Extensive research conducted in Australia, Canada, the United Kingdom, the United States and elsewhere has found that nonadherence with asthma therapy is widespread, and is a significant risk factor for asthma morbidity and mortality. Because of the limited sensitivity and specificity of self-reported measures of adherence (5), some of the most convincing studies have used objective measures, such as pharmacy databases, medication measurement and electronic medication monitors to assess adherence behaviour.
Conservative estimates indicate that almost half of the prescription medications dispensed yearly are not taken as prescribed (6). The real-life response to a clinician's prescription of preventive therapy will include a range of undesirable patient behaviours, including a failure to fill the initial prescription, erratic use or under-use of therapy, and premature discontinuation of therapy. Studies indicate that primary nonadherence (not filling initial prescriptions) ranges from 6 - 44% (7 - 12).
Even when patients fill prescriptions for asthma medications, studies of secondary nonadherence (rates of medication use) suggest that long-term rates of adherence to preventive therapies (e.g. controller or preventer medications) among adult patients are often poor. Spector et al. (13), one of the first investigative teams to use an electronic medication monitor to examine adherence to MDI-delivered medications, followed 19 adult asthmatic patients using an anti-inflammatory drug for 12 weeks. Patients adhered to the four-times-daily regimen for a mean of 47% of the days, with a range of 4.3% to 95%. Patients were also asked to maintain asthma diaries as part of this study, and a comparative analysis of electronic data and diary data found that subjects over-reported their appropriate use of medication in their diaries more than 50% of the time. In a similar study, Mawhinney et al. (14) studied adherence in adult asthmatic patients over a 3 - 4 week period. Adherence to the medication as prescribed was observed, on average, for 37% of the days, and under-use on more than 38% of the days monitored. Yeung et al. (15) used an electronic monitor to follow patients' use of inhaled corticosteroids over a period of 2 - 3 weeks. When patients were aware that they were being monitored, 60% of them were fully adherent, 20% were partially adherent (taking just 70% of the prescribed dose) and 20% were totally nonadherent. However, when patients were unaware of the monitoring, 6 out of 11 took between 30% and 51% of the prescribed doses.
Several studies have suggested that patients from low-income, ethnic-minority groups (primarily African American) in developed countries may have lower rates of adherence to asthma therapy. Celano et al. (16)examined adherence to anti-inflammatory medication delivered by MDI in low-income, urban, primarily African American children with asthma. Adherence to treatment administered by MDI was estimated by weighing canisters and calculating the ratio of the number of puffs used over the study period to the number of puffs prescribed. Estimated MDI adherence in this study was 44% for all participants and only 12% of the children had rates above 75%. In a group of 80 asthma patients, treated under the Medicaid scheme, who were repeat users of the emergency department or overnight hospitalization, only 46% had been prescribed ICSs and only 43% had a written action plan (17). Less than half of children with asthma living in Tennessee, receiving treatment funded by Medicaid, had a prescription for oral corticosteroids filled following an emergency department visit or a period of hospitalization for asthma (18).
Low rates have also been reported from studies that used different measurement systems. Coutts, Gibson and Paton (19) in the United Kingdom published the first study to examine children's adherence to anti-inflammatory therapy using an electronic medication monitor that recorded and stored the date and time of each use. Children (aged 9 - 16 years) were monitored for 2 - 6 months and asked to maintain asthma diaries as well as to use the monitored inhaler. Despite symptomatic asthma, underuse of the inhaled corticosteroids was observed on 55% of the study days. In a second study from the United Kingdom, Gibson et al. (20) used electronic monitoring to evaluate the adherence of preschool children to inhaled prophylactic medication. Median adherence was 100% on 50% of study days, and an overall median of 77% of the prescribed doses were taken during the average 2-month monitoring period. It is important to realize that the poor adherence observed occurred in the children of a group of parents who had a clear understanding that adherence was being monitored, and who had been provided with careful explanations of the importance of adherence to prophylactic medications. The authors noted that this poor adherence might reflect persistent misunderstandings or concerns about the side-effects of the medications.
Jonasson et al. (21) reported from Sweden on adherence to inhaled budesonide administered with a breath-driven asthma inhaler in 163 children (aged 7 - 16 years) with mild asthma who were participating in a randomized, double-blind clinical trial. Mean daily diary-card adherence was 93% over the 12- week study, whereas inhaler dose-counting recorded only 77% adherence. Milgrom et al. (22), in the United States used electronic monitors to study the adherence of school-aged children to inhaled corticosteroids. The participants were unaware of the function of the electronic device. Diary-card data showed that patients reported taking all doses on a median of 54% of study days and at least one dose on 97% of study days. However, electronic records of inhaled corticosteroid use showed a median of only 5% of study days on which all inhaled corticosteroid doses were taken and a median of 58% of days on which at least one dose was taken. The participants skipped all inhaled corticosteroid doses on a median of 42% of days and almost half of them missed their inhaled corticosteroids completely for more than a week at a time.