The most up-to-date, comprehensive, and accurate source of data. Your organization can access profiles of every active provider in the U.S.—over 3.5 million.
See how we’ve helped leading healthcare organizations achieve significant cost savings, improve data accuracy, and enhance patient care. Here, you will find our results, research, reports, and everything else our scientists are testing in the Veda Lab – no lab coat required.
At Veda we understand that every data point is an opportunity to improve the healthcare experience. And we can see the potential when data is no longer a barrier.
HealthX Ventures Blog: How Veda Is Aiming to Fix Healthcare’s Broken Provider Directories
Q&A with Veda CEO Meghan Gaffney and HealthX Ventures on how Veda leverages automated data science and machine learning solutions to provide healthcare payers and networks with the most accurate and comprehensive data available.
When you or someone you love needs care, it would make sense to seek help from a directory of providers who are in-network and available to help. Unfortunately, this is one of the most painful parts of America’s broken healthcare system. One of the most shocking and significant barriers for patients seeking care in today’s healthcare landscape is the vast amount of inaccurate provider directory data. This inaccurate data creates frustration, delays, and inefficiencies that negatively impact both the patient experience in finding care, and also contributes to negative overall patient outcomes. Now, imagine loved ones dealing with memory loss, a behavioral health crisis, or long Covid—and finding the right provider becomes an even greater challenge.
Veda, co-founded by Meghan Gaffney, addresses this critical issue by leveraging automated data science and machine learning solutions to provide healthcare payers and networks with the most accurate and comprehensive data available. By curating data from over 300,000 unique sources, Veda offers real-time, precise profiles of more than 3.5 million healthcare providers, facilitating better decision-making and improved patient experiences.
Imagine this: you use your health insurance company’s online directory to find a doctor at your preferred location and needed specialty, only to be met with incorrect phone numbers, wrong addresses, or no information at all. You’re stuck making phone calls that lead to nowhere, and wasting precious hours navigating bad information while experiencing delays in actually getting the care you need most. After years of diving deep to the complexities around this problem, Veda is at the forefront of clean data and better access to care. Veda’s tools correct inaccurate data and make it possible for members to get what they truly need: the ability to easily and quickly make an appointment with a healthcare provider.
One of Veda’s earliest supporters was HealthX Ventures. With a mission to improve access to care, HealthX found Veda working at the forefront of health equity and contributing to better outcomes for all.
Q&A with Bob Lindner on why sustainably-fed AI models are the path forward
As an AI company powered by our proprietary data training AI models, the article, “When A.I.’s Output Is a Threat to A.I. Itself,” in the New York Times caught our eye. Illustrating exactly what happens when you make a copy of a copy, the article lays out the problems that arise when AI-created inputs generate AI-created outputs and repeat…and repeat.
Veda focuses on having the right sources and the right training data to solve provider data challenges. A data processing system is only as good as the data it’s trained on; if the training data becomes stale—or, is a copy of a copy—inaccurate outputs will likely result.
We asked Veda’s Chief Science & Technology Officer, Bob Lindner, PhD, for his thoughts on AI-model training, AI inputs, and what happens if you rely too heavily on one source.
Veda doesn’t use payers’ directories as inputs in its AI and data training models. Why not?
At Veda, we use what we call “sustainably-fed models.” This means we use hundreds of thousands of input sources to feed our provider directory models. However, there is one kind of source we don’t use: payer-provided directories.
Provider directories are made by health plans that are spending millions of dollars of effort to make them. By lifting that data directly into Veda’s AI learning model, we would permanently depend on ongoing spending from the payers.
We aim to build accurate provider directories that allow the payers to stop expensive administrative efforts. A system that depends on payer-collected data isn’t useful in the long term as that data will go away.
The models will begin ingesting data that was generated by models and you will experience quality decay just like the New York Times article describes. We use sustainably sourced inputs that won’t be contaminated or affected by the model outputs.
Veda does the work and collects first party sources that stand independently without requiring the payer directories as inputs.
