...

• identity of a person, system, sensor, equipment or device that collected, generated, or updated the data.

• source of the data

• the date and time

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.

• Manually (handwritten) data must be clearly legible.

  • this applies to uploaded handwritten data that may accompany the captured study data in ClinSpark such as Medical Records or Informed Consent procedures

• 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.

Contemporaneous

The essential recording of the actual time and date activities or actions take place.

• timestamping 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 - Documentation - Confluence (atlassian.net)main idea is that data is not updated at some point in the future. When data is updated after the initial collection, this can increase the chance of mistakes, incomplete data and inaccurate data collection/ recording. ).

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 (whether handwritten or digitally captured) 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.

Accurate

All data should reflect the reality of the activity or action and be error free.

• no edits of initial data collection, and if there are edits or corrections necessary the original data should be documented in a way that the original data can be referred back to. (blocked out/ removed or deleted)

• Electronic data collection systems must have built in accuracy checks and verification controls

• equipment used for measurements should be calibrated regularly

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.

Consistent

Date and timestamp is expected to be in sequence, including the audit trail.

Enduring

Data must be durable enough to last long after it is recorded, decades in some situations.

• a robust and well tested data reliable backup system

• data recovery system

• disaster recovery plans

• uninterruptable power supplies

• cybersecurityprocesses

• resilient infrastructure

• cybersecurity

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

References

...

1. Guidance for Industry Electronic Source Data in Clinical Investigations

   1. https://www.fda.gov/media/85183/download