Introduction
Healthcare-associated infections (HAIs) and antimicrobial resistance pose an increasing threat to global health. Reference Cassini, Högberg and Plachouras1,2 Point prevalence surveys are a key surveillance tool that can inform hospital infection prevention and control (IPAC) programs on the overall burden and trends of HAIs and antimicrobial-resistant organisms (AROs) and influence associated IPAC mitigation practices. They are a useful low-cost alternative to active surveillance and provide valuable benchmarking data across healthcare settings and jurisdictions. Reference Llata, Gaynes and Fridkin3
Several countries 4–Reference Grae, Singh and Jowitt9 perform repeated point prevalence surveys to compliment prospective data and describe evolving IPAC practices. While direct comparison of results across countries is challenging due to the variability of survey methodology and differences in healthcare systems, Reference Magill, Berg, Brooks and Jarvis10 common findings have been identified. These include a higher prevalence of HAIs among patients in intensive care, high utilization of invasive devices and device-associated infections as well as the emergence of resistant pathogens, highlighting the need for vigilant and improved infection control strategies.
The Canadian Nosocomial Infection Surveillance Program (CNISP) previously performed three point prevalence surveys of HAIs, AROs, and antimicrobial use (AMU). The prevalence of patients with at least one HAI increased from 9.9% (2002) to 11.3% (2009) then declined to 7.9% (2017). Reference Mitchell, Taylor and Rudnick11 Results showed that one in 12 patients (and one in eight patients admitted to intensive care) had a HAI and that AROs were responsible for 9% of these infections. Reference Johnstone, Garber and Muller12
In 2024, CNISP conducted a fourth national point prevalence survey. We provide an updated estimate of the prevalence of HAIs, AROs, and isolation burden among adult inpatients in Canadian acute care hospitals following the COVID-19 pandemic and identify trends across the four surveys.
Methods
Study design
Point prevalence surveys were conducted in CNISP acute-care hospitals in 2002, 2009, 2017, and 2024. Survey methodology has been published elsewhere. Reference Mitchell, Taylor and Rudnick11,13
Study population
CNISP is a collaboration between the Public Health Agency of Canada, the National Microbiology Laboratory, the Association of Medical Microbiology and Infectious Disease Canada, and sentinel hospitals across Canada that conduct standardized surveillance of HAIs and AROs. 14
Patients of any age who were admitted to a participating CNISP hospital for 48 hours or longer were eligible for inclusion, including admitted patients in the emergency department. Patients who had been admitted for less than 48 hours but were admitted within the last month to the survey hospital were also included. We excluded patients admitted to long-term care, maternity, mental health, day surgery, or rehabilitation units.
Definitions
National Healthcare Safety Network (NHSN) HAI definitions were used for pneumonia, urinary tract infection (UTI), surgical site infection (SSI), and viral gastroenteritis. 15 Bloodstream infection (BSI), Clostridioides difficile infection (CDI), and viral respiratory infection (VRI) were defined using CNISP definitions. 16,17
We considered a HAI to be present if the patient was symptomatic of, or was receiving antimicrobial therapy for, an infection on the day of the survey and the onset of symptoms was on day 3 (≥48 hours) or later of hospital admission. For healthcare-associated (HA) VRI and CDI, symptom onset was on day 4 (≥72 hours) or later of hospital admission. If a SSI was present on the day of the survey or the patient was being treated for a SSI on the day of the survey, and SSI definition criteria were met, this was captured as a HAI.
Training, data collection, and validation
Survey methodology and standardized forms were developed and piloted by an expert working group in 2002 and updated accordingly for each survey year. The surveys were conducted in February or March of each survey year to limit the influence of seasonal variation in HAIs and to permit comparison between surveys. Training on the survey methodology was provided to IPAC leads within each CNISP hospital. IPAC leads trained their frontline IPAC staff on definitions, data collection, and data entry procedures. IPAC staff reviewed the medical records of eligible patients (patients with or without a HAI) and data were submitted electronically through a secure online web-based platform or by a standardized Excel template.
We collected de-identified patient-level data, including demographics, additional precautions, and the presence of HAI, ARO, and AMU. AMU data will be reported separately. In 2024, a subset of hospitals (n = 32) collected data on the utilization of invasive devices. Data variables collected in each survey are outlined in Supplementary Table 1. CNISP collects hospital-level data (e.g., number of beds and specialized services provided) annually using a standardized hospital profile form. We extracted hospital profile data for CNISP hospitals that participated in the four surveys and included these data in the analyses.
