Should Transport Ventilators Be Used in Times of Crisis? The Use of Emergency Authorized Nonconventional Ventilators Is Associated With Mortality Among Patients With COVID-19 Acute Respiratory Distress Syndrome.

OBJECTIVES: Nonconventional ventilators (NCVs), defined here as transport ventilators and certain noninvasive positive pressure devices, were used extensively as crisis-time ventilators for intubated patients with COVID-19. We assessed whether there was an association between the use of NCV and higher mortality, independent of other factors. DESIGN: This is a multicenter retrospective observational study. SETTING: The sample was recruited from a single healthcare system in New York. The recruitment period spanned from March 1, 2020, to April 30, 2020. PATIENTS: The sample includes patients who were intubated for COVID-19 acute respiratory distress syndrome (ARDS). INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The primary outcome was 28-day in-hospital mortality. Multivariable logistic regression was used to derive the odds of mortality among patients managed exclusively with NCV throughout their ventilation period compared with the remainder of the sample while adjusting for other factors. A secondary analysis was also done, in which the mortality of a subset of the sample exclusively ventilated with NCV was compared with that of a propensity score-matched subset of the control group. Exclusive use of NCV was associated with a higher 28-day in-hospital mortality while adjusting for confounders in the regression analysis (odds ratio, 1.41; 95% CI [1.07-1.86]). In the propensity score matching analysis, the mortality of patients exclusively ventilated with NCV was 68.9%, and that of the control was 60.7% ( p = 0.02). CONCLUSIONS: Use of NCV was associated with increased mortality among patients with COVID-19 ARDS. More lives may be saved during future ventilator shortages if more full-feature ICU ventilators, rather than NCVs, are reserved in national and local stockpiles.

Application of the RE-AIM framework to evaluate the implementation of telehealth pulmonary rehabilitation in a randomized controlled trial among African-American and Hispanic patients with advanced stage Chronic Obstructive Pulmonary Disease.

BACKGROUND: Pulmonary rehabilitation (PR) decreases rehospitalization for people with COPD. However, less than 2% receive PR, partly due to lack of referral and sparsity of PR facilities. This disparity is particularly pronounced in African American and Hispanic persons with COPD. Telehealth-provided PR could increase access and improve health outcomes. METHODS: We applied the RE-AIM framework in a post-hoc analysis of our mixed methods RCT comparing referral to Telehealth-delivered PR (TelePR) versus standard PR (SPR) for African American and Hispanic COPD patients hospitalized for COPD exacerbation. Both arms received a referral to PR for 8 weeks, social worker follow-up, and surveys administered at baseline, 8 weeks, 6, and 12 months. PR sessions were conducted twice a week for 90 min each (16 sessions total). Quantitative data were analyzed using 2-sample t tests or nonparametric Wilcoxon tests for continuous data and χ2/Fisher exact tests for categorical data. Logistic regression-estimated odds ratios (ORs) were used for the intention-to-treat primary outcome. Qualitative interviews were conducted at the end of the study to assess adherence and satisfaction and were analyzed using inductive and deductive methods. The goal was to understand Reach (whether the target population was able to be enrolled), Effectiveness (primary outcome was a composite of 6-month COPD rehospitalization and death), Adoption (proportion of people willing to initiate the program), Implementation (whether the program was able to be executed as intended, and Maintenance (whether the program was continued). RESULTS: Two hundred nine people enrolled out of a 276-recruitment goal. Only 85 completed at least one PR session 57/111 (51%) TelePR; 28/98 (28%) SPR. Referral to TelePR compared to SPR did not decrease the composite outcome of 6-month COPD-readmission rate/death (OR1.35;95%CI 0.69,2.66). There was significant reduction in fatigue (PROMIS® scale) from baseline to 8-weeks in TelePR compared to SPR (MD-1.34; ± SD4.22; p = 0.02). Participants who received TelePR experienced improvements from baseline in several outcomes (ie, before and after 8 weeks of PR) in the following: COPD symptoms, knowledge about COPD management, fatigue, and functional capacity. Among the patients who had 1 initial visit, adherence rates were similar (TelePR arm, 59% of sessions; SPR arm, 63%). No intervention-related adverse events occurred. Barriers to PR adoption included difficulty or reluctance to complete medical clearances and beliefs about PR efficacy. Notably, only 9 participants sustained exercise after program completion. Maintenance of the program was not possible due to low insurance reimbursement and sparsity of Respiratory Therapists. CONCLUSIONS: TelePR can reach COPD patients with health disparities and can be successfully implemented. The small sample size and large confidence intervals prevent conclusion about the relative effectiveness of participating in TelePR compared to SPR. However, improved outcomes were seen for those in TelePR as well as in SPR. Increasing adoption of PR and TelePR requires consideration of comorbidity burden, and perception of PR utility, and must facilitate medical clearances. Given the sparsity of SPR locations, TelePR can overcome at least the barrier of access. However, given the challenges to the uptake and completion of PR - many of the additional barriers in PR (both in TelePR and SPR) need to be addressed. Awareness of these real-world challenges will not only inform implementation of TelePR for clinicians seeking to adopt this platform but will also inform study designers and reviewers regarding the feasibility of approaches to patient recruitment and retention.

