Differentiating impacts of non‐pharmaceutical interventions on non‐coronavirus disease‐2019 respiratory viral infections: Hospital‐based retrospective observational study in Taiwan

Abstract Background Physical distancing and facemask use are worldwide recognized as effective non‐pharmaceutical interventions (NPIs) against the coronavirus disease‐2019 (COVID‐19). Since January 2020, Taiwan has introduced both NPIs but their effectiveness on non‐COVID‐19 respiratory viruses (NCRVs) remain underexplored. Methods This retrospective observational study examined electronic records at a tertiary hospital in northern Taiwan from pre‐COVID (January–December 2019) to post‐COVID period (January–May 2020). Patients with respiratory syndromes were tested for both enveloped (eg, influenza virus and seasonal coronavirus) and non‐enveloped RVs (eg, enterovirus and rhinovirus) using multiplex reverse transcription polymerase chain reaction assays. Monthly positivity rates of NCRVs among adult and pediatric patients were analyzed with comparison between pre‐ and post‐COVID periods. Results A total of 9693 patients underwent 12 127 multiplex RT‐PCR tests. The average positivity rate of NCRVs reduced by 11.2% (25.6% to 14.4%) after nationwide PHIs. Despite the COVID‐19 pandemic, the most commonly identified enveloped and non‐enveloped viruses were influenza virus and enterovirus/rhinovirus, respectively. Observed reduction in NCRV incidence was predominantly contributed by enveloped NCRVs including influenza viruses. We did not observe epidemiological impacts of NPIs on non‐enveloped viruses but an increasing trend in enterovirus/rhinovirus test positivity rate among pediatric patients. Our data were validated using Taiwan's national notification database. Conclusions Our frontline investigation suggests that the current NPIs in Taiwan might not effectively control the transmission of non‐enveloped respiratory viruses, despite their protective effects against influenza and seasonal coronavirus. Health authorities may consider using hydrogen peroxide or chloride‐based disinfectants as additional preventative strategies against non‐enveloped respiratory viruses in the post‐COVID‐19 era.


| INTRODUC TI ON
and RhV). Without implementing a combination of timely testing, accurate diagnosis, effective treatment, and non-pharmaceutical interventions, countries' healthcare systems could be heavily exhausted by these viral RTIs compounded with the COVID-19 pandemic. [3][4][5][6] Physical distancing and face mask use have been worldwide recognized as effective non-pharmaceutical interventions (NPIs) to mitigate the spread of COVID-19. Mathematical models have forecast that an 80% coverage of face mask use among populations can effectively reduce the transmission and mortality of SARS-CoV-2 by 17%-45%. 7 A Chinese study using real-world data reported that face mask use in COVID-19 patients and their close contacts resulted in a 79% risk reduction of SARS-CoV-2 transmission. 8 A global systematic review reported that face mask use and physical distancing could reduce risks of SARS-CoV-2 infections by 85% and 82%, respectively. 9 Using N95 respirators among medical practitioners could reduce the infection rate of COVID-19 by 95%. 10

| Study design and patient recruitment
We presented a retrospective cross-sectional study using hospitalbased surveillance data from Taipei Veterans General Hospital (TVGH), one of the biggest medical centers in Taiwan. We examined medical records on patients presenting respiratory symptoms from January 2019 to May 2020, defining two periods as pre-COVID (January 2019 to December 2019) and post-COVID (January 2020 to May 2020). The study was approved by the institutional review board of TVGH (reference number: 2019-06-022CC).

| Respiratory examinations
Patients who presented respiratory symptoms were queried for traveling, occupation, contact, and cluster (TOCC) history followed by physical examinations and chest radiogram. One set of nasopharyngeal swab sample was collected from patients and sent to a The ban also discouraged unnecessary travels to reduce potential contacts. Most public activities were postponed or canceled after the announcement of the ban.

| Data analysis
Given the average incubation periods and serial intervals of NCRVs span from one to two weeks, 13-17 those who were tested more than once in any 14-day period were only counted as one test. Overall positivity rates of NCRVs each calendar month were calculated.
The positivity rate of each NCRV was calculated to examine trend changes with various NPIs. We also categorized NCRVs into enveloped and non-enveloped viruses based on their virologic characteristics. Positivity rates of these two virological groups were examined against different NPIs in a temporal sequence. Due to the limited sample size, we regrouped AdV, hMPV, hBoV, and PIV as other viruses. National surveillance data from TCDC on influenza virus infection were applied to examine data validity and representativeness.

