Comparison of static and rolling logistic regression models on predicting invasive mechanical ventilation or death from COVID-19-A retrospective, multicentre study.

INTRODUCTION: COVID-19 virus has undergone mutations, and the introduction of vaccines and effective treatments have changed its clinical severity. We hypothesized that models that evolve may better predict invasive mechanical ventilation or death than do static models. METHODS: This retrospective study of adult patients with COVID-19 from six Michigan hospitals analysed 20 demographic, comorbid, vital sign and laboratory factors, one derived factor and nine factors representing changes in vital signs or laboratory values with time for their ability to predict death or invasive mechanical ventilation within the next 4, 8 or 24 h. Static logistic regression was constructed on the initial 300 patients and tested on the remaining 6741 patients. Rolling logistic regression was similarly constructed on the initial 300 patients, but then new patients were added, and older patients removed. Each new construction model was subsequently tested on the next patient. Static and rolling models were compared with receiver operator characteristic and precision-recall curves. RESULTS: Of the 7041 patients, 534 (7.6%) required invasive mechanical ventilation or died within 14 days of arrival. Rolling models improved discrimination (0.865 ± 0.010, 0.856 ± 0.007 and 0.843 ± 0.005 for the 4, 8 and 24-h models, respectively; all p < 0.001 compared with the static logistic regressions with 0.827 ± 0.011, 0.794 ± 0.012 and 0.735 ± 0.012, respectively). Similarly, the areas under the precision-recall curves improved from 0.006, 0.010 and 0.021 with the static models to 0.030, 0.045 and 0.076 for the 4-, 8- and 24-h rolling models, respectively, all p < 0.001. CONCLUSION: Rolling models with contemporaneous data maintained better metrics of performance than static models, which used older data.

Retrospective study evaluating telehealth antenatal anesthesia consults for high-risk obstetric patients.

BACKGROUND: Telehealth has gained popularity, particularly in the COVID-19 era. The use of telehealth is now being applied to preoperative evaluation clinics in an effort to overcome barriers to antenatal anesthesia assessment of high-risk obstetrical patients. OBJECTIVES: The objective of this study is to determine if the quality of antenatal anesthesia telehealth consults of high-risk obstetric patients is comparable to in-person encounters. This is determined by assessing if telehealth consults are feasible and meet the standards of care, as well as the level of patient satisfaction and ease of use as reported by providers. STUDY DESIGN: This retrospective study assessed patients prior to delivery who completed a video-telehealth anesthesia consultation (51 cases) from November 1st, 2019 to November 30th, 2020 and all of those for patients receiving an in-person anesthesia consultation (171 controls) from November 2017 through October 2019. Our primary hypothesis was that telehealth and in-person consultations would not result in different standards of care. The primary outcome was an indicator of meeting the standard of care, and the difference in proportions between the telehealth and in-person consultation was tested by Fisher's exact test. Our secondary hypotheses were that patients reported high levels of satisfaction and could use telehealth easily and providers could use the platform easily. Secondary outcomes were assessed by using the Consultation and Relational Empathy (CARE) and the Telehealth Usability Questionnaire (TUQ) surveys, respectively. RESULTS: For the primary outcome, 94.1% (48/51) of telehealth and 89.5% (153/171) of in-person visits met the standard of care, indicating no significant difference between groups (p-value = .4204). The CARE score was 46 [41,50] {median [interquartile range]}, (p-value < .0001), indicating patient satisfaction with telehealth. The use-average scores on the TUQ for the patient and provider were 6.67 [6.33, 7] and 6 [5.33, 7] respectively, indicating great system usability. CONCLUSION: This study demonstrates no significant difference in the standard of care between in-person and telehealth visits. Furthermore, telehealth consultation was feasible and associated with high patient satisfaction and platform usability. Preoperative consultation of high-risk obstetric patients using telehealth visits should be routinely considered in clinical practice.Condensation: There is no significant difference in the standard of care between in-person and telehealth antenatal anesthesia consultations, and patients report high satisfaction and platform usability.Telehealth is gaining popularity, but its role in antenatal anesthesia consultation of high risk obstetrical patients has not yet been defined with respect to standard of care, patient satisfaction, and platform usability.There was no significant difference in standard of care between in-person and telehealth antenatal anesthesia consultations, and patients reported high satisfaction and platform usability.Telehealth should be considered as an alternative to in-person antenatal anesthesia consultation of high risk obstetrical patients. It is a particularly attractive alternative to in-person consultation due to cost-savings, increased patient accessibility, and ease of use.

Clinically applicable approach for predicting mechanical ventilation in patients with COVID-19.

BACKGROUND: Patients with coronavirus disease 2019 (COVID-19) requiring mechanical ventilation have high mortality and resource utilisation. The ability to predict which patients may require mechanical ventilation allows increased acuity of care and targeted interventions to potentially mitigate deterioration. METHODS: We included hospitalised patients with COVID-19 in this single-centre retrospective observational study. Our primary outcome was mechanical ventilation or death within 24 h. As clinical decompensation is more recognisable, but less modifiable, as the prediction window shrinks, we also assessed 4, 8, and 48 h prediction windows. Model features included demographic information, laboratory results, comorbidities, medication administration, and vital signs. We created a Random Forest model, and assessed performance using 10-fold cross-validation. The model was compared with models derived from generalised estimating equations using discrimination. RESULTS: Ninety-three (23%) of 398 patients required mechanical ventilation or died within 14 days of admission. The Random Forest model predicted pending mechanical ventilation with good discrimination (C-statistic=0.858; 95% confidence interval, 0.841-0.874), which is comparable with the discrimination of the generalised estimating equation regression. Vitals sign data including SpO2/FiO2 ratio (Random Forest Feature Importance Z-score=8.56), ventilatory frequency (5.97), and heart rate (5.87) had the highest predictive utility. In our highest-risk cohort, the number of patients needed to identify a single new case was 3.2, and for our second quintile it was 5.0. CONCLUSION: Machine learning techniques can be leveraged to improve the ability to predict which patients with COVID-19 are likely to require mechanical ventilation, identifying unrecognised bellwethers and providing insight into the constellation of accompanying signs of respiratory failure in COVID-19.