Efficacy and Safety of Sarilumab in Hospitalized Patients With Coronavirus Disease 2019: A Randomized Clinical Trial.

BACKGROUND: Open-label platform trials and a prospective meta-analysis suggest efficacy of anti-interleukin (IL)-6R therapies in hospitalized patients with coronavirus disease 2019 (COVID-19) receiving corticosteroids. This study evaluated the efficacy and safety of sarilumab, an anti-IL-6R monoclonal antibody, in the treatment of hospitalized patients with COVID-19. METHODS: In this adaptive, phase 2/3, randomized, double-blind, placebo-controlled trial, adults hospitalized with COVID-19 received intravenous sarilumab 400 mg or placebo. The phase 3 primary analysis population included patients with critical COVID-19 receiving mechanical ventilation (MV). The primary outcome was proportion of patients with ≥1-point improvement in clinical status from baseline to day 22. RESULTS: There were 457 and 1365 patients randomized and treated in phases 2 and 3, respectively. In phase 3, patients with critical COVID-19 receiving MV (n = 298; 28.2% on corticosteroids), the proportion with ≥1-point improvement in clinical status (alive, not receiving MV) at day 22 was 43.2% for sarilumab and 35.5% for placebo (risk difference, +7.5%; 95% confidence interval [CI], -7.4 to 21.3; P =.3261), a relative risk improvement of 21.7%. In post hoc analyses pooling phase 2 and 3 critical patients receiving MV, the hazard ratio for death for sarilumab vs placebo was 0.76 (95% CI, .51 to 1.13) overall and 0.49 (95% CI, .25 to .94) in patients receiving corticosteroids at baseline. CONCLUSIONS: This study did not establish the efficacy of sarilumab in hospitalized patients with severe/critical COVID-19. Post hoc analyses were consistent with other studies that found a benefit of sarilumab in patients receiving corticosteroids. CLINICAL TRIALS REGISTRATION: NCT04315298.

Prognostic models for COVID-19 needed updating to warrant transportability over time and space.

