Search

Use of the electronic health record (EHR) has become a routine part of perioperative care in the United States. Secondary use of EHR data includes research, quality, and educational initiatives. Fundamental to secondary use is a framework to ensure fidelity, transparency, and completeness of the source data. In developing this framework, competing priorities must be considered as to which data sources are used and how data are organized and incorporated into a useable format. In assembling perioperative data from diverse institutions across the United States and Europe, the Multicenter Perioperative Outcomes Group (MPOG) has developed methods to support such a framework. This special article outlines how MPOG has approached considerations of data structure, validation, and accessibility to support multicenter integration of perioperative EHRs. In this multicenter practice registry, MPOG has developed processes to extract data from the perioperative EHR; transform data into a standardized format; and validate, deidentify, and transfer data to a secure central Coordinating Center database. Participating institutions may obtain access to this central database, governed by quality and research committees, to inform clinical practice and contribute to the scientific and clinical communities. Through a rigorous and standardized approach to ensure data integrity, MPOG enables data to be usable for quality improvement and advancing scientific knowledge. As of March 2019, our collaboration of 46 hospitals has accrued 10.7 million anesthesia records with associated perioperative EHR data across heterogeneous vendors. Facilitated by MPOG, each site retains access to a local repository containing all site-specific perioperative data, distinct from source EHRs and readily available for local research, quality, and educational initiatives. Through committee approval processes, investigators at participating sites may additionally access multicenter data for similar initiatives. Emerging from this work are 4 considerations that our group has prioritized to improve data quality: (1) data should be available at the local level before Coordinating Center transfer; (2) data should be rigorously validated against standardized metrics before use; (3) data should be curated into computable phenotypes that are easily accessible; and (4) data should be collected for both research and quality improvement purposes because these complementary goals bolster the strength of each endeavor.

Background Heart failure with reduced ejection fraction (HFrEF) is a condition imposing significant health care burden. Given its syndromic nature and often insidious onset, the diagnosis may not be made until clinical manifestations prompt further evaluation. Detecting HFrEF in precursor stages could allow for early initiation of treatments to modify disease progression. Granular data collected during the perioperative period may represent an underutilized method for improving the diagnosis of HFrEF. We hypothesized that patients ultimately diagnosed with HFrEF following surgery can be identified via machine-learning approaches using pre- and intraoperative data. Methods Perioperative data were reviewed from adult patients undergoing general anesthesia for major surgical procedures at an academic quaternary care center between 2010 and 2016. Patients with known HFrEF, heart failure with preserved ejection fraction, preoperative critical illness, or undergoing cardiac, cardiology, or electrophysiologic procedures were excluded. Patients were classified as healthy controls or undiagnosed HFrEF. Undiagnosed HFrEF was defined as lacking a HFrEF diagnosis preoperatively but establishing a diagnosis within 730 days postoperatively. Undiagnosed HFrEF patients were adjudicated by expert clinician review, excluding cases for which HFrEF was secondary to a perioperative triggering event, or any event not associated with HFrEF natural disease progression. Machine-learning models, including L1 regularized logistic regression, random forest, and extreme gradient boosting were developed to detect undiagnosed HFrEF, using perioperative data including 628 preoperative and 1195 intraoperative features. Training/validation and test datasets were used with parameter tuning. Test set model performance was evaluated using area under the receiver operating characteristic curve (AUROC), positive predictive value, and other standard metrics. Results Among 67,697 cases analyzed, 279 (0.41%) patients had undiagnosed HFrEF. The AUROC for the logistic regression model was 0.869 (95% confidence interval, 0.829-0.911), 0.872 (0.836-0.909) for the random forest model, and 0.873 (0.833-0.913) for the extreme gradient boosting model. The corresponding positive predictive values were 1.69% (1.06%-2.32%), 1.42% (0.85%-1.98%), and 1.78% (1.15%-2.40%), respectively. Conclusions Machine-learning models leveraging perioperative data can detect undiagnosed HFrEF with good performance. However, the low prevalence of the disease results in a low positive predictive value, and for clinically meaningful sensitivity thresholds to be actionable, confirmatory testing with high specificity (eg, echocardiography or cardiac biomarkers) would be required following model detection. Future studies are necessary to externally validate algorithm performance at additional centers and explore the feasibility of embedding algorithms into the perioperative electronic health record for clinician use in real time.

Background Accurate anesthesiology procedure code data are essential to quality improvement, research, and reimbursement tasks within anesthesiology practices. Advanced data science techniques, including machine learning and natural language processing, offer opportunities to develop classification tools for Current Procedural Terminology codes across anesthesia procedures. Methods Models were created using a Train/Test dataset including 1,164,343 procedures from 16 academic and private hospitals. Five supervised machine learning models were created to classify anesthesiology Current Procedural Terminology codes, with accuracy defined as first choice classification matching the institutional-assigned code existing in the perioperative database. The two best performing models were further refined and tested on a Holdout dataset from a single institution distinct from Train/Test. A tunable confidence parameter was created to identify cases for which models were highly accurate, with the goal of at least 95% accuracy, above the reported 2018 Centers for Medicare and Medicaid Services (Baltimore, Maryland) fee-for-service accuracy. Actual submitted claim data from billing specialists were used as a reference standard. Results Support vector machine and neural network label-embedding attentive models were the best performing models, respectively, demonstrating overall accuracies of 87.9% and 84.2% (single best code), and 96.8% and 94.0% (within top three). Classification accuracy was 96.4% in 47.0% of cases using support vector machine and 94.4% in 62.2% of cases using label-embedding attentive model within the Train/Test dataset. In the Holdout dataset, respective classification accuracies were 93.1% in 58.0% of cases and 95.0% among 62.0%. The most important feature in model training was procedure text. Conclusions Through application of machine learning and natural language processing techniques, highly accurate real-time models were created for anesthesiology Current Procedural Terminology code classification. The increased processing speed and a priori targeted accuracy of this classification approach may provide performance optimization and cost reduction for quality improvement, research, and reimbursement tasks reliant on anesthesiology procedure codes. Editor’s Perspective What We Already Know about This Topic What This Article Tells Us That Is New

