key: cord-0834008-07dz9glc
authors: Sayed, Samir; Diwadkar, Avantika R.; Dudley, Jesse W.; O’Brien, Janielle; Dvorin, Donald; Kenyon, Chén C.; Himes, Blanca E.; Hill, David A.; Henrickson, Sarah E.
title: COVID-19 Pandemic-Related Reductions in Pediatric Asthma Exacerbations Corresponded with an Overall Decrease in Respiratory Viral Infections
date: 2021-11-14
journal: J Allergy Clin Immunol Pract
DOI: 10.1016/j.jaip.2021.10.067
sha: 568d6ed2e495266789f9d2e1e7c321f096f48888
doc_id: 834008
cord_uid: 07dz9glc

Background Respiratory viruses, air pollutants, and aeroallergens are all implicated in worsening pediatric asthma symptoms, but their relative contributions to asthma exacerbations are poorly understood. A significant decrease in asthma exacerbations has been observed during the COVID-19 pandemic, providing a unique opportunity to study how major asthma triggers correlate with asthma activity. Objective To determine whether changes in respiratory viruses, air pollutants, and/or aeroallergens during the COVID-19 pandemic were concomitant with decreased asthma exacerbations. Methods Health care utilization and respiratory viral testing data between January 1st, 2015 and December 31st, 2020 were extracted from the Children’s Hospital of Philadelphia (CHOP) Care Network’s electronic health record. Air pollution and allergen data were extracted from U.S. Environmental Protection Agency public databases and a National Allergy Bureau-certified station, respectively. Pandemic data (2020) were compared to historical data. Results Recovery of in-person asthma encounters during phased re-opening (June 6th – November 15th, 2020) was uneven: primary care well and specialty encounters reached 94% and 74% of pre-pandemic levels, respectively, while primary care sick and hospital encounters reached 21% and 40% of pre-pandemic levels, respectively. During the pandemic, influenza A and influenza B decreased to negligible frequency when compared to pre-pandemic cases, while RSV and rhinovirus infections decreased to low (though non-negligible) pre-pandemic levels, as well. No changes in air pollution or aeroallergen levels relative to historical observations were noted. Conclusions Our results suggest that viral respiratory infections are a primary driver of pediatric asthma exacerbations. These findings have broad relevance to both clinical practice and the development of health policies aimed at reducing asthma morbidity.

aeroallergens are implicated in worsening pediatric asthma symptoms, the interplay 122 between them and asthma exacerbations is not well understood. Asthma exacerbations 123 decreased significantly during the COVID-19 pandemic allowing for the investigation of 124 these major asthma triggers to asthma activity. 125

infections and acute asthma activity we observed during the COVID-19 pandemic 127 supports a strong link between respiratory virus infections and pediatric asthma 128

How does this study impact currently management guidelines? Our Short-acting beta-agonist 156

Systemic Steroid 157 INTRODUCTION exposure to respiratory viruses, air pollution, and aeroallergens 2 . In addition, asthma 160 patients have more frequent, severe, and longer-lasting symptoms with respiratory viral 161 infections than people without asthma 3, 4 . Exposure to air pollutants, including particulate 162 pollution (PM2.5, PM10), ozone, and nitrogen dioxide (NO2), has been associated with 163 increased risk of asthma development, exacerbations, and hospitalizations [5] [6] [7] [8] [9] . In children 164 with atopic asthma, aeroallergens are also a cause of asthma exacerbations 10, 11 . 165

Public health interventions to mitigate the spread of SARS-CoV-2, the causative 166 virus of Coronavirus Disease 2019 , included social distancing, mask-167 wearing, and quarantining of sick or exposed individuals [12] [13] [14] [15] . While it was initially 168 suspected that asthma might be a COVID-19 risk factora concern that may have 169 increased preventive health behaviors among those with asthma, 16, 17 subsequent studies 170

showed that people with asthma who contract SARS-CoV-2 were at lower risk for adverse 171 outcomes 18, 19 . The institution and modulation of COVID-19-related public health 172 measures offer a unique opportunity to study their effects on health outcomes beyond 173 those directly infected, including asthma 20,21 . Asthma symptoms and exacerbations 174 decreased during the early stages of the COVID-19 pandemic 22 -25 , along with factors 175 that impact asthma, such as respiratory viral infections 23, [26] [27] [28] [29] . In a prior publication, we 176 found that during the first two months following public health interventions in Philadelphia, 177 in-person asthma visits and steroid prescriptions decreased by more than 80%, with a 178 concomitant decrease in rhinovirus infections and no change in air pollution compared 179 with historical data for the years 2015-2019 23 . Given that public health measures changed 180 throughout 2020 and health systems resumed more in-person services, we sought to 181 determine whether the decreased patterns of asthma activity we initially observed were 182 maintained throughout 2020, and whether levels of respiratory viral infections, air 183 pollution, and aeroallergens mirrored asthma activity. 184

