Data management
the persistent issue of siloed operations within and across organisations hampers progress. Data management teams, biostatisticians, and external vendors often work independently, with limited integration of processes or technology. This isolation is compounded by the increasing complexity of trial designs and the sheer volume of data being generated.
The silo effect in clinical trials arises from organisational structures, varied technological capabilities, and the inherent complexity of managing diverse data types. “Data management is a very critical component of the trial life cycle,” explains Bazgha Qutab, who leads drug development in Europe as principal at ZS Associates, a management consulting and technology firm that partners with companies to bring together data, science, technology, and innovation for better outcomes. “The data management space itself in the past 10 years has been inundated from how do we actually build a robust data management infrastructure to how we do data management in an automated way and in scale.”
Data management focuses on primary trial data collection and preparation for submission, while biostatisticians work on analysis and interpretation. Meanwhile, external vendors such as Contract Research Organisations (CROs) often operate using their proprietary systems, which may not easily integrate with sponsor systems. These independent operations create barriers to collaboration and hinder the seamless flow of information. Adding to the challenge is the increasing reliance on innovative trial designs. Adaptive trials, synthetic data integration, and decentralised trials require the collection and analysis of more diverse data types. The current infrastructure struggles to support these designs, further highlighting the need for transformation.
Breaking down silos: The path forward To bridge these silos, the industry must address both cultural and technological barriers. Collaboration between data management, biostatistics, and external vendors should be underpinned by robust data standards, modernised platforms, and streamlined processes. “There’s a lot more data modalities coming in now as you start to integrate digital endpoints, real world data, omics, and eventually programs experimenting with synthetic control data arms. As you get that pulled in, there’s just a quantity of data and then continuing to figure out the quality of that data,” explains Mike Martin, who leads global drug development practise area as principal at ZS Associates.
32
Standardisation is a cornerstone of efficient data management. Initiatives like Clinical Data Interchange Standards Consortium (CDISC) standards, including ADaM and SDTM, have brought significant progress in harmonising data for regulatory submissions. “Data standards take a pretty critical role now; of course, data standards also get vague as new data types come in, especially the digital data, synthetic data and simulated data, as well as we see pharma companies maintain multiple versions of data standards so governance needs to play a critical role in data management” adds Qutab. However, as these new data types become integral to trials, these standards must evolve. Ensuring that emerging data types align with established frameworks will facilitate integration and maintain data quality across the trial life cycle. At the same time, digital transformation is revolutionising clinical trials, providing opportunities to enhance transparency and efficiency. Advanced platforms consolidate multiple functions – data collection, querying, analysis, and submission – into a unified system. This eliminates the need for disparate systems and reduces the risk of errors and data loss during transitions. “Digital comes in the early design side of the development life cycle and helps us create the whole experience around the design,” says Qutab. Platforms such as Medidata, Veeva, and modernised electronic data capture (EDC) systems are leading the charge. By enabling seamless integration of electronic health records (EHR) and EDC systems, automated data capture and data collection, these platforms support end-to-end data management and promote collaboration. For example, patient data collected through wearables can be directly integrated into trial databases, ensuring real-time insights and minimising manual data handling. “That is where digital with the patient interaction is going to become a game changer and where device data, wearables, digital data collection, essentially, is changing the whole space of trials and how we run trials,” she continues.
AI and automation hold immense potential to streamline data management processes as well. Tasks such as data cleaning, query resolution, and statistical analysis setup can be automated, freeing resources for higher-value activities. Real-time data validation, supported by AI, ensures data quality at the point of collection, reducing the reliance on manual interventions during database lock. “I do believe that AI and these technologies will help us move that to scale, and not make these one-day improvements that we’ve been seeing in the past years – it’s going to have a massive impact
Clinical Trials Insight /
www.worldpharmaceuticals.net
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37
