Electronic Submission of Surgical Site Infection Data to the National Healthcare Safety Network Using Clinical Document Architecture File Generation and Upload

Background:  Traditionally, healthcare associated infection (HAI) data submission to the National Healthcare Safety Network (NHSN) requires clinicians to manually enter HAI data into online forms via the NHSN web application. The data collection and submission process, particularly for surgical site infections (SSI) and surgical procedure denominators (SPD), can be very resource intensive.

Objective:  Recently, the NHSN embarked on the development of a clinical document architecture (CDA) solution for SSI data submission.  Clinical document architecture is a Health Level Seven (HL7) electronic document standard that enables automated surveillance systems to generate electronic files for upload to the NHSN; thereby eliminating the need for manual data entry. Herein we demonstrate how a nonprofit, acute-care hospital utilized an automated surveillance system (SafetySurveillor®, Premier, Inc, Charlotte, NC) to successfully generate SSI and SPD files for upload to the NHSN.

Methods: The CDA file generation and upload functionality was utilized to submit SSI and SPD data from April 2010 to September 2010.  The SSI event forms housed within the automated surveillance system are electronic forms compatible with the NHSN event forms and are pre-populated with demographic, admission, microbiology, and surgical procedure data leaving only some clinical questions to be answered. Data elements necessary to generate SPD files were sent from the facility’s surgery documentation and patient registration systems to the automated surveillance database.  The data was then normalized and made available for further completion by clinicians.  At the end of each monthly reporting period, CDA files containing all SSI event forms and SPD data were generated and subsequently uploaded to the NHSN. 

Results:  Approximately 360 SPD files were generated and uploaded to the NHSN during the six month study period.  Similarly, SSIs were uploaded in months when an event occurred.  Approximately 70% of the data elements required for submission of SSI events and SPD data were available within the automated surveillance system; whereas, the remaining 30% of elements were completed by the clinician.

Conclusions: To our knowledge, this is the first description of successful submission of SSI and SPD data to the NHSN using the CDA file generation and upload functionality.  This process reduces manual data submission burden which enhances opportunities for HAI prevention.  In addition, this enables healthcare facilities to meet regulatory mandates for SSI reporting.   In order to achieve the goal of 100% complete data from automated sources, healthcare facilities should improve standardization and completeness of documentation within their surgical procedure data systems.  Ultimately, electronic health records may present a better solution for obtaining data for this purpose.