Development and initial testing of Asthma Predictive Index for a retrospective study: an exploratory study.

Area of Science:

• Pediatric Pulmonology
• Epidemiology
• Clinical Research Methodology

Background:

• The Asthma Predictive Index (API) is established for prospective asthma prediction.
• Its utility in retrospective studies for diagnosing asthma has not been previously validated.
• Retrospective cohort studies often require reliable methods for assessing historical disease status.

Purpose of the Study:

• To develop and validate a modified Asthma Predictive Index (API) suitable for retrospective asthma diagnosis.
• To assess the criterion and construct validity of the retrospective API.

Main Methods:

• A cross-sectional study utilizing data from a prior retrospective cohort.
• Operationalized retrospective API using wheezing episodes and original API criteria.
• Assessed validity against Predetermined Asthma Criteria (PAC) using kappa statistics and agreement rates.
• Examined associations between the retrospective API and known asthma risk factors.

Main Results:

• The study included 105 children (52.4% male, 85.7% Caucasian, mean age 5.8 years).
• Retrospective API identified 14.3% of children with asthma, while PAC identified 31.4%.
• High agreement (89.5%) and significant kappa (0.66, p<0.01) were found between retrospective API and PAC.
• Atopy, lower parental education, no breastfeeding, and family history of asthma were significant risk factors.

Conclusions:

• The modified Asthma Predictive Index (API) is suitable for ascertaining asthma status in retrospective studies.
• Findings suggest the retrospective API is a valid tool for asthma assessment.
• Further validation with larger sample sizes is recommended.