Distinguishing phenotypes of childhood wheeze and cough using latent class analysis

Area of Science:

• Pediatric Pulmonology
• Respiratory Medicine
• Clinical Epidemiology

Background:

• Childhood airway diseases are diverse, with limited phenotypic classification.
• Existing approaches often overlook multiple disease dimensions.
• Accurate phenotyping is crucial for understanding disease trajectories and treatment.

Purpose of the Study:

• To identify distinct phenotypes of childhood wheeze and chronic cough using a multidimensional statistical model.
• To compare identified phenotypes based on symptoms, lung function, and airway responsiveness.
• To evaluate the predictive value of these phenotypes for school-age outcomes.

Main Methods:

• Latent class analysis applied to a population-based cohort of 1,650 preschool children (319 with wheeze/cough).
• Data included parent-reported symptoms, skin-prick tests, lung function, and airway responsiveness from two surveys.
• Phenotype outcomes were assessed at school age for subsequent wheeze, cough, and inhaler use.

Main Results:

• The model identified three distinct phenotypes of childhood wheeze and two of chronic cough.
• Significant differences in subsequent wheeze, chronic cough, and inhaler use were observed across the five phenotypes at school age.
• The identified phenotypes partially reconcile discrepancies in existing classification systems.

Conclusions:

• A novel multidimensional approach effectively identifies clinically relevant phenotypes in childhood airway diseases.
• These phenotypes demonstrate predictive value for long-term respiratory outcomes.
• This methodology holds potential for improving phenotype definition in both pediatric and adult obstructive airway diseases.