Beyond the data integrity problems, if you are using payers’ directories to power directory cleaning for other payers, you are effectively lifting the hard work from payer 1 and using it to help payer 2, potentially running into data sharing agreement problems. This is another risk of cavalier machine learning applications—unauthorized use of the data powering them.
Can you give us an analogy to describe how problematic this really is?
Imagine we make chocolate and we are telling Hershey that they should just sell our chocolate because it’s way better than their own. We tell them, “You could save a lot of money by not making it yourselves anymore.”
However, we make our chocolate by buying a ton of Hershey’s chocolate, remelting it with some new ingredients, and casting it into a different shape.
In the beginning, everything is fine. Hershey loves the new bar and they’re saving money because we’re doing the manufacturing. Eventually, they turn off their own production. Now, with the production turned off, we can’t make our chocolate either. The model falls apart and in the end, no one has any chocolate. A real recipe for disaster.
CMS 2025 Final Rule: New Behavioral Health Requirements for MA Plans
Mental Health Awareness Month and Summary of New CMS Final Rule
Fitting for Mental Health Awareness Month, the Centers for Medicare & Medicaid Services (CMS) recently released its 2025 Final Rule that, among other things, aims to improve access to behavioral health providers for Medicare Advantage members.
Ready to learn about the CMS 2025 Final Rule and Veda’s strategic approach to its behavioral health network requirements?
The CMS 2025 Final Rule significantly expands the behavioral health network requirements for Medicare Advantage (MA) health plans. As reported by Fierce Healthcare, all Medicare Advantage plans will likely see increased administrative burdens due to the behavioral health network expansion requirements.
Not only is Veda a proven and trusted partner for achieving compliance with CMS requirements, Veda’s solutions are unrivaled in their ability to help health plans verify, expand, improve, and map their behavioral health networks.
Here are the behavioral health requirements covered in the Contract Year 2025 Medicare Advantage and Part D Final Rule and Veda’s approach:
New “Outpatient Behavioral Health” Category Added to Network Adequacy Evaluations
Veda’s curated dataset includes accurate information for even the hardest-to-validate specialties, credentials, and facility types— requiring no additional burden on health plans and health systems.
Wide Range of Specialists Included in “Outpatient Behavioral Health” Category
More specialties and outpatient care classifications were added to solve behavioral health provider shortages. The specialists in the new “Outpatient Behavioral Health” category include MHCs, MFTs, Opioid Treatment Program providers, Community Mental Health Centers, addiction medicine physicians, nurse practitioners (NPs), physician assistants (PAs), and clinical nurse specialists (CNSs).
What Can Veda Do?
Veda offers current and complete outpatient behavioral health provider profiles, on each of the specialties above, containing the necessary details to contract with providers and meet adequacy requirements.
Skill Sets of Certain Behavioral Providers Must Be Verified
To address concerns that NPs, PAs, and CNSs have the requisite skills and training to address the behavioral health needs of plan members, MA plans must independently verify that the providers they are adding to their network have furnished (or will furnish) behavioral health services to at least 20 patients within a 12-month period using information such as (i) MA plan’s claims data, (ii) prescription drug claims data, (iii) EHRs, or (iv) similarly reliable data.
What Can Veda Do?
Veda partners with customers to develop effective ways to complete the requisite independent verifications, utilizing the best data available in the industry. Our proprietary Vectyr database is updated 24/7 to include the most up-to-date, complete, and accurate provider data in the behavioral health space. This data can help MA plans verify behavioral health providers and comply with CMS’s requirements.
“Outpatient Behavioral Health” Facility Added to Time & Distance Standards and Telehealth Specialty Requirements
CMS now includes the “Outpatient Behavioral Health” facility specialty in the list of specialty types that will receive a 10% credit toward meeting time and distance standards. Additionally, MA plan’s networks must include at least one telehealth provider within the Outpatient Behavioral Health specialty.
What Can Veda Do?
Veda’s nationwide database can identify telehealth providers, flag whether a provider accepts patients at their practice location, and help source providers to meet the new behavioral health requirements. We can also help MA plans ensure compliance with CMS’s time and distance standards.
