Development of a model to diagnose influenza in travelers using data on the number of influenza cases and symptoms

• Published in: Journal of Global Antimicrobial Resistance Vol. 39; p. 34
• Main Authors: Asai, Yusuke, Yamamoto, Kei, Nomoto, Hidetoshi, Nakagawa, Hidenori, Sahara, Toshinori, Yamato, Masaya, Sakamoto, Naoya, Hase, Ryota, Shinohara, Koh, Yoshimura, Yukihiro, Nagasaka, Atsushi, Ichikawa, Takahiro, Imakita, Natsuko, Miyawaki, Hiroshi, Yokota, Kyoko, Yamamoto, Yoshihiro, Itoh, Naoya, Okumura, Nobumasa, Yoshimi, Yusuke, Ohmagari, Norio
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.12.2024; Elsevier
• Subjects: diagnosis; epidemiology; Infectious Disease; Influenza; mathematical model; mathematical model; epidemiology; diagnosis; Influenza
• ISSN: 2213-7165
• DOI: 10.1016/j.jgar.2024.10.107

To develop a statistical model to diagnose influenza in international travelers using data on symptoms and the number of influenza cases in each country. The importation of infectious diseases has increased dramatically in recent years. Diagnosis of infectious diseases is traditionally based on symptoms and blood and biochemical tests. Although it is known that cases of influenza are common throughout the year in tropical and subtropical regions, and that cases are also common among travelers from these regions, the number of infected cases in each country has not been fully utilized for diagnosis. This study included cases registered in the Japan Registry for Infectious Diseases from Abroad (https://jrida-jprecor.ncgm.go.jp/en/j-rida/index.html) with influenza test results and information on travel destination and duration. Multivariate logistic regression was used for the diagnostic model, and the explanatory variables were symptoms and the infection probability score calculated from information on the number of cases in each country and the incubation period. Analysis showed significant differences in sore throat (p<0.001), shivering (p<0.001), diarrhea (p=0.001), muscle pain (p=0.002), fever (p=0.011) and runny nose (p=0.013). The infection probability score derived from the number of infected cases and incubation period information was also significantly different, p=0.009. The estimated AUCs were 80.2 and 82.2 for the symptom-only diagnostic model and the model including the infection probability score, respectively. The use of different epidemiological information improves the accuracy of diagnosis. The active use of epidemiological information, including understanding the risk of infection among international travelers, is desirable.