Introduction
Data integrity is essential to all validation processes in pharmaceutical and medical device manufacturing facilities. Understanding and following the ALCOA+ principles and how ClinSpark supports the ALOCOA+ principles are essential for understanding requirements for data integrity.
Data integrity is so interwoven with product quality, patient safety, and regulatory compliance, following the ALCOA+ principles is a high priority for the entire ClinSpark team.
Data Integrity is crucial in the pharmaceutical research industry and with advances in technology, the increased levels of automation, globalization of life sciences, and the use of contract manufacturers have made data integrity more important than ever.
ALCOA
Attributable
In order for data to be attributable, the following must be recorded:
identity of a person, system, sensor, equipment or device that collected, generated, or updated the data.
source of the data
the date and time
How ClinSpark Helps
ClinSpark captures and stores the identity of the user, with a date and time stamp, in the audit trail
Many instrument integrations additionally include the serial number of the device
It is possible to configure ClinSpark to capture a barcode scan of a device in the clinical records, particularly useful if serial numbers are not available as metadata
Legible
This principle is less of an issue with electronically captured data; however it is still relevant so that all data can be read and understood years after capture.
All data collected must be able to be clearly understood years after the study has been conducted.
Consistent straightforward language is required.
Data collected, generated, or updated must be permanent.
Any and all data updates/ changes/ archival are accompanied by a timestamp as well as an Audit record providing a reason for change.
How ClinSpark Helps
eSource data is not subject to handwriting proficiency or clarity
Data captured electronically is not subject to e.g. fading (ink on paper)
ClinSpark supports the use of configurable selection lists and code lists which can improve consistency
Collected data are retained by the investigator so long as ClinSpark is in use
All data changes are accompanied by a full audit trail
Contemporaneous
The essential recording of the actual time and date activities or actions take place.
time-stamping is essential to ensure data is not held in a queue, and recorded using system clocks that are accurate and record time zones (see ClinSpark Timing Variables).
How ClinSpark Helps
Where properly configured by the user, all protocol required record creation and editing / modifying actions can be tracked with a date and time stamp (to the nearest second)
Data may be changed at a later date, but this is obvious to the user in the interface and audit trail
Original
Data records should be original and not copies of data. Although this principle is less of an issue with electronically captured data; it is still relevant to ensure there are technical and procedural processes to ensure the original data collected cannot be inadvertently changed by a copying process.
initial recording of data should be the main record.
crucial to ensure there are technical and procedural processes to ensure the original data cannot be changed.
any analysis, reports, or calculations based on generated or collected data must be traceable to its original source.
How ClinSpark Helps
By definition, data displayed to health care professionals / investigators in ClinSpark can be regarded as source
This includes medical device data that is transferred automatically and not copied to paper and transcribed
It can also include laboratory safety data or other data transferred from analysers and LIMS/LIS systems external to ClinSpark, where these records are signed electronically by investigators
Accurate
All data should reflect the reality of the activity or action and be error free.
Electronic data collection systems must have built in accuracy checks and verification controls
equipment used for measurements should be calibrated regularly
How ClinSpark Helps
ClinSpark can be configured to include both simple (range checks for minimum and maximum acceptable values) and sophisticated (programmed in Javascript) for all collected data
ClinSpark supports internal and internal QC querying workflows
ClinSpark includes simple graphical visualisation tools to support trend analysis and the spotting of outliers
ClinSpark supports data exports in multiple formats to allow for detailed analysis and comparison using third party tools
Equipment calibration is outside the scope of this article
ALCOA +
Complete
This principle pertains to the original recording, the metadata, retest data, analysis data, etc. and requires audit trail tracking for any changes made to the data.
Note
ClinSpark does not allow for data to be dissembled or separated. It is permanently retained with the audit trail and any changes
Consistent
Date and timestamp is expected to be in sequence, including the audit trail.
Note
Collected data are always associated with a date and timestamp and, are typically displayed in timeline order. If exported, data can be sorted in any order
Enduring
Data must be durable enough to last long after it is recorded, decades in some situations.
a robust and reliable backup system
data recovery processes
resilient infrastructure
cybersecurity
Note
ClinSpark continuously streams data to remote storage in a different availability zone
Data restoration processes are automated
ClinSpark is hosted by AWS, the largest hosting provider in the world
The ClinSpark platform has been designed and implemented with security at its core
Available
Data must not only exist but must be accessible for audits, reviews and inspections.
data must be searchable
data must be properly indexed and labelled
readable at any time during the retention period
Notes
ClinSpark supports filtering of data in the user interface
Data can be exported for filtering, viewing and searching in third-party tool
System owners have full control over the visibility of data in the retention period
References
Guidance for Industry Electronic Source Data in Clinical Investigations