Veda is Equipped to Meet The Needs Introduced By The 2025 CMS Rule Changes
Veda excels at helping health plans and systems connect members with behavioral health services, treatment facilities, and telehealth providers. The 2025 MA rule changes are an opportunity for health plans and health systems to explore how Veda can help them expand, improve, and map their behavioral health networks and verify the claims data for behavioral health providers.
Veda’s solutions can help connect members to quality behavioral health services more quickly, efficiently, and at less cost than the traditional methods relied on in the past. Armed with the most accurate provider data available, Veda’s solutions contribute to positive member experiences while helping people find the right care for their behavioral health needs.
Provider Data Accuracy Verified By Third-Party Audit
Independent Audit Says Veda’s AI Precision Exceeds 90%, Solving Ghost Networks and Payer Network Attestation Challenges
Veda is paving the way for improved data quality, transparency, and accountability in healthcare through continuous validation and testing of our AI-powered solutions. Our scientific rigor commits us to continuously testing our methodology and we believe all AI-powered data solutions should undergo an unbiased independent validation of their data. Learn about our approach to healthcare data challenges and the third-party study performed by Erdős Institute proving our data accuracy.
Provider data management is inherently flawed. The information found in provider directories is often manually updated, the scope of required information keeps expanding and information changes often; practices move, physicians change practices, and contracts between practices and health plans expire. Multiple industry reports state between 20% and 30% of directory information changes annually.
With provider data accuracy rates of 90+%, the study highlights the potential of automation and machine learning in achieving high levels of data accuracy
Reinforcing Commitment to Accountable AI-Powered Solutions in Healthcare
To prove our products and approach to provider directory accuracy are best-in-class, the Erdős Institute conducted a blind third-party audit of Veda’s Vectyr product.
Vectyr is Veda’s curated dataset that powers all our provider data products. With Vectyr’s continuously monitored and validated data, Veda customers can quickly find correct provider information with the confidence of Veda’s optimal accuracy. Many back-office workflows—like directory management, credentialing, and claims—need access to accurate, up-to-date provider information. Our Vectyr product curates data from over 300,000 sources including NPPES, DEA, and state licensing organizations.
Why validate with impartial analysis?
As a leader in the industry, armed with proprietary solutions, we understand the key to solving complex healthcare problems lies in innovative technology. We’re taking the lead again, this time via rigorous third-party validation. By subjecting our technology to impartial analysis we’re taking a step forward in the evolution towards greater transparency.
Employing rigorous methodologies and independent sampling techniques, ensuring an unbiased assessment of provider directory accuracy? Sounds like our scientific approach to everything we do. Additionally, having an independent institute like Erdős conduct the validation study safeguards against potential conflicts of interest and ensures the credibility and integrity of the findings.
Erdõs’ Method for Proving Accuracy
To gauge provider data accuracy, Erdős and Veda simulated a Centers for Medicare & Medicaid Services “secret shopper” audit. Callers attempted to make an appointment on behalf of a patient and collect information that would be necessary to do so: the phone number and location of the provider (which are typically major areas of inaccuracies). The audit provided a measure for the main appointment information and manual research was used to measure more detailed provider information.
To reduce bias in the CMS simulation, Erdős created a sample of NPIs for the measurement. The selection of the sample was aimed to be representative with respect to geographic categories of rural and urban and focused on stratified sampling by specialty. In the total sample, 184 NPIs were called. Of these 184 NPIs, 92 phone numbers and 118 locations could be assessed.
The call-based outcomes found phone and address accuracy are consistent at 90%. Moreover, Erdős found that additional Vectyr fields were also highly accurate. For example, the Vectyr database demonstrated accuracy levels of 99% for fields such as languages spoken. With provider data accuracy rates of 90+%, the study highlights the potential of automation and machine learning in achieving high levels of data accuracy.
