Accuracy versus reliability-based modelling approaches for medical decision making.

Forecasting in the medical domain is critical to the quality of decisions made by physicians, patients, and health planners. Modeling is one of the most important components of decision support systems, which are frequently used to simulate and analyze under-studied systems in order to make more appropriate decisions in medical science. In the medical modeling literature, various approaches with varying structures and characteristics have been proposed to cover a wide range of application categories and domains. Regardless of the differences between modeling approaches, all of them aim to maximize the accuracy or reliability of the results in order to achieve the most generalizable model and, as a result, a higher level of profitability decisions. Despite the theoretical significance and practical impact of reliability on generalizability, particularly in high-risk decisions and applications, a significant number of models in the fields of medical forecasting, classification, and time series prediction have been developed to maximize accuracy in mind. In other words, given the volatility of medical variables, it is also necessary to have stable and reliable forecasts in order to make sound decisions. The quality of medical decisions resulting from accuracy and reliability-based intelligent and statistical modeling approaches is compared and evaluated in this paper in order to determine the relative importance of accuracy and reliability on the quality of made decisions in decision support systems. For this purpose, 33 different case studies from the UCI in three categories of supervised modeling, namely causal forecasting, time series prediction, and classification, were considered. These cases were chosen from various domains, such as disease diagnosis (obesity, Parkinson's disease, diabetes, hepatitis, stenosis of arteries, orthopedic disease, autism) and cancer (lung, breast, cervical), experiments, therapy (immunotherapy, cryotherapy), fertility prediction, and predicting the number of patients in the emergency room and ICU. According to empirical findings, the reliability-based strategy outperformed the accuracy-based strategy in causal forecasting cases by 2.26%, classification cases by 13.49%, and time series prediction cases by 3.08%. Furthermore, compared to similar accuracy-based models, the reliability-based models can generate a 6.28% improvement. As a result, they can be considered an appropriate alternative to traditional accuracy-based models for medical decision support systems modeling purposes.

Early Prediction of Severe COVID-19 in Patients by a Novel Immune-Related Predictive Model.

During the progression of coronavirus disease 2019 (COVID-19), immune response and inflammation reactions are dynamic events that develop rapidly and are associated with the severity of disease. Here, we aimed to develop a predictive model based on the immune and inflammatory response to discriminate patients with severe COVID-19. COVID-19 patients were enrolled, and their demographic and immune inflammatory reaction indicators were collected and analyzed. Logistic regression analysis was performed to identify the independent predictors, which were further used to construct a predictive model. The predictive performance of the model was evaluated by receiver operating characteristic curve, and optimal diagnostic threshold was calculated; these were further validated by 5-fold cross-validation and external validation. We screened three key indicators, including neutrophils, eosinophils, and IgA, for predicting severe COVID-19 and obtained a combined neutrophil, eosinophil, and IgA ratio (NEAR) model (NEU [109/liter] - 150×EOS [109/liter] + 3×IgA [g/liter]). NEAR achieved an area under the curve (AUC) of 0.961, and when a threshold of 9 was applied, the sensitivity and specificity of the predicting model were 100% and 88.89%, respectively. Thus, NEAR is an effective index for predicting the severity of COVID-19 and can be used as a powerful tool for clinicians to make better clinical decisions. IMPORTANCE The immune inflammatory response changes rapidly with the progression of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and is responsible for clearance of the virus and further recovery from the infection. However, the intensified immune and inflammatory response in the development of the disease may lead to more serious and fatal consequences, which indicates that immune indicators have the potential to predict serious cases. Here, we identified both eosinophils and serum IgA as prognostic markers of COVID-19, which sheds light on new research directions and is worthy of further research in the scientific research field as well as clinical application. In this study, the combination of NEU count, EOS count, and IgA level was included in a new predictive model of the severity of COVID-19, which can be used as a powerful tool for better clinical decision-making.

Severe Acute Respiratory Syndrome Coronavirus 2 Normalized Viral Loads and Subgenomic RNA Detection as Tools for Improving Clinical Decision Making and Work Reincorporation.

