Thank you for Subscribing to Life Science Review Weekly Brief
Data will continue to disrupt Regulatory processes and systems. There are several factors that drive this disruption such as granular data requirements from health authorities, financial pressure to do more with less, and organizational pressure to improve efficiency and quality. From compliance activities to process improvements, companies will need to unify their data to meet these demands.
This should not be a surprise since the data and information that sponsor companies share with health authorities has become increasingly granular over the past 20 years. Initiatives like Structure Product Labeling (SPL), electronic Common Technical Document (eCTD), Study Data Tabulation Model (SDTM), Extended Medicinal Product Dictionary (xEVMPD), and Identification of Medicinal Products (IDMP) are examples of standards, guidance and regulations that drive the capture and exchange of structured drug and biologic data. These standards prompted the creation of databases, systems, and tools to store and submit data in a compliant manner to support narrative information delivered in documents.
Now is the time for companies to consider data unification. Data unification is the process of turning data from multiple sources into a unified set of records that can be trusted and relied upon in every process, system, or service in use.
"Companies that embrace and prioritize unifying their data can meet compliance requirements and utilize trusted, curated data to unlock digital technology to improve processes and systems."
Seeing this you would think that there is a complete set of processes, systems, and standards for data intake, curation, storage, and transmission between sponsors, third party vendors and health authorities. Unfortunately, the truth is that there is still a long way to go. Information provided to health authorities (and internally at sponsor companies) is still primarily captured in documents – often in PDF format. Data sits in separate systems without common data definitions or governance models. Data is duplicated from system to system due the lack of a data model or interfaces between systems and databases. The lack of data governance and data sharing leads to companies spending time and resources on quality checks and validation (and re-validation) of data.
You may ask “How did we get here?” In many cases companies did not implement systems and databases with the end in mind. Additionally, companies did not (or could not) know about each new system or piece of data that would eventually be needed. Or companies just relied on vendor solutions to solve a specific business case without considering the larger context of how the data in a solution fit into the overall architecture (for systems and data).
Defining and identifying master data in our organizations is one aspect of data unification. When master data is applied across an organization, the data is captured and verified once in the source system and then referenced in other databases, systems, processes, etc. This results in one source for data input and any changes to data in the source are reflected in the referencing databases, systems, and processes. Another benefit is the elimination of redundant quality checks and revalidation of data since it was input and verified at the source.
A potential pitfall of data unification or a data governance program within an organization is that it becomes siloed in the same manner that our systems are siloed today. This needs to be avoided. An example is when data governance is implemented at a functional or departmental level versus at an organizational level. Unfortunately, this will only create a larger departmental silo that includes data governance and systems. While it’s true that quick wins can occur at the functional level, if data are not connected organizationally, we will not achieve data unification.
Organizations need to treat data as an asset and not an afterthought and invest in the people, processes, and systems to unlock the benefits of unified data. Achieving a unified data framework requires a corporate mandate, prioritization, and funding to establish data governance, master data (and data) mapping, and standards across the organization. Cross-functional data governance bodies need to be empowered to be stewards for the data unification framework and enforce data reviews for new systems, upgrades, and changes.
Having unified data will allow companies to comply with increasingly granular data requirements from health authorities. A unified data model will reduce redundant activities that occur today by eliminating the need to duplicate data from system to system and eliminate the need for quality checks since the data is verified in the source system. Unified data also opens opportunities to use digital tools like Artificial Intelligence (AI), Machine Learning (ML), and Robotic Process Automation (RPA) to improve process efficiency and quality along with building algorithms to analyze data as inputs into research and development or regulatory strategies.
Companies that embrace and prioritize unifying their data can meet compliance requirements and utilize trusted, curated data to unlock digital technology to improve processes and systems. Unified data can eliminate redundant processes and streamline data flows which can reduce the time to deliver life changing medicines to patients.