There is a clinic owner in Oklahoma who, until recently, had not taken a vacation in five or six years. Not because she lacked the desire, or the means, but because if she stepped away from her desk—even briefly—payroll might not make it to her employees. She was cashing checks by hand. The revenue cycle of her autism therapy practice, that sprawling apparatus connecting rendered services to actual dollars in the bank, depended entirely on her physical presence. The system, such as it was, could not survive her absence.
Nathan Lee, who co-founded Camber with a vision to fix healthcare reimbursement, tells this story not as an exceptional case but as an illustration of a broader pathology. The American healthcare system, he observes, is perhaps the only industry in which you can deliver a service and have no idea whether you’ll actually get paid for it.
For providers of Applied Behavior Analysis therapy—the evidence-based treatment that has become the standard of care for children with autism—this uncertainty is particularly acute. The billing is uniquely complex, requiring weekly submissions across multiple service locations, ongoing reauthorizations, and navigation of payer-specific requirements that vary not only by state but by individual insurance company.
Enter Camber
Camber, which emerged from stealth mode in early 2025 with $50 million in funding from Andreessen Horowitz, Craft Ventures, and Y Combinator (amongst others), is attempting to transform this chaos into something resembling order. Strip away the jargon, and the question they’re asking is obvious: what if the process of getting paid for healthcare services worked more like swiping a credit card at a coffee shop? The company now processes over $2.5 billion in annual claims and serves 90,000 patients across 42 states, from small owner-operator clinics to national providers with thousands of clinicians.
The founding team emerged from Harvard (Lee and Celina Qi), Columbia (Christophe Rimann), and Tufts (Jade Chan), eventually crossing paths through consulting work in San Francisco. The quartet launched Camber in late 2020, before the generative AI boom, built on a thesis about predictive modeling: rather than submitting claims and scrambling to work through denials after the fact, they would build systems sophisticated enough to know, before submission, whether a claim would be paid.
Years later, back in the “Golden City” for the 2026 JPM Healthcare Conference, Lee sat down with Acuity Media Network and described Camber’s mission with the measured precision of someone who has spent years thinking about supply chains and data flows. “Providers don’t start their clinics because they want to be experts in finance or business,” Lee told me. “Camber automates the entire process of making sure they get paid, so they can focus on delivering care.”
Traditional billing operations see first-pass payment rates—the percentage of claims paid on the first submission—as low as 70 cents on the dollar. Best-in-class organizations might achieve 95 to 98 percent, but often only after months of reworking denied claims. Meanwhile, Camber reports a first-pass pay rate above 95 percent, typically within 14 to 30 days. For a small clinic operating on a payroll-to-payroll basis, the difference between waiting 18 months to collect 80 percent of revenue and collecting 95 percent within a month is existential.
Consider a case Lee shared after our interview: a provider whose accountant called in December, puzzled. “What did you do differently this year?” the accountant asked. “You have way more cash in the bank—about $84,000 more than usual.” The answer was Camber. With his previous biller, the provider had been getting around an 80 percent first-pass pay rate with an 85 percent collection rate. After switching, those numbers jumped to 98.5% and 98.8%, respectively.
A Fragile Architecture
The architecture of American healthcare billing is, by any reasonable standard, absurd. There are clearinghouses—essentially electronic post offices that route claims between providers and payers—and there are portals, each with its own idiosyncrasies, many of them operated by local funders insufficiently sophisticated to integrate with modern systems. There are authorization requirements that differ by payer and by state. Coding rules change quarterly. And there is, always, the phone: billing staff spending hours waiting for someone at an insurance company to pick up, to resolve a claim denial that might have been caused by a single misplaced modifier.
When the Change Healthcare cyberattack struck in February 2024, this fragile ecosystem came close to collapse. Change Healthcare, a subsidiary of UnitedHealth Group, processes 15 billion healthcare transactions annually, touching one in every three patient records. The ransomware attack by the Russian group ALPHV BlackCat encrypted significant portions of the company’s functionality, effectively shutting down the medical claims clearing process for a substantial portion of the American healthcare sector. Nearly 94 percent of hospitals reported financial impact. One-third reported that the attack disrupted more than half of their revenue. Small practices—already operating at thin margins—faced the prospect of bankruptcy.
“Our technical teams are nimble, and the infrastructure we’ve built is flexible enough that something like rerouting claims to a different clearinghouse can be done overnight,” Lee said of Camber’s response. The incident illustrated both the vulnerability of concentrated infrastructure and the potential value of systems designed with redundancy and adaptability.
For ABA providers specifically, the billing challenges are compounded by the nature of the treatment itself. A child receiving therapy might need services several times per week for months or years, with authorization requirements that must be continually renewed. Initial claim denial rates for ABA therapy can range from 15 to 30 percent, significantly higher than other healthcare services. The specialized CPT codes, place-of-service variations, and credential requirements create a unique challenge that generalist billing companies frequently mishandle.
Technology First
Camber’s approach inverts the traditional revenue cycle management model. Where legacy operations begin with human teams and augment them with technology, Camber built technology infrastructure first and supplements it with humans only for genuinely complex situations. The company’s composition reflects this priority: the team is heavily technical, mostly engineering-focused. The founders spent three years building before going to market.
