We found a higher incidence in densely populated urben areas which changed to a slight cluster after vaccination. There was a significant decrease in HFMD morbidity during the pre-vaccination period 2012 to 2016 (215.22/105) compared to the post vaccination period 2017 to 2020 (179.81/105). Boys, children under 5 years and Scattered children were more likely to develop HFMD compared with other individuals; the proportions decreased after vaccination except for Scattered children. The incidence of the commonest pathogenic enterovirus (EV-A71), causing HFMD in Hefei was significantly reduced after EV-A7 vaccination with a further significant reduction in severe HFMD. There was an increase in Other Enteroviruses causing HFMD, especially CV-A6.
We have shown that the four districts located in the center of Hefei had higher incidence than the other areas, and the relative distance from the city center the lower the incidence. After vaccination, the random distribution of HFMD changed to slight cluster among Districts. The regional distribution of HFMD in Sichuan Province also conforms to this feature [23]. It may be the higher population density and mobility in the four districts enabled easier for enterovirus to spread [24, 25], which indicates that prevention and control of transmission routes should be a focus of attention in populated areas.
However, in remote rural area or economically underdeveloped areas, the health awareness of people may not be high enough resulting in patients with HFMD not seeking the health service, as indicated by studies which showed a negative correlation between socioeconomic development and the incidence of severe illness suggesting that cases in less economically developed areas were more likely to develop severe illness [25, 26].
Lujiang County usually had the lowest incidence but the morbidity escalated in 2014. This could be due to previous decreased health vigilance in this area andpossibly people with low antibody levels making it difficult to respond to outbreak of HFMD [27]. Furthermore, following vaccination, the areas of higher incidence were usually the center of Hefei, and the random distribution of HFMD changed to slight cluster among districts, suggesting that the center of Hefei or urban are the key areas for prevention and control of HFMD.
According to previous studies [6], HFMD in northern China presents a single-peak pattern in spring and summer, and a double-peak pattern in southern China, while both patterns can occur in Hefei as shown by an autumn peak from September to November in 2017 and 2020. This may be due to the position of Hefei being in middle of China resulting in a semitropical climate. The HFMD epidemic in Guangxi Province had the same autumn peak in 2017 after vaccination [28]. The extremely low incidence in Hefei from February to July 2020 may be due to the COVID-19 outbreak in China in late 2019 [29]. Hefei suspended school and work during the period, strengthened respiratory transmission prevention and control efforts, effectively reduced human-to-human transmission and respiratory transmission [30], which resulted in respiratory precautions playing an effective role in HFMD prevention and control.
After EV-A71 vaccination, the seasonality and periodicity of HFMD incidence became atypical, with no epidemic peak being noted for the first time in the spring of 2017. A study using the Distributed Lag Nonlinear Model found that HFMD incidence in children had a nonlinear relationship with the daily mean temperature, and there was a significant lag effect in Nanjing, China [31], suggesting that HFMD seasonality was related to climate, which also showed in our study. The seasonal HFMD patterns has implications for seasonal and periodic HFMD prevention programs. Therefore, in combination with the cyclical nature of the virus transmission, the analysis of the HFMD temporal characteristics before and after vaccination may be helpful to anticipate the development of epidemics timeously. However, the impact of socioeconomic factors and other health conditions on the temporal characteristics of HFMD requires further analysis.
Children under 5 years especially males had a higher proportion of HFMD in Hefei, which was consistent with other areas [31,32,33]. The morbidity decreased in in males and females following vaccination, with males remaining at higher risk of HFMD, which were consistent with studies in Guangxi and Zibo cities of China [28, 34]. There was a slight decrease in the group aged 0–4 years old, especially in children aged less than 1 year and aged 1 year old. Meanwhile, boys and kids under 1 year old were more vulnerable to temperature variations [31], suggests that more effectual protection are needed to be adopted for the key groups of HFMD.
Scattered children were much more affected by HFMD compared to Kindergarten children and students, with HFMD increasing in Scattered children after EV-A71 vaccination. These changes were consistent with Guangxi [28], and the EV-A71 vaccine reduced 42.9% of HFMD cases among Scattered children in Guangdong [35]. This may be due to the large proportion of Scattered children under 5-year age group, and it is more difficult to implement primary, secondary and tertiary prevention among these children.
