Accuracy of the Traditional COVID-19 Phone Triaging System and Phone Triage-Driven Deep Learning Model.

OBJECTIVES: During the COVID-19 pandemic, a quick and reliable phone-triage system is critical for early care and efficient distribution of hospital resources. The study aimed to assess the accuracy of the traditional phone-triage system and phone triage-driven deep learning model in the prediction of positive COVID-19 patients. SETTING: This is a retrospective study conducted at the family medicine department, Cairo University. METHODS: The study included a dataset of 943 suspected COVID-19 patients from the phone triage during the first wave of the pandemic. The accuracy of the phone triaging system was assessed. PCR-dependent and phone triage-driven deep learning model for automated classifications of natural human responses was conducted. RESULTS: Based on the RT-PCR results, we found that myalgia, fever, and contact with a case with respiratory symptoms had the highest sensitivity among the symptoms/ risk factors that were asked during the phone calls (86.3%, 77.5%, and 75.1%, respectively). While immunodeficiency, smoking, and loss of smell or taste had the highest specificity (96.9%, 83.6%, and 74.0%, respectively). The positive predictive value (PPV) of phone triage was 48.4%. The classification accuracy achieved by the deep learning model was 66%, while the PPV was 70.5%. CONCLUSION: Phone triage and deep learning models are feasible and convenient tools for screening COVID-19 patients. Using the deep learning models for symptoms screening will help to provide the proper medical care as early as possible for those at a higher risk of developing severe illness paving the way for a more efficient allocation of the scanty health resources.

Use of the McIsaac Score to Predict Group A Streptococcal Pharyngitis in Outpatient Nurse Phone Triage and Electronic Visits Compared With In-Person Visits: Retrospective Observational Study.

BACKGROUND: The McIsaac criteria are a validated scoring system used to determine the likelihood of an acute sore throat being caused by group A streptococcus (GAS) to stratify patients who need strep testing. OBJECTIVE: We aim to compare McIsaac criteria obtained during face-to-face (f2f) and non-f2f encounters. METHODS: This retrospective study compared the percentage of positive GAS tests by McIsaac score for scores calculated during nurse protocol phone encounters, e-visits (electronic visits), and in person f2f clinic visits. RESULTS: There was no difference in percentages of positive strep tests between encounter types for any of the McIsaac scores. There were significantly more phone and e-visit encounters with any missing score components compared with f2f visits. For individual score components, there were significantly fewer e-visits missing fever and cough information compared with phone encounters and f2f encounters. F2f encounters were significantly less likely to be missing descriptions of tonsils and lymphadenopathy compared with phone and e-visit encounters. McIsaac scores of 4 had positive GAS rates of 55% to 68% across encounter types. There were 4 encounters not missing any score components with a McIsaac score of 0. None of these 4 encounters had a positive GAS test. CONCLUSIONS: McIsaac scores of 4 collected during non-f2f care could be used to consider empiric treatment for GAS without testing if significant barriers to testing exist such as the COVID-19 pandemic or geographic barriers. Future studies should evaluate further whether non-f2f encounters with McIsaac scores of 0 can be safely excluded from GAS testing.

Gender Difference in Perceived Symptoms and Laboratory Investigations in Suspected and Confirmed COVID-19 Cases: A Retrospective Study.

BACKGROUND: Evaluating gender-specific effects of COVID-19 is important to develop effective therapeutic strategies. The aim of this study was to explore gender difference in perceived symptoms and laboratory investigations in suspected and confirmed cases. METHODS: This is a retrospective study that included data from suspected COVID-19 patients during the first wave of the pandemic. Participants using the phone triaging system at Kasralainy outpatient clinics were included. The analyzed data included patient history and results of nasopharyngeal swab and laboratory data. RESULTS: Out of 440 COVID-19 suspected cases, 56.36% were females. The perceived COVID-19 symptoms showed no significant gender difference in suspected cases while in confirmed cases females were 4 times more likely to complain of cough [OR (95% CI) 3.92 (1.316-11.68), P-value .014] and 5 times more likely to experience loss of smell or taste [OR (95% CI) 4.84 (1.62-14.43), P-value .005]. Laboratory markers revealed high levels of aspartate aminotransferase, alanine aminotransferase, blood urea, serum creatinine, creatine kinase, and serum ferritin in males and this was statistically significant (P-value <.001) in suspected and confirmed cases. Females confirmed with COVID-19 were 80%, 97%, and 97% less likely to have high levels of ALT, creatin kinase, and serum ferritin [OR (95% CI) 0.20 (0.07-0.54), 0.07 (0.01-0.38), and 0.07 (0.01-0.90), P-value .002, .002, and .041, respectively]. CONCLUSION: Gender differences were found in laboratory markers in COVID-19 suspected and confirmed cases and in perceived symptoms in confirmed cases.