RESUMEN
RATIONALE: Objective adherence to inhaled therapy by patients with chronic obstructive pulmonary disease (COPD) has not been reported. OBJECTIVES: To objectively quantify adherence to preventer Diskus inhaler therapy by patients with COPD with an electronic audio recording device (INCA). METHODS: This was a prospective observational study. On discharge from hospital patients were given a salmeterol/fluticasone inhaler with an INCA device attached. Analysis of this audio quantified the frequency and proficiency of inhaler use. MEASUREMENTS AND MAIN RESULTS: Patients with COPD (n = 244) were recruited. The mean age was 71 years, mean FEV1 was 1.3 L, and 59% had evidence of mild/moderate cognitive impairment. By combining time of use, interval between doses, and critical technique errors, thus incorporating both intentional and unintentional nonadherence, a measure "actual adherence" was calculated. Mean actual adherence was 22.6% of that expected if the doses were taken correctly and on time. Six percent had an actual adherence greater than 80%. Hierarchical clustering found three equally sized well-separated clusters corresponding to distinct patterns. Cluster 1 (34%) had low inhaler use and high error rates. Cluster 2 (25%) had high inhaler use and high error rates. Cluster 3 (36%) had overall good adherence. Poor lung function and comorbidities were predictive of poor technique, whereas age and cognition with poor lung function distinguished those with poor adherence and frequent errors in technique. CONCLUSIONS: These data may inform clinicians in understanding why a prescribed inhaler is not effective and to devise strategies to promote adherence in COPD.