BACKGROUND: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) reverse-transcription polymerase chain reaction (RT-PCR) provides a highly variable cycle threshold (Ct) value that cannot distinguish viral infectivity. Subgenomic ribonucleic acid (sgRNA) has been used to monitor active replication. Given the importance of long RT-PCR positivity and the need for work reincorporation and discontinuing isolation, we studied the functionality of normalized viral loads (NVLs) for patient monitoring and sgRNA for viral infectivity detection. METHODS: The NVLs measured through the Nucleocapsid and RNA-dependent-RNA-polymerase genes and sgRNA RT-PCRs were performed in 2 consecutive swabs from 84 healthcare workers. RESULTS: The NVLs provided similar and accurate quantities of both genes of SARS-CoV-2 at 2 different timepoints of infection, overcoming Ct-value and swab collection variability. Among SARS-CoV-2-positive samples, 51.19% were sgRNA-positive in the 1st RT-PCR and 5.95% in the 2nd RT-PCR. All sgRNA-positive samples had >4 log10 RNA copies/1000 cells, whereas samples with ≤1 log10 NVLs were sgRNA-negative. Although NVLs were positive until 29 days after symptom onset, 84.1% of sgRNA-positive samples were from the first 7 days, which correlated with viral culture viability. Multivariate analyses showed that sgRNA, NVLs, and days of symptoms were significantly associated (P < .001). CONCLUSIONS: The NVLs and sgRNA are 2 rapid accessible techniques that could be easily implemented in routine hospital practice providing a useful proxy for viral infectivity and coronavirus disease 2019 patient follow-up.

Schizophrenia population health management: perspectives of and lessons learned from population health decision makers.

BACKGROUND: Despite therapeutic advances for patients with schizophrenia, improving patient outcomes and reducing the cost of care continue to challenge formulary decision makers. OBJECTIVES: To (1) understand the perspectives of formulary decision makers on challenges to optimal schizophrenia population management and (2) identify best practices and recommendations for mitigating these challenges. METHODS: This mixed-methods study, conducted in a double-blind manner, comprised in-depth telephone interviews with formulary decision makers from February through May 2020, and a web-based follow-on survey that was sent to all participants in October 2020. US-based formulary decision makers were recruited if they were directly involved in schizophrenia drug formulary or coverage decision making for national or regional payers, health systems, or behavioral health centers. Formulary decision makers' perceptions of challenges, policies, and programs related to schizophrenia population health management were assessed generally and in the context of the COVID-19 pandemic. RESULTS: 19 formulary decision makers participated in the interviews and 18 (95%) completed the survey. Participants reported a spectrum of patient- and payer-driven challenges in schizophrenia population health management, including medication nonadherence, high pharmacy and medical costs, and frequent hospitalizations and emergency department visits. Participants noted that COVID-19 had worsened all identified challenges, although patient unemployment (mean score of 2.00 on a scale of 1 [made much worse] to 5 [made much better]) and reduced access to psychiatric care (mean score, 2.12) were most negatively affected. The most common strategies implemented in order to improve schizophrenia population health management included case management (89%), telemedicine (83%), care coordination programs (72%), strategies to mitigate barriers to accessing medication (61%), and providing nonmedical services to address social determinants of health (56%). Participants noted that, ideally, all treatments for schizophrenia would be available on their formularies without utilization management policies in place in order to increase accessibility to medication, but cost to the health plans made that difficult. Whereas 61% of respondents believed that long-acting injectable antipsychotics (LAIs) were currently underused in their organizations, only 28% represented organizations with open access policies for LAIs. Participants believed that among patients with schizophrenia, LAIs were most beneficial for those with a history of poor or uncertain adherence to oral medications (mean score of 4.50 on a scale of 1 [not at all beneficial] to 5 [extremely beneficial]) and those with recurring emergency department visits and inpatient stays (mean score, 3.94). Study participants reported slightly increased use of LAIs (mean score of 3.17 on a scale of 1 [negatively impacted] to 5 [positively impacted]) among their patients with schizophrenia in response to the COVID-19 pandemic; 29% of participants reported easing access restrictions for LAIs. CONCLUSIONS: Participants described persisting challenges and various approaches intended to improve schizophrenia population health management. They also recommended strategies to optimize future health management for this population, including expanding programs to address social determinants of health and mitigating barriers to accessing treatment. DISCLOSURES: This study was funded by Janssen Scientific Affairs, LLC. Roach, Graf, Pednekar, and Chou are employees of PRECISIONheor, which received financial support from Janssen Scientific Affairs, LLC, to conduct this study. Chou owns equity in Precision Medicine Group, the parent company of PRECISIONheor. Lin and Benson are employees of Janssen Scientific Affairs, LLC. Doshi has served as a consultant, advisory board member, or both, for Acadia, Allergan, Boehringer Ingelheim, Janssen, Merck, Otsuka, and Sage Therapeutics and has received research funding from AbbVie, Biogen, Humana, Janssen, Novartis, Merck, Pfizer, PhRMA, Regeneron, Sanofi, and Valeant.

How Coronavirus Disease 2019 Changed Dermatology Practice in 1 Year Around the World: Perspectives from 11 Countries.