CNISP epidemiologists evaluated the data for quality and completeness and followed up with hospitals as needed. To assess the level of agreement between primary and validation data collectors, a validation survey was piloted among 472 patients in 41 hospitals. Validation methods are further described 18 and full results will be reported separately.
Statistical analysis
Analyses were conducted using R statistical software version 4.3.2. Reference Core Team19 We compared the characteristics of participating hospitals and patients who were surveyed, the prevalence of HAIs and AROs causing infection using χ 2 tests, Fisher’s exact tests for categorical variables, or Kruskal–Wallis tests for continuous variables. The Mann-Kendall test was used to test trends. Agreement between the data collections was analyzed using kappa (κ) statistics (0.81–1.00 very good, 0.61–0.80 good, 0.41–0.60 moderate, 0.21–0.40 fair/marginal, and <0.2 poor). We considered a 2-sided P value ≤ 0.05 to be statistically significant.
Analyses were restricted to adult patients (≥18 years). We calculated the prevalence of HAIs as the percentage of patients with at least one HAI on the survey day over the total number of patients surveyed and 95% confidence intervals (CIs) were calculated. We previously reported that characteristics of the participating hospitals were similar over the 2002, 2009, and 2017 surveys. Reference Mitchell, Taylor and Rudnick11 To account for changes in the hospital characteristics between the 2017 and 2024 surveys, we conducted a sensitivity analysis restricted to the 39 hospitals that participated in both of these surveys.
Ethical considerations
These surveys were either considered exempt as quality assurance projects or approved by the research ethics boards at participating hospitals.
Results
Hospital and patient characteristics
In 2024, 62 of 109 CNISP hospitals participated in the point prevalence survey (response rate 56.9%), which included 10,473 eligible adult patients.
The characteristics of the participating hospitals were compared across the four surveys (Supplementary Table 2). Compared to previous surveys, in 2024 we observed a decrease in the median bed size, the proportion of teaching hospitals, and the proportion of hospitals which provide the following specialized services: intensive care, burn, and solid organ transplant. This change in hospital characteristics is primarily reflective of the addition of small, community hospitals in recent years to improve the representativeness of the CNISP network.
The characteristics of adult patients surveyed were compared across all four surveys (Supplementary Table 3). A difference in patient age, sex, and location in hospital were reported, though the magnitude of these differences was small from a clinical perspective.
Prevalence of HAIs
In 2024, among 10,473 patients surveyed, 844 (8.1%) were reported to have at least one HAI, with a total of 913 HAIs reported. Of those, 779 (7.4%) patients had one HAI, 61 (0.6%) had two HAIs, and 4 (<0.1%) had three or more HAIs.
The prevalence of patients with at least one HAI increased from 10.4% (95% CI: 9.6%–11.2%) in 2002 to 12.4% (95% CI: 11.7%–13.2%) in 2009 followed by a decline to 8.4% (95% CI: 7.8%–9.0%) in 2017. HAI prevalence remained stable in 2024 at 8.1% (95% CI: 7.6%–8.6%). After excluding COVID-19 infections in 2024, the HAI prevalence remained similar (7.8%, 95% CI: 7.3%–8.3%). In an analysis restricted to the 39 hospitals which participated in both the 2017 and 2024 surveys, the prevalence of infection remained stable between the 2017 (8.3%, 95% CI: 7.6%–8.9%) and 2024 (8.0%, 95% CI: 7.4%–8.6%) surveys. The validation results found a 93.2% agreement (κ = 0.62) between the primary and validation results for the presence of a HAI. The level of agreement ranged from κ = 0.58 for HA-pneumonia to κ = 0.72 for HA-CDI.
From 2002 to 2024, the prevalence of patients with a SSI or UTI declined. The prevalence of patients with pneumonia or CDI decreased from 2009 to 2017 and remained stable in 2024. The prevalence of patients with a primary BSI significantly declined from 1.7% (95% CI: 1.4%–2.0%) in 2009 to 1.0% (95% CI: 0.8%–1.2%) in 2017 then significantly increased to 1.5% (95% CI: 1.3%–1.8%) in 2024 (p = 0.002). The prevalence of patients with a VRI while low, significantly increased from 0.3% (95% CI: 0.2%–0.4%) in 2017 to 0.6% (95% CI: 0.5%–0.8%) in 2024 (p < 0.001) (Figure 1, Supplementary Table 4).
