The current Clinical Data Entry sub process contains seven tasks before redesign of the sub process occurred:
- Worksheets received in-house for data entry (mail-room)
- Worksheets sorted by study and placed on rack for pick up (manual)
- Worksheets picked up for entry and sorted by patient/visit by data enterer
- Data Entered (Clinical Data Entry application)
- Data Entry completed for patient/visit and reviewed for discrepancies (errors) (automated and manual)
- Discrepant data verified and corrected (automated and manual)
- Data marked for review by patient visit (automated)
The current Clinical Data Review sub process contains five tasks before redesign of the sub process occurred:
- Data marked for review by patient visit extracted from database via a set of queries used to identify critical patient information based on a set of data review guidelines submitted to the review team by the Clinical research teams. These reports are used to identify discrepant data (manual and automated)
- Data reviewed for discrepancies (automated)
- Discrepant data investigated based on discrepancy type. If the data is identified as discrepant due to internal error, the error is remediated immediately and rectified using the current data entry tool. If the data is found to be questionable based on the data collected by the investigator, a Discrepancy Notification form is generated and faxed to the investigative sight for remediation (manual and automated)
- Data received back from investigative site, reviewed and updated in the database (manual and automated)
- Data marked for medical review by patient visit (automated)
The goal of the process reengineering effort is to strategically achieve a significant percentage increase in the overall FTE productivity and data quality throughout the data management process (specifically data entry and review) while tactically aligning the reduction in resource cost with such gains. A workflow application already in place can be improved to enable the improved processes to reduce the cycle time of the overall processes and improve the quality of the data gathered during the data entry and review activities.
The functional areas of data entry and review have studied the relationship between cycle times and data quality and the areas of business process and technology to conclude that an improvement in both disciplines would make a significant impact to the strategic goals of the overall CDM process. While analyzing the SIPOC model (Supplier, Input, Process, Output, Customer) for the area of data entry and review, it was determined that one of the root causes of customer dissatisfaction was increased cycle times and poor data quality being delivered by the sub processes of data entry and review. Each attribute that made up the overall data entry and review sub processes were examined using the SIPOC model: worksheet tracking, workflow management, data entry, data proofing and error management (components of data review). It was revealed that by providing additional training and development of resources, an improvement in business processes and an improvement in the technology of the workflow management system could provide an additional gain in productivity and data quality. The point of saturation in gains for FTE productivity reached a plateau fairly quickly and required the addition of resources to meet the outstanding productivity needs thereby increasing the cost ratio against FTE productivity for data entry.
Further analysis on the actual data entry and review technology was completed through a series of exercises. One particular exercise revealed that changes to the data entry and review technology could provide additional projected gains in productivity meeting the overall strategic goals of an increase in FTE productivity by 15% and reducing the resource cost for a fiscal year.
Listed below are cross sections of problems that have been identified as impediments to the areas of cycle time and data quality in relation to the data entry and review sub processes with the usage of current data entry and review technology:
- Examples of Cycle Time Impediments
- Timely drill down required to access data entry screens to perform inherent task of data entry
- Screens have no relation to the worksheets making navigation on screens difficult
- Use of mouse at all junctures of data entry
- Lengthy data processing times
- Proofing using a paper process-CRF that does not represent the worksheet from which the comparison of data making it difficult to make the one to one comparisons in a timely manner
- No indication of data errors on the Case Report Form which requires further drill down into the system to find and fix errors
- Examples of Data Quality Impediments:
- Navigational misrepresentation on screens often time produce data entry errors
- No correlation between worksheet and Case Report Form makes it difficult to complete 100% proofing effectively
The realization has been made that without implementing a technology enabler to facilitate expedited data entry and review, the leverage gained by providing resource training and development and improving business processes cannot provide the increase in productivity numbers and data quality that are desired by the business organization while reducing cost. The technology, as it is implemented with the immediate facilities to expedite data entry and proofing coupled with training and improved business processes can meet the goal for the current percentage increase gains long-term without reaching saturation as rapidly. The increased scope of worksheet tracking and workflow management implemented would project higher FTE productivity gains and decreased cost over time.
