S84 Can clinicians accurately predict non-adherence to medication in patients with difficult asthma? A comparison between clinical judgement and prescription issue data. (2nd December 2011)

• Record Type: Journal Article
• Authors: Murphy, A C; Belingheri, S; Brightling, C E; Wardlaw, A J; Pavord, I D; Bradding, P; Green, R H
• Is Part Of: Thorax. Volume 66(2011)Supplement 4
• Page Start: A40
• Page End: A41
• Publication Date: 2011-12-02
• Subjects: Chest -- Diseases -- Periodicals; Thorax; Chest -- Diseases; Periodicals; Periodicals; 617.54
• DOI: 10.1136/thoraxjnl-2011-201054b.84
• Languages: English
• ISSNs: 0040-6376

Abstract

Abstract : Background: It is increasingly recognised that sub-optimal adherence to prescribed medication regimes is common in patients with difficult asthma and may be associated with adverse clinical outcomes. Simple reliable methods of measuring non-adherence have not been developed and clinicians' judgement is often relied upon as the only assessment of adherence. We have previously shown that checking prescription data are a useful method of investigating non-adherence in this patient population. We aimed to determine how well a clinician's judgement of adherence correlated to adherence as measured by prescription issue data. Methods: Adult asthma patients attending a difficult asthma clinic during July/August 2010 were included. GP retrospective prescription issue data and hospital dispensing data for asthma medications, patient demographics and clinical outcome data were collated. The medication adherence score was calculated as the number of doses issued/number of doses prescribed × 100 for a mean duration of 12 months. Clinicians with an expertise in the management of difficult asthma were asked to stratify each patient according to their perceived adherence, classified as adherent (=80% adherence score; partially adherent 50%–79%; non-adherent <50%). Agreement between clinician judgement and adherence score was analysed using a weighted κ coefficient. Logistic regression was performed to determine whether any clinical features could predict better agreement. Results: Data from 63 consecutive patients was included in the analysis (41 female, median (range) age 51.2(61.3)). Clinicians suspected non-adherence in 11/63 (17.4%) of patients, partial non-adherence in 23/63 (36.5 %) and good adherence in 29/63 (46%). Agreement between clinician assessment and prescription issue data were seen in only 23/63 (36.5%) of cases and overall agreement using a weighted κ coefficient was poor (weighted agreement 63.5%, expected agreement 58.8%, κ 0.11, SE 0.1, z=1.16, p=0.12). There was no relationship between patients' age, gender, Juniper asthma control score, prescribed inhaled corticosteroid dose or FEV1 percent predicted and the chances of agreement or disagreement between the two methods. Conclusions: Clinical judgement alone appears to be a poor predictor of adherence to medication in patients with difficult asthma. The assessment of non-adherence requires objective measurements. Prescription issue data are one such measurement; but further work is required.