Reinforcing Commitment to Accountable AI-Powered Solutions in Healthcare
With a commitment to accountable AI-powered solutions and five approved patents in the industry, we believe all healthcare data vendors need to think more rigorously about what “correct” data means. Attestation does not create quality data. Patients don’t need attested data, they need correct data. Therefore, data vendors must measure performance the same way patients do: making an appointment on the first try, with the correct information. We promise accuracy, and we deliver.
ViVE 2024 brought together health tech innovators, vendors, investors, and media. What did Veda bring? Engaging discussions around Medicare Advantage, AI in healthcare, ghost networks, and rural health. Not to mention some heavy media coverage announcing our Humana partnership. Plus, a little fun with a certain famous car to showcase Veda’s ghost network-busting abilities. (We were in Hollywood after all.)
Rural Healthcare and Data
“The way for rural healthcare to succeed is to make it easy to find a doctor, to be a doctor and to pay a doctor,” said Dr. Bob Lindner, Chief Science & Technology officer at Veda during a panel on the challenging rural healthcare landscape. “All of these three things have challenges that can be traced back to the data.”
Bob presented data as a tool that rural health systems can utilize to access accurate information on traveling doctors, a unique offering in rural areas. As rural health faces more hits and challenges, data has the answers.
Veda will use its patented automation technology to analyze, verify, and standardize Humana’s data to ensure the information is accurate and comprehensive, along with real-time scoring of data quality.
Whether you were building a mini-fig or entering to win an ECTO-1 replica, plenty of fun was found at the Veda booth. With a page taken from a beloved movie franchise, Veda showed off its ‘ghost-busting’ abilities and talked ghost networks.
Ghost networks are provider networks that appear robust and full of available providers but actually contain bad data and thus, much more limited availability and unreachable providers. These “ghosts” are no longer practicing, not accepting new patients, are not in-network, or have errors in their contact information.
Veda’s accurate data eliminates ghost networks, improves member satisfaction, and stays ahead of emerging regulations.
CMS Directory Accuracy Audits and Sanctions: Achieving True Directory Accuracy
The Centers for Medicare & Medicaid Services (CMS) regularly audits health plan programs and provider directories. All health plans providing services to Medicare and Medicare Advantage members are nearly guaranteed to be audited by CMS. By definition, the CMS directory accuracy audits aim to improve patient access and experience. Additionally, many standards for provider directories and network adequacy are developed based on CMS regulations.
Veda works with health plans to prepare for CMS audits and then interpret and address their audit results.
Unfortunately, health plans’ directory accuracy claims may not match with CMS’s findings—in the case of lower accuracy discovered, the plans may receive CMS sanctions and fines. Why are the directory accuracy rates differing and what can be done to reconcile the accuracy rates?
Why do accuracy rates determined by CMS and health insurance providers differ?
Many factors determine accuracy rates in provider directories. CMS zeroes in on specific fields (such as name, address, and phone number) for determining accuracy while insurance providers may go further in determining accuracy (such as specialty fields). Here are the reasons why updating directories while maintaining high accuracy levels—is an uphill battle:
Many systems rely on heavily manual workflows, causing delays in data updates. Human error degrades data quality
Provider abrasion and long turnaround times are present when constantly attesting to information
Phone calls, even when used for verification, have a 20% variability rate. Meaning, if your call center has two people call the same provider twice in one day, you’ll get a different answer 20% of the time
Why Does CMS Audit Provider Directories?
A few years ago, a CMS Online Provider Directory Review Report looked at Medicare Advantage directories and found that 52% had at least one inaccuracy. The areas of deficiency included such errors as:
The provider was not at the location listed,
The phone number was incorrect, or
The provider was not accepting new patients when the directory indicated they were.
And, despite provisions in the 2021 No Surprises Act legislation, new research has shown that directories remain inconsistent, one study citing “of the almost 450,000 doctors found in more than one directory, just 19% had consistent address and specialty information.” (Let alone complete accurate information including phone numbers.) The audits continually find inaccuracies as the years go on.
How Do Health Plans Prepare for CMS Audits?
Traditional approaches to audit preparation include phone calls and mock audits.