Although the decreased incidence of severe HFMD may be mainly attributed to the improvement in health conditions, the incidence of severe HFMD decreased as the incidence of EV-A71 also dropped, demonstrating the clinical significance of prevention and control of EV-A71 following vaccination. It was also consistent with other cities in China where have implemented EV-A71 vaccination, for example, the proportion of EV-A71 among severe cases dropped from 37.9% to 14.1% following vaccination in Nanchang [36].
The main cause of HFMD used to be EV-A71 but showed a significant decline in 2015 and has been controlled at a very low level since 2018. The 2015 decline may be related to the improvement of health conditions and the increased awareness of infection prevention and control in of the community following an HFMD epidemic in 2014 [37, 38]. The decline in 2018 may be related to the gradual increased rate of EV-A71 vaccination, which began in the second half of 2016 [32] and to the higher level of EV-A71 serum neutralizing antibody in healthy people [39].
We have shown an upward trend in EV-A71 vaccination rates and a downward trend in EV-A71 causing HFMD. Furthermore, the reported three main enterovirus types accounted for a balanced proportion during the spring peak, while the proportion of Other Enteroviruses stood out in the autumn peak. However, following vaccination, the proportion of EV-A71 accounted for less HFMD cases, but the proportion of Other Enteroviruses changed significantly, contributing much more to autumn cases. These changes in seasonal patterns has been reported in Nanchang city of China that the proportion of CV-A6 was prominent at autumn following vaccination, and particularly among severe cases [36]. Furthermore, these findings have major implications for prevention of HFMD in different seasons, such as emphasizing EV-A71 vaccination in spring and summer while increasing protection against Other Enteroviruses in autumn and winter.
The EV-A71 vaccine has no cross-immunity against CV-A16 and Other Enteroviruses [39], thus increasing the proportion of Other Enteroviruses causing HFMD, especially the sharp increase in CV-A6 we have shown. The increasing prevalence of CV-A6 may result in future HFMD outbreaks, however it may be missed as a cause due to its complex clinical manifestations [40]. Furthermore, it was also shown that CV-A6 accounted for 62.33% of severe HFMD in Kunming in 2018 [32]. The prevention and control of HFMD should be diversified and the Other Enteroviruses should also be monitored. EV-A71 vaccination may also reduce morbidity and mortality in HFMD cases since these patients usually have a higher proportion of CNS diseases and increased mortality [41].
We have shown high newborn-completion-vaccination rates (71.33% in 2018) in Hefei while Guangzhou had a 15.11% rate in 2018 [42]. In Ningbo, the newborn-completion-vaccination rate was 44.23% [43], and the incidence density was as high as 5 945.50/105 person-years [33]. Countries such as Japan and Singapore also have a high prevalence of HFMD, but did not use the EV-A71 vaccine, although EV-A71 prevalence dropped in 2015 but subsequently increased thereafter [44, 45]. The EV-A71 vaccine implementation is indeed important, but the cyclical nature of the virus transmission should also be considered during implementation.
Since the implementation of EV-A71 vaccination in Hefei, the overall vaccination rates have shown a trend of initially rising and then decreasing, with the highest rate in 2018. However, the completion rates of 2-dose vaccination were not high, with a cumulative rate of 46.70%, which will reduce the overall effectiveness of the vaccine. Therefore, future attention should be focused to continuously improve maintenance of vaccination. Notably, the guidelines recommend completion of EV-A71 with two doses before 12 months of age suggesting timeliness was considered to be an indicator of the effectiveness of EV71 vaccine [43, 46].
Our study has several limitations. Firstly, we were unable to associate the seasonality of HFMD morbidity following vaccination since monthly vaccinations are not done. We thus used a combination of annual vaccinations and HFMD incidence to observe the effect of vaccination on the trend of HFMD which did not account for delayed effects. Secondly, we could not describe the HFMD morbidity in vaccinated people since individual vaccination details were not available, so we could only use the EV-A71 vaccination status of the whole population to make inferences. Therefore, we were unable to investigate the effects of EV-A71 vaccine on different enteroviruses directly. Thirdly, we are unable to analyze the difference between individual people who complete the two doses of vaccine schedule and those who did not complete it.