Semiautomated surveillance of deep surgical site infections after colorectal surgeries: A multicenter external validation of two surveillance algorithms

The augmented value of using clinical notes in semi-automated surveillance of deep surgical site infections after colorectal surgery

BACKGROUND

In patients who underwent colorectal surgery, an existing semi-automated surveillance algorithm based on structured data achieves high sensitivity in detecting deep surgical site infections (SSI), however, generates a significant number of false positives. The inclusion of unstructured, clinical narratives to the algorithm may decrease the number of patients requiring manual chart review. The aim of this study was to investigate the performance of this semi-automated surveillance algorithm augmented with a natural language processing (NLP) component to improve positive predictive value (PPV) and thus workload reduction (WR).

METHODS

Retrospective, observational cohort study in patients who underwent colorectal surgery from January 1, 2015, through September 30, 2020. NLP was used to detect keyword counts in clinical notes. Several NLP-algorithms were developed with different count input types and classifiers, and added as component to the original semi-automated algorithm. Traditional manual surveillance was compared with the NLP-augmented surveillance algorithms and sensitivity, specificity, PPV and WR were calculated.

RESULTS

From the NLP-augmented models, the decision tree models with discretized counts or binary counts had the best performance (sensitivity 95.1% (95%CI 83.5-99.4%), WR 60.9%) and improved PPV and WR by only 2.6% and 3.6%, respectively, compared to the original algorithm.

CONCLUSIONS

The addition of an NLP component to the existing algorithm had modest effect on WR (decrease of 1.4-12.5%), at the cost of sensitivity. For future implementation it will be a trade-off between optimal case-finding techniques versus practical considerations such as acceptability and availability of resources.

External validation of semi-automated surveillance algorithms for deep surgical site infections after colorectal surgery in an independent country

BACKGROUND

Automated surveillance methods that re-use electronic health record data are considered an attractive alternative to traditional manual surveillance. However, surveillance algorithms need to be thoroughly validated before being implemented in a clinical setting. With semi-automated surveillance patients are classified as low or high probability of having developed infection, and only high probability patients subsequently undergo manual record review. The aim of this study was to externally validate two existing semi-automated surveillance algorithms for deep SSI after colorectal surgery, developed on Spanish and Dutch data, in a Swedish setting.

METHODS

The algorithms were validated in 225 randomly selected surgeries from Karolinska University Hospital from the period January 1, 2015 until August 31, 2020. Both algorithms were based on (re)admission and discharge data, mortality, reoperations, radiology orders, and antibiotic prescriptions, while one additionally used microbiology cultures. SSI was based on ECDC definitions. Sensitivity, specificity, positive predictive value, negative predictive value, and workload reduction were assessed compared to manual surveillance.

RESULTS

Both algorithms performed well, yet the algorithm not relying on microbiological culture data had highest sensitivity (97.6, 95%CI: 87.4-99.6), which was comparable to previously published results. The latter algorithm aligned best with clinical practice and would lead to 57% records less to review.

CONCLUSIONS

The results highlight the importance of thorough validation before implementation in other clinical settings than in which algorithms were originally developed: the algorithm excluding microbiology cultures had highest sensitivity in this new setting and has the potential to support large-scale semi-automated surveillance of SSI after colorectal surgery.