Surveillance of cardiac surgical site infections (SSIs) improved significantly when registry and infection control surveillance data were linked with electronic health records, a retrospective analysis showed.
Over the course of a 47-month period starting in 2011, Vaidehi Nayar of the Children’s Hospital of Philadelphia and her coinvestigators launched a quality improvement initiative at their institution that linked administrative databases with their clinical registry, allowing caregivers to more accurately monitor and assess SSIs and provide more effective adjudication and treatments thereafter. The investigators chose to link their hospital’s electronic health record (EHR) billing information and reporting from the infection surveillance database for the National Healthcare Safety Network with data from the Society of Thoracic Surgeons Congenital Heart Surgery Database (STS-CHSD).
Vaidehi Nayar
To further facilitate the flow and interpretation of data, the investigators used a visualization tool to analyze the STS-CHSD for case ascertainment; to resolve discrepancies among STS-CHSD, infection surveillance, and billing SSI cases; and to assess the impact of the hospital’s quality improvement protocols. These protocols consisted of wound alert reports from the EHR, bedside reviews for SSI adjudication, inpatient and outpatient SSI prevention bundles, prophylactic antibiotic dosing changes, removal of steroids from the bypass circuit, and biller education on SSIs.
Control charts in the data visualization tool allowed for statistical monitoring of SSI rate changes, and SSI case discrepancies across the databases were reviewed to ensure that differences were the result of variations in SSI reporting criteria for each database, not inaccurate surveillance population ascertainment or inaccurate SSI identification, according to Ms. Nayar and her colleagues,
“Workflow changes, including the wound alert report and bedside reviews, facilitated communication among providers and improved adjudication of suspected SSIs,” she said in presenting the J. Maxwell Chamberlain Memorial Paper for Congenital Heart Surgery at the annual meeting of the Society of Thoracic Surgeons in San Diego earlier this year.
In total, 156 SSIs were identified via the STS-CHSD, 79 via the infection surveillance database, and 433 via billing. There was a significant decrease in the rolling 12-month SSI rate from 2.48% (21/848 in January 2013) to 0.76% (11/1,442 in January 2014), based on the STS-CHSD data, although Ms. Nayar pointed out that this decline could not definitely be attributed to reporting improvements or to the specific quality improvement initiatives that were implemented. Ms. Nayar also explained that there was a “general improvement in reporting, as shown by the stark drop in billing” and “a general alignment of all three data sources.”
“Accurate assessment of morbidity data, including [SSIs], has implications for public reporting, benchmarking, assessment of [quality improvement] impact, and possible denial of payments,” said Ms. Nayar. “In summary, we achieved our two simultaneous goals of improving SSI reporting – or decreasing the data errors – and decreasing SSI incidents by improving overall patient care.”
During discussion, Ms. Nayar elaborated on the study’s generalizability and potential application to other facets of congenital heart disease treatments, saying that such application is, in fact, possible.
“Yes, this is very generalizable, but one key important fact is relevant: As long as there is a source of truth for case ascertainment, this methodology can be used in several different areas,” she explained. “For example, we are currently working at our institution to integrate and link the [Pediatric Cardiac Critical Care Consortium] database to be able to better monitor any critical care–related morbidity information and ultimately use similar methodology to assess the impact of outcomes in the critical care field.”
Ms. Nayar said that she did not have any relevant financial conflicts of interest to disclose.