Accuracy of an audiovisual screening kiosk for acute respiratory infections at a Berlin travel outpatient clinic, 2023: A retrospective data analysis (Preprint)

BACKGROUND Acute respiratory illnesses (ARIs) in travel medicine clinics pose a particular risk, as they may be the first point of care for returning travelers with respiratory viruses, possibly infecting healthcare workers and leading to further healthcare-associated spread. At the Charité Travel Clinic, a video-based, contactless screening and registration terminal (SRT) was piloted to screen and separate acute patients with an ARI. OBJECTIVE The primary objective of this study was to determine the accuracy of the SRT in identifying patients with an ARI. The secondary objective was to describe the demographics, symptoms, and travel history of the patients according to their screening classification status. METHODS From October 4 to November 22, 2023, 189 unscheduled patients were screened by medical assistants using the audiovisual intercom of the SRT prior to entering the clinic and subsequently received a final diagnosis from a physician. After exclusions, 160 patients were included in the analysis. Screening accuracy was assessed using a confusion matrix comparing SRT screening results with physicians’ final diagnoses. Demographic and symptom data were analyzed by classification status. RESULTS Among the 160 patients included in the analysis, the median age was 34 years (IQR 28–45.5), and 60% (96/160) were female. ARIs accounted for 14.4% (23/160) of final diagnoses. The SRT demonstrated a sensitivity of 87% (95% CI 66%–97%) and a specificity of 84% (95% CI 77%–90%). The positive predictive value was 48% (95% CI 32%–64%), and the negative predictive value was 97% (95% CI 93%–99%). Cough was the most frequent symptom among true positives (17/20, 85.0%) and the least frequent symptom among false positives (3/22, 13.6%). All three false negative cases were female, reported cough without other ARI symptoms, and were born outside Germany. CONCLUSIONS The SRT was able to reliably rule out ARIs, with a slight tendency to err on the side of caution. In this limited study population, it performed well when compared to algorithm-based digital interventions. Symptom-based, audiovisual screening may help to reduce ARI exposure for healthcare workers and other patients and thus may be a relevant tool for screening in epidemic response.