دورية أكاديمية

Clinical trials are becoming more complex: a machine learning analysis of data from over 16,000 trials

التفاصيل البيبلوغرافية
العنوان: Clinical trials are becoming more complex: a machine learning analysis of data from over 16,000 trials
المؤلفون: Nigel Markey, Ben Howitt, Ilyass El-Mansouri, Carel Schwartzenberg, Olga Kotova, Christoph Meier
المصدر: Scientific Reports, Vol 14, Iss 1, Pp 1-6 (2024)
بيانات النشر: Nature Portfolio, 2024.
سنة النشر: 2024
المجموعة: LCC:Medicine
LCC:Science
مصطلحات موضوعية: Medicine, Science
الوصف: Abstract The past decade has seen substantial innovation in clinical trials, including new trial formats, endpoints, and others. Also there have been regulatory changes, increasing competitive pressures and other external factors which impact clinical trials. In parallel, trial timelines have increased and success rates remain stubbornly low. This has led many observers to question whether clinical trials have become overly complex and if this complexity is always needed. Here we present a large-scale analysis of protocols and other data from over 16,000 trials. Using a machine learning algorithm, we automatically assessed key features of these trials, such as number of endpoints, number of inclusion–exclusion criteria and others. Using a regression analysis we combined these features into a new metric, the Trial Complexity Score, which correlates with overall clinical trial duration. Looking at this score across different clinical phases and therapeutic areas we see substantial increases over time, suggesting that clinical trials are indeed becoming more complex. We discuss drivers of increasing trial complexity, necessary or helpful (‘good’) complexity versus unnecessary (‘bad’) complexity, and we explore mechanisms of how sponsors of clinical trials can reduce trial complexity where appropriate.
نوع الوثيقة: article
وصف الملف: electronic resource
اللغة: English
تدمد: 2045-2322
العلاقة: https://doaj.org/toc/2045-2322Test
DOI: 10.1038/s41598-024-53211-z
الوصول الحر: https://doaj.org/article/b18bc256f3d6493485e5add98342bcf3Test
رقم الانضمام: edsdoj.b18bc256f3d6493485e5add98342bcf3
قاعدة البيانات: Directory of Open Access Journals
الوصف
تدمد:20452322
DOI:10.1038/s41598-024-53211-z