Context

Culturally diverse older adults may prefer varying control over medical decisions. Decision control preferences (DCPs) may profoundly affect advance care planning (ACP) and communication.

Objectives

To determine the DCPs of diverse, older adults and whether DCPs are associated with participant characteristics, ACP, and communication satisfaction.

Methods

A total of 146 participants were recruited from clinics and senior centers in San Francisco. We assessed DCPs using the control preferences scale: doctor makes all decisions (low), shares with doctor (medium), makes own decisions (high). We assessed associations between DCPs and demographics; prior advance directives; ability to make in-the-moment goals of care decisions; self-efficacy, readiness, and prior asked questions; and satisfaction with patient-doctor communication (on a five-point Likert scale), using Chi-square and Kruskal-Wallis analysis of variance.

Results

Mean age was 71 ± 10 years, 53% were non-white, 47% completed an advance directive, and 70% made goals of care decisions. Of the sample, 18% had low DCPs, 33% medium, and 49% high. Older age was the only characteristic associated with DCPs (low: 75 ± 11 years, medium: 69 ± 10 years, high: 70 ± 9 years, P = 0.003). DCPs were not associated with ACP, in-the-moment decisions, or communication satisfaction. Readiness was the only question-asking behavior associated (low: 3.8 ± 1.2, medium: 4.1 ± 1.2, high: 4.3 ± 1.2, P = 0.05).

Conclusion

Nearly one-fifth of diverse, older adults want doctors to make their medical decisions. Older age and lower readiness to ask questions were the only demographic variables significantly associated with low DCPs. Yet, older adults with low DCPs still engaged in ACP, asked questions, and reported communication satisfaction. Clinicians can encourage ACP and questions for all patients, but should assess DCPs to provide the desired amount of decision support.

BACKGROUND: The impact of albumin use during major surgery is unknown, and a dearth of evidence governing its use in major noncardiac surgery has long precluded its standardization in clinical guidelines. OBJECTIVE: In this study, we investigate institutional variation in albumin use among medical centers in the United States during major noncardiac surgery and explore the association of intraoperative albumin administration with important postoperative outcomes. METHODS: The study is an observational retrospective cohort analysis performed among 54 U.S. hospitals in the Multicenter Perioperative Outcomes Group and includes adult patients who underwent major noncardiac surgery under general anesthesia between January 2014 and June 2020. The primary endpoint was the incidence of albumin administration. Secondary endpoints are acute kidney injury (AKI), net-positive fluid balance, pulmonary complications, and 30-day mortality. Albumin-exposed and albumin-unexposed cases were compared within a propensity score-matched cohort to evaluate associations of albumin use with outcomes. RESULTS: Among 614,215 major surgeries, predominantly iso-oncotic albumin was administered in 15.3% of cases and featured significant inter-institutional variability in use patterns. Cases receiving intraoperative albumin involved patients of higher American Society of Anesthesiologists physical status and featured larger infused crystalloid volumes, greater blood loss, and vasopressor use. Overall, albumin was most often administered at high-volume surgery centers with academic affiliation, and within a propensity score-matched cohort (n=153,218), the use of albumin was associated with AKI (aOR 1.24, 95% CI 1.20-1.28, P <0.001), severe AKI (aOR 1.45, 95% CI 1.34-1.56, P <0.001), net-positive fluid balance (aOR 1.18, 95% CI 1.16-1.20, P <0.001), pulmonary complications (aOR 1.56, 95% CI 1.30-1.86, P <0.001), and 30-day all-cause mortality (aOR 1.37, 95% CI 1.26-1.49, P <0.001). CONCLUSIONS: Intravenous albumin is commonly administered among noncardiac surgeries with significant inter-institutional variability in use in the United States. Albumin administration was associated with an increased risk of postoperative complications.

Background

Accurate, pragmatic risk stratification for postoperative delirium (POD) is necessary to target preventative resources toward high-risk patients. Machine learning (ML) offers a novel approach to leveraging electronic health record (EHR) data for POD prediction. We sought to develop and internally validate a ML-derived POD risk prediction model using preoperative risk features, and to compare its performance to models developed with traditional logistic regression.

