2024 Volume 47 Issue 5 Pages 924-929
The region-to-region spread of human infectious diseases is considered to be dependent on the human mobility flow (HMF). However, it has been hard to obtain the evidence for this. Since the onset of the coronavirus disease 2019 (COVID-19) pandemic in Japan 2020, the government has enforced countermeasures against COVID-19 nationwide, namely the restriction of personal travelling, universal masking, and hand hygiene. As a result, the spread of acute respiratory infections had been effectively controlled. However, COVID-19 as well as pediatric respiratory syncytial virus (RSV) infections were not well-controlled. The region-to-region spread of pediatric RSV infections in 2020–2021 was recognizable unlike those in 2018 and 2019. In this study, we investigated the correlation between the trend of regional reports of the pediatric RSV infections and the HMF based on cellular phone signal data. Upon closer examination of both epidemiological trend and HMF data, the spread of pediatric RSV infection from one region to another was logically explained by HMF, which would serve as the evidence of the dependence of regional transmission on HMF. This is the first solid evidence where this correlation has been clearly observed for the common respiratory infections. While social implementation of infection control measures has successfully suppressed the droplet-mediated respiratory infections, such as influenza, but not the airborne infections, it was suggested that the aerosol transmission and adult asymptomatic carrier were involved in the transmission of RSV akin to COVID-19.
Respiratory infections are transmitted from person to person mainly through droplets as commonly experienced at the individual levels in settings such as households, schools, and workplaces. At regional levels, the spread of these infections is suspected to be influenced by human mobility flow (HMF). To gain insight into the correlation between the region-to-region spread of respiratory infections and the HMF, we took the pediatric infection of respiratory syncytial virus (RSV) as a model.
RSV infection primarily affects infants and is spread through droplets and contact transmission.1) It was also reported that RSV could be transmitted via aerosols.2) The RSV infection typically manifests as cold-like symptoms, with individuals having underlying conditions having a risk of developing severe conditions. There are no specific treatments against RSV infection, and general preventive measures are recommended, including standard precautions, with an emphasis on hand hygiene in daily life. For the prevention of RSV infection, a vaccine against RSV infection has been developed recently. As of 2023, a vaccine against RSV infection for the elderly has been approved in Japan.
In accordance to the infectious disease law, RSV infection is classified as a Category V infectious disease in Japan. Reports of RSV infections are collected from approximately 3000 pediatric sentinel medical institutions throughout the country and recorded in the National Epidemiology Surveillance of Infectious Diseases (NESID) system. This data includes information about the geographic locations where the pediatric RSV infections were diagnosed. In Japan, an annual outbreak typically occurrs from autumn to winter.3)
During the pandemic of coronavirus disease 2019 (COVID-19), the Japanese government announced the declaration of state of emergency 4 times and quasi-state of emergency twice and tried to implement a series of infection control measures in daily life, including the following: avoiding the “3 Cs” (closed spaces, crowded places, and close-contact settings) in daily life, workplaces, and restaurants; encouraging remote works; restricting mass gatherings; shortening the open hours of shops and restaurants; mandating mask-wearing; providing free testing and vaccination; refraining from personal traveling; and promoting the hand hygiene. This was so-called “voluntary lockdown.”4) It was also important to note that international travel was restricted. The number of travelers visiting Japan decreased by more than 95% from April 2019 to 2021 compared to the same month in 2019 according to Japan National Tourism Organization (JNTO).5) Business traveling and the transportation of daily necessities remained.
The social implementation of countermeasures to COVID-19 was effective against pediatric infections, including influenza, pharyngoconjunctival fever, group A streptococcal pharyngitis, acute infectious gastroenteritis, herpangina.6) However, although slowed down, the endemics of RSV infection as well as COVID-19 were not stopped by the voluntary lockdown.
In a typical year, the number of cases with pediatric RSV infections starts to increase in Okinawa, the southernmost region of Japan, before the summer. Subsequently, RSV outbreaks happen almost simultaneously across Japan. However, in 2021, the trend of pediatric RSV infection reports exhibited a distinct feature compared to previous years. The expansion of infections from one region to another was apparent.
