Asthma severity categorization using a claims-based algorithm or pulmonary function testing.

Area of Science:

• Pulmonary Medicine
• Health Informatics
• Epidemiology

Background:

• Asthma severity classification is crucial for effective patient management and resource allocation.
• Administrative databases offer a scalable method for analyzing large patient populations.
• The accuracy of administrative data for asthma severity assessment compared to clinical measures like pulmonary function tests (PFTs) requires evaluation.

Purpose of the Study:

• To determine if pulmonary function test (PFT) results significantly alter asthma severity categorization when using algorithms based on administrative data.
• To compare asthma severity classifications derived from claims data versus PFT results.
• To analyze the impact of different categorization methods on healthcare costs.

Main Methods:

• A cohort of asthma patients aged 6-64 with available PFT data was identified from a central Massachusetts claims database (1999-2005).
• Asthma severity was categorized using a claims-based algorithm and by PFT results (percent predicted forced expiratory volume in 1 second [FEV1] or peak expiratory flow [PEF]).
• Gamma rank correlation assessed the association between methods; healthcare costs were compared across categories.

Main Results:

• A significant ordinal association was found between the claims-based and PFT-based severity categorizations (p = 0.0002).
• The PFT method resulted in more mild and fewer moderate/severe asthma categorizations compared to the claims-based algorithm.
• PFTs altered the severity category in only 10.9% of patients; both methods showed higher costs for more severe asthma, with similar overall costs except in moderate categories.

Conclusions:

• The claims-based algorithm generally classified asthma as more severe than the PFT-based approach.
• PFT results would have appreciably changed asthma severity categorization in a small patient subset.
• Findings support the utility of administrative database analyses for evaluating asthma care in large populations.