Methods

This was a retrospective analysis of preoperative EHR data from 24,885 adults undergoing a procedure requiring anesthesia care, recovering in the main post-anesthesia care unit, and staying in the hospital at least overnight between December 2016 and December 2019 at either of two hospitals in a tertiary care health system. One hundred fifteen preoperative risk features including demographics, comorbidities, nursing assessments, surgery type, and other preoperative EHR data were used to predict postoperative delirium (POD), defined as any instance of Nursing Delirium Screening Scale ≥2 or positive Confusion Assessment Method for the Intensive Care Unit within the first 7 postoperative days. Two ML models (Neural Network and XGBoost), two traditional logistic regression models ("clinician-guided" and "ML hybrid"), and a previously described delirium risk stratification tool (AWOL-S) were evaluated using the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, positive likelihood ratio, and positive predictive value. Model calibration was assessed with a calibration curve. Patients with no POD assessments charted or at least 20% of input variables missing were excluded.

Results

POD incidence was 5.3%. The AUC-ROC for Neural Net was 0.841 [95% CI 0. 816-0.863] and for XGBoost was 0.851 [95% CI 0.827-0.874], which was significantly better than the clinician-guided (AUC-ROC 0.763 [0.734-0.793], p < 0.001) and ML hybrid (AUC-ROC 0.824 [0.800-0.849], p < 0.001) regression models and AWOL-S (AUC-ROC 0.762 [95% CI 0.713-0.812], p < 0.001). Neural Net, XGBoost, and ML hybrid models demonstrated excellent calibration, while calibration of the clinician-guided and AWOL-S models was moderate; they tended to overestimate delirium risk in those already at highest risk.

Conclusion

Using pragmatically collected EHR data, two ML models predicted POD in a broad perioperative population with high discrimination. Optimal application of the models would provide automated, real-time delirium risk stratification to improve perioperative management of surgical patients at risk for POD.

OBJECTIVES: To develop proof-of-concept algorithms using alternative approaches to capture provider sentiment in ICU notes. DESIGN: Retrospective observational cohort study. SETTING: The Multiparameter Intelligent Monitoring of Intensive Care III (MIMIC-III) and the University of California, San Francisco (UCSF) deidentified notes databases. PATIENTS: Adult (≥18 yr old) patients admitted to the ICU. MEASUREMENTS AND MAIN RESULTS: We developed two sentiment models: 1) a keywords-based approach using a consensus-based clinical sentiment lexicon comprised of 72 positive and 103 negative phrases, including negations and 2) a Decoding-enhanced Bidirectional Encoder Representations from Transformers with disentangled attention-v3-based deep learning model (keywords-independent) trained on clinical sentiment labels. We applied the models to 198,944 notes across 52,997 ICU admissions in the MIMIC-III database. Analyses were replicated on an external sample of patients admitted to a UCSF ICU from 2018 to 2019. We also labeled sentiment in 1,493 note fragments and compared the predictive accuracy of our tools to three popular sentiment classifiers. Clinical sentiment terms were found in 99% of patient visits across 88% of notes. Our two sentiment tools were substantially more predictive (Spearman correlations of 0.62-0.84, p values < 0.00001) of labeled sentiment compared with general language algorithms (0.28-0.46). CONCLUSION: Our exploratory healthcare-specific sentiment models can more accurately detect positivity and negativity in clinical notes compared with general sentiment tools not designed for clinical usage.

Background

Data suggest that pregnant women are not at elevated risk of acquiring severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection or developing severe disease compared with nonpregnant patients. However, management of pregnant patients who are critically ill with coronavirus disease 2019 (COVID-19) infection is complicated by physiologic changes and other pregnancy considerations and requires balancing maternal and fetal well-being.

Case

We report the case of a patient at 28 weeks of gestation with acute respiratory distress syndrome (ARDS) from COVID-19 infection, whose deteriorating respiratory condition prompted delivery. Our patient's oxygenation and respiratory mechanics improved within hours of delivery, though she required prolonged mechanical ventilation until postpartum day 10. Neonatal swabs for SARS-CoV-2 and COVID-19 immunoglobulin (Ig) G and IgM were negative.

Conclusion

We describe our multidisciplinary management of a preterm pregnant patient with ARDS from COVID-19 infection and her neonate.