The Clinical Data Management (CDM) process is like clinical trials, which required a fully planned and integrated scheme from the beginning to the end. The outcome of the process is considered first and the whole process of designing the system is made by considering the final deliverable in mind. The main purpose of an effective clinical data management process is to deliver a valid, error free and statistically relevant database.
The clinical data management process starts early, even before the study protocol is finalized. The first step in CDM is the review and the finalization of the documents of the study, which gives an idea about how to design a database which will be later used for data collection. The CDM personnel responsible for database designing will review aspects of the protocol like scheduled visits, the basic demographic data of the patient which needs to be entered, dose calculations etc. To ease out this process, a Case Report Form or the CRF as it is known in the industry is designed by the team and this constitutes the first step in translating the protocol specific activities into data being generated. It has to be kept in mind that the data fields are not repetitive, rather clear and concise.
Based on all the above, a Data Management Plan or the DMP is developed which serves as a very essential document to support the team during conflicts and problems arising during the process. The DMP usually consists of database entry and design, data tracking methods and guidelines, SAE reporting and narration, SAE reconciliation, discrepancy management, quality control and final database locking guidelines. After the DMP has been developed, comes the designing of the database. A database is a software application which helps the user to carry out various functions during the process of CDM. An important aspect of data base designing is to validate the system which ensures data security and accuracy. The components of the study which are taken into consideration during this phase are sites, investigators, patients, objectives etc. The data entered by the user in the front end i.e. on their computer screens gets stored in the database.
The data collection can be done by using case report forms which can be either in paper or electronic format. The tracking of the CRFs is monitored by a Clinical Research Associate or the site monitor for completion and accuracy. In case of any inaccurate or nonsensical data entered, a “query” is generated during the quality check procedure in data management. The resolution of these queries is known as “Discrepancy Management”. This is one of the most critical activities in the CDM process so it is advised to handle this particular aspect with utmost care. The final step in a CDM process is Database Locking wherein all data management activity should be completed prior to this. Once the database is locked, no further changes can be made to it. The data is then extracted for statistical analysis. This outlines the various steps involved in Clinical Data Management.
At the James Lind Institute, one can find courses related to Clinical Data Management. For more information on the same, please visit our website www.jli.edu.in