Assumptions: For the business-case, a default study with:
- 10 modules per visit
- 10 visits per patient
- 500 patients per trial
Has been used as the benchmark for quantifying gains based on the technology improvements made to facilitate expedited data entry.
The overall success criteria will be demonstrated with increased productivity – more visits per person overall for trial OR realized by a savings in time which could result in fewer persons to complete units of work. Both point to a reduction in cost to conduct the defined trial due to fewer resources used, or in reduced cycle time with same resources used, or in the ability to add on additional studies without adding extra resource.
Main Objectives
The objective of the newly designed sub processes addresses the determined inefficiencies of how the tasks of the data entry and review sub processes were being completed. Each as-is sub process was reviewed and hand-offs were identified between each sub process. The redesign effort has instituted new milestones and those milestones are to be integrated into the workflow management system. Technology enablers have also been reviewed at this juncture and a series of requirements have been drafted to help support the new process changes. It should be noted that a suite of tools are already in production to support the old process and change requests are constructed in tangent with the redesign process to improve on the current systems.
To meet the goal of decreasing the cycle time of this process will be to provide a technology enabler that will expedite the data management sub-process of data entry and review. By doing so, the reduction in the cycle time of the data entry and review sub processes will decrease and thereby lower the overall cycle time of the data management process. At the same time, the application will provide mechanisms to validate the data during the data entry process to reduce errors and provide “Clean” data for review. The implementation of such a mechanism would meet the goal of increasing productivity while decreasing cycle times of the process and reducing the error rate generated during data entry and review. Furthermore, the decrease in cost of resources will be realized.
The system objectives clearly include enabling the new and improved data entry and review sub processes. The redesigned sub processes identify tasks and will occur in the following order:
- Worksheets received in house for data entry (mail-room)
- Worksheets sorted by study, patient/visit and entered by one or two data enterers (based on volume of the worksheets) into the Data Entry system to track the date of the arrival of the worksheets, number of worksheets, patient and visit information and was then stored as batches so the data entry staff could finish a batch of worksheets and move on to the next batch. The data entry staff could monitor (through the use of the new enhancements to the data entry tool), the number of worksheets by study and patient/visit and based on their assigned study, select those batches of worksheets relevant to their assigned workload (automated)
- Data Entered by data entry staff for the batch of worksheets collected and sorted by patient visit (automated)
- Data Entry completed for each patient/visit and marked for review (automated)
- Data Review team receives automated reports to monitor the status of the data marked for review and based on available data, selects those patient visits allocated by study for patient visit and conducts data review (automated)
- Data Review notifies investigative sites of discrepancies in data via email notification setup internally through system within 24 hours of finding discrepant data. All other discrepancies are remediate by data reviewer and updated within system (automated)
- Data Review received updates from investigative sites within 72 hours and updates system accordingly (automated)
- Data Review marks data for medical review (automated)
The redesigned sub processes eliminated 4 tasks and will use the existing data entry and review tool with additional enhancements to support the new sub processes. The overlap of responsibilities was addressed wherein the data entry staff was solely responsible for the entry of all trial data and the data review team was responsible for reviewing, verifying and fixing all erroneous data and was granted access to the data entry system to make the changes.
The stakeholders directly and indirectly affected by the implementation of any improvement efforts to the technology that facilitates the Data Entry and Review sub processes include all data entry, data quality and encoding personnel, under the functional area of Operations, and the authors of the CSR and WMA reports downstream. The stakeholders in the area of data entry and review have unequivocally voiced the following areas of concern that impede the increase in FTE productivity desired by the organization’s management:
- Workflow Management: The inability to manage the worksheets received by the investigative sites to devise a streamlined workflow. The lack of such a mechanism does not empower the functional and project managers in the area of data entry to effectively allocate work in a timely manner. There is no way to project workflow against the current workload received in-house and projected incoming work without a manual process that is timely and costly which inevitably leads to reduced productivity.
- Time Management: The timeliness in completing the tasks of data entry, verification and error management. The current data entry and review technology does not facilitate expedited data entry, verification and error management. The functions of data entry and review require additional contingent resources to manage the entry and review of outstanding work to meet tight deadlines regardless of the complexity of a study.
