Effectiveness of Montelukast on asthma control in infants: methodology of a French claims data study

Manon Belhassen, Gérard de Pouvourville, Laurent Laforest, Jacques Brouard, Jacques de Blic, Brigitte Fauroux, Valérie Laigle, Céline Chanut-Vogel, Liliane Lamezec, Eric Van Ganse

This pilot study, conducted on a 1/97th representative sample of French claims data, prepared a project to assess the effectiveness of Montelukast (MTL-4) as add-on therapy for asthma in infants (6-24 months) compared to inhaled corticosteroids (ICS), based on real-world data. Due to the very recent opening of French claims data for effectiveness research, and the complex structure of this data source, we first tested the feasibility of identifying infants with asthma and outcome criteria, and the ability to perform relevant comparisons. We identified a cohort of infants with uncontrolled asthma and receiving ≥2 consecutive dispensations of any respiratory drug (R03 ATC classification) during a 6-month period. Uncontrolled asthma was identified either from exacerbations or from markers of acute loss of asthma control; date of occurrence of an event (exacerbation and/or acute loss of asthma control) was defined as index date. The study groups comprised infants receiving MTL-4 +/- ICS (MTL-4 group) or ICS without MTL-4 (ICS group) at index date. These two groups were matched on gender, age, quarter of index date, therapy before index date, past asthma-related hospitalization (ever), and were followed for 6 months. The outcome was the rate of infants with uncontrolled asthma, defined as above. This pilot cohort study included 1,149 infants with asthma (mean age 14.1 months, 64% boys). Of these, 51 and 768 were assigned to the MTL-4 and ICS groups, respectively. Uncontrolled asthma occurred in 78.8% and 78.4% of infants in these groups, respectively (oral corticosteroids were dispensed to 49% and 61%, respectively). Assessment of uncontrolled asthma, exposure to MTL-4 and ICS, and occurrence of outcomes were achieved. For the development of matching criteria, we defined a new marker of severity (therapeutic typologies). These data support the feasibility of the final project, to be conducted on claims data from the whole French population. We also showed that, with appropriate methodology and by using valid criteria, French claims data are an adequate resource for conducting comparative effectiveness studies in pediatric asthma. Finally, the algorithm used to identify infants with asthma could be applied to other studies using claims data.