Lower respiratory tract infections (LRTI) are a leading cause of morbidity and mortality in children.1 In Spain, the global hospitalization rate for severe acute respiratory infection ranges from 10 to 30 cases per 100,000 inhabitants. For children under 5 years, the rate varies between 10 and 170 cases per 100,000 inhabitants annually.2 These infections are primarily viral, involving agents such as RSV, Rhinovirus/Enteroviruses and others. Severity could be influenced by environmental, epidemiological, microbiological, and host susceptibility factors.

The onset of the SARS-CoV-2 pandemic in March 2020, raised concerns about its overlap with seasonal respiratory infections, leading to potential care burdens.3 In response, health authorities worldwide implemented community strategies such as hand hygiene, face covering, and social distancing, which were associated with a reduction in respiratory virus transmission and pediatric admissions due to viral infections.4 This reduction remained consistent across countries, despite the different pandemic trajectories.5

This study aims to assess the impact of these measures and their relaxation on respiratory virus circulation in children before and during the SARS-CoV-2 pandemic (January 2017-December 2022).

A total of 2,991 respiratory virus tests were performed in children during the study period. 1,251 tests were performed during the pre-pandemic phase, 213 during confinement and mandatory mask use, and 1,527 tests during the relaxation of NPI (reopening schools and ending mask mandates). The median age of the patients tested was 17.2 months (interquartile range [IQR]: 3.0-78.4), and statistically significant differences were found between the different periods, with older ages during confinement and the period of mandatory mask use. The average number of tests per week was 9.5 (8.8-10.2), being higher during the period of school reopening and the period of confinement, 11.1 (9.9-12.3) and 16.6 (14.2-18.9), respectively (Table 1). In relation to the rate of positivity by virus, globally RV was the most detected virus (0.39 [0.37-0.42]); followed by RSV and AdV, with rates of 0.15 (0.12-0.17) and 0.08 (0.07-0.10), respectively. Other less frequent viruses were HMPV, non-SARS-CoV-2 hCoV, Influenza A/B, and other detections. Positivity rates for each virus, considering the different phases, are detailed in Table 1.

Co-circulation of respiratory viruses

The correlation between the simultaneous weekly detection of different viruses is represented in Fig. 1.

Fig. 1.

Heatmap showing correlations (Pearson correlation coefficient) between weekly simultaneous detections (co-circulation).

Not significant correlations (p≥0.05) are struck through.

RV/EV, Rhinovirus/Enteroviruses; RSV, Respiratory Syncytial Virus; AdV, Adenovirus; HMPV, human Metapneumoviruses; VIA/B, Influenza A/B viruses.

(0.25MB).

Co-circulation was observed between RV, AdV, and other detections; RSV and Influenza viruses; and AdV, HMPV and other detections. On the other hand, a negative correlation in co-circulation was observed between RV and RSV.

Viral detections in relation to non-pharmacological intervention measures

There was a significant decrease in viral detections following population confinement, which was the most restrictive NPI (Fig. 2, red line). However, this decrease was not immediately observed except for RV, AdV, and other less common viruses. Since this measure coincided with the end of RSV, HMPV, and Influenza seasons, the true impact of this measure on the transmission of these three viruses might be underestimated. The trend for all viral detections remained stable even after children's deconfinement (pink line). The implementation of compulsory masks (green line) immediately changed the trend of detections of RSV, HMPV and Influenza A/B. Despite maintaining this measure, the initiation of school activities (brown line) was followed by a slow and steady increase in RV, AdV, and less common viruses. Seasonal peaks of RSV, HMPV, and Influenza occurred later than expected. Influenza detections significantly increased after the compulsory mask period ended (blue line). Additionally, a drop in the detection rates of RV and RSV was interrupted at that time, and detections slightly increased.

Fig. 2.

Weekly detection rates of each virus, presented with deseasonalized time-series data graphics.

Red line: confinement; pink line: children's deconfinement; green line: implementation of compulsory masks; brown line: initiation of school activities; blue line: end of compulsory masks.

(0.92MB).

The results of the interrupted time-series analysis are presented in Table 2. The results showed a significant decrease in RV and HMPV detections following confinement, as well as a tendency to a significant reduction in AdV. During the period of children's deconfinement and the introduction of the compulsory mask, a lower detection rate of all viruses was observed, which was statistically significant for RV and other detections. The start of the 2020-21 school year did not lead to an immediate increase in viral detections. Lastly, the end of mandatory masks was linked to an increase in influenza detections, while RSV detection slightly increased after this measure.

Table 2.

Average absolute effect on the weekly virus detection rates, as determined by an interrupted time-series analysis conducted up to 2 months after the implementation of NPI measures.

