Harmonized system of coding of surgical procedures and analysis of surgical site infections (ISO) of five European countries | BMC Medical Research Methodology

To facilitate the conduct of ongoing clinical trials analyzing the incidence and impact of SSIs internationally, we have developed a coding system harmonizing the country-specific procedure codes used in some European countries.

Surveillance by hospital infection control programs often depends on unproven screening strategies to identify patients with possible SSIs, such as readmission screening, review of daily microbiological results, and self-assessment by the surgeon. While clinical scoring systems have been validated for the detection of SSI [11, 12], these scores are not systematically documented in EHRs and are not easily calculated retrospectively from available data. Although a specialized coding system is important for accounting purposes in each country, an international coding system is necessary to harmonize data in large clinical trials and to facilitate the inclusion of clinical and administrative data from routine. The use of an international code of procedure has the potential to improve the detection and monitoring of SSIs and allows for more standardized inter-hospital comparison internationally.

By using the coding systems presented, future prospective trials, retrospective analyzes and routine surveillance efforts can streamline international collaboration by avoiding the need for manual translation of surgical procedure codes. Although our coding system was developed in the context of ISO research, it can be used in all international tracks comparing aspects of surgery, eg indications, outcomes, costs, etc. Future efforts could also expand to code both horizontally, i.e. mapping other country-specific codes onto the SALT code, or vertically, i.e. adding new codes. If needed, granularity could be added with minimal modification by expanding the namespace while preserving compatibility (eg, VCH1-A, VCH1-B, etc.).

The international comparison of social and health systems is increasingly important, as the current SARS-CoV-2 pandemic highlights. National and international harmonization efforts include computer modeling and decision support systems to guide evidence-based medicine. However, to be effective, the mutual coding not only of procedures, but also of other tasks from the perspective of the whole system, including health, social, housing, employment, education and justice, must be targeted. New approaches including these aspects aim to create harmonized codes by automated computerized techniques [13, 14]. Automated code harmonization and machine learning models, currently being validated, could also be the subject of future trials in the context of surgery.

Our approach has several limitations. As it was created manually, the implementation of the inclusion and exclusion criteria is prone to error: the national codes contain several procedures which are clearly not surgeries, for example haemodialysis. We believe our iterative approach with multiple rounds of independent validation by ID physicians and surgeons minimizes the potential for human error. In addition, the basis of our codes was an export of surgeries performed in participating centers. The code may therefore not be universal or complete. However, based on the number of patients included (>170,000), we believe we have covered all quantitatively relevant procedures. In addition, this code excludes eye surgery, as well as pediatric surgery and is limited to France, Germany, Italy, Spain and the United Kingdom.

The current edition of SALT and future updated editions will be accessible online in machine-readable format. Further editions of the coding system proposed here should include the full spectrum of codes in all countries as well as other countries. It can not only be used for testing in infectious diseases and infection control, but also for a wider range of scientific research questions in various medical fields.

Christine E. Phillips