METHODS

Analytic overview

A model-based cost-effectiveness analysis was conducted to estimate the ratio of incremental costs and health benefits between SCIT + ICS (intervention strategy) versus ICS (comparator strategy). A hypothetical cohort (1,000 patients per strategy) of pediatric patients (8-year-old at baseline) with a diagnosis of moderate persistent AA (sensitized to HDM with clinically relevant symptoms) without AR was defined as the target population to be simulated in the base case scenario. The SCIT+ICS strategy consisted of a monthly administration scheme for three years (a period in which the effect of SCIT is expected to be perceived) followed by ICS within a 10-year time horizon (13). The comparator strategy consisted in treatment with ICS + symptomatic medications during the overall time horizon. The 10-year time horizon was defined as a period in which the differences in long-term effects of SCIT+ ICS vs ICS in health outcomes and costs would be observed. Different scenario analyses were performed considering allergic rhinitis (AR) as a comorbid condition in the base-case population and a cohort of adult patients (18 years old at baseline) with and without AR to evaluate potential differences in cost-effectiveness estimations. The perspective of the health care system was adopted, and only direct-medical costs were used.
Benefits associated to the evaluated strategies were expressed as quality-adjusted life-years (QALYs). Costs and QALYs were discounted at 5% per year according to the Colombian guidelines for conducting economic evaluations (14). An exchange rate of $3,250 Colombian pesos (COP) per one American dollar (USD) was used to convert from 2018 COP to USD. A willingness-to-pay threshold (WTP) of $18,125 USD (i.e., three per capita Gross Domestic Product (GDP) of Colombia) per additional QALY was defined as the criteria for evaluating cost-effectiveness (14). The decision model was built using MS Excel 2018 (Microsoft, Redmond, WA).

Measures of effectiveness

The effect of the evaluated therapies was measured in terms of reduction in the probability of a moderate or severe exacerbation (as an event of emergency department visit – ED, or hospitalization, respectively), and the reduction or discontinuation of asthma medications (as an indicators of reduced disease severity and risk of exacerbations) (15). These outcomes reflect an improvement of health-related quality of life and a reduction in costs, and are thus considered robust metrics for relating clinical effectiveness to costs and QALYs (9,16,17).

Model structure

A Markov model was developed based on the stepwise approach proposed by the Global Strategy for Asthma Management and Prevention of the Global Initiative of Asthma (GINA) in which treatment steps are associated to different levels of asthma severity (18) (Figure 1). This conceptual structure was used to create a flexible and reproducible model accounting for relevant outcomes in clinical practice such as ED visits, hospitalizations and medication-step down that have not been widely used in previous HDM SCIT cost-effectiveness studies despite its recognized relevance for policy makers and clinicians (17,19). Furthermore, this model can be used to account for the different recommended treatment schemes within each GINA Step according to the target population to be simulated (12,18).