| Statistical analysis
Descriptive statistics were applied to describe patient characteristics. Categorical variables were analyzed by Pearson's chi-squared test. The significance level was set at 0.05. All analyses were conducted using RStudio ® statistical software (version: 1.3.959).

| RE SULTS
Of all 9693 patients undergoing 12 127 multiplex RT-PCR tests, 4855 were tested from January to December 2019 and 4838 from January to May 2020. Table 1  The proportion of patients infected with more than one NCRVs during pre-COVID period were higher than those visited during post-COVID period (3.7% versus 1.4%).  pre-COVID and post-COVID periods, except for March 2020 given a relatively small sample size. 11 Specifically, only one case suspected with severe complicated influenza infection was notified at TVGH in March 2020 with subsequent RT-PCR confirmation, resulting in a 100% case positivity rate ( Figure 1B).

| Overall positivity rates of NCRVs
The

| Positivity rates of enveloped and nonenveloped respiratory viruses
The ( Figure 4D).

| Positivity rates of influenza virus
During the early phase of COVID-19 epidemic in January 2020, the positivity rate of influenza in TVGH was comparable to that in 2019.
As the government started strengthening the intervention meas- rate from January to May in 2020 was significantly lower than that in 2019 (P T <0.001, Figure 5A). There was no influenza pediatric case observed after March 2020, and the overall positivity rate in 2020 was significantly lower than that in 2019, which was similar to adult patients (P T =0.02, Figure 5E).

| Positivity rates of Enterovirus/Rhinovirus (EnV/RhV)
Regardless of NPIs, the positivity rates of EnV/RhV during the COVID-19 epidemic did not reduce as compared to the previous year. In adults, the positivity rate of EnV/RhV remained above 6% in January and February 2020, and then gradually declined after March 2020, but the monthly positivity rates remained similar to those of the previous year. The overall positivity rate of adult patients in 2020 (4.9%) was not significantly different than that in 2019 (5.5%) (P T =0.48, Figure 5B). However, in pediatric patients, even with stepwise NPIs, we did not observe a significant decline in the positivity rates of EnV/RhV during the epidemic period. On the other hand, the positivity rate increased above 30% since April 2020 and became significantly higher in May 2020 (44.1%) as compared to May 2019 (23.7%) (P =.03, Figure 5F). The overall positivity rate of pediatric patients in 2020 (29.0%) was also not significantly different than that in 2019 (27.6%) (P T =0.97).

| Positivity rates of seasonal coronaviruses (sCoVs)
A high level of positivity rates was observed during the early phase of COVID-19 epidemic since January 2020. In adult patients, the positivity rates of sCoVs in 2020 were higher than that in the same

| Positivity rates of other respiratory viruses
With respect to all other NCRVs, compared with the same period in 2019, the positivity rates in adult patients rose in January 2020 then gradually declined after March 2020, and the positivity rate in April 2020 (1.2%) was significantly lower than that in April 2019 (5.5%) (P <.001). However, the overall positivity rate in 2020 (2.6%) was not significantly different than that in 2019 (3.4%) (P T =0.2, Figure 5D).
Similarly, the positivity rate in pediatric patients in January 2020 was much higher than that of the previous year (23.1% to 5.0%, P <.001), but it declined since April 2020 to a value lower than that in 2019; and, the overall positivity rate in 2020 (12.9%) was significantly lower than of that in 2019 (24.3%) (P T <0.001, Figure 5H).

| D ISCUSS I ON
Our study revealed a three-fold increase in the numbers of patients tested for NCRVs after the COVID-19 outbreak in January 2020.
Regarding types of NCRVs, influenza virus and enterovirus/rhinovirus (EnV/RhV) were the most commonly reported enveloped and Our study bears several limitations. Firstly, our research was conducted in a medical center in northern Taiwan   in Taipei, Taiwan.

CO N FLI C T S O F I NTE R E S T
No conflict of interest has been declared by the authors.

PEER R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/irv.12858.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to ethical restrictions.