BACKGROUND: Supporting decisions for patients who present to the emergency department (ED) with COVID-19 requires accurate prognostication. We aimed to evaluate prognostic models for predicting outcomes in hospitalized patients with COVID-19, in different locations and across time. METHODS: We included patients who presented to the ED with suspected COVID-19 and were admitted to 12 hospitals in the New York City (NYC) area and 4 large Dutch hospitals. We used second-wave patients who presented between September and December 2020 (2137 and 3252 in NYC and the Netherlands, respectively) to evaluate models that were developed on first-wave patients who presented between March and August 2020 (12,163 and 5831). We evaluated two prognostic models for in-hospital death: The Northwell COVID-19 Survival (NOCOS) model was developed on NYC data and the COVID Outcome Prediction in the Emergency Department (COPE) model was developed on Dutch data. These models were validated on subsequent second-wave data at the same site (temporal validation) and at the other site (geographic validation). We assessed model performance by the Area Under the receiver operating characteristic Curve (AUC), by the E-statistic, and by net benefit. RESULTS: Twenty-eight-day mortality was considerably higher in the NYC first-wave data (21.0%), compared to the second-wave (10.1%) and the Dutch data (first wave 10.8%; second wave 10.0%). COPE discriminated well at temporal validation (AUC 0.82), with excellent calibration (E-statistic 0.8%). At geographic validation, discrimination was satisfactory (AUC 0.78), but with moderate over-prediction of mortality risk, particularly in higher-risk patients (E-statistic 2.9%). While discrimination was adequate when NOCOS was tested on second-wave NYC data (AUC 0.77), NOCOS systematically overestimated the mortality risk (E-statistic 5.1%). Discrimination in the Dutch data was good (AUC 0.81), but with over-prediction of risk, particularly in lower-risk patients (E-statistic 4.0%). Recalibration of COPE and NOCOS led to limited net benefit improvement in Dutch data, but to substantial net benefit improvement in NYC data. CONCLUSIONS: NOCOS performed moderately worse than COPE, probably reflecting unique aspects of the early pandemic in NYC. Frequent updating of prognostic models is likely to be required for transportability over time and space during a dynamic pandemic.

Trajectories of hypoxemia and pulmonary mechanics of COVID-19 ARDS in the NorthCARDS dataset.

BACKGROUND: Understanding heterogeneity seen in patients with COVIDARDS and comparing to non-COVIDARDS may inform tailored treatments. METHODS: A multidisciplinary team of frontline clinicians and data scientists worked to create the Northwell COVIDARDS dataset (NorthCARDS) leveraging over 11,542 COVID-19 hospital admissions. The data was then summarized to examine descriptive differences based on clinically meaningful categories of lung compliance, and to examine trends in oxygenation. FINDINGS: Of the 1536 COVIDARDS patients in the NorthCARDS dataset, there were 531 (34.6%) who had very low lung compliance (< 20 ml/cmH2O), 970 (63.2%) with low-normal compliance (20-50 ml/cmH2O), and 35 (2.2%) with high lung compliance (> 50 ml/cmH2O). The very low compliance group had double the median time to intubation compared to the low-normal group (107.3 h (IQR 25.8, 239.2) vs. 39.5 h (IQR 5.4, 91.6)). Overall, 68.8% (n = 1057) of the patients died during hospitalization. In comparison to non-COVIDARDS reports, there were less patients in the high compliance category (2.2% vs. 12%, compliance ≥ 50 mL/cmH20), and more patients with P/F ≤ 150 (59.8% vs. 45.6%). There is a statistically significant correlation between compliance and P/F ratio. The Oxygenation Index is the highest in the very low compliance group (12.51, SD(6.15)), and lowest in high compliance group (8.78, SD(4.93)). CONCLUSIONS: The respiratory system compliance distribution of COVIDARDS is similar to non-COVIDARDS. In some patients, there may be a relation between time to intubation and duration of high levels of supplemental oxygen treatment on trajectory of lung compliance.

The Value of Thromboelastography (TEG) in COVID-19 Critical Illness as Illustrated by a Case Series.

