Flagler Health
Flagler Health sells an AI operating system for musculoskeletal practices that runs inside the electronic medical record the practice already uses. Revenue cycle tools handle eligibility, prior authorization, coding, billing and denials, with a claimed 99% clean claim rate. An eFax agent works each inbound referral through to scheduling, and the AI Call Center acts as a 24 hour front desk for intake, appointments and reminders. Patient Engagement reads the chart to flag patients eligible for procedures and other care and brings them back in, while Care Management follows patients between visits through remote therapeutic monitoring, collecting pain, mood, sleep and mobility reports with a bilingual team behind it. An analytics dashboard sits across the lot.
Practices can start with one product and add others with no new login, and Flagler connects to athenahealth, Epic, MEDITECH, AdvancedMD, eClinicalWorks, NextGen, ModMed, Prompt and TruBridge. Contracts are performance based, with targets agreed before implementation, no upfront cost and results promised within 60 days. The company counts more than 10,000 providers and 100 partners and reports $164,000 in added revenue per provider on average. At Coastal Health, interventional pain physician Christopher Netzel uses it to track patient outcomes remotely.
CEO Albert Katz, a former CFO of Spine and Wellness Centers of America, founded the New York company in 2022 with chief medical officer Leon Anijar and chief technology officer Will Hu. Bessemer Venture Partners led its $50 million Series B in August 2026, bringing total funding to $63 million.
Capability Axes
An AI Health Index grade measures what a buyer can verify from public sources on the date shown. It is not a rating of how good the product is. A vendor can build an excellent system and grade low on an axis because it publishes nothing an outsider can check. How grades read
Models drive several core modules. Patient Engagement reads records, imaging reports, billing codes and the practice's own patterns to flag patients who could benefit from a procedure or service, the eFax agent turns each inbound referral into a scheduled visit, and the AI Call Center answers patients around the clock for intake and appointments. Prior authorization, coding and billing also run on record data with AI assistance.
Around that software sits work people do: a bilingual care team for follow up between visits, implementation staff and practice support. The learned components are the engine of the referral, call and candidate modules, while billing support, monitoring outreach and analytics would carry value with less automation behind them.
The agents are sold to run on their own. The eFax agent handles each inbound referral for scheduling, the AI Call Center takes intake and appointments around the clock without putting a patient on hold, and Coastal Health's physician describes the whole as a hands free, turnkey service that runs in the background. Oversight is asserted at a few points.
In remote monitoring, providers can elect to receive summaries of patient responses, and the monitoring terms make the provider responsible for checking what is entered into the chart. A bilingual team handles patient follow up. Nothing sets an approval gate before a referral is booked, a threshold at which a call passes to staff, or a review step before a candidate flag or care suggestion reaches a patient.
The platform combines electronic medical record data, imaging reports and billing codes with each practice's history, and the company says it trained its models on musculoskeletal patient data to surface recommendations inside the record system a practice already uses. The call center, the eFax agent and the revenue cycle tools apply automation to different steps of the same patient journey, and an investor describes the platform as getting smarter with every patient it handles.
Inputs and use cases are clear. What produces the output is not: no model family, voice model or language model is named, nothing describes how models are updated or retrained as new patients arrive, and the training population behind the musculoskeletal models is not characterized.
The trust center lists the processors behind the service: Amazon Web Services for cloud, Databricks for data and analytics, Twilio for communications, Datadog for monitoring and MongoDB Atlas as a database. The remote monitoring terms allow transfers to providers or their designees for care coordination. That names the hosting, data and messaging layers, and Twilio's place on the list matters for a call center that carries patient conversations. The model layer is missing.
No inference provider is named for the recommendation, voice or fax agents, and nothing states which third party receives call audio, transcripts or referral documents for processing, or how long it keeps them.
The company reports $164,000 in added revenue per provider on average, four days of administrative work saved a month and 99% clean claims, with product figures beside them: 81% growth in same day booking from eFax referrals, up to 25% lower call center costs, 40% better patient engagement and 93% patient satisfaction in Care Management. It also says 87% of patients in its experience improved in at least one area such as pain, mood, sleep or mobility.
At Coastal Health, Christopher Netzel describes tracking outcomes remotely without a workflow change, a qualitative account rather than a measurement. None of the figures names the practices behind it or gives a baseline, cohort or window, and the homepage's independently validated badge is tied to no named study.
Remote monitoring participants can send messages, photos, videos and symptom reports, and the program terms authorize collecting, transmitting and analyzing them and sharing them with care providers or their designees. The company can enter monitoring information into the electronic health record. The website privacy policy excludes business associate data and describes what happens if protected information arrives through a site form, while clinical data runs under client agreements.
That describes the clinical data path in general terms. The questions AI raises go unanswered: how long call recordings, transcripts and referral documents are kept, whether patient data trains the musculoskeletal models the company says it built, and how an error in a patient message is handled.
Flagler identifies itself as a business associate for clinical services and puts protected data under client agreements and business associate agreements, and the homepage carries a HIPAA compliant badge. The remote monitoring consent sets out what a participant may submit, its transmission to the practice and its use in the electronic health record, and clinical customers are directed to secure channels set up under their agreements.
The marketing site runs under a separate privacy policy outside that service scope. The business associate agreement itself is not available before contracting, and its coverage across the six products, from the call center to revenue cycle, and the obligations it passes to processors such as Twilio and Databricks are settled in each contract.