Phone Calls
Pricey and oftentimes inconsistent, call campaigns amount to hundreds of thousands of phone calls being made every day to check data.
Mock Audits
Mimics the audit experience with sample sets of small amounts of data but are not reflective of the overall directory.
These approaches are not sufficient for achieving successful audit results.
What Is CMS Looking for in Audits of Directories?
Not all information is equally important during an audit. The scoring algorithm assigns different weights for fields so if you’re starting somewhere, Veda recommends starting with the key areas of focus: Name, Address, Phone, Speciality, and Accepting Patients.
Addressing the most important data elements with quality validated data will move a health plan towards audit success.
How Veda’s Solutions Interpret and Address CMS Audit Results
CMS performs audits to advocate for members and better outcomes so interpreting audit results is the perfect place to get started with directory updates. Our research shows that when it comes to what members care about it is pretty simple: Choice, Accuracy, and Accessibility —meaning the ability to schedule, with their preferred provider, easily and quickly. On the first try.
Where to Start For CMS Audit Success
Many health plans are realizing that achieving directory accuracy and audit success is not a one-and-done. An ongoing surveillance approach is needed to confidently prepare and ultimately, achieve success in an audit.
Veda’s approach consistently evaluates the directory to provide ongoing insights. For example, we leverage technology to identify and prioritize providers for updates who haven’t attested recently, to ensure they have a data trail that supports their current status and information in a directory. By prioritizing bad data, this audit strategy is efficient and effective.
AI for Amateurs: Questions Answered by Veda’s AI Experts
Everything You Always Wanted to Know About AI But Were Afraid to Ask
Feeling overwhelmed by AI? You’re not alone. At Veda, we take complicated data and we make it simple. Now we’re here to explain the basics of the complicated topic of artificial intelligence. If you’re feeling like everyone is diving into the deep end with AI knowledge but you’re still in the kiddy pool, this is for you.
Impossible to miss, 2023 is synonymous with the year AI debuted to the masses. AI capabilities have brought up questions in every industry, including healthcare. Your organization will likely find itself navigating the risks and rewards associated with healthcare AI in the coming year.
But, let’s start with a question you’re too afraid to ask at the company meeting: What is AI? Like, really. We’ve found a lot of false information out there and we’re here as a trustworthy source you can pull information from.
Why is Veda a Trusted Source?
As pioneers who have used AI technology since our founding, we’re passionate advocates for AI and want to ensure everyone else feels comfortable with it too.
Want our credentials? Our technology and data science team has 80 years of collective AI experience. Veda co-founder and Chief Science and Technology Officer, Bob Lindner, is the author of five technology patents on AI, entity resolution, and machine learning. Bob also has over 16 years of experience writing and publishing scientific and academic papers in the artificial intelligence field.
Backed by extensive experience and science, we’re the AI experts.
What is AI?
OFFICIAL ANSWER: Artificial intelligence Is a field of study that focuses on how machines can solve complex problems that usually involve human intelligence.
AI is not one specific tool. It is a field of study. With AI’s computing power, computers can make decisions and predictions, and take actions. An algorithm recommending which movie you should watch next is an AI action.
VEDA’S TAKE: So why does this matter? Why is AI important? By freeing up human resources, AI can reduce manual and often error-prone tasks. Freeing up people so they have the time to do the things they do best, that’s the power of AI.
What is machine learning?
OFFICIAL ANSWER: Machine learning is a sub-field of AI and focuses on algorithms that train models to make predictions on new data without being explicitly programmed. Meaning, the machine learns the way humans do, with experience.
Note: In recent years, some organizations have begun using the terms artificial intelligence and machine learning interchangeably.
Instead of learning step by step, computers using machine learning can learn through trial and error and lots of practice. What does machine learning practice on? Lots and lots of data. The data can be things like images, video, audio, and text. When fed loads of data, machine learning will recognize patterns and make predictions based on these patterns.
AI is not one specific tool. It is a field of study. With AI’s computing power, computers can make decisions and predictions, and take actions.
