Post by Mathias Poensgen
First, it was mandatory to submit results to ClinicalTrials.gov. Then, from July 2014, it also became mandatory for companies to report their results to EudraCT. Combine this with the requirement of companies needing to submit to additional registries, we can see why trial disclosure has become increasingly difficult, despite being valuable to the industry.
The challenges of manually handling disclosure today boil down to these key points, all of which are costly, time-consuming, and lead to data quality issues:
Difficult to keep control of data
Duplication of data entry in multiple registries
Ensuring data consistency across these registries
For an efficient and high-quality process, a central area is needed that supports the complete disclosure process. This involves the planning of registrations, collection of the data from the different source systems, validation of data, uploading of data to the specific registries and tracking of the process. This approach is purposely designed to minimize the risk of publishing inconsistent data in different registries. This centralized, single platform should automate the process as much as possible, removing the risk of human error.
For example, the data can be imported electronically from the source system. Automatic validation rules like edit checks in EDC support the quality control process for high data quality. Integrated planning and tracking capabilities renders the usage of spreadsheets unnecessary and help keeping the process under control.
In order to benefit from this system, there are five key steps that the disclosure process should follow:
Plan – As studies need to be registered to different registries, you need to decide for each one which registries apply and to plan the initial registrations for each registry. For some registries, you will need to update the data (e.g., the enrollment status) on a regular time-frame (e.g., quarterly basis). Companies need to plan for these updates in order to ensure they are not missing any of them. For many studies, within one year after the end of the trial, results need to be published at ClinicalTrials.gov and EudraCT. As results are dealing with very sensitive data, careful planning is required.
Collect – In order to ensure data consistency and quality, as well as minimize manual efforts, companies should import as many data points from the source as possible. Some of this data will need to be transcribed (the source data will have most of the information in a scientific language). As registries are addressing the general public, this information also needs to be given in lay language. Another example is translations; the source system will most likely have the information in English. For some registries, certain information may need to be translated to a local language.
Validate – Now that the data has been collected and consolidated, it can be used for registrations at multiple registries. To this end, registry specific datasets need to be generated automatically, and therefore should be validated against registry specific validation rules. Registries like ClinicalTrials.gov or EudraCT have hundreds of edit checks in place, so these should be checked against these edit checks first before submitting the data to ensure clean data is submitted. In addition, you may want to implement company specific edit checks. The data to be submitted should be reviewed and approved beforehand.
Upload – Submitting to the registries is straight forward, especially if the registry supports XML files. However, some registries do not and require manual data entry. In these cases, a report should show the data in the same order as required by the registries for easy copy and paste.
Track – Tracking serves to keep the process under control, allowing managers to easily get answers to questions such as when was a specific study registered, what is the status of each registration, what is coming up, are there any overdue registrations, what feedback did we get from the registries and did we address these comments, etc.
By following these five key steps, the problems of manually handling disclosure are resolved through benefits that save time and reduce cost:
Ensures compliance with global clinical trial registries
Maintains consistent data across multiple registries
Allows reuse of data across multiple registries
Our cloud-based trial disclosure solution, agDisclosure™, is specifically designed to automate the submission of clinical trial results data to key registries by following the five key steps mentioned above. You can find out more about this solution by visiting our website. We do have customers that have used the solution to help them download and transform data from CTGOV and push it onto EudraCT.