Identification of asthmatic children using prescription data and diagnosis

Area of Science:

• Pediatric Respiratory Medicine
• Health Services Research
• Pharmacoepidemiology

Background:

• Asthma is a prevalent chronic respiratory condition in children, requiring accurate identification for effective management.
• Accurate identification of pediatric asthma cases is crucial for public health surveillance and targeted interventions.
• Leveraging electronic health records and prescription data offers a scalable approach to population-level health assessment.

Purpose of the Study:

• To develop and validate a method for identifying children aged 6-14 years with asthma using prescription data for anti-asthmatic medications.
• To optimize a model that maximizes the inclusion of correctly diagnosed asthmatic children while minimizing false positives.

Main Methods:

• A register-based study analyzed prescription data for 125,907 Danish children aged 6-14 years.
• Asthma diagnoses were validated using hospital discharge information and general practitioner questionnaires.
• Various models combining anti-asthmatic drug prescriptions and different time windows were evaluated for optimal performance.

Main Results:

• The optimal model identified children with at least one anti-asthmatic prescription (excluding specific beta-agonist or steroid formulations) within a 12-month period.
• This model achieved a specificity of 0.86 and a sensitivity of 0.63.
• Extending the observation period did not improve specificity and significantly decreased sensitivity.

Conclusions:

• Register-based prescription data can effectively identify school-aged children with asthma.
• This validated method is valuable for health services research and proactive management of pediatric asthma.
• The findings support the use of prescription data as a tool for epidemiological studies and clinical care improvement.