VEDA’S TAKE:Veda uses machine learning, and therefore, AI for the healthcare industry, every day. For what exactly? To power our provider information. Veda uses machine learning to:
Determine correct addresses and phone numbers
Transform provider rosters from one format to another
Simulate an experience a member may have when booking an appointment
With a patented training data approach, our machine learning can make predictions on a wide variety of new data (that it has never seen before in the training set).
Feeling good about AI and machine learning? Further your AI understanding with these blogs:
Is Your Data Partner Ready for New AI Regulations?
How to Find the Right Provider Data Vendor Partnership
It seems that almost every week we see new vendor offerings within the provider data management ecosphere—each claiming to offer a revolutionary way of visualizing your data and making impactful improvements for healthcare. The provider data accuracy challenges that health plans and provider organizations face are vast, so it’s not surprising that new health tech companies are quick to capitalize. Caution: all health tech companies may not be keeping new and developing AI regulations in mind when developing their technology.
All health tech companies may not be keeping new and developing AI regulations in mind when developing their technology.
Looking at the Bigger Provider Data Picture
Quality provider data can help the healthcare industry tackle some of the biggest challenges health plans and provider networks face, like onboarding, credentialing, roster creation, and referrals.
A commitment to improving your provider data can have great ROI potential and long-term impacts on your business and patient populations. But, with all the buzzwords and capability promises from the masses, how do you find the best vendor fit for your business needs?
Necessary Topics to Cover When Vetting Provider Data Vendors
To set you up for partnership success, here are discussion ideas for data vendor conversations:
Find out the vendor’s practices for responsible data collection, storage, and validation. For example, does the vendor keep their data in the U.S.? Veda’s Take: You want to ensure any vendor interacting with critical data is not utilizing offshore, 3rd party data processing centers that can unnecessarily expose data and therefore, providers’ information. Consider this: If patient data is protected from offshore processing vendors via HIPAA and other regulations, shouldn’t the same protections be afforded to provider data? Veda offers comprehensive security and data protection and is HITRUST-certified.
Understand the vendor’s business policies regarding AI. For example, how does an AI vendor utilize Language Learning Models? Veda’s Take: Avoid risk by future-proofing your AI policies. Machine Learning can be a powerful tool, but should be approached thoughtfully and aligned with current and future U.S. legislative requirements such as President Biden’s executive order for the Development and Use of Artificial Intelligence. Warning: fewer companies comply with the proposed regulatory requirements than you may think.
Ask about their reporting and measurement. How does the vendor define accurate data and measure it during and after delivery? Veda’s Take:Data can be compliant but inaccurate and unusable for health plan members who depend on the data to get care. Be prepared to discuss your business goals and thresholds for accuracy—consider going beyond meeting regulatory guidelines. We think healthcare data should truly be considered “accurate” if it meets a member’s needs when accessing care.
Discuss what is necessary for your business success. Can the vendor offer the necessary tools to reach your goals and eliminate what is unproven? Veda’s Take: Don’t be distracted by tools that may not deliver results or provide value for your goals. For example, APIs are often necessary to save time on automation. Therefore, many businesses focus on the ability to have an API connection and the API integration above even the results the product delivers. Or, in another example, the UX and the interface of a product can become a focal point above the actual functionality of the product. If you can’t trust the data and know how to interpret and use it, then connectivity and appearances don’t matter.
Determine what happens first. Can a vendor partner prioritize your specific business requirements? Veda’s Take: All businesses have different objectives and these goals greatly impact priorities. Your vendor should clearly articulate what needs to happen first, upon implementation of the product, to realize immediate value and reach success. There is no one-size-fits-all solution when working with a provider data vendor. Before integration and during the initial conversations is the perfect time to establish an approach to prioritization within business rules.
Get familiar with the training process. What does the implementation, training, and delivery process look like for your AI data vendor? Veda’s Take: How a provider data vendor plans to work with you, and how they plan to train others in your organization, is key to partnership success. Beyond day-to-day use of the tools, how does the vendor recommend using the data and applying the findings? Who should be trained on what tasks? Clear and concise preparation will ready everyone in your organization.
