What is Clinical Data Management?
Every clinical trial generates copious amounts of data. This includes patient reports, their medical history, the trial design, the dose regimen, the number of subjects, their age and gender, informed consent forms, laboratory notes, pharmacy dispensing records, transcription copies and so on. For this data to be able to be used for the analysis of the drug’s potency it needs to be organized. This is achieved through the use of databases which are designed for the particular set of trials. In simple words, a clinical database is a set of data stored in a rows and columns format. After entering the data into the database, it is checked for any discrepancies and corrected if any. The organized and validated data is then analysed using statistical analysis methods before presenting to the concerned authorities.This collection, integration and validation of the research data is called clinical data management.
Why is it required?
The approval of a drug by the licensing authorities depends on the quality of data generated in the clinical and pre-clinical studies carried out by the concerned pharmaceutical company. To obtain that, it is necessary to adhere to quality standards and practices. Hence, it is important that all staff involved in the trials are trained and qualified in data management tasks. Training the staff to follow quality standards will ensure minimal errors during data entry. That alone would save a lot of time and efforts spent for the clean-up process.
The Flow of the Process
The process of clinical data management can be classified into 3 phases: Start-up phase, Conduct phase and Close-out phase. The start-up phase involves the designing of the case report forms or CRFs followed by database design. A case report form or CRF is a set of questionnaire set by the sponsor for the subject/patient. The sponsor sets the questionnaire according to the data he/she requires for the trial. A CRF can be a tangible paper CRF or an eCRF where the data is entered directly into the system. Use of eCRF saves time and resources and is also a more eco-friendly option. After this the database is designed and built based on the data required for the trial. Where eCRF is used the database design and CRF design are closely linked. Data entry involves feeding of the data from the CRF into a relational database- direct or real time in case of eCRF. This is followed by validation by edit checks and making notes of queries. Queries are any non-sensical or questionable data which usually arise as a result of typographical errors. Every query has to be attended by one member each of the research team and the administration. Data entry, query resolution and medical coding are ongoing procedures during the Conduct phase. Most commonly used dictionaries for medical coding are MedDRA and WHODD. The Close-out phase begins with severe adverse event (SAE) reconciliation, which is basically a comparison between the data management database (collected via CRF) and the pharmacovigilance database (collected through SAE forms). This comparison is done to ensure consistency of data in the two systems. After this a database lock is done to ensure data is not manipulated during the final analysis stage and is done after all management activities are completed. By now the clean database is ready for analysis and presentation.
General flow of data in CDM Process
The increasing demands from pharma companies to fast-track the drug development process have evolved the CDM to generate high quality data for drug evaluation. The advancements on the technological front too have had a positive impact on CDM resulting in highly improved speed and quality of data generation.