	RV	RSV	AdV	HMPV	Influenza	Others
ConfinementWeek 12-21, 2020	-0.27(-0.46 - 0)p=0.047**	0.03(-0.06 - 0.11)p=0.294	-0.08(-0.18 - 0.02)p=0.059*	-0.15(-0.22 - -0.09)p=0.001**	0.02(-0.02 - 0.06)p=0.167	0.01(-0.01 - 0.02)p=0.142
Children deconfinementWeek 17-24, 2020	-0.42(-0.63 - -0.25)p=0.001**	-0.02(-0.23 - 0.07)p=0.367	-0.03(-0.13 - 0.08)p=0.266	-0.01(-0.05 - 0.03)p=0.344	-0.01(-0.04- 0.32)p=0.343	-0.02(-0.04- -0.01)p=0.016**
Compulsory masksWeek 22-30, 2020	-0.32(-0.51 - 0.14)p=0.003**	-0.02(-0.10 - 0.08)p=0.386	-0.05(-0.14 - 0.05)p=0.163	-0.02(-0.10 - 0.07)p=0.348	-0.01(-0.05 - 0.03)p=0.345	-0.02(-0.04 - 0.01)p=0.015**
Schools reopeningWeek 38-45, 2020	-0.01(-0.21 - 0.17)p=0.437	-0.08(-0.16 - 0.02)p=0.051*	0.01(-0.08 - 0.11)p=0.373	-0.01(-0.08 - 0.08)p=0.464	-0.01(-0.45 - 0.03)p=0.368	-0.01(-0.03 - 0.01)p=0.112
End of compulsory masksWeek 16-25, 2022	0.01(-0.16 - 0.18)p=0.473	0.09(-0.01 - 0.22)p=0.080*	0.03(-0.06 - 0.12)p=0.249	-0.02(-0.09 - 0.06)p=0.315	0.03(0.01 - 0.06)p=0.033**	0.01(-0.01 - 0.02)p=0.372

Average absolute effect (95% confidence interval)

*

p<0.10,

**

p<0.05.

RV, Rhinovirus; RSV: Respiratory Syncytial Virus; AdV: Adenovirus; HMPV, Human Metapneumoviruses.

The recent SARS-CoV-2 pandemic provided a unique opportunity to evaluate the impact of non-pharmaceutical interventions on limiting the spread of infections. This study expands on the current literature by examining the effects of confinement and strict social control measures on the spread of respiratory viruses, as well as the changes following the lifting of these measures. Unlike previous studies, this research employs deseasonalized time-series analysis to quantify the change in viral detection rates after each measure's implementation or removal.

Rhinovirus was the most frequently detected virus, followed by RSV and AdV. This is consistent with previous research highlighting the role of RV and RSV as major contributors to pediatric respiratory infections.10 In the pre-pandemic phase, RV and AdV exhibited a year-round virus pattern. In contrast, RSV, Influenza, and HMPV showed clearly epidemic periods during the winter months, as is typically described in temperate regions.11–13

With the emergence of the highly transmissible SARS-CoV-2,14 international authorities urged the implementation of strong isolation measures, previously used in earlier pandemics. For instance, during the H1N1 2009 influenza pandemic, social distancing, and early school closure effectively moderated the peak case incidence in some settings.15,16 During the lockdown, the most stringent of the NPI measures, there was a decrease in the positivity rate of all respiratory viruses, as reported in other European countries.11,12,17,18 Some studies documented a reduction in daily emergency room visits17 and hospital admissions due to respiratory tract infections coinciding with the introduction of confinement,17,19 when children and adults were practicing social distancing. Particularly, during this period, we observed a significant decrease in RV and HMPV detections, as well as a tendency for AdV detections in the short-term.

After the end of confinement for children, between the 17th and the 29th week of 2020, the trend for viral detection in them remained stable for all viruses. In the short-term analysis, we observed a significant decrease in RV and HMPV detections, as well as tendencies to statistically significant results in AdV and other detections. It should be noted that there was some overlap between children's deconfinement and confinement period, which may have led to delayed effects from the earlier measures observed during the subsequent period. However, the expected changes in trends following the release of children from confinement, which was a relaxation measure after a strict one, were not observed. This suggests that children's deconfinement had no short-term impact on the transmission of viruses. This pattern was also identified in other European countries, such as the United Kingdom,11 Finland,18 and in countries in the southern hemisphere, such as New Zealand.19