Patient-level demographic characteristics of our study population are shown in Table E1 . 187

We extracted asthma patient data corresponding to January 1 2020. The first lockdown occurred March 18 th to June 5 th , 2020, while the phased re-197 opening consisted of an initial re-opening between June 6 th and June 26 th , 2020, and a 198 further re-opening between June 27 th and November 15 th , 2020. The second lockdown 199 was instituted on November 16 th and lasted through January 4 th , 2021. A more thorough 200 description of these timeframes may be found in the Online Repository Text. 201 202

For each encounter, its type (i.e., primary-well, primary-sick, specialty [Allergy and 204 Pulmonary], ED, inpatient, and ICU) and date were extracted, along with the patient's 205 gender, race, ethnicity, date of birth, and payor type. Race was based on self-or 206 parent/guardian-selection of one of the following categories: "White", "Black", "Asian or 207

Asthma-related drug prescription data for all outpatient asthma-related encounters (both 209 primary and secondary diagnosis) and inpatient asthma-related encounters (primary 210 diagnosis only) were obtained from CHOP prescription records (Table E2) AirNow (an air quality data management system that reports real-time and forecast air 233 quality estimates) 31 . Historical data from 2015-2019 for these pollutants was downloaded 234 from Air Data (an EPA resource that provides quality-assured summary air pollution 235 measures collected from outdoor regulatory monitors across the United States 32 ). AirNow 236 did not provide historical data for pollutants considered, and Air Data did not contain 2020 237 data, as its data is released months after the data is reported in AirNow. For regulatory 238 monitors included in our study, AirNow and AirData measures were obtained at the same 239 monitoring sites. During the phased re-opening June 6 th to November 15 th , there was a gradual 291 return of non-acute asthma-related outpatient encounters to historical and pre-pandemic 292 levels. Specifically, primary-well and specialty outpatient encounters rose to 94% (456 vs 293 486 encounters/week) and 74% of historical levels (317 vs 429 encounters/week), 294 respectively ( Figure 1A) . In contrast, primary-sick encounters increased to 21% of 295 historical levels (66 vs 319 encounters/week; Figure 1A) , and hospital encounters 296 increased to 40% of historical levels (75 vs 188 encounters/week; Figure 1B) .

prescription patterns at CHOP were similar to 2015-2019 historical averages. Comparison 299 of CHOP prescription patterns during the pre-lockdown to the first lockdown period found 300 ( Figure 2B) of these drug classes, their levels remained similar from January to June according to 307 historical data. We did not observe an asthma medication shortage in Philadelphia during 308 the early stages of the pandemic. 309

When examining the phased re-opening period, prescription patterns for all 310 medications showed a recovery towards historical and pre-pandemic levels, but all 311 remained lower through December 2020 (Figure 2) averages. Positive rhinovirus tests, while non-zero, also remained significantly lower 335 than antecedent averages (Figure 4) . 336

Air pollution and aeroallergen trends did not substantially change during the 337 pandemic compared to historical or expected seasonal data 11, 28 , respectively (Figure 5) . as well as a closer look at the seasonality of the viruses when compared to historical 358 averages due to the increased duration of the observation period. Allergen data during 359 this span has also been included in the current analysis. Additionally, analysis of well vs 360 sick outpatient encounters was performed, as was a more detailed analysis of asthma 361 medication prescriptions during this timeframe. 362

The COVID-19 pandemic has resulted in a substantial decrease in respiratory viral 363 infections 23,26,29,38 , including influenza 23,28,39-40 . Consistent with these reports, our results 364

show that the number of positive virus tests decreased and remained lower than the 365 historical average. Specifically, in the ED and satellite care centers between September 366 2020 to March 2021, when peaks in the number of IFV-A, IFV-B and RSV infections were 367 addition, the number of positive cases of rhinovirus, a key virus linked to asthma 369 exacerbations 41,42 , remained lower than historical averages. These trends may not be 370 solely due to behavioral responses or public health interventions, in that after the major 371 U.S. 2020 fall and winter holidays that were accompanied by ill-advised gatherings (e.g., 372

Thanksgiving, Christmas and New Year's Eve), the number of respiratory viruses, other 373 than SARS-CoV-2, remained very low even as COVID-19 rates increased. Given the 374 known role of respiratory viral infections as a trigger of asthma exacerbations, it is likely 375 that the sustained decrease in respiratory virus levels strongly contributed to the decrease 376 in asthma encounters in 2020. 377

