No abstract

A well-known effect size (ES) indicator is Cohen’s d. Cohen defined d measures of small, medium, and large ES as 0.2, 0.5, and 0.8, respectively. This approach has been criticized because practical and clinical importance depends on the context of research. The aim of the study was to examine physicians’ perception of ES using iron deficiency anemia treatment as an example and observing the effects of pretreatment level and duration of treatment on the magnitude of ES. We prepared a questionnaire describing four different clinical studies: (1) 1 month of treatment of anemia in a group of patients with a mean hemoglobin (Hb) of 10 g/dL; (2) 3 months of treatment at an Hb level of 10 g/dL; (3) 1 month of treatment at an Hb level of 8 g/dL; and (4) 3 months of treatment at an Hb level of 8 g/dL. In each scenario, respondents were required to evaluate six various levels of Hb improvement as being very small, small, medium, large, or very large effect: 0.1 g/dL, 0.3 g/dL, 0.7 g/dL, 1.1 g/dL, 1.7 g/dL, and 2.8 g/dL. The responses of 35 physicians were evaluated. For 10 mg/dL, the Cohen's d for small, medium, and large ES was 0.5, 0.8, and 1.2 respectively, for 1 month of treatment. In terms of 3 months of treatment, the Cohen's d was 0.8, 1.2, and 2, respectively. Two separate pretreatment Hb levels (8 g/dL and 10 g/dL) demonstrated a minor difference. Determination of ES during the planning phase of studies requires thorough evaluation of specific clinical cases. Our results are divergent from the classic Cohen’s d values. Additionally, duration of treatment affects ES perception.

Purpose: The purpose of the study is to analyze the neurological manifestations and to determine the association between these symptoms and mortality in hospitalized patients with coronavirus disease 2019 (COVID-19). Materials and Methods: Five hundred and forty-seven hospitalized patients with positive reverse transcriptase-polymerase chain reaction tests for severe acute respiratory syndrome coronavirus in a nasopharyngeal swab were included in this study. The demographic features, laboratory data, and radiologic imaging, neurological symptoms of hospitalized patients with COVID-19 were collected. Results: Of 547 hospitalized COVID-19 patients, the median age was 61 (range 18–93), 61.4% were male. Three hundred and forty-seven (63.4%) patients had a severe infection and 200 (36.6%) patients had a mild infection. Eighty-eight patients (16.1%) died during hospitalization. One hundred and fifty-four (28.2%) patients had at least one neurological symptom. Thirty-five (6.4%) patients manifested with only neurological symptoms at hospital admission. The most frequent neurological symptoms were headache (15.2%), taste and smell disorders (9.1%), and myalgia (6.6%). The other initial neurological manifestations were acute cerebral ischemic stroke, impaired consciousness, epileptic seizure, and posterior reversible encephalopathy. The late-onset neurological complications were autoimmune encephalitis and Guillain-Barre syndrome. The neurological manifestation was linked to the severity of disease (P = 0.005) but not correlated with mortality (P = 0.137). Conclusion: Neurological symptoms were frequent in COVID-19 patients. The neurological symptoms can be the initial symptoms or can be late-onset complications of COVID-19.

BACKGROUND The early prediction of antibiotic resistance in patients with urinary tract infection is important to guide appropriate antibiotic therapy selection. OBJECTIVE In the present study, we aimed to predict antibiotic resistance in patients with urinary tract infection. Additionally, we aimed to interpret the machine learning models we developed. METHODS We used admission, diagnosis, prescription, and microbiology records of patients who underwent urine culture tests in Yongin Severance Hospital, South Korea. We developed 5 sub-models to classify urinary tract infection pathogens as either sensitive or resistant to cephalosporin, piperacillin/tazobactam, trimethoprim/sulfamethoxazole, fluoroquinolone, and carbapenem. To analyze how each variable contributed to the machine learning model’s predictions of antibiotic resistance, we used the SHapley Additive exPlanations method. Finally, we proposed a prototype machine learning based clinical decision support system to provide clinicians the resistance probabilities for each antibiotic. RESULTS The area under the curve values ranged from 0.710 to 0.826 in the training set and 0.642 to 0.812 in the test set for predicting antibiotic resistance. The administration of drugs before infection and exposure time to these drugs were found to be important variables for predicting antibiotic resistance. CONCLUSIONS The study results demonstrated the potential of machine learning to predict antibiotic resistance in patients with urinary tract infection. Machine learning can assist clinicians in making decisions regarding the selection of appropriate antibiotic therapy in patients with urinary tract infection.

No abstract