The death of data and analytics is nigh. Long live data and analytics. 

2023 will mark a significant turning point in data science, data and analytics and how an organization uses its data for competitive advantage and improvement. We are no longer at the foothills, we are facing the sheer mountain face, planning our vertical assent and finding our first handholds. Generative artificial intelligence (AI) has been an active area of research since the 1960s, born from the Massachusetts Institute of Technology – those were some of the first times a human could have a conversation with a machine. Only in the last six to twelve months have we seen a global acceleration of understanding and use of generative AI capabilities through large language models producing tools, such as ChatGPT (OpenAI/Microsoft) and Bard (Google). Our initial collective dip in the pool included playful prompts that wrote “Dr. Seuss poems” and quickly challenged students and academia in essay writing. We are now moving toward the deep end and truly exploring where value lives. In healthcare, this will introduce data-driven processes that will, among others: 

• improve clinical workflows;
• improve patient experience and support health system navigation;
• expand the use and effectiveness of digital therapeutics;
• discover drugs faster and cheaper; and
• broaden the reach of expert advice and diagnostics to rural and low-resource environments.

In this innovative time, we are also seeing an emergence of voices around regulation, oversight and adoption of responsible AI practices and principles. It may all appear jarring and cause panic for some. It may seem extreme and may appear to collectively slow us down. It comes, however, with the territory when transformative technology is discovered and becomes scalable. 

As Winston Churchill once said, “Where there is great power, there is great responsibility.”  As we journey through the discovery of generative AI capabilities and value, we will also need to regulate and provide oversight. Given the nature of the technology and the impact it will have on the people we serve, this will fall on all of us in healthcare to ensure that we are continually reminding ourselves that “the needs of the patients come first” and “do no harm”. Our North Star helps us keep our goal in perspective and the reason why we exist.

As things settle around generative AI, regulations will find their groove, disrupted industries will make headline news touting efficiency, convenience and new customer experiences. New industries will emerge – industries that coordinate services, people and work in ways we could never imagine. In healthcare, we may still be looking through the windowpane of desire. The well-known window shows us the exciting potential of innovative technology with the longing for adoption and diffusion into scalable value for the patient and the provider. Will the global adoption of generative AI break the curse of lag in the adoption of technology in healthcare? Will healthcare experience rapid advancement in innovation in lockstep with other industries? Time will tell. However there are three key areas of focus that, regardless of the speed of adoption, healthcare will need to advance to generate value from this disruptive technology (see Figure 1).

Figure 1. Three key areas of focus for data and analytics as we move into generative AI for healthcare

Patient and provider experience: There is no better time to apply Deming’s system of profound knowledge (SoPK). Deming introduced us to a system that taught us to understand the system, the variation in it, the human psychology within it and the knowledge produced by it. As we journey into the world of generative AI and other machine-driven collaborative technology, we must understand the experience of all the players, how they relate to the entire process and the potential variation (bias) that may exist. Being able to profoundly see and understand patient and provider experience will be at the core of data and analytics in healthcare – it will become the fuel that lights our North Star.

Adaptive processes: The days of mapping out a process diagram and generate swim lane diagrams are short-lived – it is near impossible to design a process map for a generative AI chatbot. We will see emerging dynamic processes (perhaps self-designed) that are rarely the same twice and focused on emerging patterns rather than pre-defined steps. All our historical understanding of “getting work done” will be disrupted. Our focus will need to become milestone driven and ultimately be guided by experience and measured by outcomes and results. Data and analytics will guide healthcare organizations in the transition from thinking in linear processes to applying adaptive processes and this guidance will be core to this transformation.

Outcomes: Our North Star remains the same: the needs of the patients come first. Our ability to ensure timely, high-quality care for all in need will become our collective challenge. The risk of a healthcare divide has never been more present, where those who are digitally savvy or can access digital pipelines to healthcare organizations are positioned to ride this transformative wave. Moreover, those who digitally contribute will bend and shape datasets to increasingly focus generative AI and algorithms based on their needs, leaving behind those underserved and increasingly marginalized individuals whose struggles will intensify as they seek care. Therefore, measuring and understanding health outcomes across populations will emerge as our primary objective. Helping our organizations see the importance of unbiased engagement, ethics, data management and equitably deploying solutions that fairly serve our patients will be job number one.

We are entering a time of uncertainty. We are all waiting to see how fast healthcare will be impacted by generative AI tools. As we look up the face of the mountain, it’s foggy.  We certainly cannot see the snow-covered peaks and, in some cases, we cannot see too many handholds ahead. While we know generative AI will be transformative and bring immense value to healthcare, we will need to purposefully embark on a paradigm shift. We will need to shift from a world of pre-planned care pathways and highly designed and tested processes to a world of adaptive processes, where outcomes are our goal with human experience guiding our way. It will be a data-driven world where, between human and machine, we know what we know and care is curated and delivered equitably, on demand and barrier-free.

Comments

Be the first to comment on this!