Impact of varying childhood asthma definitions on incidence and clinical outcomes.

BACKGROUND

The absence of a standardized asthma definition in epidemiologic studies undermines the consistency of incidence estimates and the comparability of clinical outcomes.

OBJECTIVE

To evaluate the impact of integrating bronchodilator response (BDR) into parent-reported asthma definitions by comparing incidence, disease severity, control, and diagnostic performance across these definitions, and to assess the individual diagnostic performance of BDR and fractional exhaled nitric oxide in identifying asthma cases.

METHODS

We used data from a prospective cohort of 919 infants with severe (hospitalized) bronchiolitis to explore 4 asthma definitions: (1) broad, any physician diagnosis of asthma by age 6 years, as reported by parents; (2) epidemiologic, definition 1 plus either asthma medication use (eg, inhaled bronchodilator, inhaled corticosteroid, systemic corticosteroid, and montelukast) or asthma-related symptoms between ages 5.0 and 5.9 years; (3) alternative strict, definition 2 plus a post-BDR increase of 8% or greater in predicted forced expiratory volume in 1 second; and (4) strict, definition 2 plus a post-bronchodilator increase of more than 10% in predicted forced expiratory volume in 1 second. Outcomes were assessed across these definitions, and their diagnostic performance was compared with a physician reviewer's asthma diagnosis (reference standard).

RESULTS

The incidence rates for the 4 definitions were 37.2%, 27.7%, 13.2%, and 9.3%, respectively, with intermittent asthma severity classification following a similar pattern (72.4%, 65.5%, 56.0%, and 52.8%). Transitioning from the first to the fourth definition improved specificity and positive predictive value but reduced sensitivity, with no consistent trends being observed for asthma control across definitions.

CONCLUSION

These findings suggest that incorporating BDR into parent-reported asthma definitions underestimates asthma incidence and identifies cases with worse clinical outcomes.