Abstract
Background: Data is not information. Information is actionable. Data, alone, is not. This distinction has significant risk management consequences when using data to support decision-making in the highly regulated pharma and biopharma manufacturing sector. A new framework is proposed to support the trustworthy use of the information generated from data, by data analytics, for effective risk-based decision-making. The emergence of a standardised approach for using such information is essential to avoid duplication and divergence of efforts across the sector. Ensuring that any proposed approach aligns with regulatory expectations for the sector is also essential.
Methods: Through an industry-academic collaboration, over 520 person hours have been contributed across interdepartmental workshops and discussion groups with pharma and biopharma manufacturing partners. Initial discussions and debate centred on data use, its risk perceptions, barriers, existing tools, and future applications. Assessment of risks to product quality and supply across two specific use cases, one involving dashboard use and another involving yield improvement from a statistical process model, were considered in shaping the framework. The regulatory and guidance documentation concerning data use in industry was reviewed. The initial version of the framework was reviewed by the Health Products Regulatory Authority (HPRA), but it is not endorsed as official regulatory guidance by the HPRA.
Results: The proposed framework is divided into three stages: FRAME, ASSESS and ASSURE. The ICH Q9 (R1) quality risk management principles are applied throughout, and the concept of intended use as highlighted in the recent Computer Software Assurance guidance issued by the FDA, is leveraged into the framework to identify the importance of the information from the outset. The stages are proposed to capture the information and its intended use, to build knowledge by identifying and assessing the hazards related to data access, data transformation and its presentation as information. Existing process and product knowledge is leveraged where appropriate to help support the intended use of the new information. Importantly, controlling for the specific hazard of unintended information use is also highlighted.
Conclusion: The proposed framework is information-centric rather than data-centric and considers the intended use of the information to identify information importance. This sets the tone for risk management considerations when turning data into actionable information. In the age Pharma 4.0, the approach proposed here may help to support the sustained end use of both data and information for decision-making, while critically, also assuring pharmaceutical product quality and product supply.
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.
Recommended Citation
O’Mahony, Marcus; Bassett, Paul; Kavanagh, Alan; Maguire, Anthony; Corrigan, Aoife; Malone, Barry; Walshe, Michael; Griffin, Siobhán; Coleman, Thomas; Holland, Margot; Courtney, Kate; Rafferty, Carl; Greene, Anne; and Warnock, Damon
(2024)
"FoReSight: Trusting and Using the Information from Data Analytics to Support Decision-Making in Pharmaceutical Manufacturing,"
Journal of Applied Pharmaceutical Regulatory Science:
Vol. 1:
Iss.
1, Article 8.
Available at:
https://arrow.tudublin.ie/japrs/vol1/iss1/8
