Pharmacy OneSource Blog

Clinical Corner: Algorithmic Detection of Healthcare-Associated Infections: The Future Direction of Surveillance

Posted on 10/09/14

healthcare_associated_infection_clinical

The Society for Healthcare Epidemiologists of America (SHEA) has published the long-awaited white paper on data requirements for electronic surveillance of healthcare-associated infections.1  The publication is timely as acute care hospital infection preventionists (IPs) are faced with federally mandated reporting of healthcare-associated infections (HAIs) in non-ICU patient populations. This additional surveillance requirement may substantially add to the 50% time effort that IPs currently allocate to HAI detection.

Surprisingly, the majority of IPs in the United States perform manual surveillance despite the availability of automated surveillance technologies for nearly three decades. More importantly, evidence shows that the subjective nature of traditional surveillance is potentially biased and has led to inter-institutional variability in the use of surveillance techniques and the interpretation of infection definitions. These inconsistencies affect the validity of publicly reported data that should be defined by consistently applied objective criteria. The National Healthcare Safety Network (NHSN) has responded to this evidence by revising HAI definitions to reduce complexity and by supporting the development of computer-based detection algorithms and the use of electronic healthcare data for HAI surveillance.

Woeltje and colleagues provided an overview of different approaches to electronic surveillance: semi-automated systems that assist the IP in performing surveillance, and fully electronic systems that conduct surveillance entirely independent of IP involvement. They provided a comprehensive description of required data elements in the table below:

Health_Associated_Infection_Algorithm

(Table 2: Key Data Elements Necessary for Electronic Surveillance of Healthcare-Associated Infections)

Facilities considering the implementation of electronic surveillance need to decide which data elements are needed from the EMR and other supporting databases and how the data should be structured. An ideal semi-automated surveillance system will have alerts that reliably eliminate data from patients who do not have HAIs (high negative predictive value) and good specificity (detection of true negative cases). It is emphasized that facilities must conduct internal validation at start-up and whenever interface changes are made to ensure that the data received is complete and accurate.

With the current focus on fully electronic surveillance systems – which perform surveillance in its entirety without case review – effective use of the data will be dependent on the IP’s skills and understanding of the strengths and limitations of output from algorithmic detection models. NHSN and the infection prevention professional societies of SHEA and the Association for Professionals in Infection Control and Epidemiology (APIC) need to offer educational programs and mentoring to address information technology knowledge deficits in the IP community as the future direction for surveillance is clear.

1 Woeltje KF, Yin MY, Klompas M, Wright MO, et al. Data Requirements for Electronic Surveillance of Healthcare-Associated Infections. Infect Control Hosp Epidemiol 2014;35(9):1083-1091.

 

Implementation of Risk Evaluation and Mitigation Strategy Programs in a Health System  

Topics: Infection Prevention

About the Author

Joan N. Hebden, RN, MS, CIC served as the Director of Infection Prevention and Control for 28 years at the University of Maryland Medical Center, Baltimore, MD. Her clinical background includes general medicine, oncology, and cardiothoracic intensive care. She has presented at national epidemiology conferences, participated in research regarding the transmission of multi-drug resistant bacteria, contributed chapters on infection control to nursing resource textbooks, and published in medical and infection control journals. Joan received her Bachelor of Science degree in nursing and a Master of Science degree in nursing education and trauma/critical care from the University of Maryland School of Nursing. She is certified in Infection Control & Epidemiology (CIC). Joan is an active member of SHEA and APIC and is currently on the Editorial Board of the American Journal of Infection Control (AJIC).