Data changes in a clinical research study using Interactive Response Technology (IRT) are a certainty. It’s important to be thoughtful and consider the consequences when requesting or making changes in a live system.
Some IRT systems have built in data change functionality that may be used to update simple data points such as Gender and Date of Birth. Even with simple data point data changes, it can be risky if these items are used as stratifying factors for randomization. It should be pointed out that stratification factors should never be updated post randomization in an Interactive response technologies system and are best documented in Electronic Case Report Forms and or notes to file.
During study startup, it’s best to lay out the rules for making IRT data changes using clinical trial software and clinical trial technology. Some considerations are if the change should be allowed to be made at all. Your team may decide that some changes are simply not worth the risk, time, and effort of a manual update when a simple note to file would be acceptable. In other cases, changes may be needed to ensure that the subject is properly dosed for future visits or is allowed to continue on in the study. These types of changes should be carefully documented and discussed before the need arises to make one. It’s always advantageous to have the treatment waiting for the subject rather than the subject waiting for the treatment.
Approval and notification details should also be worked out for the types of changes that may arise during a study. A simple data point like gender or DOB may not be something that the sponsor team wants or needs to approve; they may opt to merely be notified that a change occurred. Some IRT data changes may affect dosing, iteration or eligibility, for example, are higher risk items that require a level of approval before any updates are made. In regards to high risk items, it’s always beneficial to understand the timeframe required to make a change. Some changes will need to accommodate subjects waiting at sites, while others may have more forgiving timeframes depending on the study design.
The end result is that being thoughtful about IRT data changes up front is always best practice. Study teams can work out what is high priority vs low priority, what requires approval vs what requires notification and in some cases IRT data won’t need to be changed at all. Having this information worked out beforehand ensures a trial in maintenance runs much smoother with fewer delays and obstacles.
