Solving the Patient Experience Dilemma through AI
Patient experience has been a central focus for hospitals for more than 20 years. However, its origins and application vary wildly across health systems, and even within a single organization.
No matter how patient experience (PX) is defined or expressed, health systems have included it in various strategies to “transform” the healthcare interaction — a term that implies something of an endgame or a perfect state of experience.
But, despite our best efforts, evolving consumer expectations continuously move the goalposts.
How consumerism raised the bar for healthcare
Consumer expectations have changed dramatically in the last 20 years. Think back to 2003. “Amazon” was either a rainforest or a novel online bookseller. “Uber” was just a German word. Blockbuster was nearing its peak, with 9,000 stores globally and nearly $6B in revenue.
And, hospitals were the last place people expected to have a concierge-like service.
As consumer expectations have shifted, hospitals have had a difficult time keeping up. The on-demand and personalized experiences of online shopping, gig services, and streaming media have proven more difficult and nebulous to adapt to a quality-driven, methodical industry like healthcare.
While there has been significant progress in areas like virtual care, online booking, and improved digital communications, there continues to be a large and growing gap between healthcare and the products / services offered by other industries.
Payors expect more, too
Not only is the consumer expecting more, but the payors are too. Satisfaction surveys and patient-reported experiences are impacting reimbursement rates and provider network composition.
This matters not only to hospitals’ bottom lines. “Patient experience” now means more than patient loyalty, community reputation, and higher reimbursement. Payors have made the connection between experience and clinical outcomes as well.
In speaking with PX leaders across the country, they see a way to leapfrog some of the gaps in technology prevalent in hospitals. I often hear about the pursuit of a truly personalized patient experience, and how personalization may be a path to meet evolving consumer expectations.
How could this become a reality when hospitals have traditionally been slow to respond? Two words: Artificial intelligence (AI).
More specifically, advanced AI. And by that I mean things like large language models (LLMs) and natural language processing (NLP) applications like those used in Chat GPT which allow patients to engage with their health in a more conversational, contextualized, and humanistic manner.
More on advanced AI later. First, let’s look at the unique challenges faced by hospitals.
A problem space for hospitals
The optimal patient experience is one where the patient and their family feel engaged and in control of their own health. Primary care doctors and family physicians can attest to this. They see it every day in patients they’ve known for years.
In the hospital, patient engagement can be more challenging. Whether it’s an unexpected trip to the emergency department (ED) or an inpatient stay following a planned surgery, this care environment can be particularly overwhelming and stressful.
For most patients, being in the hospital means a loss of control. What’s wrong with me or my loved one? What happens next? What should I do? What resources are available to me? These questions and more are part and parcel of the experience. (If you work in pediatrics, read my colleague’s empathetic and moving article on the additional challenges faced by children’s hospitals.)
Not all technology is adopted
To address the patient experience in hospitals, more and more health systems are turning to technology as a core strategy for better engagement. Patient utilization rates of those solutions, however, have been varied, often preventing the expected return on investment from being fully realized.
Why is this the case? Even with all the various strategies in place — including the use of technology — I believe there are three themes hospitals must address when evaluating solutions in 2023 and beyond:
Complex patient experiences are not conducive to a one-size-fits-all approach
Most solutions are not personalized enough to be relevant
Patients still do not feel involved or that they have control over their own health care journey.
So, where do we go from here in providing the optimal patient experience?
Enter AI
I firmly believe that incorporating advanced AI into care delivery and care management is the next step hospitals must take in our industry’s ongoing efforts to improve PX. And I’m not the only one who thinks so. As Thomas Davenport and Ravi Kalakota explain in their excellent Future Healthcare Journal article:
Patient engagement and adherence has long been seen as the ‘last mile’ problem of healthcare — the final barrier between ineffective and good health outcomes. The more patients proactively participate in their own well-being and care, the better the outcomes — utilization, financial outcomes and member experience. These factors are increasingly being addressed by big data and AI.
Finally, forward-thinking providers are starting to embrace consumerism in a similar fashion to the Amazons, Netflixs, and Ubers of the world.
