With patient access top of mind for execs, how can AI help?

Healthcare systems can’t get very far without their patients, yet many struggle to ensure that their communities have access to necessary care in a timely and affordable manner. The state of patient access across the US is grim, with one recent scorecard from the PAN Foundation giving the nation a solid C on its ability to adequately fulfil the needs of patients (without sending them into financial ruin).  

In that survey, close to half (48%) of participants reported logistical barriers to care, such as trouble getting appointments and difficulty with health-related finances. The struggles increased for people in traditionally underserved communities, with about 40% saying that some aspect of their identity contributed to their inability to get the best possible care. 

While access woes peaked during COVID and are now slightly easing in many areas of the country, healthcare leaders are generally aware that patients are still facing enormous barriers to getting into health systems in the first place, let alone navigating the overwhelming complexity of the typical care journey. 

In one recent poll by the Healthcare Financial Management Association (HFMA) and Guidehouse, 45% of executive respondents selected patient access as their second-highest priority for investment, just behind artificial intelligence tools for automating operational and financial tasks within the revenue cycle. 

And The Center for Connected Medicine, in conjunction with KLAS Research, similarly found that leaders are laser focused on using technology, including AI, to reduce patient access concerns while improving the quality of care for those who can actually get into the system. 

There’s no question that there’s something to this idea of pairing AI with efforts to improve patient access. With its ability to ingest and analyze massive volumes of information, AI has the potential to uncover hitherto unseen patterns in patient flow, predict trends in peak utilization, and optimize resources to ensure efficiency in scheduling and staffing. 

How can AI tools make it easier and more affordable for patients to receive needed care? 

Optimizing scheduling in the clinic and the operating room

Just a few years ago, a lack of insight into OR scheduling left WakeMed Health and Hospitals in North Carolina with chunks of unused time scattered throughout the day, frustrating surgeons, patients, and financial leaders alike.  

Bringing an AI platform into the equation helped leadership visualize the bottlenecks in the scheduling process, including personal preferences among surgeons that were contributing to the problem.  

As a result, on-time surgery starts increased from an average of 70% to 92%, and the health system has increased overall capacity by 22%, or an additional 60 hours of time per work week. 

Other health systems are actively engaged in using AI to reduce no-shows via behavior prediction and automated reminders, get to grips with fragmented scheduling systems that leave holes in the calendar throughout the day, and use smart chatbots or other AI-driven tools to make it easier for patients to schedule appointments online at their convenience rather than going back and forth with a front desk employee. 

Supporting the hospital at home to expand access to inpatient-level care

COVID showed just how easily the inpatient setting can be overwhelmed with high-needs patients, and how vital it is to have an alternative for individuals who need enhance monitoring but are stable enough to be safe outside of the hospital setting. 

Hence the rise of the hospital at home: a strategy that relies on the most advanced remote patient monitoring and communication technologies to create a hospital-like atmosphere in a patient’s own residence, freeing up hospital beds for those most in need.  

AI is crucial to the success of these well-received programs, which have equally captured the attention of patients and health system leaders. Smart monitors and virtual visit technologies, powered by AI, create the right conditions for reducing costs without sacrificing on outcomes. 

Recently, Johns Hopkins reported cost savings of up to 30% compared to traditional inpatient care, while Milliman has projected potential savings of up to 50%. Some large health systems, like Mass General Brigham and Atrium Health, anticipate freeing up 10% of inpatient beds by enrolling suitable patients in a hospital at home program. 

Connecting patients with scarce mental healthcare resources 

Mental and behavioral healthcare are among the hardest types of care to find, with a huge number of therapists reporting long waitlists for critical services.  AI-driven virtual care is stepping in to help. 

From platforms that use AI to transcribe and manage therapy notes to basic AI chatbots offering assistance with anxiety or depression to fully immersive virtual reality settings with large language models for therapists, leaders in the space are exploring a variety of ways to ease bottlenecks in access for the millions of people in need of support. 

It will be especially interesting to watch how generative AI will be used in this space to augment – or even attempt to replace – human therapists, which could dramatically expand access, but may have significant implications to the way people build intimate therapeutic relationships with other flesh-and-blood beings. 

Communicating with patients outside of traditional appointments

Expanding provider capacity is key to improving patient access, and AI has a role to play in freeing up physician bandwidth. One striking example is a new GenAI feature within the Epic Systems EMR platform that lets providers use GPT-4 to draft replies to patients’ digital messages. 

A study from UC San Diego Health found that this strategy reduced cognitive fatigue among providers who tried the feature, because it allowed them to edit or add to an existing message instead of having to create the communication from scratch.  

While the study didn’t find any time savings associated with the strategy yet, a less fatigued provider may be more able to work efficiently with patients, which in turn improves the quality of care, reduces the likelihood of care gaps, and potentially avoids the need for additional services down the line. 

It seems likely that providers will take advantage of these types of tools in the near future to communicate a larger number of patients than they could with manual techniques, essentially multiplying their bandwidth and speeding up conversations so patients can quickly take the next steps in their care. 

Artificial intelligence still has a long way to go before it can truly solve for the many patient access problems plaguing the industry, but these and other use cases show that there is exciting promise for integrating automation into broader solutions to make it easier, more affordable, and more timely to get access to needed care.  

Jennifer Bresnick is a journalist and freelance content creator with a decade of experience in the health IT industry.  Her work has focused on leveraging innovative technology tools to create value, improve health equity, and achieve the promises of the learning health system.  She can be reached at [email protected].