Clinical Trial Data Management in Different Stages Of CRO Clinical Trials

0

CRO clinical trials are generally divided into three phases, the Start-Up phase, Conduct phase and Closeout phase. Clinical Trial Data Management is different in each stage and in this article, we break down how data is managed in different stages of CRO clinical trials.

 Clinical Trial Data Management in the Startup Stage

Once the protocol from a sponsor, the CRO begins the process of designing a Case Report Form (CRF).

A CRF helps in collecting data patient data from the investigational sites. The main purpose of this data reporting document used in a clinical study is to collect the study data in a specific format that allows for efficient data analysis.

All data collected using the CRF should be according to what has been specified in the protocol and anything which is not mentioned in the protocol should be excluded in the CRF.

While designing the CRF, guidelines need to be set for collecting the data with all users in mind because there will be multiple persons accessing the CRFs. Collecting CRO clinical trials data required by the regulatory agencies for further analysis should be clear, concise and free from duplication.

It should also capture all information required by the protocol and should meet the needs of different clients and end-users. The CRF needs to be finalised and available before an investigator starts recruiting patients into the CRO clinical trials study because as soon as they enrol the patients into the study they will begin capturing the patient data.

Once the draft is ready it needs to be sent to the sponsor for approval and any modifications suggested by the sponsor need to be carried out.

 If there are no findings and no comments received from the sponsor then the CRF can be finalised and the next step is to design the database.

Designing the database for CRO clinical trials requires using clinical data management systems such as Oracle clinical data management system. Programmers design the database and create the exact replica of CRF pages in the database. This enables the data entry person to accurately enter the data recorded on the CRF. The pages that are entered into the database should be according to the data entry guidelines.

Once the database is designed, a data manager needs to test it before it can be activated.

In case any issues are found, the database designer needs to fix these and once everything has been resolved the database can be activated.

The next step is the process of validation and derivation using automated edit checks. These are required to flag errors or discrepancies in the data present on the CRF or in the database, so they can be corrected

The programmer writes the validation processes and the data reviewer enters clean and unclean data to test whether these checks are firing or not in a process commonly referred to as user acceptance testing.

User Acceptance Testing (UAT) has to be performed for all the checks and this is the final step before releasing the application or database to sites for capturing the clinical trial. It is the last step in the setup and the objective is to get zero errors so there’s no unclean data. The database then gets activated and is ready to accept the production data.

CRF designers and Database Designers are responsible for the programs, validation and derivation processes.

Data managers check data and activate the database. They are required at every step in the setup process while designing the CRF.

Only after data managers have reviewed the CRF, tested and validated the database, can it be activated.

Data managers also perform the UAT before activation of the database to make sure all the edit checks are working fine.

Clinical Trial Data Management in the Conduct Stage

Once the Database is activated, in the case of paper studies, CRFs will be sent to the data managers and the data entry person has to enter the data in the database.

In the case of Electronic Data Capture (EDC) studies, investigational sites enter subject data directly into the database.

Both, the CRF and external data such as third-party vendor data in a trial protocol is visible here. Once data is entered, discrepancy management is performed by the Data Manager.

Discrepancy management involves identifying and managing likely problems based on data collected during the study. For example, empty fields, unrelated items, incorrect ranges, lab ranges, dates not in a logical sequence and any missing visits or pages all come under discrepancy management.

 Apart from this, coders perform coding of the data such as the medical terms and drug names which are required to be entered in the database.

Simultaneously, safety data reconciliation is also performed. This involves a comparison of the clinical database and safety database because the majority of the data includes the safety terms and adverse event terms.

Performing discrepancy management, coding and safety data reconciliation identify several issues out of these activities so a query should be generated while performing these activities.

The queries would be generated in two ways through automated checks described in the setup process through validation and derivation procedures which flag any errors.

When data is entered, processes in the back end begin running and start to generate queries in the database. In the case of manual checking for Quality Control, the data manager has to conduct a manual review.

In case of query generation where automated checks in place, a query should be generated and sent to site for clarification to see if the site comes back with some resolution or if any modification is required.

 Clinical Trial Data Management in the Closeout Stage

The closeout stage is the last stage in the clinical data management system of a CRO clinical trial where the database is locked. When all clinical trial data has been reviewed, queries resolved and issues addressed then the database will be locked.

The database cannot be changed in any manner after locking and it’s important to ensure all pre-lock activities are checked off.

All queries should have been resolved by this stage and the database should also have been updated after performing quality control checks.

Approvals from the investigator in the form of a sign-off on the CRF needs to be obtained before locking of the database. They have to verify the CRF data or whatever the data entered in the database and compare it with the source document.

Once it has been confirmed that the entire data is clean and nothing is outstanding, an approval from the sponsor on the source document is required.

The final step is the investigator approval, where the investigator signs off on the CRFs and once locked, no modifications can take place.