In pursuit of personalization
If there’s anything that my 20+ years in this field have taught me is that the care experience must be highly personalized and highly relevant for patients to engage. The patient’s experience must be created precisely based on what’s known about them, both clinically and personally. In short, the patient experience must be based on precision engagement.
When done right, AI in patient-facing technologies do exactly that. Nivedita Nayak could not have said it better in her Classic Informatics article.
One of the most significant ways artificial intelligence is transforming healthcare is by improving the patient experience. Patients are now able to receive personalized care that is tailored to their individual needs, thanks to AI. The technology is helpful to predict patient needs, identify potential health risks, and even personalize treatment plans. This means that patients receive more personalized care, which can lead to better outcomes and a more positive experience overall.
Enhancing PX at scale
Like many of my colleagues who focus on patient experience, I’m proud of my background in public health. We know how important data is for addressing population health — an increasingly important objective for value-based care models.
Today’s advanced AI — in particular LLMs and NLP — uses data to address a population of one. When trained right, they crunch dozens or even hundreds of data points to create a highly personalized patient story that guides care plans, automates interventions, and responds to patient data in real time.
That story — the one that PCPs and family physicians build over months and years with their patients — can be ascertained in mere seconds by LLMs combing through clinical data, provider notes, test results, patient-provided data, and more, much of which is already housed in your electronic health records (EHR) system.
For the hospital, this is a significant advantage considering that their contact with each patient is largely episodic. (You can have an appendectomy exactly once.) But for the patient, advanced AI is a powerful ally in delivering the kind of patient experience that influences engagement all along the healthcare journey of each and every patient, and thousands more like them.
How’s that for population health?
Applying AI to the most vital points in care
Patients engage in their healthcare journey in different ways. A common theme, however, is that the journey is often complex and unfamiliar.
Information, instructions, and interpreted results are hard to understand. This is why the vast majority of patient experiences are imprecise and not relevant enough to prompt true patient engagement.
Since joining Vital in 2022, I’ve had the privilege to work with data scientists who are making great strides in helping hospitals advance their efforts to improve the patient experience. These innovators design and build machine learning models to simplify the commonly complex care process, create greater understanding of what’s important for the patient to know, and personalize the interaction with the patient based on what’s known about them.
Each day, I am beyond impressed with our team’s ability to create highly personalized and therefore, relevant experiences for the patient and their family. Whether it’s:
Informing and managing expectations about wait times;
Automating precise patient education recommendations;
Translating physician notes into plain language;
Summarizing test results;
Keeping the patient informed and engaged with their daily schedule;
Informing the patient and their family about their health progress and intelligently setting goals for the patient to improve their status and outcomes;
Helping patients make service requests with just a tap on their mobile device; or
Simplifying discharge instructions for patients leaving the hospital.
The promise of a precise, personalized patient experience
Because these models are trained on data elements from millions of healthcare encounters each year, the precision of interventions that rely on each patient’s data is superior. So, when used to support and augment the care process, Vital’s software solutions create that precision engagement needed to take the next steps in PX.
Further, because the content and interventions provided to the patient are so relevant, we see very high patient utilization and engagement. This results in better patient outcomes and significant improvements in patient satisfaction. The ROI is significant and demonstrable.
PX + AI = true healthcare transformation
Amidst the current conversation and debate about the fitness and efficacy of AI in care delivery, there’s clear evidence that artificial intelligence, especially LLMs and NLP, can be used quite effectively to improve and even transform the care experience.
“Transform.” There’s that word again. With consumerism advancing so quickly in other industries, I guess it’s not surprising that this word continues to come up in healthcare.
While I don’t think hospitals are ever going to go the way of Blockbuster, competition remains fierce. Those who embrace AI to deliver a better patient experience will have the advantage moving forward.
More importantly, it’s fundamental to our shared purpose—the reasons many of us got into health care in the first place. We like caring for others. And AI can help.
Learn more
Vital is radically improving patient experience with software that gives more control, clarity, and predictability to emergency department visits and hospital stays. Using AI, Vital moves complex health record data into easy-to-use, highly personalized interfaces that inform and engage over one million patients per year. Find out how Vital’s AI-powered solutions can help you serve patients and drive revenue. .