Figure 1. Prevalence of healthcare-associated infection types among adult patients surveyed in 2002, 2009, 2017, and 2024. Note: BSI = bloodstream infection, CDI = Clostridioides difficile infection, SSI = surgical site infection, UTI = urinary tract infection, VRI = viral respiratory infection.
For all four surveys combined, the prevalence of HAIs was significantly higher in patients admitted to ICU, where 24.7% (95% CI: 22.9%–26.7%) had at least one HAI compared with 8.6% (95% CI: 8.2%–8.9%) in all other units combined (p < 0.001). We observed a decline in the prevalence of HAIs in patients in the ICU from 35.5% (95% CI: 30.1%–41.3%) in 2002 to 29.6% (95% CI: 25.6%–34.0%) in 2009, 20.2% (95% CI: 17.1%–23.8%) in 2017, then stabilization in 2024 (20.6%, 95% CI: 17.7%–23.8%). The prevalence of ICU patients with a UTI, SSI, and pneumonia significantly declined over time. The prevalence of ICU patients with CDI remained unchanged. The prevalence of ICU patients with a BSI declined from 2002 to 2017 followed by a significant increase in 2024 (Supplementary Table 4).
Prevalence of invasive devices and device-associated infections
Among the subset of 32 hospitals which collected data on the use of invasive devices in 2024, 38.0% (2,526/6,644) of patients surveyed had at least one invasive device on the survey day. Indwelling urinary catheter (18.1%, n = 1204) and central venous catheter were the most common (17.7%, n = 1176) followed by inserted tubes and drains (15.0%, n = 998) and endotracheal intubation (4.4%, n = 295). All of the above invasive devices were more common among patients in ICU compared to patients in coronary care, medical, and surgical units (Supplementary Table 5).
Among patients surveyed, the prevalence of catheter-associated infection, ventilator-associated pneumonia, and SSI associated with a prosthetic implant remained similar between the 2017 and 2024 surveys. However, the prevalence of central line-associated BSIs while low, nearly doubled between the 2017 (0.4%) and 2024 (0.7%, p = 0.02) surveys (Supplementary Table 4).
Prevalence of additional precautions
The prevalence of patients on additional precautions significantly increased from 2017 (15.4%, 95% CI: 14.7%–16.2%) to 2024 (21.7%, 95% CI: 20.9%–22.5%, p < 0.001). This increase was mainly driven by an increase in contact precautions from 14.0% (95% CI: 13.3%–14.7%) in 2017 to 20.2% (95% CI: 19.5%–21.0%) in 2024 (p < 0.001) and to a lesser extent an increase in the proportion of patients on droplet/contact precautions (2.7%, 95% CI: 2.4%–3.1% to 4.3%, 95% CI: 3.9%–4.7%, p < 0.001).
Among patients surveyed in 2024, the most common reason for additional precautions was methicillin-resistant Staphylococcus aureus (MRSA) (6.6%, 95% CI: 6.2%–7.1%), followed by VRI (5.6%, 95% CI: 5.1%–6.0%) of which 2.4% (95% CI: 2.1%–2.7%) were COVID-19, vancomycin-resistant Enterococci (VRE) (3.2%, 95% CI: 2.9%–3.6%), CDI (1.9%, 95% CI: 1.7%–2.2%), and carbapenemase-producing organisms (CPO) (1.7%, 95% CI: 1.5%–2.0%). In addition, based on risk assessment upon admission three patients were isolated for Candida auris in 2024. Between 2017 and 2024, we observed a significant increase in the prevalence of patients isolated for CPO (0.1% to 1.7%, p < 0.001) and VRI (2.1% to 5.6%, p < 0.001) (Figure 2).
Fig. 2. Prevalence of adult patients on additional precautions, 2002, 2009, 2017, and 2024. Note: MRSA = methicillin-resistant S. aureus, VRI = viral respiratory infection, VRE = vancomycin-resistant Enterococci, CDI = Clostridioides difficile infection, CPO = carbapenamase-producing organisms, TB = Tuberculosis.
Prevalence of antimicrobial-resistant organisms
Table 1 summarizes select AROs that cause HAIs. Overall, AROs remain an uncommon cause of HAIs across all survey years. The most common resistant organism was MRSA which was present in 5.8% of pneumonia infections, 5.6% of BSIs, 4.0% of SSIs, and 1.2% of UTIs. Carbapenemase-producing Enterobacterales (CPE) were identified in only five infections in the 2017 and 2024 surveys.
