Assessing the feasibility of manufacturing sites for remote good manufacturing practice inspection

Abstract

Background: The COVID-19 pandemic caused unprecedented disruption to traditional on-site regulatory inspection models due to travel restrictions. In response, regulatory authorities rapidly implemented remote inspection methodologies using digital technologies as one of the mitigation measures. While these approaches proved valuable, their application was often ad hoc, with limited structure or harmonised criteria guiding when remote inspections were appropriate. The absence of a standardised framework risked inconsistency in regulatory decision-making and potential variability in inspection outcomes. This study aimed to develop a structured, risk-based assessment tool to support competent authorities in determining the feasibility and appropriateness of conducting remote Good Manufacturing Inspections (GMP) while maintaining confidence in regulatory oversight. Method: The risk assessment tool was developed through a comprehensive literature review and stakeholder engagement. Existing regulatory guidance, published studies, and inspection practices were analysed to identify key risk factors influencing remote inspection feasibility. In parallel, feedback was collected through a structured questionnaire distributed to regulatory and industry experts. The tool was designed to evaluate whether a remote inspection could provide a level of assurance comparable to an on-site inspection. Key considerations included the robustness of IT infrastructure, data accessibility and integrity, site complexity, and prior knowledge of the manufacturing facility from previous inspections or regulatory interactions. Results: A semi-quantitative risk assessment tool was developed and subsequently validated by two independent expert groups. The tool comprises eight domains, each containing four indicators, resulting in a total of 32 criteria for assessment. The domains include: product and patient risk; GMP compliance history; inspection and regulatory background; manufacturing and process complexity; quality management system maturity; change management and external signals; data integrity and computerized systems; and technical feasibility for remote inspection. Each indicator is assigned a score of 1, 3, or 5, with higher scores representing greater risk severity. Domain-specific weightings are applied to reflect their relative importance, and weighted scores are calculated for each domain. These are then aggregated to produce an overall risk score. The final score is categorised into three ranges: low risk, where remote inspection is considered appropriate; medium risk, where remote inspection may be conditionally appropriate; and high risk, where an on-site inspection is recommended. The tool demonstrated consistency and usability across expert groups, supporting its potential application in regulatory settings. Conclusion: The COVID-19 pandemic demonstrated that digital technologies can effectively support continuity of regulatory oversight; however, remote inspections cannot fully replicate the depth of insight, contextual understanding, and trust established through on-site inspections. The proposed risk assessment tool introduces a structured and harmonised approach to decision-making, enabling competent authorities to assess the suitability of remote inspections in a consistent and transparent manner. While remote methodologies offer advantages, particularly in situations where physical access is restricted, challenges related to technology, cybersecurity, human factors, and limitations in observing non-verbal cues remain. Future work should focus on refining remote inspection frameworks, enhancing digital infrastructure, and integrating hybrid inspection models, ensuring that innovation complements rather than compromises the rigour and reliability of established regulatory practices.