Watch the videocast, here
Sandeep Dadlani, Executive VP and Chief Digital and Technology Officer at UnitedHealth Group (UHG), the world’s largest healthcare enterprise with diversified businesses, has a long and deep technology experience across multiple industries on both the supply and demand side. His early days at UHG coincided with the explosion of GenAI on the global stage and has been shaping some of the thinking and doing for him.
In speaking with Sandeep, it is clear about the methodical and structured approach he is driving at UHG could define how healthcare leverages the latest technology miracle.
Our most recent candid interview – as part of our GenAI Leaders Series – is to learn how Sandeep fashions GenAI’s use and ways to realize its potential in healthcare and potentially beyond.
Pragmatic, excited, and responsible: the steps to get it done
While UHG is the world’s largest commercial healthcare enterprise with ~$360B (Sep 30, 2023) in revenues, it is also amongst the largest enterprises by workforce of some 440,000 clinicians, technologists, and market-facing professionals. An enterprise of this size with even greater implications for the health and well-being of over 150 million people must be deliberate when exploring new technologies. That is precisely the approach that is being considered while the world is abuzz with GenAI.
A pragmatic approach is to find the problem(s) to solve as the first critical step in being able to address it with any new technology. At UHG there is a bottom-up effort at identifying use cases that have led to piloting some 500 use cases while the top-down identified some 14 use cases. The approach to identifying the top-down use cases was an enterprise celebratory event called Tech-Tank involving tens of thousands of employees. While ideation and spitballing are part of the effort, UHG took a hard look at the business case and the ability to scale those use cases in their selection. Given the size of UHG, scaling means very different, and early indications are very encouraging.
The use cases are generally in the administrative realm of the value chain, which historically has accumulated suboptimal processes and is a rich target for technology transformation. These low-hanging fruits include processes used by thousands of call center agents to summarize their interactions with United’s members.
“…call summarization is a simple thing but has eluded the industry for a while but really eases the work for our call center advocates and has them focus on caring for the person who is calling” – Sandeep Dadlani
Never mind the cape GenAI wears, just focus on its superpowers
“…great synthesis and data extraction from structured and unstructured fantastically well, content generation very well and automates code writing…” Sandeep Dadlani
UHG’s selection of use cases keeps clinicians in the loop to ensure that they can practice at the top of their license and not replace them. This extends to all processes that may or may not include clinicians, that a human is always in the loop to help improve the outcomes and it is done responsibly.
And so, the notion of responsible AI does not have a stronger motivator than the use of GenAI in healthcare. In the context of life and death implications, be it for diagnosis, choice of therapies, or care delivery, responsible AI must become table stakes in action vs. narrative. There must be added urgency to ensuring fairness, eliminating bias, and clear explanations of results.
GenAI’s iPhone moment is more impactful than the Kodak moment
IT services are experiencing a flat revenue trajectory in 2023 after a quarter of a century of sequential growth. As a result, most of them are investing in GenAI to fuel the next era of growth. However, the philosophy of investments in healthcare could have long-term implications. There are two schools of GenAI investments in the context of the triple aim of care (reducing the cost of care, improving health outcomes, and enhancing the experience of care);
- Positively improving the triple aim of care by empowering clinicians to practice at the top of their license, incorporating ambient tech to be virtual caregivers, or accelerating drug discovery. This philosophy will take longer to pay off but will be sustainable and result in strong growth.
- Maintaining the status quo by following legacy paradigms, including labor arbitrage, could see an immediate improvement but is unlikely to be sustainable.
The potential of GenAI is like the launch of iPhones in 2007 and the realization that it could not only replace the 36 pictures of a Kodak film role, but one could store thousands of pictures on the device. The notion of experimentation became common because one did not need to be precise in the shooting of a picture, photography expanded to everyone with a smartphone, and functionality expanded beyond pictures. In a similar vein, expect GenAI to deliver more technology faster with better outcomes.
Yet before IT service providers run the idea to the banks, it is important to address the improved productivity and how that will be shared. Early indicators suggest that we should expect 30-70% productivity gains, and enterprises expect that the productivity gains will be shared with them by service providers. Providers who figure out how they realize productivity gains and find an equitable way to share them with their employees and clients will likely prosper.
The Bottom-Line: A future of elevating work beyond the mundane, learning continuously and faster, while GenAI becomes a copilot aiding in better decision-making and improving outcomes many times over.
GenAI opens the door to interrogating data differently and smartly, leading to using data (structured, unstructured, images, audio, etc.) in ways perhaps only imagined. In a future where we are going to experience an acute shortage of clinicians, GenAI, being an able aid to a clinician, will help with speed and accuracy of diagnosis, reduce administrative burden, ensure gaps in care are addressed by engaging with health consumers, and the list goes on. The sky is the limit with GenAI, and that has some extraordinary possibilities in healthcare…assuming we make the right choices and deploy GenAI against the right problems.
Posted in : Artificial Intelligence, Buyers' Sourcing Best Practices, ChatGPT, GenAI, GenAI Leaders Series, Generative Enterprise, GPT-4, Healthcare, Healthcare and Outsourcing
