AI and analytics might be more advanced than ever before, but… Why do we still struggle to apply them to the clinical care sector?

Back in October, Alex Maiersperger from SAS, Tony Jurek from Deloitte and Jim Conyers from Enlitic joined Health Tech Forward 2022 to examine how organisations can become data fluent when it comes to healthcare.

Achieving data fluency isn’t a walk in the park.

Doctors’ trust in algorithms and unfamiliar tools wavers without understanding how these technologies work. Data scientists struggle to provide frontline workers with relevant data at the right time. The lack of a common data model between sectors hampers mutual understanding.

Despite these hurdles, simple rules can guide us toward data fluency. By addressing these challenges head-on and embracing simplicity, we can truly unlock the potential of data in healthcare. So how do we get there?

DO: Work with doctors while building the algorithm

Building trust between data scientists and doctors is crucial when developing algorithms in healthcare. When medical professionals have an understanding of how the technology operates, they are more likely to trust its outputs. However, if collaboration is lacking, problems arise. Alex Maiersperger from SAS says that excessive focus on “pretty dashboard design” can lead to oversimplified results that doctors find difficult to trust.

Luckily, a positive shift is underway as organisations start to recognise the value of working closely with doctors to foster data fluency and even embrace AI-based hospital strategies. 

DON’T: Overlook the clinical workflow

There aren’t a lot of organisations out there that are focused on data fluency from a healthcare standpoint. The main obstacle for them is delivering the right piece of information to the front-line workers at the right time. 

“How do we convert data so that it has relevancy? Does it have context in that relevancy?” asks Jim Conyers from Enlitic. One thing is clear: we have to solve the clinical workflow problem first to achieve data fluency.

Jim Conyers, CEO of Enlitic

DO: Create a common data model between medical workers and data scientists

Consistency within clinical trial data remains an ongoing challenge. 

Jim Conyers highlights the importance of establishing a common ontology-based structure, which will serve as a foundation for fostering data literacy and facilitating widespread adoption. Simply put, content devoid of context is just raw data. But by integrating relevant context, we can transform the data into actionable information that unveils valuable insights.

DON’T: take away individual responsibility for what’s going into the algorithms

As consumerism gained prominence, new titles like Chief Customer Officer emerged, which, according to Alex Maiersperger, brought both benefits and challenges to organisations. 

Surely, they created a perception that there is a designated person responsible for customer-centric initiatives. But they also undermined organisational dynamics by absolving individuals of their inherent responsibility to contribute to the core nature of their roles.

Ideally, data literacy should be a collective endeavour where everyone within the organisation takes ownership. And it is essential for data fluency to permeate across departments, transcending hierarchical boundaries. 

By instilling a culture where everyone embraces data literacy in their roles, organisations can unlock the potential of data-driven decision-making and foster a collaborative environment that propels the business forward.


Watch the full panel discussion here

About Health Tech Forward: Health Tech Forward is a two-day annual conference that hosts top global digital health investors, entrepreneurs, and government leaders for exclusive sessions and compelling discussions about the future of digital health.