Interestingly, the trend graphics in the deseasonalized analysis showed that the detection rates for RSV, Influenza, and HMPV did not start to decline until after the mandatory mask requirement was implemented in the 23rd week of 2020. It was also associated with a short-term decrease in RV and other virus detections. As previously mentioned, the most stringent measure of the NPI was confinement. Therefore, a delayed effect of confinement on the circulation of these viruses may explain this finding. Strict public health restrictions can significantly reduce respiratory virus transmission, resulting in a decrease in childhood morbidity, including hospital and ICU admissions.20,21 In our environment, the use of face masks in adults could provide protection against viral infections in children under 6 years of age, who were not obliged to wear them. While systematic reviews have raised, some controversy about the effectiveness of face masks in protecting against viral infections:22,23 some articles in the literature support their use in controlling the source of infection.24,25 There is evidence that masks are effective in the community setting if they are used early.26 Low adherence to the implementation of social isolation measures within households may favour intrafamilial transmissibility. The study of Otomaru et al. analyzed intrafamilial transmission of RSV and found that most transmission events occurred in the week following exposure within the home.26,27

Our study, in line with others,18 found that the start of the 2020-21 school year did not lead to an immediate increase in viral detections in the short term. As the school year progressed, coinciding with increased interactions among children in a school setting and other relaxation of lockdown rules, there was a progressive increase in RV determinations, as described in other countries around the world.11,19,28,29 However, detection levels remained lower than before the pandemic. Lumley and Mansuy et al. reported a similar pattern of RV detection in their studies and noted an increased detection of AdV,11,30 which is consistent with our findings. Huang and Takashita et al. suggested that the increased incidence of RV could be due to the relaxation of isolation measures. Indeed, transmission is facilitated by factors such as cohabitation and the virological characteristics of the RV, a non-enveloped virus that is less susceptible to inactivation by hand washing or disinfectants and can survive on surfaces for extended periods.19,29 Haapanen et al. study showed that the incidence of RV in children increased during summer vacation, coinciding with the lifting of social restrictions, and before the reopening schools. This suggests that day care centers and schools might not be the main drivers of RV spread.18

Interestingly, the peak of RSV, HMPV, and Influenza was delayed, potentially due to the influence of NPI measures. In our region, RSV incidence remained low since the 20th week of 2021, like other European and Southern Hemisphere countries, until an inter-seasonal spike in RSV incidence during the summer of 202129,31–34 particularly among older children.35,36 This spike might be attributed to a lack of pre-existing immunity due to decreased exposure in the preceding years. The inter-seasonal spike could be attributable to increased social interaction due to the reopening of leisure centers or parents returning to work.11 In contrast, countries such as France and Iceland, which gradually eased restrictions between November 2020 and February 2021 while keeping schools open, experienced a seasonal spike in the winter of 2020/21, starting several weeks later than usual.35

Following the discontinuation of compulsory mask usage, we observed a short-term increase in Influenza detections, even though, the flu season had not finished when this decision was taken. This resurgence suggests that mask mandates helped to supress Influenza transmission. Additionally, a drop in the detection rates of RV and RSV was interrupted at that time, and detections slightly increased, but these changes were not statistically significant in the short-term interrupted analysis. The differential impact of mask usage on different viruses highlights the complexity of viral interactions and the need for multifaceted approaches to manage their transmission.

Viral interference seems to be another factor influencing the presence or absence of viruses in the community.13,37 A negative association between RSV and RV detection was observed in our study although the number of cases detected was low in absolute values.38 Achten et al. reported that negative association was consistently present across RSV seasons, calendar month, disease severity spectrum, and geographical regions.39 Traditionally, a competitive relationship has been described between RSV and Influenza. During the influenza H1N1 pandemic, the change in influenza activity was associated with a shift in seasonal RSV activity, suggesting viral interference between both viruses.40

This study has several limitations. Firstly, the findings are specific to a single center's population and should be understood within the local epidemiological context. Secondly, the study's focus on patients in the PICU and those with chronic comorbidities may lead to the underrepresentation of healthy children, potentially limiting the generalizability of the viral detection rates to the broader population. However, these groups are instrumental in effectively evaluating NPI measures. Despite the inclusion of high-risk outpatients with mild respiratory diseases, which does expand the representation and reflects the majority of infection types prevalent in the overall pediatric population, restricting testing to these populations resulted in a comparably small final sample size to the regional pediatric reference population. Finally, results were not stratified by patient age, yet statistically significant differences were observed in the initial descriptive analysis due to inclusion of respiratory samples from patients meeting criteria regardless of age.

In conclusion, this study highlights the significant impact of NPIs in reducing respiratory virus transmission among children, even when not tailored exclusively to preschool-aged children. Rhinovirus, RSV, HMPV, and Influenza appear particularly responsive to these interventions, both upon implementation and cessation. These findings enhance our understanding of respiratory virus transmission dynamics, offering insights into the potential effectiveness of NPIs in curbing their spread.

No funding was received to perform this study.

The authors report there are no competing interests to declare.

The authors state that they have no conflict of interests.