Adapting and Expanding the Algorithmic Software Tool to Help Manage Asthma (ASTHMAXcel) for Pediatric Use

Abstract

RATIONALE: Asthma control determination relies on accurate reporting of SABA utilization; however, it is often limited by recall bias. We therefore assessed the utility of self-reported SABA use. METHODS: Data were collected from 1,062 adults (2016-2018) with self-reported asthma and > _4 weeks of EMM-recorded SABA use prior to the completion of a mobile Asthma Control Test (ACT). ACT Question 4 (Q4) assessed self-reported rescue inhaler/nebulizer medication use during the prior 4 weeks with responses: 1) not at all, 2) < _1x/week; 3) 2-3x/week; 4) 1-2x/day; and 5) > _3x/day. Four-week mean EMM-recorded SABA puffs was binned into comparable responses: 1) 0/week; 2) 1x/week; 3) >1-3x/ week; 4) >3-14x/week; and 5) >14x/week. Asthma was considered ''uncontrolled'' with responses of 3), 4), or 5), then sensitivity, specificity, and area under the receiver operating curve (AUC) was calculated to evaluate the utility of Q4 on EMM-defined ''uncontrolled'' asthma using logistic regression. Lastly, prevalence of under-reported SABA use was quantified as the proportion of patients with a Q4 response coinciding with lower EMM-recorded use. RESULTS: The median Q4 response was 3 (IQR:2-4) and EMMs recorded on average 8 (SD:16) puffs/week. Approximately 60.5% and 64.9% of patients had uncontrolled asthma according to Q4 and EMMs, respectively. The Q4 sensitivity, specificity, and AUC was 82%, 63% and 0.72, respectively, with 35% of patients under-reporting SABA use. CONCLUSIONS: The utility of self-reported vs. objectively-recorded SABA use was good. Under-reporting was prevalent, highlighting the potential value of EMM data to more precisely identify uncontrolled asthma with higher sensitivity.