Adverse event and complication tracking in anaesthesiology: dependence on self-reporting despite implementation of electronic health records

Use of natural language processing method to identify regional anesthesia from clinical notes

INTRODUCTION

Accurate data capture is integral for research and quality improvement efforts. Unfortunately, limited guidance for defining and documenting regional anesthesia has resulted in wide variation in documentation practices, even within individual hospitals, which can lead to missing and inaccurate data. This cross-sectional study sought to evaluate the performance of a natural language processing (NLP)-based algorithm developed to identify regional anesthesia within unstructured clinical notes.

METHODS

We obtained postoperative clinical notes for all patients undergoing elective non-cardiac surgery with general anesthesia at one of six Veterans Health Administration hospitals in California between January 1, 2017, and December 31, 2022. After developing and executing our algorithm, we compared our results to a frequently used referent, the Corporate Data Warehouse structured data, to assess the completeness and accuracy of the currently available data. Measures of agreement included sensitivity, positive predictive value, false negative rate, and accuracy.

RESULTS

We identified 27,713 procedures, of which 9310 (33.6%) received regional anesthesia. 96.6% of all referent regional anesthesia cases were identified in the clinic notes with a very low false negative rate and good accuracy (false negative rate=0.8%, accuracy=82.5%). Surprisingly, the clinic notes documented more than two times the number of regional anesthesia cases that were documented in the referent (algorithm n=9154 vs referent n=4606).

DISCUSSION

While our algorithm identified nearly all regional anesthesia cases from the referent, it also identified more than two times as many regional anesthesia cases as the referent, raising concerns about the accuracy and completeness of regional anesthesia documentation in administrative and clinical databases. We found that NLP was a promising alternative for identifying clinical information when existing databases lack complete documentation.

Implementation and evaluation of an event reporting system in an academic anaesthesia department

BACKGROUND

Reporting adverse clinical events is essential to a culture of safety in healthcare. However, self-reporting such events is generally not widely prevalent in a typical academic anaesthesia department.

METHODS

We set out to create a self-reporting tool to securely accept data from multiple anaesthesia service locations, including data linked to our electronic anaesthesia record, and combine them into an accessible database.We created a web-based database module for incident reporting integrated into the department's intranet. The system was also designed to actively prompt anaesthesia providers for reports following each day of clinical work.

RESULTS

478 events were recorded in the database in the first year of implementation. There were 33 347 anaesthesia encounters in that period, translating to a reporting rate of 1.43% (95% CI 1.31% to 1.57%). In the second year, which coincided with the second phase of implementation, 608 events were reported out of 45 985 anaesthesia encounters, for a reporting rate of 1.32% (95% CI 1.22% to 1.43%). Approximately 40% of events entered into the database occurred in a non-operating room location. The annual reporting rates for 2014, 2015, 2016, 2017, 2018 and 2019 were 1.26% (95% CI 1.16% to 1.37%), 1.15% (95% CI 1.05% to 1.25%), 1% (95% CI 0.9% to 1.1%), 0.6% (95% CI 0.53% to 0.68%), 0.5% (95% CI 0.44% to 0.57%), 0.4% (95% CI 0.3% to 0.5%), respectively.

CONCLUSIONS

Our incident reporting system facilitated reporting of events within and outside the operating room. The system captured event data valid for quality improvement within the anaesthesia department.

Neuroanesthesiology Quality Improvement Reporting Patterns: A Tertiary Medical Center Experience

BACKGROUND

Understanding quality improvement (QI) reporting patterns is important for practice-based improvement and for prioritizing QI initiatives. The aim of this project was to identify major domains of neuroanesthesiology QI reports at a single academic institution with 2 hospital-based practice sites.

METHODS

We retrospectively reviewed institutional QI databases to identify reports from neuroanesthesia cases between 2013 and 2021. Each report was categorized into one of the 16 primary predefined QI domains; the QI report domains were ranked by frequency. Descriptive statistics are used to present the analysis.

RESULTS

Seven hundred three QI reports (3.2% of all cases) were submitted for the 22,248 neurosurgical and neuroradiology procedures during the study period. Most of the QI reports across the institution were in the domain of communication/documentation (28.4%). Both hospitals shared the same 6 top QI report domains, although the relative frequency of each domain differed between the 2 hospitals. Drug error was the top QI report domain at one hospital, representing 19.3% of that site's neuroanesthesia QI reports. Communication/documentation was the top domain at the other hospital, representing 34.7% of that site's reports. The other 4 shared top domains were equipment/device failure, oropharyngeal injury, skin injury, and vascular catheter dislodgement.

CONCLUSIONS

The majority of neuroanesthesiology QI reports fell into 6 domains: drug error, communication/documentation, equipment/device failure, oropharyngeal injury, skin injury, and vascular catheter dislodgement. Similar analyses from other centers can guide generalizability and potential utility of using QI reporting domains to inform the development of neuroanesthesiology quality measures and reporting frameworks.

A Survey-Based Assessment of the Practices Governing Morbidity and Mortality Conferences and the Effects of the COVID-19 Pandemic

BACKGROUND

Morbidity and mortality (M&M) conferences are essential components for resident education and provide a valuable tool to improve patient safety and quality of care. M&M conferences help identify important gaps in safety and reduce avoidable events in future patient care. Active methods to improve the utilization of M&M conferences have been shown to enhance their educational value for residents, faculty and multidisciplinary teams in healthcare institutions.

OBJECTIVE

The purpose of this study was to use a survey-based methodology to assess how morbidity and mortality conferences are conducted in residency programs, including characteristics such as frequency, involvement of personnel and the effects of COVID-19.

METHODS

From February to October 2021, a validated 19 question survey was electronically distributed to residency program directors in anesthesiology, emergency medicine and general surgery, after a search for email addresses in the ACGME database. The survey was created and hosted on Google Forms.

RESULTS

A total of 125 of 713 program directors (17.5%) responded to the survey. Eighty-three percent of respondent programs reported mandatory participation for residents, with residents providing most of the presentations. Case presentations utilized various formats including SBAR, adverse event analysis and root cause analysis as the most common modalities. Though most programs reported no change in frequency of M&M conferences due to COVID-19, most respondents reported a shift to a virtual or hybrid platform.

CONCLUSION

M&M conferences are an important educational and quality improvement modality, and many residency directors changed practice to incorporate virtual platforms due to the COVID-19 pandemic to maintain uninterrupted educational sessions. Nonetheless, significant variation still exists in how these conferences are conducted between different institutions.