Identifying patient-specific beliefs and behaviours for conversations about adherence in asthma

• Published in: Internal medicine journal 2012-06, Vol.42 (6), p.e136-e144
• Main Authors: Foster, J. M., Smith, L., Bosnic-Anticevich, S. Z., Usherwood, T., Sawyer, S. M., Rand, C. S., Reddel, H. K.
• Format: Article
• Language: English
• Subjects: adherence; Adult; Aged; Asthma - therapy; behaviour; belief; compliance; Cross-Sectional Studies; Disease Management; Drug Monitoring - methods; Factor Analysis, Statistical; Female; Humans; inhaled corticosteroid; Male; medication; Middle Aged; Multivariate Analysis; Patient Compliance - statistics & numerical data; Principal Component Analysis; Prospective Studies; Self Report; therapy
• ISSN: 1444-0903; 1445-5994
• DOI: 10.1111/j.1445-5994.2011.02541.x

Background:  Asthma guidelines advise addressing adherence at every visit, but no simple tools exist to assist clinicians in identifying key adherence‐related beliefs or behaviours for individual patients. Aims:  To identify potentially modifiable beliefs and behaviours that predict electronically recorded adherence with controller therapy. Methods:  Patients aged ≥14 years with doctor‐diagnosed asthma who were prescribed inhaled corticosteroid/long‐acting β2‐agonist (ICS/LABA) completed questionnaires on medication beliefs/behaviours, side‐effects, Morisky adherence behaviour score and Asthma Control Test (ACT), and recorded spirometry. Adherence with ICS/LABA was measured electronically over 8 weeks. Predictors of adherence were identified by univariate and multivariate analyses. Results:  99/100 patients completed the study (57 female; forced expiratory volume in 1 s mean ± standard deviation 83 ± 23% predicted; ACT 19.9 ± 3.8). Mean electronically recorded adherence (n= 85) was 75% ± 25, and mean self‐reported adherence was 85% ± 26%. Factor analysis of questionnaire items significantly associated with poor adherence identified seven themes: perceived necessity, safety concerns, acceptance of asthma chronicity/medication effectiveness, advice from friends/family, motivation/routine, ease of use and satisfaction with asthma management. Morisky score was moderately associated with actual adherence (r=−0.45, P < 0.0001). In regression analysis, 10 items independently predicted adherence (adjusted R2= 0.67; P < 0.001). Opinions of friends/family about the patient's medication use were strongly associated with poor adherence. Global concerns about ICS/LABA therapy were more predictive of poor adherence than were specific side‐effects; the one‐third of patients who reported experiencing side‐effects from their steroid inhaler had lower adherence than others (mean 62% vs 81%; P= 0.015). Conclusions:  This study identified several specific beliefs and behaviours which clinicians could use for initiating patient‐centred conversations about medication adherence in asthma.