infections. The most frequently reported HAIs were skin, soft-tissue, or mucosal infections (44.1%) and RTIs (38.7%) (Table 1). Of the 109 residents with an infection, 32 (29.4%) had a clinical specimen collected for microbiological testing.
TABLE 1 Healthcare-Associated Infections in Victorian RACF Residents (2014)
a All infections were defined according to revised McGeer criteria.Reference Stone, Ashraf, Calder, Crnich, Crossley and Drinka 6
b Two residents had two infections (109 residents with 111 HAIs).
Urinary Tract Infections
In total, 14, 26, and 5 residents met case definitions A, B, and C, respectively. The majority of these residents (97.8%) were prescribed antibiotic therapy for a UTI (Table 2). Case definition A accounted for 31.1% of all UTIs. Of those that fulfilled the clinical criteria alone (case definition B), 53.9% (14 of 26) did not undergo microbiological testing, 19.2% (5 of 26) were tested but organism(s) cultured did not meet the microbiologic criteria threshold, 15.4% (4 of 26) were tested and cultured negative, and 11.5% (3 of 26) were tested but results were unavailable. As a composite clinical measure, combining case definitions A and B resulted in an estimated crude prevalence of 3.6% (135 of 3,741; 95% CI, 3.0%–4.3%).
TABLE 2 Comparison of Urinary Tract Infections (UTI) Identified by 3 Case Definitions
DISCUSSION
Our study is the largest in Australia to apply standardized methodology to estimate HAI prevalence in RACFs. This study supports the national agenda concerning monitoring of antimicrobial use and resistance to improve safety and clinical outcomes in the aged-care setting. 9 HAI prevalence was 2.9%, which is lower than but still comparable to estimates provided in 2013 by Victorian and international studies: 3.3% (unpublished data) and 3.4%, 10 respectively. Using revised McGeer definitions, skin, soft-tissue, and mucosal infections and RTIs were most frequently identified.
The estimated prevalence of UTIs is influenced significantly by the surveillance case definition employed. In particular, we observed that a large proportion of presumed UTIs (68.9%) did not meet the revised McGeer definition, mostly the microbiologic criteria (57.8%). If the objective of surveillance is to monitor the burden of illness, including impact upon resources and antimicrobial prescribing, a broader case definition would be more appropriate to capture cases of confirmed and presumed UTI.
We have demonstrated the feasibility of an annual PPS tool in Australian RACFs using internationally accepted methodology. Given the need for enhanced monitoring of HAIs in non-acute healthcare settings, we propose that this tool could provide quality indicators for Australian RACFs. Historically, HAIs have been regarded as important performance indicators within acute-care facilities, 11 and a similar framework could be adopted within RACFs. Standardized reporting of HAIs allows identification of potential processes for policy development and formulation of targeted clinical guidelines. Furthermore, standardized reporting would facilitate comparison of baseline and post-intervention outcomes for individual RACFs implementing prevention programs.
The strengths of our study include the application of internationally accepted HAI definitions to facilitate international comparisons and longitudinal evaluation by participating RACFs.
Some limitations of our study include the potential seasonality of some HAIs. The PPS was performed during winter months, which may have particularly influenced RTI prevalence. Although a large number of public RACFs participated, this study did not include non–government-operated RACFs and therefore may not be generalizable to other RACF types. Given that not all eligible facilities participated in the study, data may be influenced by selection bias. Data quality was dependent upon documentation in resident records. We acknowledge that incomplete and/or missing clinical details may contribute to the underestimation of HAIs.
Given the low uptake of microbiologic testing in Australian RACFs, it may be beneficial for future PPSs to measure the utility of microbiologic testing and to incorporate the assessment of clinically significant HAIs to ensure that UTI prevalence is not underestimated. Antibiotic prescribing practices for UTIs could also be used to apply a more sensitive case definition, and therefore an estimated prevalence that is more closely aligned with clinical practice.
ACKNOWLEDGMENTS
First, we wish to thank the Rural Infection Control Practice Group and the Victorian Healthcare Associated Infection Surveillance System Coordinating Center who coordinated the 2014 Point Prevalence Study. Additionally, we wish to thank the infection control practitioners and clinical staff employed at the participating Victorian RACFs who collected and submitted the data.
Financial support: The VICNISS Coordinating Centre is funded by the Victorian Department of Health.
Potential conflicts of interest: All authors report no conflict of interest or financial disclosures.