The trust center lists ISO 27001:2022 and shows SOC 2 Type I and Type II entries, and the homepage displays an AICPA SOC badge. The report links open an access request, so a buyer reads the attestations only after approval. The portal also gives an overview of operational controls and the processor list covering communications, cloud, data services and monitoring. The recognized frameworks are named with their types. The attestation period, the products in scope and any exceptions sit inside reports supplied on request, and nothing states which of the six products the audited system covers.
The billing, scheduling and referral products are administrative. Two parts reach further. Patient Engagement reads chart information to flag procedure candidates and match a patient to a provider or service, and the remote monitoring terms describe sending care and wellness suggestions to patients on the practice's behalf. The monitoring terms leave diagnosis and treatment with the patient's provider, and the website privacy policy separates the public site from clinical care.
That allocation describes the relationship with the practice. No FDA position, pathway or exemption is stated for patient specific candidate flags, call center triage or the care suggestions sent between visits.
Records and practice histories drive candidate identification across pain management, orthopedics and primary care, and the investor framing is that the system improves with every patient it sees. The controls described sit with the practice: providers check monitoring entries against what the patient reported, and staff support comes with the service. Nothing describes how the models are governed.
There is no evaluation across specialties, payer mixes or patient groups, no check on whether candidate flags fall differently by age, sex, language or insurance, and no process for testing the call center across the languages its patients speak.
The remote monitoring terms make the provider responsible for checking chart entries against the patient's account, because errors can occur in transmission, and patients can withdraw from monitoring at any time. Those terms disclaim warranties of accuracy and completeness and exclude liability for categories including personal injury and consequential damages. On the commercial side, Flagler sells on performance targets agreed before implementation and says it wins when the practice does.
That ties its fees to results, but no published term gives a remedy if a target is missed, and nothing commits Flagler to correct a wrong referral, a missed call or a mistaken chart entry, or to answer to the patient affected.
Supported record systems are athenahealth, Epic, MEDITECH, AdvancedMD, eClinicalWorks, NextGen, ModMed, Prompt and TruBridge, and the implementation team configures the connection so the practice works from its current record with no extra portal or login. The monitoring terms describe entering patient supplied information into the EHR for the provider to verify, and candidate identification and triage read chart data.
That names many systems and describes reading and writing at the workflow level. Which of the nine connections run in both directions in production, at which practices, and which record types each can change are not set out, and no marketplace listing is cited for any of them.
The service runs beside a practice's current electronic medical record rather than replacing it. The deployment team handles setup once the practice grants record system access and names a point of contact, products switch on one at a time, and the company promises results within 60 days. The trust center names Amazon Web Services for cloud, MongoDB Atlas for the database, Databricks for analytics and Datadog for monitoring, which establishes a hosted service on third party infrastructure.
Where patient records, call recordings and agent processing sit is not stated: no region, no backup location, no description of how one practice's data is kept apart from another's, and no option to run the platform in a practice's or network's own environment.
Contracts are performance based. Flagler agrees targets with the practice before implementation, charges no upfront cost, promises results within 60 days and lets a practice start with one product, such as revenue cycle or referral management, before adding others. That shows how the deal is shaped and what drives the fee, though not the share of revenue or savings payable to Flagler or the formula behind it.
The remote monitoring terms make the provider financially responsible for those services. Exact fees, thresholds and the payment calculation are left to each agreement, and no rate appears for any of the six products.
Flagler is built for musculoskeletal care: orthopedic practices, interventional pain and pain management clinics, primary care groups and multisite MSK networks, in a market its lead investor sizes at $400 billion a year. The platform covers the front desk, documentation and reimbursement inside the practice, plus patient communication between visits. Coastal Health is the one named practice, an interventional pain example for remote outcome tracking.
The provider count and outcome figures pool every customer, and the 87% improvement figure combines different kinds of outcomes. Nothing is measured separately by specialty or setting, so an orthopedic referral office and a pain clinic running the monitoring workflow carry the same evidence, and no result covers primary care at all.
Alternatives to Flagler Health
Other products that do the same kind of work as Flagler Health, and what each one does.
Ember AI is an AI revenue integrity platform for specialty physician practices, surgery centers and health systems, sold on preventing denials upstream and recovering them downstream.
Patient access and revenue cycle automation that combines AI agents, rules-based automation agents and a staff of more than 6,000 onshore and offshore specialists inside a platform the company calls its Healthcare Revenue Operating System.
AI orchestration platform for provider side revenue cycle, founded 2022, positioned as an overlay on the systems of record rather than a replacement.
Clearwave sells front office automation to specialty practices, multi site groups and health systems: scheduling, call handling, insurance eligibility, pre registration, clinical intake, check in, payments and patient reminders.
Arintra sells an enterprise revenue assurance platform built on autonomous medical coding. When a provider closes an encounter, its generative AI engine reads the whole chart, assigns specialty specific codes and writes them back into the record system.
AI native operating system for behavioral health clinics, combining named AI workers with an integrated EHR, CRM and revenue cycle management in one platform, with e-prescribing and telehealth inside it.
Pricing
Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.
| Entry Price | Pricing Basis | BAA Tier | Implementation | Source |
|---|---|---|---|---|
|
Custom quote
|
Performance based contracts with targets set before implementation and no upfront cost. Practices can start with one product and add others. The fee formula and rates are set in the agreement. | Clinical services run under business associate agreements. The agreement and its coverage by product are settled in contracting. | No upfront cost. Flagler's team manages the medical record connection and ongoing support, with results promised within 60 days. | Vendor Published |
The platform is sold to practices and MSK networks. The remote monitoring terms put financial responsibility for those services on the provider. Each contract sets how performance, the products chosen and ongoing service affect charges.