medwireNews: Adherence to first-line diabetes medication, ascertained from prescription records, is the only routine clinical variable that is a strong indicator of people’s adherence to second-line treatment, say researchers.
“Many factors relating to non-adherence to diabetes medications are not captured in routine data,” observe Andrew Farmer (University of Oxford, UK) and co-researchers in Diabetes, Obesity and Metabolism.
“Nevertheless, it is possible that patterns of previous prescriptions issued could provide a practical indicator that clinicians could use to determine those where further assistance may be offered to improve adherence.”
The team found highly significant associations between other routine clinical predictors – including age, glycated hemoglobin, and diabetes duration – and adherence to first-line diabetes medication, with nonadherence defined as obtaining 80% or less of the medication required to meet treatment needs.
But they attribute the strength of the associations to the large study population, which consisted of 67,882 people with type 2 diabetes identified in the UK Clinical Practice Research Datalink. Despite the significant associations, the factors in combination had only 60.1% accuracy for distinguishing between the 81% of people who were adherent to their first-line medication and the 19% who were not.
Poor adherence to first-line medication was by far the strongest predictor of poor adherence to second-line medication; 32% of those who were non-adherent to the first medication were also nonadherent to the second, compared with just 9% of those who had adhered in the past.
Again, there were other clinical predictors associated with adherence to second-line medication, but these “added little” to the discriminatory ability of adherence to the first-line treatment, say the researchers.
Farmer and team note that all factors combined, including adherence to first-line treatment, only had 72.4% accuracy for adherence to second-line medication.
“This is likely to relate to the fact that many predictors of adherence are social or psychological and therefore not captured by routine data,” they say.
They add that depression and ethnicity had only small effects in their study, despite being cited by other researchers as important factors.
“It is possible with the limited data we had that we may have underestimated their effects in our population, and in some settings these factors may be important discriminators,” the team concludes.
medwireNews is an independent medical news service provided by Springer Healthcare. © 2019 Springer Healthcare part of the Springer Nature group