Data Quality Sustainability

By Preeya Beczek – Independent Regulatory Affairs Expert

The life sciences industry is under increased pressure and scrutiny to have accurate real-time data, as more of it becomes public and continuously under regulators lens. At the same time, more internal functions are connecting their key information to gain operational efficiency and obtain regulatory compliance with initiatives such as IDMP / SPOR and EUDAMED.

Progressive organizations view data quality as an organisation competency which requires data governance and processes (see figure below) to be operationalized and a culture of quality to be institutionalised – meaning people need to understand the why AND the how when it comes to data quality, for it to be sustainable over time.

So here is the rub! Nearly every function works with data, we all manage data in some shape or form, and this is increasing exponentially. Whether that is content in documents or registration data for any given product. Thus, data is the intellectual property and asset. So, when we think about the current organisation, the overall functional community should already be good ‘data managers’ and ‘data stewards’. They are generating, writing, reviewing, entering, changing, validating, and submitting data on a day-to-day basis. So, they are responsible to ensure data is accurate and right first time with the relevant validation steps. The functional heads are the data owners and accountable for ensuring data quality. As we implement and enhance the end-to-end technology capabilities (and sometimes automation) there is little room for error. So, naturally there is a need for building new resource capabilities or some upskilling, when it comes to data quality and a strong data mindset.

However, there is a notion that to have good data quality, it takes significant extra resources; not necessarily true! A small team of dedicated data quality stewards constantly check the data to identify any defects. They would monitor and trend data defects and feedback to the data quality council. Regulatory Information Management (RIM) platforms may also have built-in trending mechanisms to show where errors are occurring, where in the workflow and by which roles. This can be helpful to identify where corrective action and additional training is required.

Often data quality frameworks / bodies are seen somewhat separate from other frameworks and teams who are already well established. A good example of this is the disconnect between Data Quality and the Quality & Compliance Function. When in fact these should be very much joined at the hip. The Compliance team undoubtedly have valuable information on corrective and preventative actions, training needs / gaps, deviations, procedural gaps for key end to end business processes and a whole host of other “data” they are trying to obtain and analyse. Additionally, Quality & Compliance are often analysing data for critical business reasons i.e., audit or inspection preparation, change control and deviation management and also addressing key leadership questions. For maintaining or increasing compliance, it’s not only about speed (how quickly data is entered) but also about accuracy (do we trust the data without many verification steps) and also standardisation (find the data in one place, entered in the same way). So, Quality & Compliance is the most obvious place for Data Quality to really partner, utilise the data, insights and resources AND create a common business target on data quality.

When setting up a data quality council / governance committee, it’s important they have the proper decision rights and focus. The council should be made up of data owners, data architects, process leads and compliance / training experts. Primary responsibilities include:

• Set and institutionalize quality standards
• Ensure proper tools & practices
• Prioritize Continuous Improvement and Data Quality Automation Opportunities

Senior leadership typically becomes onboard when they understand the business value to the investment of resources for the data quality council and day to day work (data management and technology strategy). One quick education step is to quantify economically the previous 5 years data remediation projects and the average time that the organization spends “verifying” information because they don’t trust the so called “authoritative source”. This tends to be an “eye opener” to the amount of wasted time, effort, and cost to inefficiently work on data quality.

Once they are onboard, they can advocate getting the data quality right the first time, ensure resources are assigned and aligned and ensure the right data quality governance framework is in place or put in place. They can also be part of the expectation setting through effective organisational communication to engage the masses. Some organisations are really good at this where their day-to-day work is data driven, their decisions are data driven and the right questions are asked to drive the right behaviours. This approach makes data highly visible, actionable at every level and central to everyone’s role.

When embarking on a data quality journey it is important to have the right operating model where all the elements connect and work together. Many organisations are setting up data quality councils / governance bodies, identifying and aligning roles and responsibilities, incorporating data quality into performance goals and objectives, as well as hiring new roles. However, before making many changes and adding layers, it’s important to think about what is already there and how it can be leveraged, whether that is the people, the process, the systems, and any governance bodies (e.g., business excellence teams, change agents, process owners / leads).

Select a handful of key data elements and work through each of the framework elements above; identify who can help, where does data already exist and can be drawn from. It’s important to deliver value quickly, ideally delivering little and often. Also test out your frameworks, processes, people and then adapt as you go. Finally, ensure you communicate broadly to how data quality is positively impacting the organisation while senior leadership advocates for a “culture of quality”.