During the COVID-19 pandemic, the use of infographics related to epidemiological data became more common. This was due to their ability to visually provide an intuitive understanding of huge amounts of data, including the timeline of the infection trend, the number of reported cases, and their relationship with countries or regions.
The measurement of HMF at a high precision level was challenging before the era of personal digital equipment, such as iPhones. The utilization of HMF data for personalized scientific research was difficult due to the cost of data acquisition and the hardware demands to handle the large datasets. However, the situation has evolved with the Ministry of Internal Affairs and Communications in Japan making HMF data publicly available along with the provision of infographic presentations7,8)
In this study, we performed an infographic analysis to explore the correlation between the outbreak pattern of pediatric RSV infection and HMF in Japan.
The number of weekly reported cases of pediatric RSV infections, influenza, COVID-19 by prefecture from 2018 to 2022 was obtained from the database provided by National Institute of Infectious Diseases, Japan.9) The HMF data were obtained from Vital Signs of Economy and Regional Society Analyzing System (V-RESAS) provided by the Cabinet Office, Government of Japan.8)
Infographic Analysis of RSV InfectionsThe number of reported cases of pediatric RSV cases was visualized by a Final Cut Pro software ver. 10.6.5 (Apple, Cupertino, CA, U.S.A.). The number of pediatric patients with RSV infection in each prefecture is represented on the map using circles, with the size of the circles indicating the quantity. These circles change over time to reflect the situation of the time. The maximum diameter of the circle was set to ensure the visibility of the circles of neighboring prefectures.
Infographic Analysis of HMFThe number of inflows across prefectural borders in 2020 is shown as a heat map at the level of local government (city/town/village) that generated by the V-RESAS system.8) The color gradation reflects the magnitude of inflows, with Yamaguchi prefecture set as the standard.
The epidemiological trend of pediatric RSV infection was presented in a movie file from 2018 to 2021 in Japan (Supplementary movie S1). In 2018 and 2019, there was an increase in the number of reported cases from Okinawa, followed by a nationwide surge from July to August. There was no order in the surge of case reporting among the prefectures during this period. The exact reason why the outbreak started from Okinawa was unknown.
The global outbreak of COVID-19 began at the end of 2019. From April 2020 onwards, the outbreak of COVID-19 began with repeated waves (Fig. 1). The Japanese Government initiated social implementation of infection control acts in daily life to delay the spread of COVID-19. In 2020, there was a notable absence of pediatric RSV infection epidemics, as seen in previous years.
The graph displays the number of reported cases for influenza (orange line, left axis) and pediatric RSV infection (blue shading, left axis) per sentinel, along with the total number of COVID-19 cases (gray line, right axis). Influenza is typically epidemic during the winter seasons, as seen in 2019 and 2019–2020 but not in 2020–21. RSV infections usually peak in autumn seasons as observed in 2019, but not in 2020. The first notable wave of COVID-19 epidemic is seen in April 2020, followed by periodic waves. The red arrows indicate the periods when the Japanese Government declared emergency alerts. The black arrow indicates the presence of a small number of RSV patients during the warm climate seasons. The black triangle indicates the time when cases with influenza, RSV, and COVID-19 were not reported. The detailed description of the RSV infection phases during from 2020 to 2021 (boxes below the graph) can be found in the main text.
