By Dr. Benyamin Deldar and David Hanbury, co-founders, Deep Medical

Picture Helen, a 77-year-old woman who lives alone in Chicago. Her doctor scheduled a routine eye exam for diabetic complications, but Helen ended up missing her appointment. Helen noticed a shadow in her vision that began to grow, so she booked another appointment. But, by the time she was seen, it was too late–the damage to her eyesight had progressed.

Helen’s case is just one of millions – a stark reminder of a widespread issue that continues to harm lives across the US and beyond.

A system that doesn’t follow up

Each year, millions of Americans miss scheduled healthcare appointments. Some forget. Some face transport issues. Some, like Helen, are let down by systems that assume a single reminder is enough.

This isn’t just an administrative headache. It’s a healthcare crisis hiding in plain sight.

One in four patients in the U.S. miss a medical appointment annually. That figure jumps to nearly 40 percent for Medicaid patients, according to NIH data. The result? NIH-supported research estimates this leads to $150 billion in wasted clinical resources each year – and lives put at risk.

Because missed appointments don’t just delay care – they often mean the difference between early intervention and a critical, even fatal, diagnosis.

For patients managing chronic illnesses or mental health conditions, the risks are severe. A BMC Health Services Research study found that missing two consecutive appointments can increase the likelihood of death within a year by up to eight times. For healthcare providers, especially those serving underserved communities, the stakes couldn’t be higher.

Automation as a starting point 

Healthcare doesn’t happen in a vacuum. Patients juggle jobs, children, caring responsibilities, transportation hurdles and, in some cases, literacy or language barriers.

A one-off reminder might not cut it. As Helen’s case shows, one missed message can have devastating consequences, which is where AI comes in.

Today’s most advanced platforms don’t just automate reminders; they can anticipate the barriers. Using a combination of behavioral data, historic trends and operational insights, AI can now predict with up to 87 percent accuracy whether a patient is likely to attend their appointment.

But prediction is only half the battle.

Personalized outreach, smarter systems

When AI is used thoughtfully, it can tailor follow-up based on individual needs. That could mean sending a text, and following up with a call. These aren’t radical intervention, but they’re the difference between someone getting care or slipping through the cracks.

In clinics using intelligent outreach platforms, no-show rates are falling. Not because reminders have increased, but because they’ve improved. Outreach is more timely, more relevant, and better aligned with the realities of patients’ lives.

This is especially powerful for community clinics and safety-net providers who face the highest no-show rates and the most stretched resources.

Bridging the gap between insight and action

We often talk about AI in healthcare as a diagnostic tool, but its greatest value might lie in its ability to listen and respond.

The most effective systems don’t just flag at-risk patients. They adapt. They act, and they create space for human intervention, at scale. Offering rescheduling that accommodates work or caregiving responsibilities can make all the difference. 

Helen just needed a timely call to remind her about her appointment. That kind of support is entirely achievable if systems are built to identify needs and respond accordingly.

Healthcare access in the U.S. is already fraught with inequality. Smarter technology won’t fix that on its own. But it can help clinics use their time better, reach patients more effectively, and reduce the number of critical health issues that could have been prevented.

The cost of a missed appointment isn’t just measured in dollars. It’s measured in deteriorating health, lost quality of life, and lives cut short. With AI-led systems that understand when, where, and how to intervene, we can close that gap.

About Dr. Benyamin Deldar MD 

Founder & Co-Chief Executive Officer 

Dr. Benyamin Deldar is the Co-CEO and business development leader at Deep Medical. His background is in medical imaging and neurointervention. He’s worked at a number of high profile institutions including Guys and St. Thomas Hospital, Epsom and St. Heliers NHS Trust and a research fellowship at Johns Hopkins University, where he co-authored the Hopkins manual of Neurointervention. He also holds a distinction in radiology from Harvard Medical School. 

About David Hanbury MEng MSc

Founder & Co-Chief Executive Officer

David Hanbury is the Co-CEO and Head of Machine Learning at Deep Medical. He’s made a huge contribution to the field of AI, winning the prestigious BP prize for best performance in Engineering and the predictive auto-ML system he developed is currently used by the United Nations’ Industrial Development Organisation (UNIDO). David Hanbury’s dedication to advancing Machine Learning (ML) prediction spans over a decade. Before starting Deep Medical, David Hanbury was the CEO and founder of OptimalAI, helping businesses integrate AI into their infrastructure, and he’s been involved in several healthcare startups including ChAI Predict and Hummingbird Technologies. David Hanbury worked as a fund manager for JP Morgan at the start of his career. He’s also an honorary staff member of the Computer Science department at UCL.