The Rideshare Revolution Comes to Healthcare

How Uber and Lyft changed the NEMT conversation — and what they can't solve alone

This is Part 1 of a three-part series on the evolution of healthcare mobility. [Part 2: The Hybrid Network Era →] [Part 3: When the Driver's Seat Is Empty →]

Every year, 3.6 million Americans miss or delay medical appointments because they can't get a ride. Not because they don't want to go, or because they forgot, but because reliable transportation simply isn't available to them. For older adults and people with disabilities, the picture is even starker: adults with disabilities are three times more likely to forgo healthcare due to transportation barriers than the general population.

This isn't a convenience problem. Missed appointments cascade into worse chronic disease management, delayed diagnoses, and higher hospital readmission rates. Patients facing transportation barriers experience 24% higher 30-day readmission rates, each one costing the healthcare system tens of thousands of dollars — and costing the patient far more than that.

For decades, the answer to this problem was NEMT: non-emergency medical transportation, a mandatory Medicaid benefit since 1966. But how that benefit gets delivered has changed dramatically over the past ten years. The catalyst? The same companies that changed how all of us get around.

The world before rideshare

Before 2016, healthcare transportation operated on a model built for predictability. A state Medicaid agency would contract with a transportation broker — large national firms like Modivcare or MTM — who would subcontract to local providers. Trips were scheduled 24 to 72 hours in advance. A dispatcher would assign a vehicle. The vehicle would show up — or sometimes wouldn't.

The system had real strengths: it could handle wheelchair-accessible vehicles, stretcher vans, and bariatric transport. It could manage recurring routes for dialysis patients or day program participants. It understood that healthcare transportation often means more than just a sedan.

But it also had deep limitations. Wait times averaged 12.5 minutes on paper but often ran much longer. Same-day and on-demand capacity was scarce. Per-trip costs for routine ambulatory rides ran north of $31 per trip. And missed appointment rates across the NEMT industry hovered around 25% — a number that should trouble anyone who cares about health outcomes.

The system was designed to schedule rides, not to respond to need. And for the growing number of healthcare programs trying to serve patients more flexibly, that gap was becoming untenable.

Enter the TNCs

In 2016, Lyft became the first major transportation network company to formalize a healthcare vertical, offering healthcare organizations a way to book rides for patients through a dedicated platform. Two years later, in March 2018, Uber launched Uber Health — a HIPAA-compliant B2B service that had over 100 healthcare organizations in beta at launch, including NYU Perlmutter Cancer Center and Yale New Haven Health. By the end of 2019, Uber Health reported 400% year-on-year growth.

Health plans and state Medicaid programs took notice. In 2019, UnitedHealthcare and Amerigroup Tennessee partnered with Lyft through TennCare on a pilot that routed Medicaid beneficiaries to healthcare appointments via rideshare. Within three months, primary care visits among participants had increased 13%. Aetna launched its own Lyft partnership in 2020. By that year, roughly 15 states had authorized or were in the process of authorizing TNCs as Medicaid NEMT providers. Meanwhile, the 2018 CHRONIC Care Act gave Medicare Advantage plans new flexibility to offer transportation as a supplemental benefit, further opening the door.

The early results were hard to argue with. CareMore's Lyft pilot cut per-ride costs 32.4% — from $31.54 to $21.32 — and saved the organization an estimated $1 million annually. Wait times dropped 30%, from 12.5 minutes to 8.8. Programs using rideshare reported missed-visit rates of just 8% compared to the industry average of 25%. An Arizona pilot saw primary care gaps decrease 50% and physician visits increase 40%. Multiple state programs achieved 99% on-time arrival rates with 7-minute average wait times, compared to 28 minutes for traditional NEMT.

For a healthy, mobile adult who needs a sedan to get to a primary care appointment, rideshare was a genuine breakthrough: faster, cheaper, more reliable, and a vastly better experience.

What rideshare can't do

But healthcare mobility serves populations whose needs go far beyond a sedan and a smartphone notification. And here, the TNC model runs into fundamental limitations.

The accessibility gap is the most visible. Standard TNC drivers have a basic background check and a valid license. NEMT drivers typically carry HIPAA certification, CPR and First Aid training, ADA compliance training, and are subject to multi-panel drug screening and ongoing testing. TNCs offer limited service for wheelchair users and essentially no capacity for stretcher transport, bariatric needs, or behavioral health accommodations. One healthcare pilot found that 8% of participating Medicaid patients required wheelchair-accessible rides that the TNC partner couldn't provide.

Beyond physical accessibility, there's a whole layer of care that disappears in the TNC model. There's no wait-and-return capability for appointments of unpredictable length. There's no same-driver consistency for passengers who depend on routine and trust — a critical need for many people with intellectual and developmental disabilities. There's no elopement monitoring for passengers at flight risk, no seizure protocol, no capacity to observe and report health changes to a care team.

Veyo, a dedicated NEMT technology company, estimates that roughly 85% of ambulatory rides require service levels above what a standard TNC can offer, once you account for HIPAA, ADA, and federal healthcare regulations.

The regulatory landscape reflects this tension. The FTA's 2016 "Dear Colleague" letter reminded transit agencies of their ADA obligations to provide equivalent service for wheelchair users when integrating TNCs into paratransit. States have taken varied approaches: North Carolina authorizes TNCs only for recovery trips, while California has established specific Medi-Cal enrollment requirements for TNC providers.

The TNC revolution was real — but it solved the straightforward trips. For the populations with the greatest transportation needs and the most to lose from a mismatch between their needs and their ride, the rideshare model alone isn't enough.

What the pandemic revealed

Then COVID-19 arrived and stress-tested everything.

NEMT utilization dropped from 3.9 million Medicaid beneficiaries in 2019 to 3.3 million in 2021. Monthly ride days remained 30% below pre-pandemic levels through 2021. Driver shortages forced agencies to cut trips and reduce service footprints. When programs eliminated shared rides to prevent disease transmission — a necessary step — operating miles in some systems increased by as much as 70%, as in New York City's Access-A-Ride. Costs rose. Capacity shrank. And the people who depended most on these services were the most exposed.

States scrambled to adapt. Indiana fast-tracked TNC provider enrollment starting in August 2020 to fill gaps in its NEMT network. Programs that had already diversified their transportation supply — blending dedicated fleets with TNCs, volunteer drivers, and multiple contracted NEMT providers — proved far more resilient than those reliant on a single source.

The pandemic didn't just disrupt healthcare mobility. It forced a fundamental rethinking of how transportation networks are built. The organizations that weathered the crisis best weren't the ones with the biggest fleet or the cheapest per-trip rate. They were the ones who had learned to assemble and manage a network — drawing on multiple types of providers, matching the right resource to the right trip, and adapting when conditions changed overnight.

That model — the hybrid network — is now the direction the entire industry is moving. And building it well turns out to be much harder than it sounds.

Next in this series: [Part 2: The Hybrid Network Era →] — Why the future of healthcare mobility is about orchestration, not just vehicles.