Medical Management Series
The medical monitor’s role in reviewing the case narratives and TLFs
The date of last data entry is approximately a couple of weeks after last patient last visit (LPLV) and the process of case narrative writing commences parallel to the data cleaning activities, while the finalization of case narratives is planned so that they can be included in the Clinical Study Report (CSR). The narrative categories required for the study are first established and then all the cases that fall into the selected criteria are extracted in the narrative listings. The narrative listings, along with clinical and safety databases, serve as the reference source for the writers to draft the initial narratives.
The medical monitor reviews the narratives and provides comments or queries based on the review outcome. The queries are handled by the data manager and operations in a similar pathway as the medical queries are issued following data review. When all the narrative queries have been answered by sites and considered closed any new information is reflected in the narratives, thereby generating revised narratives, which can then be finalized after a second medical review.
Our recommended approach for reviewing narratives is to establish a three-step process of verification, medical review, and quality control (QC). The verification step is to ensure the completeness and correctness by cross-checking narrative content with the source of information in the clinical and/ or safety database. Medical review is performed to confirm the plausibility and medical logic within the narrative content. Finally, a QC by a second medical reviewer checks that the narrative content is ready for finalization.
In a Phase III study where the volume of case narratives can be high, the workflow needs to be planned in advance so that timelines for different sets of narratives are organized to coincide with the data cleaning timelines and the finalized narratives are delivered before CSR drafting. During these reviews, the scientific integrity of the study data should be the underlying objective to govern the actions to be taken for any observation or discrepancy identified by the medical monitor.
Tables, listings and figures (TLFs)
Once the data cleaning measures have been completed, database lock (DBL) takes place and data is transferred to the statistics team in order to generate TLFs as detailed in the Statistical Analysis Plan (SAP). More often than not, the SAP provides drafts for the TLF shells, which are essential for presenting the study findings in a clear and concise manner, and in a way that allows for the best interpretation of study results. Where available therefore, pre-planned TLF shells are to be employed for dry runs by the statistical team. Together with the statisticians, a review of the mock output by the medical monitor is then conducted to guarantee its conformity to study objectives and endpoints, and, where required, adapt or create additional TLFs, leading to subsequent approval of the TLF shells.
Following the production of the actual data output and its validation by the responsible statistician, the produced TLFs will once again be closely reviewed by the medical monitor. A Statistical Output Comments Document may be used by the team members to list the review findings by the medical monitor, and the relevant actions taken by the responsible statistician. A Data Review Meeting (DRM) can be scheduled after the review of TLFs to discuss the findings. Any discrepancies that have been identified will be addressed at this time and the most appropriate course of action will be agreed upon.
For minor errors or inconsistencies in the generated output, footnotes might be used in the presentation of the TLFs and explanatory notes added to the CSR when required. It is important to bear in mind, however, that these footnotes should provide adequate clarifications concerning the issue at hand. Additionally, wherever required, a Note to File (NTF) can be documented and referenced to clarify any unclear situations such as discrepancies and missing data.
Major inconsistencies, on the other hand, present a considerably bigger problem and relevant standard operating procedures (SOPs) or guidelines should be followed while handling these cases. Major issues will, inevitably, require unlocking of the database for necessary corrections to be made. This is a very crucial step, and needs to be adequately implemented, following laid-down specifications and guidelines in order to maintain data integrity. For the specific issue in question, and when required, the site is responsible for providing needed clarifications or making any necessary corrections. These changes are all tracked and reviewed again by the medical monitor to guarantee consistency. Subsequently, the database is locked again and the TLFs are re-run and analyzed to see if the desired consistency and structure of data sets have been achieved. The medical monitor works closely with the statistician, along with the core team, to finalize the TLFs and to allow drafting of the CSR.
When applicable, a top line results meeting can be scheduled to share available results with the core study, and program management teams, to gather ideas on how the data can be presented within the CSR most adequately and utilized in subsequent steps of the development program.
Mohamed El Malt MD, PhD, Vijayanand Rajendran, MD, Kelechi K. Olu MD, MSc (2016) Medical management in clinical trials: A roadmap to operational excellence (Part 4 – Close-out Phase). Journal for Clinical Studies 8(6): 22-25.
Medical Management in Clinical Trials:
A Roadmap to Operational Excellence
Europital is a medical CRO operating worldwide and has provided Medical Management for over 10 years. If you would like to receive a free copy of our book, Medical Management in Clinical Trials: A Roadmap to Operational Excellence, or you have a question regarding Medical Management in Clinical Trials please contact us. Your question will be answered by our Chief Medical Officer.