Facilities enrolled in the NHSN PSC are required to submit a PSC Annual Survey for their facility type (i.e., hospital, LTACH, or Inpatient Rehabilitation Facility [IRF]) at the beginning of each year, reporting data from the previous year, hereafter referred to as the survey year. The PSC Annual Hospital, LTACH, and IRF Surveys collect facilitylevel data about facility characteristics, standard operations, and infection prevention practices, including yeast identification, AFST practices, and C. auris screening practices (https://www.cdc.gov/nhsn/forms/57.103_pshospsurv_blank.pdf, https://www.cdc.gov/nhsn/forms/57.150_LTACFacSurv_BLANK.pdf). The NHSN application incorporates standardized business rules and automated data checks for PSC Annual Surveys to ensure completeness, consistency, and data quality.

We analyzed NHSN PSC Annual Survey data for survey year 2024 for both ACH and LTACH. IRF surveys were excluded from the analysis. Facilities were considered active if they submitted a PSC Annual Facility Survey in 2024 or 2023. The number of enrolled facilities is higher and includes facilities that have not reported a survey in recent years. We examined responses regarding yeast identification practices (matrix-assisted laser desorption/ionization time-of-flight mass spectrometry [MALDI-TOF MS], chromogenic agar, molecular methods, and referral testing), AFST practices (methods used, antifungals tested, and triggers for testing), and C. auris screening (whether routine screening is performed, screening triggers, and testing methods used) (https://www.cdc.gov/nhsn/psc/locations.html). We performed descriptive analyses, stratifying by ACH and LTACH. Responses were excluded from the denominator for analyses specific to a given survey item if hospitals indicated they did not have yeast identification or AFST performed, marked the item as “not applicable,” or had missing data for that item. These exclusions were applied at the item level; facilities were included in other analyses for which they provided applicable data. Data were frozen on 16 May 2025 and analyzed using SAS version 9.4 software (SAS Institute, Cary, North Carolina).

Our analysis provides the first national description of yeast identification practices in U.S. ACHs and LTACHs. Encouragingly, the vast majority of facilities reported having Candida identified to the species level for bloodstream and other normally sterile site isolates, which is consistent with IDSA guidelines (5). Species-level identification was substantially less common for urine and respiratory specimens compared with bloodstream and other sterile site isolates. Although yeast recovered from nonsterile sites often represents colonization, species-level identification in selected clinical contexts might help monitor emerging resistance patterns. Additional studies could improve understanding of the clinical impact of species-level identification for non-sterile site isolates. MALDI-TOF MS was the predominant identification method (ACH: 71%, LTACH: 63%), reflecting substantial progress toward faster, more accurate species-level diagnosis since this technique was first approved by the U.S. Food and Drug Administration (FDA) in 2013 (11). However, a substantial proportion of hospitals reported relying exclusively on a single platform, most often MALDI-TOF MS (ACH: 36%, LTACH: 29%) or Vitek 2 (ACH: 10%, LTACH: 14%), which can only identify organisms already represented in their reference databases. Sole reliance on Vitek 2 is particularly concerning, given known challenges in accurately identifying rarer Candida species such as Candida blankii and Candida famata (7, 12). In addition, sole reliance on MALDI-TOF MS may be limited by the completeness of its spectral reference database; rare or emerging yeasts might be underrepresented, potentially leading to misidentification or failure to identify uncommon species (13). The low percentage of facilities routinely using chromogenic agar for primary specimens (ACH: 27%, LTACH: 32%) suggests an opportunity to improve laboratory practices as this low-cost and practical method can help differentiate Candida species and because bloodstream infections can have >1 species present (7, 12, 14).

Approximately two-thirds of ACHs and LTACHs had AFST available on-site or at an affiliated center, a higher proportion than reported in 2015 NHSN data on general ACHs (28%) (8). Compared with the prior study, automatic AFST without a clinician’s order was also more common, especially for bloodstream isolates (59% of ACHs and 55% of LTACHs vs <33% in 2015) (8). These increases are encouraging as reliance on off-site testing for AFST and requiring a clinician’s order (as opposed to using reflexive AFST) might slow turnaround, which could potentially delay appropriate therapy or limit opportunities for timely de-escalation from empiric echinocandin therapy to fluconazole (5, 8, 15). Although fluconazole remains the most commonly tested agent, we observed broader AFST coverage compared with the 2015 study, with more frequent inclusion of echinocandins and newer triazoles (8). This expansion might reflect the growing awareness of emerging resistance among Candida species.

Regarding C. auris screening, our findings complement those from a 2022 survey of Emerging Infections Network (EIN) members. However, direct comparison is challenging because the EIN survey represented a convenience sample of infectious diseases clinician responses from the same healthcare facilities could not be deduplicated, and participation was voluntary (16). In our study, PCR predominated over culture as the testing method (ACH: 52% vs 47%; LTACH: 73% vs 32%), contrasting with the prior study where culture was more than twice as common compared with PCR for healthcare facilities with in-house testing (67% vs 31%) (16). This finding could reflect differences in the survey design. It might also reflect increased PCR use due to its faster turnaround time compared with culture, which facilitates timely isolation and infection-prevention measures, as well as the availability of FDA-approved PCR assays for C. auris identification (16–18). The higher uptake of PCR and frequent use of surveillance testing upon admission in LTACHs (86%) might reflect the high risk of transmission in this setting and the need for rapid detection to prevent transmission among a medically complex patient population (2, 18). Because these data largely reflect practices before widespread availability of an FDA-cleared commercial PCR assay for C. auris colonization screening (cleared July 2024) (19, 20), future studies could evaluate whether such assays further influence implementation and standardization of screening practices.

Study limitations include that the NHSN PSC Annual Survey data were self-reported and could be subject to reporting or misclassification bias. In addition, we could not verify laboratory practices directly, assess test performance or turnaround times, or capture patient-level outcomes.

Nonetheless, we provided new benchmark data on national yeast identification practices and C. auris screening practices and updated information on AFST practices at ACHs and LTACHs. This information could support facilities in evaluating and strengthening their yeast identification, AFST, and C. auris screening practices to optimize patient care and enhance public health preparedness for emerging antifungal-resistant Candida species.