Coronavirus disease 2019 (COVID-19) brought the world to its knees. As each nation grappled with launching an effective response while simultaneously minimizing repercussions on health care systems, economies, and societies, the medical and scientific landscape shifted forever. In particular, COVID-19 has challenged and transformed the field of dermatology and the way we practice. In this article, dermatologists from 11 countries share insights gained from local experience. These global perspectives will help provide a better framework for delivering quality dermatologic care and understanding how the field has evolved during this medical crisis.

Treatment of moderate to severe respiratory COVID-19: a cost-utility analysis.

Despite COVID-19's significant morbidity and mortality, considering cost-effectiveness of pharmacologic treatment strategies for hospitalized patients remains critical to support healthcare resource decisions within budgetary constraints. As such, we calculated the cost-effectiveness of using remdesivir and dexamethasone for moderate to severe COVID-19 respiratory infections using the United States health care system as a representative model. A decision analytic model modelled a base case scenario of a 60-year-old patient admitted to hospital with COVID-19. Patients requiring oxygen were considered moderate severity, and patients with severe COVID-19 required intubation with intensive care. Strategies modelled included giving remdesivir to all patients, remdesivir in only moderate and only severe infections, dexamethasone to all patients, dexamethasone in severe infections, remdesivir in moderate/dexamethasone in severe infections, and best supportive care. Data for the model came from the published literature. The time horizon was 1 year; no discounting was performed due to the short duration. The perspective was of the payer in the United States health care system. Supportive care for moderate/severe COVID-19 cost $11,112.98 with 0.7155 quality adjusted life-year (QALY) obtained. Using dexamethasone for all patients was the most-cost effective with an incremental cost-effectiveness ratio of $980.84/QALY; all remdesivir strategies were more costly and less effective. Probabilistic sensitivity analyses showed dexamethasone for all patients was most cost-effective in 98.3% of scenarios. Dexamethasone for moderate-severe COVID-19 infections was the most cost-effective strategy and would have minimal budget impact. Based on current data, remdesivir is unlikely to be a cost-effective treatment for COVID-19.

Outcomes of Vascular and Endovascular Interventions Performed During the Coronavirus Disease 2019 (COVID-19) Pandemic.

OBJECTIVE: The aim of the COVER Study is to identify global outcomes and decision making for vascular procedures during the pandemic. BACKGROUND DATA: During its initial peak, there were many reports of delays to vital surgery and the release of several guidelines advising later thresholds for vascular surgical intervention for key conditions. METHODS: An international multi-center observational study of outcomes after open and endovascular interventions. RESULTS: In an analysis of 1103 vascular intervention (57 centers in 19 countries), 71.6% were elective or scheduled procedures. Mean age was 67 ±â€Š14 years (75.6% male). Suspected or confirmed COVID-19 infection was documented in 4.0%. Overall, in-hospital mortality was 11.0% [aortic interventions mortality 15.2% (23/151), amputations 12.1% (28/232), carotid interventions 10.7% (11/103), lower limb revascularisations 9.8% (51/521)]. Chronic obstructive pulmonary disease [odds ratio (OR) 2.02, 95% confidence interval (CI) 1.30-3.15] and active lower respiratory tract infection due to any cause (OR 24.94, 95% CI 12.57-241.70) ware associated with mortality, whereas elective or scheduled cases were lower risk (OR 0.4, 95% CI 0.22-0.73 and 0.60, 95% CI 0.45-0.98, respectively. After adjustment, antiplatelet (OR 0.503, 95% CI: 0.273-0.928) and oral anticoagulation (OR 0.411, 95% CI: 0.205-0.824) were linked to reduced risk of in-hospital mortality. CONCLUSIONS: Mortality after vascular interventions during this period was unexpectedly high. Suspected or confirmed COVID-19 cases were uncommon. Therefore an alternative cause, for example, recommendations for delayed surgery, should be considered. The vascular community must anticipate longer term implications for survival.

Imaging, Paternalism and the Worried Patient: Rethinking Our Approach.

The Covid pandemic has taught many lessons, including the importance of mental health. The value of the radiologist in holistic patient care may be underestimated and underresearched. Barriers to the acceptance of imaging as an important component in reassurance may be rooted in old ideas minimizing the importance of mental health.

Effect of the COVID-19 Pandemic on Delayed Skin Cancer Services.

The COVID-19 pandemic has presented a unique set of challenges to cancer care centers around the world. Diagnostic and treatment delays associated with lockdown periods may be expected to increase the total number of avoidable skin cancer deaths. During this unprecedented time, dermatologists have been pressed to balance early surgical interventions for skin cancer with the risk of viral transmission. This article summarizes evidenced-based recommendations for the surgical management of cutaneous melanoma, keratinocyte cancer, and Merkel cell carcinoma during the COVID-19 pandemic. Additional long-term studies are required to determine the effect of COVID-19 on skin cancer outcomes.

Use of alternative consultation forms in Danish general practice in the initial phase of the COVID-19 pandemic - a qualitative study.