The timecourse of the outbreaks of RSV infections in Japan was analyzed at prefectural levels from 2020 to 2021 (Fig. 2, Supplementary movie S1). In July 2020, there was an increase in the reported number of RSV patients in Okinawa and Kagoshima, the southernmost regions of Japan (Phase 1, since the epidemiological week 30th). In late November 2020, the reported number started to grow in Miyazaki, the neighboring prefecture to Kagoshima (Phase 2, since the epidemiological week 48th). In January 2021, the reported number significantly increased in Kumamoto, Nagasaki, Fukuoka, and Saga prefectures in Kyushu island almost simultaneously while the number of cases decreased in Kagoshima (Phase 3). The state of emergency was announced by the Japanese Government at that time. In March 2021, the reported number of cases increased in Oita and Yamaguchi prefectures (Phase 4). At the same time, the number of cases started to grow in prefectures remote from the Kyushu island, including Osaka (Phase 4). In April 2021, the reported number began to increase in Hiroshima and the neighboring prefectures to Osaka, including Nara, Kyoto, and Shiga (Phase 5). In May 2021, the number started to grow in Okayama and Hyogo prefectures positioning next to Hiroshima and Osaka (Phase 6). In the end of May, the number began to grow in the prefectures in Shikoku island (Phase 7). Following, the cases with RSV infection were reported almost simultaneously from the most of the prefectures across Japan (Phase 8). Lastly, the number of cases increased in Tottori and Nagano (Phase 9).
The geographic map indicates the epidemiologic trend of pediatric RSV infection from 2020 to 2021. (A) The 44th week (late October) of 2020 during Phase 1. Kagoshima prefecture is the southernmost of Kyushu island where the outbreak initiated in 2020 (arrowhead). (B) The 50th week (early December) of 2020 during Phase 2. Miyazaki prefecture is adjacent to Kagoshima prefecture (arrowhead). (C) The 8th week (late February) of 2021 during Phase 3. Oita prefecture positions at the middle-right of Kyushu (arrowhead). The outbreak is not observed even other Kyushu prefectures are in epidemic. (D) The 16th week (mid April) of 2021 during Phase 5. Yamaguchi prefecture is located at the westernmost of the mainland Japan (arrowhead). (E) The 26th week (late May to early July) of 2021 during Phase 9. Tottori and Nagano prefectures are indicated by arrowheads. Okinawa prefecture is shown at the bottom right. The number of reported cases was shown as the red circle by prefecture.
It was notable that the spread of RSV infection was sufficiently slow, allowing us to chase the region-to-region spread starting from the south to north of Japan. The region-to-region spread was more apparent at the early to middle of the endemic. Such trend was not observable in 2018 and 2019. It took approximately four months to spread from Kagoshima to Miyazaki. It took one month from Phase 2 to Phase 3 despite the longer transmission distance than Phase 1 to Phase 2. The speed of region-to-region spread became higher after Phase 4. In the early phase, the spread speed was estimated approximately 1 km/d. In the late phase, it was estimated approximately 4 km/d.
Correlation of Endemic Data with HMF DataNext, each phase of the endemic was analyzed in detail in reference to the HMF data. The HMF was reduced to 38% compared to that in 2019, suggesting the effectiveness of the voluntary lockdown.
The heat map of the HMF was shown in Fig. 3, representing the inflow of people who moved across prefectural borders (Fig. 3). In phase 1, the reported pediatric patients with RSV infection should be living in the Kagoshima city area (Fig. 3A, labeled as 1 on the figure). Kagoshima city area was high in HMF. By examining the data, it is possible to determine from which prefectures the inflow came from. The people living in Kagoshima visited Miyazaki prefectures the most frequently (30% of the outflows) and the people in Miyazaki vice versa (33.4%). This HMF was consistent to the spread of RSV infections from Kagoshima to Miyazaki in phase 2 (Fig. 3A, labeled as 2 on the figure). These two prefectures were separated from the middle and north part of Kyushu by the low HMF areas (Fig. 3A, horizontal line). This observation was consistent with the delay in the transmission of RSV infection from the southern region to the central and northern part of Kyushu island.
The number of inflows across prefectural borders in 2020 is shown as a heat map at the level of local government (city/town/village). The number represents the phase of epidemic as described in the text. Lines indicate the low HMF areas that separate HMF regions. The color gradation reflects the magnitude of inflows. (A) The human mobility flow (HMF) of Kyushu area during phases 1 to 4, with notable HMF directions (arrows). The black arrow represents the HMF from southern Kyushu to Fukuoka city. The white arrows indicate the HMF between Fukuoka city area (circle) and Kumamoto, Saga, and Nagasaki prefectures. The dotted arrow is the HMF from Fukuoka to Kitakyushu city within Fukuoka prefecture crossing low HMF areas. Oita prefecture is surrounded by low HMF areas as highlighted by an arc. Yamaguchi prefecture is located at the western end of mainland Japan facing Kitakyushu city across the strait. (B) The HMF of the western mainland of Japan during phases 4 to 9. Osaka prefecture is circled. Tottori prefecture is surrounded by an arc.
