Effect of the COVID-19 Lockdown on Asthma Biological Rhythms

Abstract

Asthma has a striking temporal character, in which time-of-day, patient age, and season each influence disease activity. The extent to which rhythms in asthma activity reflect exposure to specific disease triggers remains unclear. In this study, we examined how virus mitigation strategies enacted during the COVID-19 pandemic ("lockdown measures") affected rhythms in asthma clinical activity in children. To this end, we retrospectively analyzed asthma clinical presentations in children aged <18 years to our regional academic medical center, comparing 4 years of medical records prior to COVID-19 lockdown measures with the 12 months immediately after the institution of such measures. We correlated these data to positive viral test results, febrile seizures, and allergic clinical surrogates (allergic reaction visits and Emergency Department [ED] antihistamine prescriptions, respectively) over the same time frame. In the 12 months following the institution of the COVID-19 lockdown, positivity rates for common respiratory viruses dropped by 70.2% and ED visits for asthma among children dropped by 62% compared to pre-COVID years. Lockdown suppressed seasonal variation in positive viral tests and asthma ED visits, while diurnal rhythms in asthma visits were unchanged. Asthma seasonality correlated most strongly with rhinovirus positivity both before and after the institution of COVID lockdown measures. Altogether, our data support a causal role for viruses in driving seasonal variability in asthma exacerbations in children.

Keywords: COVID-19; asthma; biological rhythms; seasonality; viruses.

Figure 1:. COIVD-19 lockdown measures suppressed common respiratory viruses in children.

(A) Number of positive viral tests/wk. from Spring 2016 to Spring 2021. (B, C, D) Seasonality of viral positivity rates (B), ED presentations for febrile seizures (C), and aeroallergen (pollen and mold) counts (D) pre- and post-COVID lockdown. Left panels, average rates pre-lockdown from Spring 2016 through Spring 2020 (blue circles, mean ± SE). Right panels, rates post-lockdown from Spring 2020-Spring 2021. JTK Cycle generated p values for seasonality are depicted. Correlation between febrile seizures and viral positivity pre-lockdown is shown in panel (B). Sp= Spring; S=Summer; F=Fall; W=Winter

Figure 2:. COVID lockdown suppressed asthma seasonality.

Weekly rates of pediatric asthma ED visits (A), regular hospital admissions (B), PICU admissions (C), and percentage of ED visits resulting in hospital admission (D). Left panels, average rates pre-lockdown from Spring 2016-Spring 2020 (blue circles, mean ± SE). Right panels, rates post-lockdown from Spring 2020-Spring 2021. JTK Cycle generated p values and best fit curves for seasonality are depicted where appropriate. Sp= Spring; S=Summer; F=Fall; W=Winter.

Figure 3:. Seasonal rhythm analysis using generalized linear modeling (GLM).

Line graphs depicting incidence rate ratios (IRR) for different clinical events (median ± 95% confidence intervals). In this visualization, statistically significant yearly periodicity (p<0.05, after adjustment for day of the week and calendar season) is signified when the confidence intervals do not cross 1 (dashed vertical line). Blue symbols, averaged Pre-COVID lockdown data from March 2016-March 2020. Red symbols, post-lockdown data collected from March 2020-March 2021. See Methods for a description of our use of GLM for time series analysis.

Figure 4:. Preserved diurnal but not weekly rhythms in ED asthma visits during COVID lockdown.

(A) Frequency of asthma visits binned by hour of ED presentation. Blue circles, pre-COVID lockdown data from 2016–2020 (mean ± SE). Red lines, ED asthma visits in the 12 months post-lockdown. Adjusted p values generated by JTK cycle are depicted. Pearson’s correlation coefficient between pre– and post-COVID data is displayed within each graph. (B) Heat map depicting the Pearson’s correlation matrix for diurnal rhythms in asthma ED visits. Each year represents a 12 month spring-spring interval. (C) Frequency of asthma visits binned by day-of-week of ED presentation. Blue circles, pre-COVID lockdown data from 2016–2020 (mean ± SE). Red lines, ED asthma visits in the 12 months post-lockdown. Significance by One-Way ANOVA is depicted (see also Table S5). (D) Heat map depicting the Pearson’s correlation matrix for weekly rhythms in asthma ED visits. Each year represents a 12 month spring-spring interval.