Medication adherence is a well-known challenge in cystic fibrosis (CF) self-management and an area for ongoing quality improvement within CF treatment centers. Across treatment centers, it is an aspect of CF care that would benefit from sharing of successful ideas, techniques, and processes.
A new study published in the Journal of Cystic Fibrosis, however, cautions that not all treatment centers calculate adherence the same way and that some formulas may in fact overestimate overall medication adherence of patients at individual centers. Even objective data obtained using digital time and date stamping of treatment can be flawed, the authors note.
“Typically, the CF community has used medication possession ratio (MPR) to estimate how much inhaled therapy is taken, which often suggests around 60% is taken,” said lead author Zhe Hui Hoo, MBChB, MRCP, MSc, University of Sheffield, United Kingdom, in an interview with Patient Care Online. “However, time and date stamped data show this to be an over-estimation,” said Hoo. “In addition, when time and date stamped data is used, unless everyone [ie, all clinic patients] is included, we can be falsely reassured about how well we are doing as clinical teams in supporting self-care.”
In addition, objectively-collected data may not show the entire picture of adherence, and could make comparisons between centers invalid, Hoo and colleagues wrote in their study. Centers should consider “both treatment appropriateness and missing data” in context with objective adherence to gain a better understanding of true adherence in this patient population.
“For this particular study, we want to test whether different methods of selecting and processing data on how much treatment people with CF take can influence the average level of adherence in a CF center,” said Hoo.
Hoo and colleagues examined data from 131 patients with CF who used the I-neb® (Respironics; Chichester, UK) medication delivery device between 2013 and 2016 at a single center in the UK.
Participants were aged 25 to 29 years, 37%-43% were female, and body mass indices ranged between 21.5 kg/m2 and 23.2 kg/m2.
Independent investigators collected data on CF patient age, gender, body mass index, % FEV1 as well as genotype, pancreatic status, CF-related diabetes, and Pseudomonas aeruginosa status during the study period.
Unadjusted vs normative adherence
Adherence was calculated based on agreed regimen (unadjusted adherence) or minimum required regimen (normative adherence). Data from I-nebs® devices not brought to clinic were downloaded during home visits. Adults not on any inhaled therapy but with chronic P. aeruginosa infection were included and their adherence was counted as “0.”
The researchers calculated unadjusted adherence rates by dividing the agreed-upon dose of medication against the number of total nebulizers as a percentage. To calculate normative adherence, researchers standardized the denominator to define the minimum required treatment regimen in consideration of P. aeruginosa status and adjusted the numerator to account for situations where patients may have had nebulizer use over 100% due to dose spacing, and when medication is taken past midnight.
They gave an example of a patient with chronic P. aeruginosa infection taking a nebulized mucolytic once daily and an antibiotic twice daily, despite agreeing to a once-daily regimen of dornase alfa, which would show a normative adherence rate of 33% for the medication.
For each adherence calculation method, researchers used 3 sampling frames to determine center-level adherence:
- Frame 1: Adults who handed in their nebulizers for analysis.
- Frame 2: Adults who used their nebulizer but where data collection was difficult.
- Frame 3: Adults with chronic P. aeruginosa, but not using CF-inhaled therapies throughout the year by assigning 0 as their adherence.
Missing data makes a difference
In general, as it became more difficult to collect data, adherence in a sampling frame either stayed the same or decreased compared with patients who readily offered their nebulizers for collection, and this association was also seen when P. aeruginosa was involved.
Overall, adherence data was available for 126 patients. The median paired difference between unadjusted adherence and normative adherence was:
- –2.6% in 2013 (95% confidence interval (CI), –1.3% to –4.1%)
- –3.0% in 2014 (95% CI, –1.7% to –5.3%)
- –5.1% in 2015 (95% CI, –3.3% to –7.0%)
- –3.9% in 2016 (95% CI, –2.1% to –5.7%)
In the sampling frames. unadjusted adherence ranged from 35.1% (sample framing 3) to 41.8% in 2013 (sampling frame 1), 44.7%-52.6% in 2014, 50.8%-57.7% in 2015, and 52.8%-59.1% in 2016.
Normative adherence rates ranged from 31.0% (sampling frame 3) to 40.0% (sampling frame 1) in 2013, 39.7%-45.2% in 2014, 44.2%-49.6% in 2015, and 50.8%-53.4% in 2016.
Overall, researchers noted that while unadjusted adherence was close to levels of normative adherence in many cases, they overestimated adherence rate. For instance, some adults with higher adherence rates showed a 40% to 100% difference in adherence, with adults who were labeled as overusing their nebulizer having a greater than 100% adherence rate.
Adherence levels based only on agreed regimen among adults who readily brought their nebulizers to clinics can over-estimate the effective adherence rate of a specific CF center.
“Our next steps will aim to understand how habit formation can be made more accessible for all people with CF looked after in a center so that nobody is left behind,” said Hoo. “This requires more research.”
Source: Hoo ZH, Curley R, Walters SJ, et al. Exploring the implications of different approaches to estimate centre-level adherence using objective adherence data in an adult cystic fibrosis centre – a retrospective observational study. J Cyst Fibros. 2019;doi:10.1016/j.jcf.2019.10.008. Accessed 26 Nov 2019.
Stay in touch with Patient Care® Online: