Unlocking access to real-world healthcare data

Geoff Hall, Consultant in Medical Oncology, Chief Clinical Data Officer at HDR-UK and CCIO at Leeds Teaching Hospitals NHS Trust, speaks to Imosphere about the health sector's data harmonization, the need for data consistency, and how OMOP will improve data quality and the impact this has on patient outcomes.

Making real-world data available at scale is fundamental to support research, as well as the advancement of patient care and better outcomes. COVID has been a notable example of how, when the world comes together, and uses its healthcare data at pace, the healthcare research community can enhance the understanding and management of diseases.

For data custodians and health researchers responding to the pandemic, there was a real need to compare and analyze vast datasets at speed and scale. However, the process of harmonizing data with other centers can often be complex and time consuming. There is recognition from the international healthcare research community that we need to work together to improve the interoperability of data. In fact, at the end of last year, Health Data Research UK released a White Paper highlighting the importance of using the most common well-described data standards in health research.

Centers heavily involved in health data research are beginning to look towards the Observational Medical Outcomes Partnership (OMOP) common data model. Converting healthcare information into the same format enables multicenter collaboration to work much more efficiently and unites centers nationally and internationally in the analysis of real-world data.

For instance, if a research center in Australia wanted to examine the impact of diabetes on the length of stay in hospital, after a hip replacement, they could write a script which analyzes OMOP data. This can then be shared with other hospitals around the world, who would run that same script against their data within days and weeks, instead of having to take months to agree a data model.

Real world data needs multicenter collaboration

The power of OMOP has already been proven. The data model was used by the Observational Health Data Sciences and Informatics network to validate international guidelines for high blood pressure. In the study, members of the team examined insurance claim data and electronic health records of 4.9 million patients across nine observational databases. That’s a transformative way to deliver healthcare research and if centers speak the same data language, it becomes amazingly easy to facilitate collaboration, quickly and at scale.

A common data model such as OMOP allows centers to combine their own data, compare it with other centers and learn from it. By analyzing how other clinical and academic centers are treating patients, it gives healthcare providers the opportunity to understand the strengths and weaknesses in their own pathways and adjust accordingly. It’s through multicenter collaboration, that you have that ability to constantly reflect and improve on the care being delivered.

Protecting patient privacy

However, it isn’t just the medical community who are interested in using data to improve the care patients receive. Commercial groups are also keen to analyze real world data and pharmaceutical companies often utilize common data models for drug safety surveillance activities. Of course, healthcare data is highly sensitive so there needs to be consideration around how to protect private information. If commercial companies produce a script that is compatible with OMOP, it enables them to write queries and receive analysis of the data in aggregate, without accessing protected healthcare information.

The script can then be shared with other centers using the OMOP model to open the possibility of clinical and academic groups collaborating with commercial organizations. Here, OMOP facilitates commercial collaboration but ultimately further enhances patient privacy.

Converting to OMOP

At Leeds Teaching Hospitals NHS Trust, funding was recently granted through the European Health Data Evidence Network to standardize patient data into OMOP and we’re working with Imosphere to do this. As a large Trust, we receive regular requests to contribute to collaborative research studies and converting our data will enable our teams to collaborate nationally and internationally with other centers.

For centers thinking about embarking on a common data model project like OMOP, it’s important to bring in the right expertise. While the data model is open information and publicly available, it’s a complicated process. Therefore, partnering with an organization with experience in this area is fundamental, especially if they can offer training to support staff to harness data management skills in the future.

Harness the potential of real-world data

Over the last two years, the value of real-world data has been pushed into the mainstream. The need to collaborate across multiple centers has become much more established and there is a sense that the public have a greater understanding of how important it is to share anonymous healthcare data.

OMOP is a solution to elevating the impact of real-world evidence and the health research community is beginning to use it in a way that will underpin this. Establishing common standards for healthcare data is fundamental to enabling and enhancing research at scale. When it comes to uncovering greater insights, research has the potential to enable discoveries that will change and improve people’s lives.


Find out more about our OMOP data harmonization services here.