Construction of a nomogram for predicting COVID-19 in-hospital mortality: A machine learning analysis.

Background and objectives: We aim to verify the use of ML algorithms to predict patient outcome using a relatively small dataset and to create a nomogram to assess in-hospital mortality of patients with COVID-19. Methods: A database of 200 COVID-19 patients admitted to the Clinical Hospital of State University of Campinas (UNICAMP) was used in this analysis. Patient features were divided into three categories: clinical, chest abnormalities, and body composition characteristics acquired by computerized tomography. These features were evaluated independently and combined to predict patient outcomes. To minimize performance fluctuations due to low sample number, reduce possible bias related to outliers, and evaluate the uncertainties generated by the small dataset, we developed a shuffling technique, a modified version of the Monte Carlo Cross Validation, creating several subgroups for training the algorithm and complementary testing subgroups. The following ML algorithms were tested: random forest, boosted decision trees, logistic regression, support vector machines, and neural networks. Performance was evaluated by analyzing Receiver operating characteristic (ROC) curves. The importance of each feature in the determination of the outcome predictability was also studied and a nomogram was created based on the most important features selected by the exclusion test. Results: Among the different sets of features, clinical variables age, lymphocyte number and weight were the most valuable features for prognosis prediction. However, we observed that skeletal muscle radiodensity and presence of pleural effusion were also important for outcome determination. Integrating these independent predictors was successfully developed to accurately predict mortality in COVID-19 in hospital patients. A nomogram based on these five features was created to predict COVID-19 mortality in hospitalized patients. The area under the ROC curve was 0.86 ± 0.04. Conclusion: ML algorithms can be reliable for the prediction of COVID-19-related in-hospital mortality, even when using a relatively small dataset. The success of ML techniques in smaller datasets broadens the applicability of these methods in several problems in the medical area. In addition, feature importance analysis allowed us to determine the most important variables for the prediction tasks resulting in a nomogram with good accuracy and clinical utility in predicting COVID-19 in-hospital mortality.

Patient perception of Breezhaler® and Ellipta® device feedback mechanisms in COPD: The ADVANTAGE Study.

OBJECTIVES: The primary objective of the ADVANTAGE study was to compare device-naïve chronic obstructive pulmonary disease (COPD) patients' perception of the Breezhaler® and Ellipta® devices' feedback mechanisms of dose delivery confirmation. The secondary objective was to assess comfort with the inhalers' mouthpiece in terms of ease to form a tight seal around the mouthpiece. These objectives were achieved by using a novel, patient perception of inhaler questionnaire developed and tested during cognitive interviews of patients by Evidera, London, United Kingdom. METHODS: Ten COPD patients were interviewed to collect feedback on the interpretation, relevance and language of the questionnaire. This questionnaire was then used in ADVANTAGE to compare patients' perception (n = 100) of both devices. Patients completed the questionnaire after a single inhalation of placebo through each inhaler. RESULTS: Using the final questionnaire, patients reported being more confident of the feedback mechanism of Breezhaler than that of the Ellipta device (mean score 4.3 versus 3.6 respectively, estimated difference [95% CI]: 0.75 [0.51, 0.99], p < .0001). Patients also reported better comfort (ease to form a tight seal with the lips) with the Breezhaler mouthpiece than the Ellipta mouthpiece (mean score 4.3 versus 3.9 respectively, estimated difference [95% CI]: 0.41 [0.21, 0.61], p < .0001). There were no safety concerns associated with either device. CONCLUSION: COPD patients showed greater preference for the Breezhaler over the Ellipta inhaler for confidence of dose delivery and comfort of the mouthpiece. TRIAL REGISTRATION: The trial is registered at ClinicalTrials.gov (ClinicalTrials.gov number NCT02551224).

Development of a simple binary response questionnaire to identify airflow obstruction in a smoking population in Argentina.

The CODE questionnaire (COPD detection questionnaire), a simple, binary response scale (yes/no), screening questionnaire, was developed for the identification of patients with chronic obstructive pulmonary disease (COPD). We conducted a survey of 468 subjects with a smoking history in 10 public hospitals in Argentina. Patients with a previous diagnosis of COPD, asthma and other respiratory illness were excluded. Items that measured conceptual domains in terms of characteristics of symptoms, smoking history and demographics data were considered. 96 (20.5%) subjects had a diagnosis of COPD according to the 2010 Global Initiative for Chronic Obstructive Lung Disease strategy document. The variables selected for the final questionnaire were based on univariate and multivariate analyses and clinical criteria. Finally, we selected the presence or absence of six variables (age ≥50 years, smoking history ≥30 pack-years, male sex, chronic cough, chronic phlegm and dyspnoea). Of patients without any of these six variables (0 points), none had COPD. The ability of the CODE questionnaire to discriminate between subjects with and without COPD was good (the area under the receiver operating characteristic curve was 0.75). Higher scores were associated with a greater probability of COPD. The CODE questionnaire is a brief, accurate questionnaire that can identify smoking individuals likely to have COPD.