FAIR data principles, what are they and how can they support compliance?

Life Sciences organisations are becoming big data enterprises, generating large amounts of data from clinical studies, lab equipment and drug development. However, this plethora of data is often produced in silos causing the need to implement good long-term data management practices to grow exponentially.

The  FAIR data principles were created to support further scientific study through good data management and rapidly became an industry standard in how data and digital assets should be stored.

Firstly, what are the FAIR principles?

These practical principles comprise of four elements. An organisation should ask themselves if they are storing their data in such a manner that it remains:

• Findable
• Accessible
• Interoperable
• Re-usable

But why should data be FAIR?

Across the industry, vastly growing amounts of research data are being produced in different formats and from different sources. The result of this is a laborious and challenging process of storing, managing, consolidating and extracting value from the data generated.

When we look at clinical studies as an example, there can be numerous partners that an organisation is working with to generate all the data needed for just one specific study. The need to implement FAIR principles in this situation is paramount to being able to re-use the data generated.

Additionally, different areas of the business have different requirements of data. Adhering to the FAIR data principles not only ensures compliance, but enables stakeholders to find and access the data they need – whenever they require it.

Can I implement FAIR and not worry about anything else?

FAIR data principles are extremely important, but they do not cover a lot of the regulatory compliance mandates that are required from life sciences organisations.

It’s understandable to see why compliance is such a crucially important matter. Adherence to regulations prove the integrity of the end product by ensuring good audit trails and accountability of research and manufacturing processes are recorded. Aligning to the above-mentioned FAIR principles supports your journey to compliance.

We should mention here that life sciences organisations must comply with data integrity guidelines, as well as ticking off the principles stipulated within ALCOA+ (you can take a look at our ALCOA+ blog series for further insight).

How do I make data findable?

The short answer is metadata. In order to ensure that your data is findable you need metadata assigned to your assets and files. This may be ‘simple metadata’, such as the source of a specific piece of data, or the subject it relates to, or it may be ‘advanced metadata’ which assigns more technical information. It’s up to your organisation whether you choose to assign this task in-house or look for a fully managed solution which encompasses this as standard.

Beware, storing data in a backup does not make it accessible

One of the most common misconceptions that we come across is that a backup or live storage system is good enough to preserve your data for the long-term. Whilst in many cases these may be simpler options, how do you know if your data is truly being preserved for the long-term?

When looking at accessibility we also need to consider file format obsolescence. Backup does not provide file format normalisation as standard. What do we mean by ‘normalisation’? Normalisation is the process of automatically maintaining a copy of a file in a long-term readable format. You could use open-source tools to do this, but you cannot guarantee access to the original metadata – an archival and preservation solution can.

Due to the nature of some of the research that happens in life sciences companies it is clear that not all data can be open and not all data can be accessed by everyone, security roles and restricted access need to be applied.

Thinking about interoperability

Data silos can cause major headaches when it comes to managing scientific data, especially in relation to personally identifiable data and clinical trials often hold many sources of data and across different systems.

So, if you’re having to look in multiple locations, or even multiple folders within the same location, this could result in your organisation losing a lot of time, simply to find a particular file or document. In order to efficiently manage your data, it is imperative to ensure it is all accessible in one platform, saving you time and financial resources.

If you’d like more in-depth content in relation to interoperability, then our eTMF Archiving and Preservation Guide may be helpful.

Re-using data

You may be able to get to your data, but can you actually use it? Is the format of that particular file readable today? Can you still easily include data from that old server that hosted the ageing device in the corner?

There is more to re-usability of data than whether it is relevant to further study. Whatever data management strategy you have in place, you must ensure it includes provision for technology and format obsolescence.

Do I really need to use the FAIR data principles?

You may be thinking by now that this all seems like a lot of work, but it doesn’t have to be. Data plays a critical role in the life sciences industry but more needs to be done to unlock its full potential.

By adopting the FAIR data principles of Findability, Accessibility, Interoperability and Re-usability into your data management strategy, life sciences organisations can unlock the long-term potential of their data for future research, improve collaboration and support alignment to compliance requirements.