BackgroundDespite the significant healthcare impact of acute kidney injury, little is known regarding prevention. Single-center data have implicated hypotension in developing postoperative acute kidney injury. The generalizability of this finding and the interaction between hypotension and baseline patient disease burden remain unknown. The authors sought to determine whether the association between intraoperative hypotension and acute kidney injury varies by preoperative risk.MethodsMajor noncardiac surgical procedures performed on adult patients across eight hospitals between 2008 and 2015 were reviewed. Derivation and validation cohorts were used, and cases were stratified into preoperative risk quartiles based upon comorbidities and surgical procedure. After preoperative risk stratification, associations between intraoperative hypotension and acute kidney injury were analyzed. Hypotension was defined as the lowest mean arterial pressure range achieved for more than 10 min; ranges were defined as absolute (mmHg) or relative (percentage of decrease from baseline).ResultsAmong 138,021 cases reviewed, 12,431 (9.0%) developed postoperative acute kidney injury. Major risk factors included anemia, estimated glomerular filtration rate, surgery type, American Society of Anesthesiologists Physical Status, and expected anesthesia duration. Using such factors and others for risk stratification, patients with low baseline risk demonstrated no associations between intraoperative hypotension and acute kidney injury. Patients with medium risk demonstrated associations between severe-range intraoperative hypotension (mean arterial pressure less than 50 mmHg) and acute kidney injury (adjusted odds ratio, 2.62; 95% CI, 1.65 to 4.16 in validation cohort). In patients with the highest risk, mild hypotension ranges (mean arterial pressure 55 to 59 mmHg) were associated with acute kidney injury (adjusted odds ratio, 1.34; 95% CI, 1.16 to 1.56). Compared with absolute hypotension, relative hypotension demonstrated weak associations with acute kidney injury not replicable in the validation cohort.ConclusionsAdult patients undergoing noncardiac surgery demonstrate varying associations with distinct levels of hypotension when stratified by preoperative risk factors. Specific levels of absolute hypotension, but not relative hypotension, are an important independent risk factor for acute kidney injury.Editor’s PerspectiveWhat We Already Know about This TopicWhat This Article Tells Us That Is New

Recent studies have investigated the effect of overlapping surgeon responsibilities or nurse to patient staffing ratios on patient outcomes, but the association of overlapping anesthesiologist responsibilities with patient outcomes remains unexplored to our knowledge.To examine the association between different levels of anesthesiologist staffing ratios and surgical patient morbidity and mortality.A retrospective, matched cohort study consisting of major noncardiac inpatient surgical procedures performed from January 1, 2010, to October 31, 2017, was conducted in 23 US academic and private hospitals. A total of 866 453 adult patients (aged ≥18 years) undergoing major inpatient surgery within the Multicenter Perioperative Outcomes Group electronic health record registry were included. Anesthesiologist sign-in and sign-out times were used to calculate a continuous time-weighted average staffing ratio variable for each operation. Propensity score-matching methods were applied to create balanced sample groups with respect to patient-, operative-, and hospital-level confounders and resulted in 4 groups based on anesthesiologist staffing ratio. Groups consisted of patients receiving care from an anesthesiologist covering 1 operation (group 1), more than 1 to no more than 2 overlapping operations (group 1-2), more than 2 to no more than 3 overlapping operations (group 2-3), and more than 3 to no more than 4 overlapping operations (group 3-4). Data analysis was performed from October 2019 to October 2021.Undergoing a major inpatient surgical operation that involved an anesthesiologist providing care for up to 4 overlapping operations.The primary composite outcome was 30-day mortality and 6 major surgical morbidities (cardiac, respiratory, gastrointestinal, urinary, bleeding, and infectious complications) derived from International Classification of Diseases, Ninth Revision and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision discharge diagnosis codes.In all, 578 815 adult patients (mean [SD] age, 55.7 [16.2] years; 55.1% female) were analyzed. After matching operations according to anesthesiologist staffing ratio, 48 555 patients were in group 1; 247 057, group 1-2; 216 193, group 2-3; and 67 010, group 3-4. Increasing anesthesiologist coverage responsibilities was associated with an increase in risk-adjusted surgical patient morbidity and mortality. Compared with patients in group 1-2, those in group 2-3 had a 4% relative increase in risk-adjusted mortality and morbidity (5.06% vs 5.25%; adjusted odds ratio [AOR], 1.04; 95% CI, 1.01-1.08; P = .02) and those in group 3-4 had a 14% increase in risk-adjusted mortality and morbidity (5.06% vs 5.75%; AOR, 1.15; 95% CI, 1.09-1.21; P .001).This study's findings suggest that increasing overlapping coverage by anesthesiologists is associated with increased surgical patient morbidity and mortality. Therefore, the potential effects of staffing ratios in perioperative team models should be considered in clinical coverage efforts.