Table 1. Selected antimicrobial-resistant organisms causing healthcare-associated infections in 2002, 2009, 2017, and 2024
MRSA = methicillin-resistant S. aureus, VRE = vancomycin-resistant Enterococci, CPE = carbapenamase-producing Enterobacterales, ESBL = extended-spectrum β-lactamase–producers, SSI = surgical site infections, BSI = bloodstream infections, UTI = urinary tract infection, n/a = not available (data not collected). * The Mann-Kendall test was used to test trends.
The prevalence of infections with extended-spectrum β-lactamase–producers significantly increased between 2002 (0.4%) and 2024 (2.6%) (p = 0.001) and were most common in patients with UTIs and BSIs. While VRE infrequently caused a HAI, the prevalence of infections associated with VRE significantly increased from 0.3% in 2002 to 1.7% in 2024 (p = 0.02).
Microbiology
In the 2024 survey, a total of 747 organisms were reported for 81.8% (913) of HAIs, mostly gram-negative organisms (48.3%), followed by gram-positive organisms (37.2%), viruses (8.7%), and fungi (5.9%). The most frequently isolated organisms for HAIs were E. coli (13.8%) and Enterococcus spp. (12.3%) both of which were most commonly identified in UTIs. S. aureus was a frequent cause of BSIs, SSIs, and pneumonia. Klebsiella spp. was most commonly identified in UTIs and Pseudomonas spp. was a common cause of pneumonia and SSIs (Table 2). An important emerging organism, SARS-CoV-2, accounted for 3.1% of all HAIs and nearly half of all VRIs. Candida auris was not identified in the 2024 survey.
Table 2. Microorganisms isolated in HAIs by infection type, 2024
Other microorganisms include: Coagulase-negative staphylococci, Stenotrophomonas maltophilia, H. influenzae, norovirus, rhinovirus, influenza, respiratory syncytial virus, adenovirus, Serratia spp., Citrobacter spp., Acinetobacter spp. etc.
With the exception of SARS-CoV-2, the distribution of the most frequently identified microorganisms was similar to the findings of the previous surveys (Supplementary Table 6).
In the analysis restricted to the 39 hospitals that participated in both the 2017 and 2024 surveys, we found that the prevalence of device-associated infections, additional precautions, AROs, and organisms were similar compared to the primary analysis which included all hospitals surveyed.
Discussion
We report the prevalence of HAIs, AROs, and isolation burden among adult patients in a network of Canadian acute-care hospitals based on findings from four repeated point prevalence surveys in 2002, 2009, 2017, and 2024. The prevalence of HAIs among patients remained stable between 2017 (8.4%) and 2024 (8.1%), despite 3.1% of HAIs being due to SARS-CoV-2 in 2024. Among infection types surveyed, the prevalence of UTIs and SSIs declined over time. However prior success in a reduction of the prevalence of pneumonia and CDI were not sustained and an increase in the prevalence of BSIs and VRIs were observed.
While we did not measure the impact of the COVID-19 pandemic on the prevalence of HAIs and AROs in our 2024 survey, a time-series analysis conducted among CNISP hospitals identified that the COVID-19 pandemic was associated with surveillance trends similar to those reported in our surveys. Reference Ioannou, Astrinaki and Vitsaxaki21 Furthermore, findings from our 2024 survey were similar to results from other surveys performed either during or post the COVID-19 pandemic. In 2021, New Zealand reported a HAI prevalence of 6.6% among adult patients.Reference Grae, Singh and Jowitt 9 In 2022, the prevalence of HAIs among all patients in GreeceReference Ioannou, Astrinaki and Vitsaxaki 21 and ItalyReference Barbadoro, Dolcini and Fortunato 22 was 10.6% and 7.4%, respectively. The ECDC reported a HAI prevalence of 8.0% in their 2022–2023 survey among 31 countries 23 and England reported a prevalence of infection of 7.6% among patients of all ages in 2023. 24 The variation in international prevalence may reflect a real difference in the prevalence of HAIs or differences in healthcare systems (e.g., patient severity and comorbidities and hospital characteristics), methodology (e.g., hospital types, inclusion criteria, and case definitions), or the timing of surveys with respect to the COVID-19 pandemic.
Overall, AROs remain an uncommon cause of HAIs across Canada. We collected data on CPE in the 2017 and 2024 surveys and identified only five infections. However, the prevalence of patients on additional precautions for CPO significantly increased between 2017 (0.1%) and 2024 (1.7%). Furthermore, national surveillance data indicates that while the incidence of CPE remains low a recent exponential increase has been identified.Reference Mitchell, Mataseje and Cayen 25 These findings all serve as a warning of the impact of increasing rates of CPE in Canadian hospitals. Nearly half of HAIs reported in 2024 were associated with gram-negative organisms from which the greatest threat of antimicrobial resistance currently exists.