- Quality Management: Currently, the data entry and review functional areas are required to proof data at a rate of 100% per patient/visit. Therefore, the process of data proofing (currently paper-based) is a timely and costly endeavor that also challenges the quality of data due to tight timelines and at certain junctures, the lack of resources attributed to the inability to estimate and project a manageable workflow for the data entry teams. The ripple affect of poor data proofing increases the workload downstream and reduces overall productivity and data quality gains in all functional areas related to trial data management.
Analysis
The current data entry and review tool is deployed worldwide and is being used to collect, process, and report on data that is gathered during clinical trials. The data entry and review subsystems are a component of the Clinical Trial System (CTS hereafter) and have been implemented as the collection and review devices for clinical trial data for trials undertaken by the organization. The application interprets trial metadata and generates data entry screens dynamically, into which clinical data is entered. The data collected through the data entry tool is stored in the CTS database and is available for review and analysis under the data management process utilizing additional subsystems as part of the suite of CTS applications.
The application must enable data entry and review in an on-line mode based on the business requirements on a trial by trial basis as defined by the organization. This tool is deployed to internal sites and is used for entering and reviewing data. The application interfaces with CTS components to build the application (generation of data entry screens), and provide input, and output mechanisms to the CTS database.
The investigative sites collect the trial data for a patient by visit and provide those paper-based worksheets to the ABC Company data entry staff. The data entry personnel log the data verbatim and submit the data for data review by setting the appropriate milestone signaling the data entry portion of the data management process has been completed. The data is then transmitted directly to the CTS database. Prior to the successful loading of the trial data to the CTS database, the data is validated and then loaded to the Core CTS tables or sent to the Suspend tables for verification. The Data Validation tool is utilized to verify the validity of the data entered in the data entry and review application before the Data Capture component of the CTS application suite performs the insert of any record into the CTS database. Once all data has been successfully stored in the CTS database via the data entry and review application, a number of additional CTS applications interact with the data entered via this tool.
The following diagram details application interfaces with the CTS database.
System Design
Data Entry Workflow Application Interface:
Initializing Patient/Visits:
Data Entry Screen:
Data Review Workflow Application Interface:
System Data Requirements
The information entered into the data entry application is stored within a Custom Table, specifically designed for the clinical study. The data are uploaded into Load Tables, which function as staging areas for the validation of the information. A validation process then matches the collected information against Metadata definitions. In other words, Metadata are information stored within separate database tables whose function is to provide format and content definition for collected data. The validation process matches collected data to the Metadata. Inconsistencies and data errors are deposited within Suspense Tables. The information in the Suspense Tables is then reviewed and corrected as necessary. The data are redirected back into the Load tables for revalidation. All data enter the CTS database via the load tables upon successful validation. There is no “back door” into the database. Upon completion of the validation process, the data are deposited into Core Tables. Core Tables function as data holders for validated data. The View Management tool initiates and populates the Extract Tables with information from the Core Tables for data review.
The following diagram represents the logical data model with table definitions required to store the collected clinical trial data for data review:
Data Entry / Data Modification Personnel:
These are the individuals responsible for entering or modifying data in the database for a particular trial. They would require “read/write/delete” access to the data.
Data Review Personnel:
These are the individuals responsible for reviewing the data in the database for a particular trial. They would require “read/write/delete” access to the data.
Access Rights:
- Read access will allow viewing of the data by any personnel part of the overall CDM process and system support staff required to trouble shoot issues with the system.
- Read/Write access will allow viewing and manipulation of the data by the data entry or review user of the system .
- Support will allow running required scripts against the data for validation.
- Read/Write/Delete will allow viewing, manipulation and deletion of the data by the data entry or review user of the system.
Summary
Drug development demands for a short time to market, as well as specific regulations make medical device studies a tough challenge. An average 80 clinical trials involving 5000+ patients are necessary for a new drug application. Shortening development time allows more drugs to enter the marketplace, offering a return on investment to shareholders.
This new system achieves the following goals:
- Reduces wait times (In-House Data Entry Staff)
- Eliminates Wait time before events
- Reduces the number of processes
- Transforms the process to events
- Minimizes number of contact points
- Make processes/Events parallel
- Improve quality at each step
- Improve cycle time
- Execute the process with minimum use of resources