SCIP 10; Are We Measuring the Right Measure?

ISSUE: Recent outbreaks of carbapenem-resistant gram-negative bacteria among hospitalized patients elicited national attention and underscored the danger of healthcare-associated infections. Whenever patients visit more than one hospital, a multidrugresistant organism (MDRO) may spread. Within hospitals, spread is also a scourge. Our goal is to build a regional system that parses microbiology culture data to make it usable for decision-support, and then alerts hospitals when a patient with a history of MDRO is admitted. PROJECT: We cull microbiology data from Health Level Seven version 2 (HL7v2) messages that hospitals send to a health information exchange. The principal informatics problem is that microbiology messages are often not structured in standard HL7v2 format by the sending hospitals. We therefore built an HL7v2 correction engine that deals with incorrect message structure and/ or content, to parse key data elements needed for infection control alerts: organism, antibiotics tested, minimum inhibitory concentrations, susceptibility interpretation, body source, and healthcare facility where drawn. These data elements can then be integrated into enhanced email alerts to hospital infection preventionists. RESULTS: In a two-month test (without sending alerts out to hospitals), we identified 437 patients with a gram-negative MDRO. Within these same two months, 31 (7%) of the 437 patients generated a test alert upon returning for inpatient or emergency department care sometime after the culture encounter. Seven of 31 returned to a healthcare institution different from where the culture was drawn. All 31 test alerts correctly identified a patient with history of MDRO. However, 2/31 included inconsistent information (because hospitals sent a preliminary result after a final result); we adjusted our procedures to address this “bug.” LESSON LEARNED: While there are many methodologic challenges in building a flexible microbiology data processing, storage, and alert system for infection preventionists, the Results may contribute to epidemiologic understanding of patterns of gramnegative (and ultimately other) MDROs across a region. We benefitted from having consulted the hospital infection preventionists during the development of the system. Our goal is for the alerts to make some contribution to reducing the emergence and spread of these dangerous bacteria.