Patients, physicians overestimate asthma treatment adherence compared with mobile app

Key takeaways:

• Objective measures of treatment adherence tend to lag subjective measurements.
• Physicians overestimated adherence in 45% of cases compared with the mobile app.
• Patients overestimated adherence in 47% of cases.

Adherence with asthma medication was lower when recorded by a mobile app compared with patient-reported and physician-reported results, according to a study published in Clinical and Translational Allergy.

Although objective measurements of adherence often are lower than subjective measurements, the use of the app itself may have influenced these results, Afonso Cachim, MD, medical resident, faculty of medicine, University of Porto, and colleagues wrote.

The secondary analysis comprised three prospective multicenter observational studies involving 61 primary and secondary care facilities in Portugal and Spain between March 2018 and October 2020.

The cohort included 193 patients aged 13 years and older (median age, 28 years; 72% female) with a diagnosis of persistent asthma and an active prescription for a daily inhaled controller medication, as well as access to a mobile device with internet capability.

These prescriptions included inhaled corticosteroids (ICS) and long-acting beta agonists (LABA) in combination therapy (n = 160; 83%), with 116 (60%) only using one inhaler.

Also, 100 (56%) had asthma classified as not well controlled, and 79 (41%) had at least one exacerbation during the prior year.

During the baseline in-person visit, patients and physicians independently assessed global adherence to asthma control inhaler use during the previous week with a visual analog scale.

Patients also received access to the InspirerMundi app, which was designed to improve adherence with therapy by gamification and social support. Once the patients and physicians agreed upon a therapeutic plan, it was registered in the app.

The app would notify patients when it was time to take their medication. It also would record when patients took their medication by processing images of their inhalers taken through the smartphone camera. The app calculated adherence by dividing the number of inhalations each day by the number of planned inhalations scheduled for each day.

The researchers assessed app-reported adherence at 1 week and 1 month after baseline. Percentages of missing values in the patient-reported and physician-reported results varied between 0% and 5%, whereas 12 patients (6%) had incomplete records.

Median adherence rates included 31% at 1 week and 18% at 1 month based on the app, compared with 82% for patient assessments (P < .001) and 80% (P < .001) for physician assessments at baseline, which the researchers called statistically significant differences.

The differences between the distribution of patient and physician data were not significant, the researchers said, nor were the correlations between adherence rates measured by the app and via the patient and physician assessments (rho = 0.118-0.156).

Using data from the app, 22% of patients were adherent at 1 week, and 13% were adherent at 1 month. Adherence rates also included 60% in the patient assessments and 62% in the physician assessments.

Compared with the app, physicians overestimated adherence in 45% of cases at 1 week and in 55% of cases at 1 month, in addition to underestimating adherence in 6% of cases at 1 week and in 4% of cases at 1 month.

Similarly compared with the app, patients overestimated adherence in 47% of cases at 1 week and in 54% of cases at 1 month, in addition to underestimating adherence in 5% of cases at 1 week and in 1% of cases at 1 month.

Although the correlation between patient and physician results was low to moderate (rho = 0.596; P < .001), the researchers said, there is still a need to integrate objective measurements of adherence into treatment.

Considering the decrease in adherence from 1 week to 1 month, the researchers suggested that patients with greater interest in gamification and social interaction may have higher rates of adherence. Lower rates of adherence, the researchers continued, may be the result of patients who do not interact with the app when they take their medication.

The researchers said they hope to improve the app’s real-time inhaler dosage counter value identification module for more objective verification of inhaler adherence as well as its gamification and social interaction to further promote medication adherence.

However, the researchers continued, such improvements will not replace additional interventions and follow-ups and later time points to improve adherence and patient monitoring.

Still, the researchers said, these findings may promote a better understanding of how different methods for measuring adherence may relate to each other and inform future studies of medication adherence in asthma control.