Exposure to air pollution and aeroallergens contribute to asthma exacerbations 5-378 11 . Consistent with our prior publication that studied the early phases of the pandemic 23 , 379

we did not observe changes to levels of air pollutants that diverged from historical trends, 380

i.e. seasonal variability in PM2.5, PM10, ozone that peaked during summer 2020, and NO2 381 that peaked during winter 2020, were consistent with 2015-2019 trends 43, 44 . Similarly, 382 seasonal peaks were observed among aeroallergens from April to June and early 383

September, according to pollen type 11,45 , but these changes were consistent with 384 historical trends. Previously, we explicitly confirmed that the seasonal decrease in levels 385 of four air pollutants in the two month period following Philadelphia's first lockdown period 386 was not statistically significant when compared to historical trends via interrupted time 387 series analysis 23 . Here, having expanded the time period of observation, there were even 388 fewer differences between pollution levels across 2020 compared with historical patterns, 389

suggesting that the implementation and relaxation of COVID-19-related public health measured with regulatory monitors. Similarly, COVID-19-related measures did not 392 influence aeroallergen levels, as illustrated by our results for weed, tree, mold and grass 393 pollens which followed expected seasonal trends. We note however, that our data has 394 limitations. First, the EPA data used comprised monitoring sites that sparsely cover of the The recovery of non-acute asthma care during the reopening phases suggests that 406 people were willing and able to access healthcare during this time. Thus, reduced access 407 was not a major driver of the persistently low acute outpatient and inpatient asthma 408 encounters. The persistently decreased prescription levels of maintenance and acute 409 management medications during phases in which routine asthma care encounters were 410 recovering further supports that a decrease in asthma exacerbations and symptoms 411 occurred. However, it is possible that fear of coming to the pediatrician's office or hospital, 412 out of concern for increased SARS-CoV-2 infection risk, drove some of the effect on acute volumetric survey of airborne pollen in Philadelphia, Pennsylvania (1991 -1997 and 571 

Vital Signs: Asthma 438 in Children -United States

Role of viral infections in the development and exacerbation of 442 asthma in children

Frequency, severity, and duration of rhinovirus infections in asthmatic and non-445 asthmatic individuals: a longitudinal cohort study

Ambient Air Temperature, and Asthma Symptoms in the Past 14 Days among Adults 448 with Active Asthma

Covid-19 Outbreak Progression in Italian 475

Regions: Approaching the Peak by the End of March in Northern Italy and First Week of 476

Emergence of a Novel Coronavirus (COVID-19): Protocol for Extending 479

JMIR Public Health Surveill

Asthma and COVID-19: Is asthma a risk factor for severe 483 outcomes? Allergy

Does asthma make COVID-19 worse?

Prevalence of asthma in hospitalized and non-hospitalized children with COVID-19

Asthma and Allergies 574 in the School Environment

The effect of 4-aminopyridine-induced 576 increased neuromuscular activity on the metabolism of developing muscles in chick 577 embryos

Sandel MT; SECTION ON ALLERGY AND 579 IMMUNOLOGY; COUNCIL ON ENVIRONMENTAL HEALTH

Control Practices and Asthma Management

Experiences of 583 lockdown during the Covid-19 pandemic: descriptive findings from a survey of families 584 in the Born in Bradford study

Trajectories of anxiety and depressive symptoms 586 during enforced isolation due to COVID-19 in England: a longitudinal observational 587 study

Symptoms of depression and anxiety during the COVID-589 19 pandemic: implications for mental health

SARS-CoV-2 and the resurgence of rhinovirus

53. CDC guidelines on preventing Respiratory Syncytial Virus (RSV) Infection

597 historical average data (dark grey lines) and +/-1SD (light grey shaded areas

influenza B, RSV, rhinovirus, and COVID-19 tests respectively. Phases of Philadelphia 623 COVID-19-related public health measures are shown

Viral respiratory testing data comparing

testing data (dark blue lines with blue markers) from the 628 ED and satellite sites are compared to the 2015-2019 historical average

grey shaded areas) during the same period. Weeks during which tests were 630 not performed are without markers

Figure 5. Levels of air pollutants and aeroallergens in Philadelphia during the COVID-19

A) Trendlines of weekly averages of daily NO2, ozone, PM10 and PM2.5 634 measures from 2020 and 2015-2019 sourced from AirNow and AirData, respectively, for 635 the period January 1st to December 31st in Philadelphia. ppb = parts per billion; µg/m3 = 636 micrograms per cubic meter

NAB)-certified 638 station near Philadelphia during the period