At the new year season in 2020–21, although the HMF was low due to the Japanese Government request compared to the pre-pandemic years, HMF was transiently increased because of the personal travelling. The outflows to Fukuoka became the highest during that time from Kagoshima and Miyazaki (23.6 and 23.1%, respectively; Fig. 3A, arrows). Fukuoka city is the largest city in the Kyushu island, and the HMF between Fukuoka and Kumamoto, Nagasaki, and Saga was very active (Fig. 3A, circle). Once RSV infection was introduced to Fukuoka city, it spread immediately to three prefectures (Fig. 3A, labeled as 3).
The HMF in Oita prefecture is unique among other Kyushu prefectures. Oita is surrounded by low HMF areas that potentially slowed the region-to-region spread of RSV infection (Fig. 3A, arc). The HMF from Oita to Fukuoka is dominant (53.0%). However, the closest city from Oita to Fukuoka prefecture is Kitakyushu city, remote from Fukuoka city (Fig. 3A, labeled as 4). There are at least two areas with low HMF between the two cities, delaying the spread of RSV infection from Fukuoka city to Kitakyushu city (Fig. 3A, short parallel lines). This was consistent to the delayed increase of reported cases of RSV infection in Oita. Similar observation was made in Yamaguchi prefecture. Shimonoseki city, Yamaguchi, positions the opposite shore to Kitakyushu city. Many Shimonoseki residents commuted to Kitakyushu city and vice versa. Once RSV infection reached Kitakyushu city, it spread to Yamaguchi prefecture, similar timing to Oita prefecture (Fig. 3B, labeled as 4).
So far, the region-to-region spread of RSV infection should affect the neighboring prefecture. However, the reported number of RSV infections increased in Osaka prefecture without any increase in prefectures between Osaka and Yamaguchi prefectures. There are several low HMF areas that divided the Yamaguchi-Osaka axis that should took time for RSV infection to spread through. The HMF between Osaka and Fukuoka was active due to the close connection by the Shinkansen train, air travel, highway buses, and logistics by truck. It was likely that such transportations helped the spread of RSV infection directly from Fukuoka to Osaka.
In phases 5 and 6, the region-to-region spread was observed as labelled as 5 and 6 in Fig. 2B (Fig. 3B, labeled as 5 and 6). The HMF was highly active between those neighboring prefectures, consistent to the notion that RSV infection spread via region-to-region in accordance to HMF. After the phase 7, the correlate of geographic spreading pattern with the HMF became blurred due to the synchronous surge of RSV reporting across Japan. In phase 9, it was notable that prefectures bearing low populations and surrounded by low HMF areas finally became epidemic for RSV infection (Fig. 3B, labeled as 9). At this phase, the next epidemic cycle seemed to start in Kyushu area.
The most straightforward evidence for region-to-region transmission of an infection is the continuous spread of an outbreak in adjacent geographic regions. However, for infectious diseases that are epidemic annually, such a pattern is not so common. When signs of an outbreak appear, the outbreak often begins almost simultaneously in multiple regions. This may be partly due to the infectious disease reporting system in which the populational coverage of the surveillance is insufficient, and reporting is not real-time. For example, in Japan, pediatric cases with RSV infections are reported from the sentinel medical institutions on a weekly basis. The dataset of 2018 and 2019 indicated that the onsets of outbreaks appeared nearly synchronized across Japan, making it unclear if there was any region-to-region spread. However, in 2020–2021, the circumstances allowed for the recognizable region-to-region spread. Two key factors contributed to this: the absence of simultaneous outbreaks among many prefectures, and the relatively slow pace of outbreak transmission.