There were no cases of HA C. auris infection reported among patients surveyed in 2024 despite the global spread of this antimicrobial resistance threat. Three patients were isolated for C. auris based on risk factors at admission; with the variability of screening practices across Canada it is possible that some colonized patients were not identified.Reference Tan, Bharat, McGill and Mitchell 26 We identified a significant increase in the prevalence of patients on additional precautions for VRIs (1.7% in 2017 to 5.6% in 2024). This reflects the high seasonal burden of respiratory viruses on Canadian hospitals.
There was no significant reduction in the prevalence of infection among patients in the ICU from 2017 (20.2%) to 2024 (20.6%); however, the burden remains high as one in five ICU patients developed a HAI in 2024. These findings are consistent with other countries which have recently reported a high prevalence of infection among patients in the ICU, ranging from 20.5% to 23%. Reference Grae, Singh and Jowitt9,Reference Barbadoro, Dolcini and Fortunato22,23 Importantly, we observed a significant increase in the prevalence of BSIs among these high-risk patients which has been identified in other studies during the same time period.Reference Ntziora and Giannitsioti 27 This is of concern as BSIs among ICU patients are associated with substantial morbidity and mortality and antimicrobial resistance further complicates the care of these patients.Reference Munro, Zilberberg and Shorr 28
In 2024, approximately one-third of all HAIs were device-associated (33%), similar to the 2017 survey (32%) and results from the 2022–2023 European survey (32%). 23 One-third of all healthcare-associated pneumonia infections and nearly half of all UTIs and BSIs were device-associated. Efforts to implement and improve compliance with evidence-based infection control interventions (e.g., bundles, surveillance, and education) to reduce device-associated HAIs should continue, yet given that two-thirds of HAIs in our survey were not associated with a device or surgical procedure, other intervention strategies are required to reduce their burden. Other strategies may include continued efforts to improve hand hygiene, improving environmental cleaning and disinfection, adopting antimicrobial stewardship principles, and improving IPAC staffing and resources. In addition, support for further research that provides evidence for prevention strategies is needed. 23,Reference Klompas, Branson and Cawcutt29–34
Limitations of this work include the potential for selection bias as only those hospitals participating in the CNISP network were eligible to participate. However, the network represents approximately 37% of acute care beds in Canada and the addition of small community hospitals has improved the generalizability of these results. 14 Minor revisions to the surveillance definitions and CDI laboratory testing practices used in previous surveys have been described.Reference Mitchell, Taylor and Rudnick 11 However, as standardized methodology was applied, these revisions had minimal impact on our ability to compare prevalence over time. Our survey findings do not account for the burden of community-associated infections on hospitals; however, these data are captured for some infections through CNISP routine incidence surveillance. Finally, these results were not corrected for changes in patient case mix and hospital characteristics. Therefore, results should be interpreted with caution as hospital participation varied over the surveys. Notably, when we restricted our analysis to only those hospitals which participated in the 2017 and 2024 surveys, we found similar results.
In conclusion, a point prevalence survey performed among a network of Canadian acute-care hospitals in 2024 following the COVID-19 pandemic identified a stable prevalence of HAIs and AROs compared to the 2017 survey, despite the emergence of SARS-CoV-2. These findings provide evidence of national progress in the prevention of HAIs and AROs and underscores the importance of continued investment in hospital infection prevention. Of concern is the increasing trend in BSIs and isolation burden due to CPOs and VRIs which have a high impact on patients and informs future direction for IPAC programs. Preventing and controlling HAIs and AROs will not only improve patient outcomes but it will also preserve the effectiveness of antibiotics and protect our ability to treat infections in the future.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/ice.2025.10259.
Acknowledgements
The authors acknowledge the significant contribution of the infection prevention and control staff in participating hospitals. The authors acknowledge the role of Olivia Varsaneux and Nicholas Papayiannakis in coordinating the funding for this task authorization. Finally, the authors are grateful for the expertise and guidance provided by the late Dr. Geoffrey Taylor to support this work over many years.
Financial support
This work was supported by the Public Health Agency of Canada.
Competing interests
The author [Susy S Hota] has received a research operations grant from Finch Therapeutics Group. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.
Open access