BACKGROUND: Attempts to manage the COVID-19 pandemic have led to radical reorganisations of health care systems worldwide. General practitioners (GPs) provide the vast majority of patient care, and knowledge of their experiences with providing care for regular health issues during a pandemic is scarce. Hence, in a Danish context we explored how GPs experienced reorganising their work in an attempt to uphold sufficient patient care while contributing to minimizing the spread of COVID-19. Further, in relation to this, we examined what guided GPs' choices between telephone, video and face-to-face consultations. METHODS: This study consisted of qualitative interviews with 13 GPs. They were interviewed twice, approximately three months apart in the initial phase of the pandemic, and they took daily notes for 20 days. All interviews were audio recorded, transcribed, and inductively analysed. RESULTS: The GPs re-organised their clinical work profoundly. Most consultations were converted to video or telephone, postponed or cancelled. The use of video first rose, but soon declined, once again replaced by an increased use of face-to-face consultations. When choosing between consultation forms, the GPs took into account the need to minimise the risk of COVID-19, the central guidelines, and their own preference for face-to-face consultations. There were variations over time and between the GPs regarding which health issues were dealt with by using video and/or the telephone. For some health issues, the GPs generally deemed it acceptable to use video or telephone, postpone or cancel appointments for a short term, and in a crisis situation. They experienced relational and technical limitations with video consultation, while diagnostic uncertainty was not regarded as a prominent issue CONCLUSION: This study demonstrates how the GPs experienced telephone and video consultations as being useful in a pandemic situation when face-to-face consultations had to be severely restricted. The GPs did, however, identify several limitations similar to those known in non-pandemic times. The weighing of pros and cons and their willingness to use these alternatives shifted and generally diminished when face-to-face consultations were once again deemed viable. In case of future pandemics, such alternatives seem valuable, at least for a short term.

Swarm Learning for decentralized and confidential clinical machine learning.

Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine1,2. Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes3. However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation4,5. Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.

Decision Making and Interventions During Interfacility Transport of High-Acuity Patients With Severe Acute Respiratory Syndrome Coronavirus 2 Infection.

OBJECTIVE: There are limited data regarding the typical characteristics of coronavirus disease 2019 (COVID-19) patients requiring interfacility transport or the clinical capabilities of the out-of-hospital transport clinicians required to provide safe transport. The objective of this study is to provide epidemiologic data and highlight the clinical skill set and decision making needed to transport critically ill COVID-19 patients. METHODS: A retrospective chart review of persons under investigation for COVID-19 transported during the first 6 months of the pandemic by Johns Hopkins Lifeline was performed. Patients who required interfacility transport and tested positive for severe acute respiratory syndrome coronavirus 2 by polymerase chain reaction assay were included in the analysis. RESULTS: Sixty-eight patients (25.4%) required vasopressor support, 35 patients (13.1%) were pharmacologically paralyzed, 15 (5.60%) were prone, and 1 (0.75%) received an inhaled pulmonary vasodilator. At least 1 ventilator setting change occurred for 59 patients (22.0%), and ventilation mode was changed for 11 patients (4.10%) during transport. CONCLUSION: The safe transport of critically ill patients with COVID-19 requires experience with vasopressors, paralytic medications, inhaled vasodilators, prone positioning, and ventilator management. The frequency of initiated critical interventions and ventilator adjustments underscores the tenuous nature of these patients and highlights the importance of transport clinician reassessment, critical thinking, and decision making.

How, when, and for whom: decisions regarding remote neuropsychological assessment during the 2020 COVID-19 pandemic.

Neuropsychological assessment provides crucial information about cognitive, behavioral, and socioemotional functioning in medical, educational, legal, and social contexts. During the 2020 COVID-19 pandemic, the Israeli Ministry of Health initially mandated that all psychological assessments be postponed. However, as referrals to time-sensitive, high-need, and high-stakes assessments began to accumulate, it became necessary to consider remote solutions. In the current paper, we describe the considerations that affected the transition to remote activity in a prominent Israeli provider of neuropsychological assessment and rehabilitation services, referring to technological and environmental conditions, cognitive requirements, and tasks, as well as to legal, regulatory, and funding issues. After discussing how assessments should be conducted to maximize feasibility and validity while minimizing risks to clients and clinicians, we propose a preliminary model for deciding whether specific referrals warrant remote administration. The model delineates key factors in decisions regarding remote assessment, emphasizing the distinct roles of the referring clinician and the neuropsychologist who conducts the assessment, and highlighting the need for collaboration between them. The abrupt need for remote assessments during the pandemic required a quick response with little preparation. The lessons learned from this process can be applied in the future, so that the need for remote services can be met with greater certainty and uniformity.