The background in which these two factors overlapped to each other is as follows: in response to the social implementation of countermeasures against COVID-19, the first state of emergency declaration in April 2020 led to a reduction in HMF by 62% compared to 2019 levels. The mask-wearing rate was high, exceeding 90%.10) Schools and childcare facilities were temporarily closed. These anti-COVID-19 acts effectively reduced the number of pediatric infections and blocked the region-to-region transmission (Fig. 1, triangle).
The conditions required for the onset of RSV infection outbreaks are the presence of susceptible individuals and a sufficient number of patients with RSV infection. Given the annual occurrence of RSV infections, a sufficient number of susceptible individuals to RSV should have been present in Japan 2020. In a typical year, RSV-infected patients would continue to exist at low levels during the warm climate seasons in almost every region across Japan (Fig. 1, arrow), which serve as a trigger of the nationwide simultaneous onset of the RSV outbreak. However, in 2020, this was not the case.
RSV infections primarily affect infants. When RSV infections spread across prefectural borders, does the transmission occur directly from one infant to another? Infants within a local community often come into contact with each other in places like nurseries and waiting rooms of medical institutions where transmission of RSV might occur. The household transmission is also possible if siblings were infected with RSV. It seems unlikely that an infant from a distant region come in contact with an infant living in a distant region during 2020–2021 when individuals displaying fever or cough are strongly encouraged to stay home and the personal traveling was avoided. Infants with symptomatic RSV infection did not have opportunities to travel during that time. In contrast, business traveling across prefectural borders by adult population was not restricted. Considering that RSV infections in adults often result in asymptomatic (textbook),1) we propose the following scenario. Parents became asymptomatic carrier (AC) when their infants were infected with RSV. These adult AC then engaged in long-distance business travel across prefectural borders. During their business trips, they unknowingly transmit the virus to other adults. These adults, upon returning to their homes, introduce the virus to their households, subsequently infecting their own infants with RSV.
If this scenario was the case, what was the mode of transmission? The transmission between parents and their infants should be contact or droplet transmission because of the close distance and no mask-wearing at home. In other hand, the adult-to-adult transmission should have been aerosol-mediated. The rate of mask waring was almost 100% among adults at business situations. For this reason, the outbreak of influenza was controlled the transmission of which mostly depends on contact and droplet (Fig. 3). Contrasting, the outbreak of COVID-19, which mainly spread via aerosol by AC,11–16) was not controlled despite implementation of infection control acts at social levels. It is reasonable to believe that RSV transmission resembles to COVID-19. Indeed, the role of aerosol transmission and the adult AC in the outbreak of RSV infection had been reported previously.2) Overall, by analyzing the epidemic dynamics, it is suggested that adult AC may play a role in the region-to-region spread of pediatric RSV infections.
A potential limitation of this study lies in the selection and the number of sentinel pediatric medical institutions that significantly affects the reporting of pediatric RSV infection cases. The number, type, and size of the sentinel pediatric medical care facilities varies between each prefecture. Additionally, the healthcare providers tend to become more proactive in suspecting and testing for RSV infections in patients as an outbreak starts in one district. Consequently, as the outbreak progresses, diagnoses are made earlier, and the number of reported cases tends to increase rapidly. This might contribute to the slow transmission in the early stages of an outbreak and faster in the later stages.
Finally, understanding the dependence of region-to-region spread of infectious diseases on the HMF can facilitate the development of strategies to delay the disease transmission between regions. Moreover, if the AC play a crucial role in the RSV transmission, this knowledge can provide insights into the prevention measures against RSV infections. If the infographics analysis was performed without expertise, and the HMF data made more accessible, the public health on infectious diseases should benefit significantly.
HS and JK collected data and analyzed the data. HS produced the infographic presentation illustrating the number of pediatric RSV cases over time. RS, MY, and JK drafted the manuscript. JK designed the research project, and finalized the manuscript. All authors reviewed and approved the content of the manuscript.
The authors declare no conflict of interest.
This article contains a supplementary material (movie file).
