Interobserver Agreement on Clinical Judgment of Work of Breathing in Spontaneously Breathing Children in the Pediatric Intensive Care Unit

RSV-CLASS -Clinical Assessment Severity Score: An easy-to-use clinical disease severity score for respiratory syncytial virus infection in hospitalized children

Respiratory syncytial virus (RSV) is the most common cause of acute respiratory tract infection in infants and young children often leading to severe disease requiring hospitalization. However, validated tools for systematic assessment of disease severity are lacking. This study aimed at creating and validating a standardized, simple-to-use disease severity score for RSV infection in children-the RSV-CLASS (Clinical Assessment Severity Score). Therefore, data from over 700 RSV-infected children over six winter seasons (2014-2020) was analyzed using univariate and multiple regression analyses for the prediction of lower respiratory tract infection (LRTI) as a proxy for a severe course of the disease. Testing a broad range of respiratory symptoms, they eventually yielded seven items. Performing stepwise selection, these were reduced to the final four items: cough, tachypnea, rales, and wheezing, each receiving one point in the proposed score named RSV-CLASS. The score was calculated for children in two cohorts A and B, one for development and one for validation, with an area under the curve of 0.90 and 0.87, respectively. With a score value of 3 or 4, 97.8% and 100% of the children, respectively, were admitted with LRTI and classified correctly. The RSV-CLASS is a disease severity score based on a neutral, analytical approach using prospective data from a large study cohort. It will contribute to systematically assessing the disease severity of RSV infection and can be used for evidence-based clinical decision-making as well as for research settings.

Enhancing Analytical Reasoning in the Intensive Care Unit

Clinical reasoning is prone to errors in judgment. Error is comprised of 2 components-bias and noise; each has an equally important role in the promulgation of error. Biases or systematic errors in reasoning are the product of misconceptions of probability and statistics. Biases arise because clinicians frequently rely on mental shortcuts or heuristics to make judgments. The most frequently used heuristics are representativeness, availability, and anchoring/adjustment which lead to the common biases of base rate neglect, misconceptions of regression, insensitivities to sample size, and fallacies of conjunctive, and disjunctive events. Bayesian reasoning is the framework within which posterior probabilities of events is identified. Familiarity with these mathematical concepts will likely enhance clinical reasoning. Noise is defined as inter or intraobserver variability in judgment that should be identical. Guidelines in medicine are a technique to reduce noise.

Interclinician agreement on the recognition of selected respiratory clinical signs in dogs and cats with abnormal breathing patterns

In humans, classification of abnormal breathing patterns (ABP) and recognition of ancillary respiratory signs are difficult, as reflected by poor-to-moderate interclinician agreement. The aims of this study were to assess interclinician agreement for respiratory sign recognition in dogs and cats and evaluate the influence of clinical experience on agreement. Dogs and cats with ABP were recruited from three hospitals. Included animals were evaluated by three clinicians at each hospital before therapeutic intervention. Consensual definitions for each respiratory clinical sign were provided to all clinicians. Interclinician agreement was measured via Fleiss' kappa and intraclass correlation coefficient statistics. Influence of clinical experience on interobserver agreement was studied via mixed-effects logistic regression. One-hundred and fifteen dogs and 49 cats with ABP were recruited. Out of 12 clinical signs evaluated, only stertor (kappa, 0.80), stridor (kappa, 0.64), attenuation of heart/lung sounds (kappa, 0.60), and goose honking (kappa, 0.84) in dogs, and stertor (kappa, 0.65) and open-mouth breathing (kappa, 0.75) in cats, were considered sufficiently reliable among clinicians. Agreement on respiratory rate estimation was good in both species (intraclass correlation coefficient, 0.75). The greater the difference in clinical experience between two clinicians, the lower the odds of agreement between the two clinicians' respiratory physical examination findings. Interclinician agreement was demonstrated to be poor for recognition of most respiratory clinical signs in dogs and cats. Teaching and clinical experience acquisition should be encouraged to improve respiratory clinical sign recognition.

[Paediatric Life Support]

The European Resuscitation Council (ERC) Paediatric Life Support (PLS) guidelines are based on the 2020 International Consensus on Cardiopulmonary Resuscitation Science with Treatment Recommendations of the International Liaison Committee on Resuscitation (ILCOR). This section provides guidelines on the management of critically ill or injured infants, children and adolescents before, during and after respiratory/cardiac arrest.

European Resuscitation Council Guidelines 2021: Paediatric Life Support

These European Resuscitation Council Paediatric Life Support (PLS) guidelines, are based on the 2020 International Consensus on Cardiopulmonary Resuscitation Science with Treatment Recommendations. This section provides guidelines on the management of critically ill infants and children, before, during and after cardiac arrest.