Justin Doods1, Caroline Lafitte2, Nadine Ulliac-Sagnes2, Johan Proeve3, Florence Botteri4, Robert Walls5, Andy Sykes6, Martin Dugas1, Fleur Fritz1. 1. Institute of Medical Informatics, University Münster, Albert-Schweitzer-Campus 1/A11, D-48149 Münster, Germany. 2. Sanofi R&D - Feasibility Management Department - 1 rue Pierre Brossolette - 91385 CHILLY MAZARIN, France. 3. Global Strategy and Development Advisor, Global Data Sciences and Analytics, Bayer Vital GmbH, BV-PH-MED-GDSA, 51368 Leverkusen, K 9, 413, Germany. 4. Novartis Pharma AG, CH-4002 Basel, Switzerland. 5. Real World Data Science, F.Hoffmann-La Roche Ltd, CH-4070 Basel, Switzerland. 6. Projects, Clinical Platforms & Sciences - Metabolic Pathways and Cardiovascular, GlaxoSmithKline, 1-3 Iron Bridge Road, Stockley ParkUxbridge, UK UB11 1BT.
Abstract
INTRODUCTION: In the last few years much work has been conducted in creating systems that support clinical trials for example by utilizing electronic health record data. One of these endeavours is the Electronic Health Record for Clinical Research project (EHR4CR). An unanswered question that the project aims to answer is which data elements are most commonly required for patient recruitment. METHODS: Free text eligibility criteria from 40 studies were analysed, simplified and elements were extracted. These elements where then added to an existing inventory of data elements for protocol feasibility. RESULTS: We simplified and extracted data elements from 40 trials, which resulted in 1170 elements. From these we created an inventory of 150 unique data elements relevant for patient identification and recruitment with definitions and referenced codes to standard terminologies. DISCUSSION: Our list was created with expertise from pharmaceutical companies. Comparisons with related work shows that identified concepts are similar. An evaluation of the availability of these elements in electronic health records is still ongoing. Hospitals that want to engage in re-use of electronic health record data for research purposes, for example by joining networks like EHR4CR, can now prioritize their effort based on this list.
INTRODUCTION: In the last few years much work has been conducted in creating systems that support clinical trials for example by utilizing electronic health record data. One of these endeavours is the Electronic Health Record for Clinical Research project (EHR4CR). An unanswered question that the project aims to answer is which data elements are most commonly required for patient recruitment. METHODS: Free text eligibility criteria from 40 studies were analysed, simplified and elements were extracted. These elements where then added to an existing inventory of data elements for protocol feasibility. RESULTS: We simplified and extracted data elements from 40 trials, which resulted in 1170 elements. From these we created an inventory of 150 unique data elements relevant for patient identification and recruitment with definitions and referenced codes to standard terminologies. DISCUSSION: Our list was created with expertise from pharmaceutical companies. Comparisons with related work shows that identified concepts are similar. An evaluation of the availability of these elements in electronic health records is still ongoing. Hospitals that want to engage in re-use of electronic health record data for research purposes, for example by joining networks like EHR4CR, can now prioritize their effort based on this list.
Authors: Philipp Bruland; Mark McGilchrist; Eric Zapletal; Dionisio Acosta; Johann Proeve; Scott Askin; Thomas Ganslandt; Justin Doods; Martin Dugas Journal: BMC Med Res Methodol Date: 2016-11-22 Impact factor: 4.615
Authors: Hans-Ulrich Prokosch; Till Acker; Johannes Bernarding; Harald Binder; Martin Boeker; Melanie Boerries; Philipp Daumke; Thomas Ganslandt; Jürgen Hesser; Gunther Höning; Michael Neumaier; Kurt Marquardt; Harald Renz; Hermann-Josef Rothkötter; Carmen Schade-Brittinger; Paul Schmücker; Jürgen Schüttler; Martin Sedlmayr; Hubert Serve; Keywan Sohrabi; Holger Storf Journal: Methods Inf Med Date: 2018-07-17 Impact factor: 2.176