g Delays often occur between CLSI and FDA revisions of antimicrobial interpretive criteria. Using our Regional Healthcare Ecosystem Analyst (RHEA) simulation model, we found that the 32-month delay in changing carbapenem-resistant Enterobacteriaceae (CRE) breakpoints might have resulted in 1,821 additional carriers in Orange County, CA, an outcome that could have been avoided by identifying CRE and initiating contact precautions. Policy makers should aim to minimize the delay in the adoption of new breakpoints for antimicrobials against emerging pathogens when containment of spread is paramount; delays of <1.5 years are ideal.
Delays often occur between the issuance of new diagnostic interpretive criteria for microbiology laboratories by standardsetting organizations, such as the Clinical and Laboratory Standards Institute (CLSI), and their adoption by the Food and Drug Administration (FDA) to inform breakpoints for the manufacturers of diagnostic tests. Delays occur due to the FDA's required regulatory processes and the necessity of generating data from pharmaceutical companies to support interpretive criteria changes. Quantifying the impact of such delays could help determine the value of addressing and rectifying their causes. A recent example is the 32-to 42-month delay (depending on the antimicrobial) between CLSI's release of moresensitive criteria for diagnosing carbapenem-resistant Enterobacteriaceae (CRE), beginning with M100-S20 issuances in 2010, and the conveyance of these new criteria to manufacturers by the FDA (1). Such a delay could result in CRE transmission if CRE carriers are missed (because old criteria are still in use) and are not placed on contact precautions to reduce CRE spread (2). This is of concern because CRE are considered an urgent public health threat by the Centers for Disease Control and Prevention (CDC) (3), and few treatment options exist for CRE infection, which can result in high mortality. Using our Regional Healthcare Ecosystem Analyst (RHEA)-generated simulation model of Orange County, CA (OC) (4), we determined the impact of this delay on estimated (i.e., potential) CRE transmission within health care facilities.
MATERIALS AND METHODSWe used our previously described Regional Healthcare Ecosystem Analyst (RHEA) software platform (4-6) to generate a detailed agent-based model of Orange County, CA, which included detailed representations of all 28 acute-care hospitals (including 5 long-term acute-care facilities [LTACs]) and 74 free-standing nursing homes serving adult patients, along with the patients flowing among these locations and the community at large. We utilized the RHEA OC model to simulate the spread of CRE (7, 8) and the impact of changing CRE breakpoints in the early stages of OC's epidemic (years 4 to 5). Our model drew from detailed 2011-2012 OC patient-level data for adult inpatient hospital and nursing home admissions (9, 10). Table 1 shows key model inputs.Briefly, the model represents each patient with a computational agent. As in real life, each virtual ...