“We care deeply about human judgment to ensure things are done right,” Lee acknowledged. “Every decision facilitated by the infrastructure has a human in the loop.” But the human role is supervisory rather than operational. The goal is a system that, like payment processing at a point-of-sale terminal, simply works—invisibly, reliably, at scale.
The data advantages compound over time. With $2.5 billion in claims flowing through the platform, Camber has accumulated what Lee describes as one of the world’s largest claims databases across specialties, including ABA, speech-language pathology, occupational therapy, and physical therapy. This corpus powers predictive models that can identify, before submission, which claims are likely to be denied and why.
“In a small billing organization—or even a large one relying on humans—all the knowledge is stuck inside people’s brains,” Lee said. “We talk with billing teams at large national organizations, and it’s always the same story: ‘That person has 20 years of experience, they know everything, and if they leave, the whole operation is in danger.’”
The platform promises to make such institutional knowledge scalable and permanent. A pattern identified in Nevada becomes, almost immediately, a pattern the system can recognize and account for in Texas. A change in TRICARE’s authorization requirements gets incorporated into the predictive model. The intelligence, as Lee puts it, is always on
Fragmentation as Foundation
The ABA industry itself is relatively young by healthcare standards. Commercial insurance reimbursement for autism services is largely a development of the past two decades, the result of advocacy by organizations like Autism Speaks and the Council of Autism Service Providers. All 50 states now have some form of autism insurance coverage mandate, though the specifics vary considerably. The result has been rapid growth: practices that started as small local operations have scaled to national reach, often without corresponding sophistication in their back-office operations.
Lee has described this landscape with a phrase that initially seems counterintuitive: “Fragmentation isn’t a flaw—it’s the foundation.” What he means is that the very messiness of the market—thousands of independent clinics, each with its own systems and payer relationships—creates the opportunity for infrastructure that can serve them all. “Independent clinics feel the most pain—and move fast to fix it,” he said. “They’re led by clinician-owners: mission-driven, deeply committed to care, but not always set up to run a business.”
“Many of these practices grew to a national scale incredibly quickly, without the depth of experience or technical know-how to operate at that level,” Lee observed. Some of these organizations—even those with hundreds of millions in top-line revenue—operate at the edge of profitability, making decisions about whether to withdraw from entire states because their collection rates cannot cover the cost of service delivery.
There are larger providers using Camber’s platform who have, Lee reported, integrated acquisitions in less than 30 days—a process that traditionally requires months of operational transition. “Nine-figure providers are acquiring eight-figure providers and turnkeying them,” he said. The system functions as plug-and-play infrastructure: connect to an existing practice management system or electronic health record, and the billing automation simply begins.
For smaller providers, the promise is different but equally significant: access to the same technological capabilities that were once available only to organizations operating at massive scale. The clinic owner in Oklahoma who couldn’t take a vacation, the practitioner in a small town who started their practice because a family member was diagnosed with autism—these operators can now, theoretically, compete on operational efficiency with national chains.
The Case for Specialization
Lee is careful to position Camber not as a competitor to practice management systems like Central Reach or Rethink but as a specialized complement. Most practice management platforms, he noted, were built by clinicians to serve clinical needs; financial optimization was secondary. “Most practice management systems were not built to be financial systems,” he said. “What we do best—what we’ve spent five years building—is purely financial revenue cycle.” Billing is hard enough to deserve dedicated infrastructure, not afterthought features bolted onto clinical software. Whether existing players in the space agree with this assessment remains to be seen.
What is harder to dispute is the fundamental dysfunction that creates the market opportunity. The Change Healthcare breach affected an estimated 192.7 million individuals—nearly two-thirds of the American population. The total response cost exceeded $2.4 billion. Practices that had done nothing wrong found themselves unable to make payroll after a third-party clearinghouse was compromised. The architecture of American healthcare billing contains single points of failure that, when they fail, cascade through the entire system.
“Providers large and small tell us they’re not stressed about this anymore—they know it just gets done,” Lee said. “That’s the goal. You shouldn’t have to think about billing. You didn’t start practicing for that.”
Keeping the Lights On
There is, in Lee’s telling, a tension in autism care between business and mission that need not exist. “Delivering high-quality care responsibly and sustainably, while treating your organization as a business—these need to go hand in hand,” he said. A practice that cannot make payroll cannot retain clinicians; a practice that cannot retain clinicians cannot maintain continuity of care. A child who might need years of therapy requires a provider stable enough to be there for years. “Business is not a dirty word here,” Lee argued. “It’s the ability to keep the lights on in a way that lets you do right by your client families and the clinicians on your payroll.”
Whether Camber represents the beginning of a broader transformation in healthcare operations or merely a successful niche play depends on factors well beyond any single company’s control: the regulatory environment, the behavior of payers, and the willingness of providers to trust critical infrastructure to external platforms. But the underlying premise seems difficult to dispute. Healthcare providers serving vulnerable populations should not be spending their energy on the baroque complexities of insurance billing. The administrative burden is not simply inefficient; it actively harms the mission it purports to support.
“With AI, I think a lot of executives know there’s a version of their organization that looks very different from what they have today,” Lee said. “We’re here to say there are partners who can help you get there. Start now and make that investment.”
