A key diagnostic criterion of asthma is bronchodilator responsiveness, traditionally defined as an improvement in the forced expiratory volume in 1 s (FEV1) or forced vital capacity (FVC) after bronchodilator administration.1 2

Diurnal variability in symptom presentation is characteristic of asthma, with previous research suggesting that a biological clock mechanism is an important factor in asthma pathogenesis.3 With respect to bronchodilator responsiveness, it is unknown if the effect of diurnal variability is restricted to only patients with asthma. Furthermore, the role of seasonal variation on bronchial hyperreactivity is unknown. We investigated the association between time of day and season of testing on the level of bronchodilator responsiveness.

We retrospectively analysed data from patients referred to Cambridge University Hospitals NHS Foundation Trust between 1 January 2016 and 31 December 2023. Data were included in this study for all patients who were 18 years or older at the time of testing, who had acceptable and reproducible prebronchodilator and postbronchodilator spirometry.4 Data were extracted directly from the testing software.

Spirometry was instructed by certified physiologists. Spirometry was performed on the MasterScreen-PFT-Pro from Vyaire Medical before, and 20 min after inhalation of 400 µg of salbutamol delivered via a metered dose inhaler and spacer. Spirometers were calibrated daily and serviced according to manufacturer guidance. GLI race-neutral reference equations were used to calculate the lower limit of normal and percent predicted values.5 We defined bronchodilator responsiveness for both the American Thoracic Society/European Respiratory Society (ATS/ERS) 2005 (change in FEV1 or FVC≥12% and ≥200 mL of the initial value),1 and ATS/ERS 2022 definitions (change of >10% relative to the predicted value for FEV1 or FVC).2

We performed multivariable logistic regression analyses to investigate the influence of time of day on the odds of having a positive bronchodilator response. We considered time of day in the context of the working hours of the lung function testing laboratory (08:30–16:30). We modelled time as a continuous (per 1-hour increment), binary (08:30–13:30 vs 13:30–16:30) and categorical predictor (08:30–11:00 vs 11:01–14:00 vs 14:00–16:30). We adjusted regression models for age (years), sex (male/female), body mass index (BMI) (kg/m2), smoking history (ever/never) and prebronchodilator FEV1/FVC. To check for effect modification, we performed stratified analyses on those referred for asthma/query asthma vs any other referral reason. We also investigated the significance of any interaction between referral reason and time of day on the odds of having a positive bronchodilator response. We compared the results from the model with the interaction term to a model with the individual predictors using the likelihood ratio test. Referral reason was based on clinical judgement and extracted from the consultant referral document. We repeated the above analyses to investigate the influence of seasonal variation on bronchodilator responsiveness. We considered results significant if the p value was less than 0.05. Analyses were performed using Stata V.18 (StataCorp).

Data from 1620 individual patients were included in our analyses (online supplemental efigure 1). 62% (1004 of 1620) were female, mean (SD) age was 53.2 (16.1) years and mean BMI was 29.4 kg/m2 (7.1). 48% (722 of 1620) of patients reported having ever smoked, with 58% (944 of 1620) referred for asthma/query asthma. Other referral reasons are described in online supplemental etable 1. 25% of patients (400 of 1620) had bronchodilator responsiveness according to the ATS/ERS 2005 definition, compared with 26% (416 of 1620) of patients when using the 2022 definition. A greater proportion of patients tested in the morning had bronchodilator responsiveness compared with those tested in the afternoon for both ATS/ERS 2005 (28% vs 22%) and ATS/ERS 2022 (28% vs 23%) definitions (table 1, figure 1a).

Characteristics of study participants overall and by testing time

Graph showing (A) the association between time of test and bronchodilator responsiveness and (B) the association between season of test and bronchodilator responsiveness. ATS/ERS, American Thoracic Society/European Respiratory Society.

The results of the multivariable logistic regression showed that per 1-hour increment in the working day, the odds of having bronchodilator responsiveness declined by 8% (OR 0.92, 95% CI 0.88 to 0.97). When the time of day was dichotomised into morning and afternoon sessions, there was a similar association, with reduced odds of bronchodilator responsiveness in the afternoon (OR 0.68, 95% CI 0.54 to 0.85) (table 2). Patients referred for asthma/query asthma had a significant association with bronchodilator responsiveness, with the magnitude of the effect similar to that of the main analysis, while there was no association in those referred for other reasons (online supplemental etables 2,3). There was also evidence of a significant interaction between referral reason and time of day (per 1-hour increment) on the odds of having a positive bronchodilator response (p value for likelihood ratio test <0.001).

Logistic regression analysis of the association between time of test and having bronchodilator responsiveness

For the seasonal analysis, proportionally more patients tested in the winter had bronchodilator responsiveness compared with other seasons (figure 1b). Compared with those tested in winter, patients tested in autumn had significantly lower odds of having bronchodilator responsiveness but only for those with bronchodilator responsiveness according to the ATS/ERS 2005 definition (OR 0.67, 95% CI 0.48 to 0.92) (online supplemental etables 4,5)

We have found that in contrast to patients tested in the afternoon, patients tested for asthma/query asthma in the morning had lower lung function and were more likely to exhibit bronchodilator responsiveness. Studies have shown that the clock protein REV-ERBα controls airway hyperresponsiveness at different times of day, and that knocking down the Rev-erbα gene abolishes the time-of-day effect.6 7 The results from our study are consistent with the recently published RADicA study, which in a relatively small sample of patients with suspected asthma found that patients self-reported their worst symptoms in the morning and night, and that patients were more likely to demonstrate bronchodilator responsiveness in the morning compared with the afternoon.8 Taken together, these findings suggest that performing diagnostic testing when patients are most symptomatic, that is, in the morning, may improve diagnostic testing. This also reinforces the need to document the timing of spirometry testing whenever possible, and that repeat testing should be done at the same time of day.

Similarly, we also found seasonal variation in bronchodilator responsiveness, with bronchodilator responsiveness more likely to be seen in the winter months. Interestingly, previous groups found that adults or children with asthma or allergic rhinitis display greater FeNO and methacholine hyperresponsiveness during pollen season, characteristically seen during summer.9 10 We saw a slight increase in the proportion with bronchodilator responsiveness in the summer compared with the spring and autumn, but still to a lesser extent than seen in the winter. This could be because previous research has focused on a younger, more atopic population where an IgE response is more predominant, compared with older adults where asthma mortality and hospital admissions are higher during winter.

Although limited by cross-sectional study design and a lack of information regarding the formal diagnosis, we show that patients with asthma/query asthma are more likely to demonstrate bronchodilator responsiveness if spirometry testing is done in the morning compared with later in the working day. We also show that bronchodilator responsiveness is more likely in the winter months. This should be considered by clinicians when interpreting the results of bronchodilator responsiveness testing. It is important that prospective studies are conducted to confirm our findings, with particular focus on repeated measurements of the same individuals at different times of the day.

Not applicable.

This study involves human participants and approval to use the data was provided by the Electronic Healthcare Record Research and Innovation (ERIN) database data access committee (Reference A096990).This was a retrospective study of anonymised routinely collected healthcare data, therefore, individual participant consent was not required.