Abstract
Objective: To compare the Inhaler Compliance AssessmentTM (INCATM), a novel audio-recording device objectively measuring timing and proficiency of inhaler use, against established adherence measures, and explore its discriminant and predictive validity.
Design: Prospective observational study; 184 chronic obstructive pulmonary disease (COPD) patients used an INCATM-enabled salmeterol/fluticasone inhaler for one-month post-hospital discharge.
Main outcome measures: INCATM (Attempted, Attempted Interval, Actual) adherence correlated with Doses Used Rate, self-reported adherence and prescription refill for concurrent validity. Discriminant validity for reason for admission, cognition and lung function; predictive validity for health status and quality-of-life.
Results: Rates of Attempted, Attempted Interval and Actual adherence were 59, 47 and 23%, respectively. Only 7% of participants had Actual adherence >80%. INCATM variables significantly correlated with Doses Used Rate but not with self-report; Attempted and Attempted Interval were weakly associated with prescription refill. Higher cognitive and lung functioning groups had better INCATM adherence. Attempted and Attempted Interval predicted health status, while Doses Used Rate predicted quality-of-life.
Conclusion: INCATM did not strongly correlate with self-report or prescription refill data. Discriminant and predictive validity demonstrated by INCATM suggests the potential utility of the INCATM as a method to identify intentional and unintentional adherence to inhaled medication and facilitate targeted intervention.
Acknowledgements
This project is a research-initiated study supported by the Royal College of Surgeons in Ireland Research Institute [grant number 1855] and hosted in Beaumont Hospital, Royal College of Surgeons in Ireland, Dublin, Ireland. The INCATM device was manufactured and supplied by Vitalograph, Ennis, Ireland and the salmeterol/fluticasone DiskusTM inhalers were provided by GlaxoSmithKline. The authors’ responsibilities were as follows: LM, FD, GM and RWC designed the research; RBR and RWC developed the electronic engineering and signal processing methodology; IS & RWC developed the INCATM adherence calculation methods; LM, FD, IS, GG and RWC advised on the statistical analysis; CM was primarily involved in patient recruitment, data collection and data analysis; KB retrieved the prescription refill data and calculated MPR and PDC variables; CM and LM wrote the manuscript; LM had primary responsibility for the final content. All co-authors edited and approved this manuscript. We would like to extend our sincerest gratitude to all participants for contributing their time and effort, and also to the staff of Beaumont hospital for their support in facilitating this research.
Notes
† Research conducted in Beaumont Hospital, Royal College of Surgeons in Ireland, Dublin, Ireland.
This article was originally published with errors. This version has been amended. Please see Corrigendum (http://dx.doi.org/10.1080/08870446.2017.1352219)