ESPE Abstracts

Background: Monitoring adherence to growth hormone (GH) treatment is important, because poor adherence can lead to suboptimal clinical outcomes. The easypod™ electromechanical injection device in combination with the web-based easypod™ connect platform electronically records and transmits accurate, objective records of the date, time and dose injected of patients receiving GH with growth disorders, allowing physicians to accurately monitor patients’ behavior.

Aim: To study patterns of suboptimal adherence to GH treatment and the effect of age, sex and regimen on treatment adherence in children living in Italy.

Patients and Methods: Data on adherence (mg injected/mg prescribed x 100) from treatment start to 24 months and background characteristics uploaded onto easypod™ connect from January 2007 to June 2019 were included. Inclusion criteria were age 4–18 y and living in Italy. Suboptimal adherence was defined as <85% adherence (>1 missed injection/week). Logistic regression models (suboptimally-adherent vs adherent) were used to study the effect of age at treatment start, sex and regimen on adherence.

Results: Data were available for 677 children for a 0–6-month interval (45% girls, 55% boys; mean age at treatment start 10.7 y; 26% GH Deficiency, 3% Small for Gestational Age, 2% Turner Syndrome, 69% not registered; 56% on a 6-day regimen and 44% on a 7-day regimen), 541 children for 0–12 months, 401 children for 0–18 months and 278 children for 0–24 months. The longer the observation period, the fewer the data available. The proportions of suboptimal adherence were 13%, 19%, 22% and 23% between 0–6, 6–12, 12–18 and 18–24 months, respectively. In total, 34% of all children who were 24 months on treatment were suboptimally adherent in at least one 6-month time interval (0–6, 6–12, 12–18 and 18–24 months): 6% at all four intervals, 6% at three intervals, 8% at two intervals and 14% at one interval. Age at treatment start, sex and regimen had no significant effect on suboptimal adherence.

Conclusions: The easypod™ connect platform with automatic adherence recording and data transmission allows proactive monitoring of adherence. Understanding adherence patterns can be helpful to target children at risk for suboptimal adherence. Further research is needed to apply more advanced statistics, such as machine learning, to predict suboptimal adherence continuously and in real-time. This prediction, together with individual reasons for suboptimal adherence, may contribute to determining when a child needs support and which personalized intervention strategy can be applied to improve adherence.