OBJECTIVE: To develop a practical thromboelastograph guided (TEG) anticoagulation protocol to guide the management of COVID-19 critically ill patients. DESIGN: An inter disciplinary team reviewed the current literature on hypercoagulability in critically ill COVID-19 patients, clinical management practices and challenges with high rates of thrombotic events despite anticoagulant therapies. SETTING: The largest tertiary care hospital within the Northwell Health System in New York. PATIENTS: COVID-19 invasively mechanically ventilated patients in Medical Intensive Care Unit Settings. METHODS: TEG was monitored in critically ill COVID-19 patients. Patterns were reviewed to guide the development of a treatment protocol leveraging TEG parameters to select anticoagulant therapy. Three patients are reported to highlight TEG profiles that led to the development of the algorithm. Clinical trajectory and treatment decisions were extracted retrospectively from the Electronic Health Record, with input from the intensivists. Anticoagulant use, laboratory and TEG values, and venous/arterial lower extremity (LE) ultrasound results were recorded. MAIN RESULTS: These patients demonstrated hypercoagulable TEG results despite prophylactic or therapeutic dosages of unfractionated heparin or low-molecular-weight heparin (LMHW). TEG surveillance identified functional fibrinogen and maximum amplitude in high-risk patients with hyper inflammatory markers. Anticoagulation assessment, TEG parameters, and LE ultrasound monitoring for venous and arterial thrombus were used to construct an algorithm to guide and escalate anticoagulant therapy. CONCLUSIONS: TEG provides patient-specific evidence for a hypercoagulable state in patients receiving all types of anticoagulant therapy. The proposed TEG algorithm guides anticoagulation management decisions to maintain or escalate anticoagulant dose and/or change choice of anticoagulant. A TEG algorithm may help negotiate the potential harm/benefit balance of full-dose anticoagulation in critically ill COVID-19 patients, by allowing for a more individualized approach that goes beyond the review of activated partial thromboplastin time (aPTT) levels.

Using community priorities and misconceptions about asthma as a vessel for community-led education among Hispanics.

OBJECTIVES: In New York City, asthma prevalence is greater in Hispanics than non-Hispanics for both children (10.9% vs. 7.4%) and adults (9.0% vs. 6.3%). Disparities in asthma management among Hispanics are found to arise, in part, from a limited education about asthma. Using elements of Community Based Participatory Research (CBPR), we worked with the community to identify asthma priorities and misconceptions among Hispanics and used that information to develop a tailored asthma educational tool-the Asthma Training Modules (ATMs). METHODS: Over the past 3 years (2016, 2017, and 2018), we conducted educational asthma workshops to collect and analyze information to develop the ATMs and a summary of the ATMs in an Asthma Educational Card (AEC). We trained 6 Asthma-Community-Leaders using the ATMs, who assembled community members for teaching sessions using the AEC. Participants completed a pre-and-post asthma knowledge questionnaire. RESULTS: We identified asthma priorities and misconceptions themed on: culturally relevant resources for Hispanics, symptom and trigger recognition, and treatments. A total of 104 participants attended the teaching sessions led by Asthma-Community-Leaders and participants' mean knowledge score increased from 64% pre-teaching to 85% post-teaching, (p < 0.01). CONCLUSION: Our community-led education, which included a tailored asthma educational tool and trained Asthma-Community-Leaders, successfully improved asthma knowledge among Hispanics. Further studies are warranted to determine whether these results are reproducible among a larger cohort and what the comparative effectiveness of our intervention as compared to other education-based interventions.

Disparities in complementary alternative medicine use and asthma exacerbation in the United States.

Background: Complementary and alternative medicines (CAM) are associated with poor asthma medication adherence, a major risk factor for asthma exacerbation. However, previous studies showed inconsistent relationships between CAM use and asthma control due to small sample sizes, demographic differences across populations studied, and poor differentiation of CAM types.Methods: We examined associations between CAM use and asthma exacerbation using a cross-sectional analysis of the 2012 National Health Interview Survey. We included adults ≥18 years with current asthma (n = 2,736) to analyze racial/ethnic differences in CAM use as well as the association between CAM use and both asthma exacerbation and emergency department (ED) visit for asthma exacerbation across racial/ethnic groups. We ran descriptive statistics and multivariable logistic regressions.Result: Blacks (OR = 0.63 [0.49-0.81]) and Hispanics (OR = 0.66 [0.48-0.92]) had decreased odds of using CAM compared to Whites. Overall, there was no association between CAM use and asthma exacerbation (OR = 0.99 [0.79-1.25]) but the subgroup of 'other complementary approaches' was associated with increased odds of asthma exacerbation among all survey respondents (1.90 [1.21-2.97]), Whites (OR = 1.90 [1.21-2.97]), and Hispanics (OR = 1.43 [0.98-2.09). CAM use was associated with decreased odds of an ED visit for asthma exacerbation (OR = 0.65 [0.45-0.93]). These associations were different among racial/ethnic groups with decreased odds of ED visit among Whites (OR = 0.50 [0.32-0.78]) but no association among Blacks and Hispanics.Conclusion: We found that both CAM use and the association between CAM use and asthma exacerbation varied by racial/ethnic group. The different relationship may arise from how CAM is used to complement or to substitute for conventional asthma management.

Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area.

Importance: There is limited information describing the presenting characteristics and outcomes of US patients requiring hospitalization for coronavirus disease 2019 (COVID-19). Objective: To describe the clinical characteristics and outcomes of patients with COVID-19 hospitalized in a US health care system. Design, Setting, and Participants: Case series of patients with COVID-19 admitted to 12 hospitals in New York City, Long Island, and Westchester County, New York, within the Northwell Health system. The study included all sequentially hospitalized patients between March 1, 2020, and April 4, 2020, inclusive of these dates. Exposures: Confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection by positive result on polymerase chain reaction testing of a nasopharyngeal sample among patients requiring admission. Main Outcomes and Measures: Clinical outcomes during hospitalization, such as invasive mechanical ventilation, kidney replacement therapy, and death. Demographics, baseline comorbidities, presenting vital signs, and test results were also collected. Results: A total of 5700 patients were included (median age, 63 years [interquartile range {IQR}, 52-75; range, 0-107 years]; 39.7% female). The most common comorbidities were hypertension (3026; 56.6%), obesity (1737; 41.7%), and diabetes (1808; 33.8%). At triage, 30.7% of patients were febrile, 17.3% had a respiratory rate greater than 24 breaths/min, and 27.8% received supplemental oxygen. The rate of respiratory virus co-infection was 2.1%. Outcomes were assessed for 2634 patients who were discharged or had died at the study end point. During hospitalization, 373 patients (14.2%) (median age, 68 years [IQR, 56-78]; 33.5% female) were treated in the intensive care unit care, 320 (12.2%) received invasive mechanical ventilation, 81 (3.2%) were treated with kidney replacement therapy, and 553 (21%) died. As of April 4, 2020, for patients requiring mechanical ventilation (n = 1151, 20.2%), 38 (3.3%) were discharged alive, 282 (24.5%) died, and 831 (72.2%) remained in hospital. The median postdischarge follow-up time was 4.4 days (IQR, 2.2-9.3). A total of 45 patients (2.2%) were readmitted during the study period. The median time to readmission was 3 days (IQR, 1.0-4.5) for readmitted patients. Among the 3066 patients who remained hospitalized at the final study follow-up date (median age, 65 years [IQR, 54-75]), the median follow-up at time of censoring was 4.5 days (IQR, 2.4-8.1). Conclusions and Relevance: This case series provides characteristics and early outcomes of sequentially hospitalized patients with confirmed COVID-19 in the New York City area.

Potential return on investment of a family-centered early childhood intervention: a cost-effectiveness analysis.

