High risk of ignoring individual patterns of medication adherence
Tuesday, March 15, 2016
Hundreds of millions of dollars are spent ensuring operational quality in clinical trials including site compliance, accuracy of data entry and enrollment targets. Meanwhile, one of the most vulnerable aspects of the process—whether or not study participants are taking the investigation compound—receives surprisingly scant attention.
Medication nonadherence is well documented in real-world settings. Approximately half of all patients do not take their medication as prescribed, leading to hundreds of billions of dollars in unnecessary healthcare costs. Less publicized is the extent of drug nonadherence in research settings. Clinical trial protocols are built around the assumption that study participants will be 80% adherent or better. A trial’s statistical power and the number of participants needed is based on this assumption. Yet adherence in clinical trials ranges from 43% to 87%, a far cry from the expected 80%.
The result of this disconnect between what is expected and what actually happens can have devastating consequences. Poor adherence can under-power a trial and drown out an efficacy signal of what might have been an effective and successful drug, resulting in an incorrect no-go decision to proceed to the next phase. Even if average adherence is at 80% or above for all participants, this still remains an average. Two participants with 60% and 100% adherence respectively would average out to the required 80%, however, not knowing exact dosing patterns of each participant will preclude a true understanding of the drug’s efficacy and safety profile.
The recent confluence of inexpensive mobile technology equipped with sufficient processing power to handle complex artificial intelligence applications is allowing us to collect and analyze data that has never been measured before. The knowledge obtained through AI, such as real-time confirmation of drug ingestion, is going to open up new possibilities and accelerate the pace in drug discovery.
Written by Guest Writer Adam Hanina. Hanina is CEO of AiCure, which uses artificial intelligence to visually confirm medication ingestion on smartphones. The AI platform is entirely software-based and ensures that the right patient is taking the right medication at the right time. Mr. Hanina has acted as a subject-matter expert on medication adherence technologies for the NIH. He is an advocate for the use of technology as a means to improve the conduct of clinical trials. For more information, please visit:
www.aicure.com or tweet @AiCureMed
This article was reprinted from Volume 23, Issue 03, of The CenterWatch Monthly, an industry leading publication providing hard-hitting, authoritative business and financial coverage of the clinical research space. The Action Items section features short columns focusing on actionable or how-to advice from clinical trial professionals. To submit an Action Item, please contact firstname.lastname@example.org. Subscribe >>