New Research Sheds Light on Intentional Non-adherence in Clinical Trials First-of-its-kind study leverages objective adherence data collected by AiCure to illustrate scope of intentional non-adherence
Study findings help inform best practices to lower non-adherence and reduce trial timelines and enrollment volume New York, N.Y. – While medication non-adherence in clinical trials has been widely recognized as a challenge in drug development, reported non-adherence rates have been largely anecdotal, isolated to individual studies, and varied widely in their estimations. A recently published study from Tufts’ Center for the Study of Drug Development (CSDD) and AiCure, an AI and advanced data analytics company focused on improving clinical trials, identified and quantified the scope of intentional non-adherence to clinical trial drugs. Published in Therapeutic Innovation & Regulatory Science (TIRS), the research is the first adherence study to obtain data from directly observing dosing patterns among participants via computer vision and AI. Leveraging insights captured by AiCure’s proprietary medication adherence platform, the researchers deciphered between intentional and unintentional non-adherence, as well as identified factors predictive of noncompliance. By presenting a baseline measurement of overall non-adherence in clinical trials, the research helps to inform best practices to lower, and ultimately prevent, deliberately non-adherent behaviors.
The study, entitled “Assessing the Scope and Predictors of Intentional Dose Non-Adherence in Clinical Trials,” consisted of nearly 260,000 anonymized dosing results captured through AiCure’s platform, which were drawn from 2,796 study volunteers participating in 23 different clinical trials of nine psychiatric, neurological and neuromuscular diseases. The study’s key findings include:
4% of all confirmed doses were intentionally non-adherent;
48% of all study volunteers had at least one intentionally non-adherence dose;
5% of study volunteers were intentionally non-adherent for more than one-third of all doses required;
14% of study volunteers had more than 10% of their total doses intentionally non-adherent over the course of their clinical trial;
Volunteers who deliberately chose not to take their first dose had an average intentional non-adherence rate five times higher than those who were adherent for their first dose.
“As the pharmaceutical industry faces intensifying pressure to optimize the drug development process, adherence and precise dosing over the course of a clinical trial are critical to achieving an accurate assessment of a treatment’s efficacy and effects as quickly and safely as possible,” said Ed Ikeguchi, M.D., CEO of AiCure. “Today trial sponsors need to recruit a greater number of patients to combat the effects of non-adherence on trial results. This study’s findings inform a path to potentially shorter, more streamlined trials, which could not only lead to cost savings but also help bring life-saving drugs to patients faster.”
Understanding Non-Adherence
The study identified several factors that were associated with – and predictive of – unreported intentional non-adherence. Volunteers in later stage phase II and III clinical trials, as well as those of longer durations, had a higher likelihood of intentional non-adherence. This finding may be a result of study fatigue, skepticism around the drug’s effects, and patient burden as participation increases. Other factors included geographical location of the study and enrollment volume at the site. Understanding the reasons behind deliberate non-adherence may help sponsors develop targeted educational resources or proactive, personalized interventions to mitigate the impact of non-adherence.
Addressing the Impact of Non-Adherence
The study’s findings have long-term implications for reducing trial timelines and enrollment volumes. Today, many drug developers typically increase the size of study populations by 15% to reduce the impact non-adherence has on a drug’s statistical relevancy, requiring longer and more expensive trials. For example, for a typical phase III oncology study, clinical research sponsors may pay an additional $5-$7 million to augment enrollment to account for potential non-adherence. Employing tactics to identify and eliminate intentionally non-adherent participants early – particularly within the trial’s first week – will play a critical role in cutting costs and lead to leaner, statically relevant trials in the future.
In addition to the publication in TIRS, Ken Getz, Deputy Director and Professor at the Tufts’ CSDD, and Marlen Rattiner, AiCure’s Vice President of Product Management, will present the study findings at DIA Global Annual Meeting on Monday, June 15th. Learn more about their presentation here.
About AiCure
AiCure is an AI and advanced data analytics company that monitors patient behavior and enables remote patient engagement in clinical trials. AiCure improves predictability of study timelines, reduces costs and accelerates timelines through remote patient engagement and assessments, including measuring digital biomarkers and real-time monitoring of patient dosing. Founded in 2010 and funded by the National Institutes of Health (NIH) and leading institutional investors, AiCure has more than 65 issued patents and works with global clients in over 30 countries. AiCure is globally recognized and is a recipient of the Scrip Award, AI 100 and Digital Health 150. For more information, please visit www.aicure.com.
Media Contact Siobhan Nguyen aicure@fleishman.com 617-986-5784
1 Gosse L, Tubach F, Dougados M, Ravaud P. Reporting of adherence to medication in recent randomized controlled trials of 6 chronic diseases: a systematic literature review. Am J Med Sci. 2007;334(4):248–54. 2 Brain C, Sameby B, Allerby K, et al. Twelve months of electronic monitoring in the Swedish COAST-study: a comparison of methods for the measurement of adherence in schizophrenia. Euro Neuropsychopharmacol. 2014;24:215–22. 3 Mathieu M. IQVIA Trends in Phase III Costs Per Patient and Enrollment Rates. Parexel Biopharmaceutical R&D Statistical Sourcebook 2019/2019; p. 223-234.