BACKGROUND: ParentCorps is a family-centered enhancement to pre-kindergarten programming in elementary schools and early education centers. When implemented in high-poverty, urban elementary schools serving primarily Black and Latino children, it has been found to yield benefits in childhood across domains of academic achievement, behavior problems, and obesity. However, its long-term cost-effectiveness is unknown. METHODS: We determined the cost-effectiveness of ParentCorps in high-poverty, urban schools using a Markov Model projecting the long-term impact of ParentCorps compared to standard pre-kindergarten programming. We measured costs and quality adjusted life years (QALYs) resulting from the development of three disease states (i.e., drug abuse, obesity, and diabetes); from the health sequelae of these disease states; from graduation from high school; from interaction with the judiciary system; and opportunity costs of unemployment with a lifetime time horizon. The model was built, and analyses were performed in 2015-2016. RESULTS: ParentCorps was estimated to save $4387 per individual and increase each individual's quality adjusted life expectancy by 0.27 QALYs. These benefits were primarily due to the impact of ParentCorps on childhood obesity and the subsequent predicted prevention of diabetes, and ParentCorps' impact on childhood behavior problems and the subsequent predicted prevention of interaction with the judiciary system and unemployment. Results were robust on sensitivity analyses, with ParentCorps remaining cost saving and health generating under nearly all assumptions, except when schools had very small pre-kindergarten programs. CONCLUSIONS: Effective family-centered interventions early in life such as ParentCorps that impact academic, behavioral and health outcomes among children attending high-poverty, urban schools have the potential to result in longer-term health benefits and substantial cost savings.

What happens to patients with COPD with long-term oxygen treatment who receive mechanical ventilation for COPD exacerbation? A 1-year retrospective follow-up study.

We performed a retrospective cohort study of patients with chronic obstructive lung disease (COPD) on long-term oxygen treatment (LTOT) who received invasive mechanical ventilation for COPD exacerbation. Of the 4791 patients, 23% died in the hospital, and 45% died in the subsequent 12 months. 67% of patients were readmitted at least once in the subsequent 12 months, and 26.8% were discharged to a nursing home or skilled nursing facility within 30 days. We conclude that these patients have high mortality rates, both in-hospital and in the 12 months postdischarge. If patients survive, many will be readmitted to the hospital and discharged to nursing home. These potential outcomes may support informed critical care decision making and more preference congruent care.

Understanding patients' and doctors' attitudes about shared decision making for advance care planning.

BACKGROUND: Although shared decision making (SDM) is the preferred model of making complex treatment decisions with patients, patients' and doctors' attitudes towards SDM for advance care planning are unknown. OBJECTIVE: We sought to: (i) gain general insights into the current practice of SDM and attitudes about patient involvement, and (ii) gain specific insights into experience with, and attitudes about, SDM for advance care planning. DESIGN: Qualitative analysis of face-to-face semi-structured interviews. SETTING AND PARTICIPANTS: Patients with chronic lung disease and their doctors at a New York City public hospital. RESULTS: Although patients described participation in decision making, many deferred the final decision to their doctors. Doctors indicated a preference for SDM but expressed barriers including perceived lack of patient understanding and lack of patient empowerment. With regard to end-of-life discussions, patients were generally open to having these discussions with their doctors, although their openness sometimes depended on the circumstance (i.e. end-of-life discussions may be more acceptable to patients for whom the chance of dying is high). Doctors reported engaging in end-of-life treatment decisions with their patients, although expressed the need for conversations to take place earlier, in advance of acute illness, and identified a lack of prognostic estimates as one barrier to engaging in this discussion. CONCLUSIONS: Doctors should explore their patients' attitudes regarding end-of-life discussions and preferences for decision-making styles. There is a need for tools such as decision aids which can empower patients to participate in decision making and can support doctors with prognostic estimates pertinent to individual patients.

Personalized estimates of benefit from preventive care guidelines: a proof of concept.

