Data integrity refers to the completeness, consistency and accuracy of data over its lifecycle. Data must be accurately captured, and that accuracy must be maintained during data storage, replication, transfer and processing – all of which have the potential to degrade or compromise the data. Data integrity is key to demonstrating compliance with pharmaceutical quality guidelines and regulations relating to manufacturing, documentation and laboratory practices. Failure to ensure data integrity is a breach of good manufacturing practice (GMP).
Data integrity: an era of regulatory reform and enforcement By Sarah Harding
In 2016, 80% of the FDA Drug Product and active pharmaceutical ingredient (API) warning letters included a reference to the lack of data integrity, mainly due to incomplete data. Also in 2016, more than 80% of Chinese clinical trial data were found to be “fabricated”, undermining the pharmaceutical market in China, which had grown over the previous two decades to become one of the country’s economic success stories. The China Food and Drug Administration (CFDA), which identified the “80% falsified” statistic as a direct result of new drug regulations introduced that year, has since brought in a new era of regulatory reform and enforcement. Raising the bar for good clinical practice, and strongly deterring those who would manipulate data for profit, have been essential steps towards re-establishing China’s reputation in the global marketplace, and enabling it to realize its significant potential in improving the world’s health through the supply of APIs and the production of affordable generic drugs.
The 1990s FDA acronym ‘ALCOA’ summarizes the principles that define the key criteria which data must fulfil to ensure data integrity in accordance with regulations. ‘ALCOA’ stands for Attributable, Legible, Contemporaneous, Original and Accurate. These principles provide a framework for achieving GMP-compliant data integrity and have been included in most guidelines relating to integrity across the world.
- Attributable: Data must be attributable to the person generating the data; who performed an action and when?
- Legible: All data recorded must be legible and permanent, even if corrected or updated.
- Contemporaneous: Data must be recorded at the time of the task or measurement. Date and time stamps should match the order of execution and data should never be back dated.
- Original: Original data, recorded for the first time.
- Accurate: Free from errors, reliable, truthful and reflective of the observation. There should be no editing without documentation/annotation of amendments.
Data integrity is applicable to both paper and electronic records, but with increasing demands from regulatory agencies across the world, it seems obvious that paper records are no longer fit for purpose. Electronic data, collected straight from the source and with built-in error checking methods and validation procedures that ensure the data remain intact and unchanged, are surely the only viable response to current regulatory demands. A digital solution, whereby all required data are dealt with automatically, saves time and also prevents transcription errors. Electronic data are more likely to be enduring (thanks to remote back-ups, for example) and they are more readily accessible and available for review, audit or inspection. This includes ‘Big Data’, which – as we have discussed before – can be converted into useful, comprehensive data through digitalized approaches in a manner that simply isn’t possible with conventional non-digitalized systems.
This is why, while other manufacturing industries discuss the application of Big Data to achieve more automated and robotic processes, pharma manufacturers and health care products suppliers tend to be more f...
(All Rights Reserved)