Sharing data from clinical trials: where we are and what lies aheadBMJ 2013; 347 doi: https://doi.org/10.1136/bmj.f4794 (Published 30 July 2013) Cite this as: BMJ 2013;347:f4794
All rapid responses
Currently, 99% of medical documentation forms are not accessible to the public, and this needs to be changed urgently.
From an informatics perspective, the next step towards sharing data from clinical trials is FREE ACCESS TO MEDICAL FORMS.
To clarify, this is about empty documentation forms, i.e. metadata, consisting of sets of data elements (also known as item catalogs), so there is no data protection risk at all. Regarding clinical trials, eligibility criteria for ~148.000 studies are available on the Internet . However, eligibility forms are only a very minor part of – on average - more than 100 case report forms (CRFs) per trial . This holds true both for commercial and investigator initiated trials. In epidemiological studies and registries, the overall number of forms per study is usually smaller (because there is no need for extensive documentation regarding potential adverse events of investigational drugs), but most CRFs are only available to project partners.
The situation is similar regarding medical forms in commercial clinical information systems.
Vendors of electronic health record (EHR) systems typically grant their customers access to sets of medical forms, but publication of these forms is not allowed. Only some sets of mandatory forms are available to the public, for example regarding quality management or tumor registration. In Europe there are more than 800 EHR vendors , and for each clinical information system in the magnitude of 100 forms are provided, covering medical history, clinical reports and various findings. Typically, these forms are customized individually for each hospital, resulting in an overarching, incompatible variety of documentation approaches.
FREE ACCESS TO MEDICAL FORMS COULD CONTRIBUTE SUBSTANTIALLY TO IMPROVED CLINICAL RESEARCH: Open metadata enables a public discussion about best practice in documentation: what data elements are really needed and how should these be collected? How can regulatory requirements for clinical trial documentation be implemented most efficiently? It is possible to learn from best practice in previous studies and successful clinical systems. This can help to design better CRFs in less time – a very important goal given the huge and continuously increasing cost of data management in clinical trials . That discussion also fosters more compatible data structures, which are urgently needed for information exchange in clinical as well as research settings. Finally, meta-analyses are enabled, because it is much easier to identify suitable data sources.
Form design requires medical and scientific expertise and is laborious, therefore CRF authors should receive academic credit. Study protocols are increasingly being published (for example, ). In the future, CRFs of these studies should also be published to enable citations and to make them available to the scientific community. There are already public repositories available for medical forms .
Free access to medical forms has also a legal and ethical dimension. For real INFORMED CONSENT, the patient needs to know what data is being collected about her or him. It makes a difference whether only a few data elements or 10 years of medical records are collected.
TRANSPARENCY and “my data, my decision” are basic principles of data protection legislation. From my perspective, the citizen has a right to know about the details of medical documentation, therefore free access to forms should become a general principle in medicine.
1. http://www.clinicaltrials.gov [accessed August 2, 2013]
2. Getz K (2008) Protocol Design Trends and their Effect on Clinical Trial Performance. RAJ Pharma 5:315-316
3. EuroRec. http://www.eurorec.org [accessed August 2, 2013]
4. Hearn J, Sullivan R (2007) The impact of the 'Clinical Trials' directive on the cost and conduct of non-commercial cancer trials in the UK. Eur J Cancer;43(1):8-13
5. BMC Trials. http://www.trialsjournal.com [accessed August 2, 2013]
6. Breil B, Kenneweg J, Fritz F, Bruland P, Doods J, et al. (2012) Multilingual Medical Data Models in ODM Format. A Novel Form-based Approach to Semantic Interoperability between Routine Healthcare and Clinical Research. ACI 3(3):276-289
Competing interests: M. Dugas receives funding from EU 7th Framework Program, Innovative Medicines Initiative (project EHR4CR, grant agreement n° 115189).