BACKGROUND: The U.S. Preventive Services Task Force (USPSTF) makes recommendations for 60 distinct clinical services, but clinicians rarely have time to fully evaluate and implement the recommendations. OBJECTIVE: To complete a proof of concept for prioritization and personalization of USPSTF recommendations, using patient-specific clinical characteristics. DESIGN: Mathematical model. DATA SOURCES: USPSTF recommendations and supporting evidence and National Vital Statistics Reports. TARGET POPULATION: Nonpregnant adults. TIME HORIZON: Lifetime. PERSPECTIVE: Individual. INTERVENTION: USPSTF grade A and B recommendations. OUTCOME MEASURES: Personalized gain in life expectancy associated each recommendation. RESULTS OF BASE-CASE ANALYSIS: Increases in life expectancy varied more than 100-fold across USPSTF recommendations, and the rank order of benefits varied considerably among patients. For an obese man aged 62 years who smoked and had hypercholesterolemia, hypertension, and a family history of colorectal cancer, the model's top 3 recommendations (from most to least gain in life expectancy) were tobacco cessation (adding 2.8 life-years), weight loss (adding 1.6 life-years), and blood pressure control (adding 0.8 life-year). Lower-ranked recommendations were a healthier diet, aspirin use, cholesterol reduction, colonoscopy, screening for abdominal aortic aneurysm, and HIV testing (each adding 0.1 to 0.3 life-years). For a person with the same characteristics plus uncontrolled type 2 diabetes mellitus, the model's top 3 recommendations were diabetes control, tobacco cessation, and weight loss (each adding 1.4 to 1.8 life-years). RESULTS OF SENSITIVITY ANALYSIS: Robust to variation of model inputs and satisfied face validity criteria. LIMITATION: Expected adherence rates and quality of life were not considered. CONCLUSION: Models of personalized preventive care may illustrate how magnitude and rank order of benefit associated with preventive guidelines vary across recommendations and patients. These predictions may help clinicians to prioritize USPSTF recommendations at the patient level.

Drivers of mortality in COVID ARDS depend on patient sub-type.

The most common cause of death in people with COVID-19 is Acute Respiratory Distress Syndrome (ARDS). Prior studies have demonstrated that ARDS is a heterogeneous syndrome and have identified ARDS sub-types (phenoclusters). However, non-COVID-19 ARDS phenoclusters do not clearly apply to COVID-19 ARDS patients. In this retrospective cohort study, we implemented an iterative approach, combining supervised and unsupervised machine learning methodologies, to identify clinically relevant COVID-19 ARDS phenoclusters, as well as characteristics that are predictive of the outcome for each phenocluster. To this end, we applied a supervised model to identify risk factors for hospital mortality for each phenocluster and compared these between phenoclusters and the entire cohort. We trained the models using a comprehensive, preprocessed dataset of 2,864 hospitalized COVID-19 ARDS patients. Our research demonstrates that the risk factors predicting mortality in the overall cohort of COVID-19 ARDS may not necessarily apply to specific phenoclusters. Additionally, some risk factors increase the risk of hospital mortality in some phenoclusters but decrease mortality in others. These phenocluster-specific risk factors would not have been observed with a single predictive model. Heterogeneity in phenoclusters of COVID-19 ARDS as well as the drivers of mortality may partially explain challenges in finding effective treatments for all patients with ARDS.

Discriminating Acute Respiratory Distress Syndrome from other forms of respiratory failure via iterative machine learning.

Acute Respiratory Distress Syndrome (ARDS) is associated with high morbidity and mortality. Identification of ARDS enables lung protective strategies, quality improvement interventions, and clinical trial enrolment, but remains challenging particularly in the first 24 hours of mechanical ventilation. To address this we built an algorithm capable of discriminating ARDS from other similarly presenting disorders immediately following mechanical ventilation. Specifically, a clinical team examined medical records from 1263 ICU-admitted, mechanically ventilated patients, retrospectively assigning each patient a diagnosis of "ARDS" or "non-ARDS" (e.g., pulmonary edema). Exploiting data readily available in the clinical setting, including patient demographics, laboratory test results from before the initiation of mechanical ventilation, and features extracted by natural language processing of radiology reports, we applied an iterative pre-processing and machine learning framework. The resulting model successfully discriminated ARDS from non-ARDS causes of respiratory failure (AUC = 0.85) among patients meeting Berlin criteria for severe hypoxia. This analysis also highlighted novel patient variables that were informative for identifying ARDS in ICU settings.

US and Dutch Perspectives on the Use of COVID-19 Clinical Prediction Models: Findings from a Qualitative Analysis.

INTRODUCTION: Clinical prediction models (CPMs) for coronavirus disease 2019 (COVID-19) may support clinical decision making, treatment, and communication. However, attitudes about using CPMs for COVID-19 decision making are unknown. METHODS: Online focus groups and interviews were conducted among health care providers, survivors of COVID-19, and surrogates (i.e., loved ones/surrogate decision makers) in the United States and the Netherlands. Semistructured questions explored experiences about clinical decision making in COVID-19 care and facilitators and barriers for implementing CPMs. RESULTS: In the United States, we conducted 4 online focus groups with 1) providers and 2) surrogates and survivors of COVID-19 between January 2021 and July 2021. In the Netherlands, we conducted 3 focus groups and 4 individual interviews with 1) providers and 2) surrogates and survivors of COVID-19 between May 2021 and July 2021. Providers expressed concern about CPM validity and the belief that patients may interpret CPM predictions as absolute. They described CPMs as potentially useful for resource allocation, triaging, education, and research. Several surrogates and people who had COVID-19 were not given prognostic estimates but believed this information would have supported and influenced their decision making. A limited number of participants felt the data would not have applied to them and that they or their loved ones may not have survived, as poor prognosis may have suggested withdrawal of treatment. CONCLUSIONS: Many providers had reservations about using CPMs for people with COVID-19 due to concerns about CPM validity and patient-level interpretation of the outcome predictions. However, several people who survived COVID-19 and their surrogates indicated that they would have found this information useful for decision making. Therefore, information provision may be needed to improve provider-level comfort and patient and surrogate understanding of CPMs. HIGHLIGHTS: While clinical prediction models (CPMs) may provide an objective means of assessing COVID-19 prognosis, provider concerns about CPM validity and the interpretation of CPM predictions may limit their clinical use.Providers felt that CPMs may be most useful for resource allocation, triage, research, or educational purposes for COVID-19.Several survivors of COVID-19 and their surrogates felt that CPMs would have been informative and may have aided them in making COVID-19 treatment decisions, while others felt the data would not have applied to them.

Informing shared decisions about advance directives for patients with severe chronic obstructive pulmonary disease: a modeling approach.

OBJECTIVE: To estimate the survival and quality-adjusted life-years (QALYs) of Full Code versus Do Not Intubate (DNI) advance directives in patients with severe chronic obstructive pulmonary disease and to evaluate how patient preferences and place of residence influence these outcomes. METHODS: A Markov decision model using published data for COPD exacerbation outcomes. The advance directives that were modeled were as follows: DNI, allowing only noninvasive mechanical ventilation, or Full Code, allowing all forms of mechanical ventilation including invasive mechanical ventilation with endotracheal tube (ETT) insertion. RESULTS: In community-dwellers, Full Code resulted in a greater likelihood of survival and higher QALYs (4-year survival: 23% Full Code, 18% DNI; QALYs: 1.34 Full Code, 1.24 DNI). When considering patient preferences regarding complications, however, if patients were willing to give up >3 months of life expectancy to avoid ETT complications, or >1 month of life expectancy to avoid long-term institutionalization, DNI resulted in higher QALYs. For patients in long-term institutions, DNI resulted in a greater likelihood of survival and higher QALYs (4-year survival: 2% DNI, 1% Full Code; QALYs: 0.29 DNI, 0.24 Full Code). In sensitivity analyses, the model was sensitive to the probabilities of ETT complication and noninvasive mechanical ventilation failure and to patient preferences about ETT complications and long-term institutionalization. CONCLUSION: Our model demonstrates that patient preferences regarding ETT complications and long-term institutionalization, as well as baseline place of residence, affect the advance directive recommendation when considered in terms of both survival and QALYs. Decision modeling can demonstrate the potential trade-off between survival and quality of life, using patient preferences and disease-specific data, to inform the shared advance directive decision.