Ensure any vendor interacting with critical data is not utilizing offshore, 3rd party data processing centers that can unnecessarily expose data and therefore, providers’ information.
Once you have a solid understanding of how a potential partner tackles the above objectives, only then can you capitalize on a business case for building a collaborative partnership with a provider data vendor.
A ghost network is a provider network that appears robust and full of available providers but actually contains bad data and thus, much more limited availability and unreachable providers. These “ghosts” are no longer practicing, not accepting new patients, are not in-network, or have errors in their contact information.
Imagine using a “Find A Doctor” online tool to pick an in-network doctor in the specialty you seek, but when you call to book an appointment, you discover the provider is no longer practicing. That’s a ghost.
“ [Ghost networks are a] pervasive issue in the American health care system. A three-phase study of the accuracy of Medicare Advantage directories, which included over 15,000 providers, found that between forty-five and fifty-two percent of provider directory listings had errors, with some individual plans having error rates as high as ninety-eight percent.”
While behavioral health networks are often cited for inaccuracy and the mental health crisis in America has brought adequate network issues to light, inaccurate directories are a systemic issue prevalent across all provider types. All directories have “ghosts.”
How do ghost networks inhibit care?
Besides adding to patient frustration, ghost networks prevent patients from accessing the care they need. Impactful stories, like this one from the Yale Law & Policy Review, shed light on the millions of people dealing with the repercussions of inaccurate provider directories each year:
“KC, who manages her brother’s mental health care, gave up on trying to find an in-network psychiatrist because calling potential providers was taking up so much of her time that it was more cost effective to pay out-of-network rates. “Ironically, I can’t imagine my brother or others in his situation being organized and effective enough to be able to make all these calls and keep track.” [Yale Law & Policy Review]
Additionally, ghost networks have inhibited care by leading to unexpected medical bills. “When Americans are purchasing and using their health insurance, they have the right to know whether their doctors are covered by that plan,” said Ron Wyden (D-Ore). “Too often, seniors and families get health care whiplash when they sign up for a plan only to find out that their preferred doctor is out-of-network, or it’s impossible to find a covered mental health care provider.”
How does the REAL Health Providers Act tackle the issue of ghost networks?
The legislation aims to combat the ghost network problem by, among other things, requiring Medicare Advantage (MA) health plans (beginning with plan year 2026) to verify their provider directory data every 90 days and, if necessary, update that information.
If a health plan cannot verify the data, the plan must indicate in its directory that the information may not be up to date.
A health plan must also remove a provider within 5 business days if the provider is no longer participating in the plan’s network.
If a patient obtains care from an out-of-network provider that a health plan’s directory indicated was in-network at the time the appointment was made, the plan may only charge that patient in-network prices.
The legislation also requires MA health plans to analyze the accuracy of their provider data on an annual basis and submit a report to HHS/CMS with the results of that analysis. HHS/CMS will use this information to publish accuracy scores for each plan’s provider directory.
How does Veda banish ghost networks?
Veda has been at the forefront of efforts to eradicate “ghost networks” in the U.S. for years. Veda understands that progress on this front is essential to connect patients to the critical care they need and ensure that individuals are not burdened with unexpected healthcare costs.
Veda takes a one-two-punch approach to eradicating ghost networks. Veda leverages its patented technology and innovative solutions to first identify where the “ghosts” reside in provider directories and then fills in the gaps left behind once the ghosts have been removed. Using a mix of AI technology, smart automation, and machine learning, Veda’s provider data is proven accurate.
Find the Ghosts
Veda’s Quantym platform identifies the errors in a provider directory
High-volume audit solution that delivers comprehensive, real-time scoring of provider data quality to identify bad data and significantly improve provider directory accuracy
Fill the Gaps with Accurate Data
Veda’s Vectyr tool supplies accurate data to replace the bad data
The most up-to-date, comprehensive, and accurate source of data on healthcare providers, groups, and facilities on the market
Veda understands that ghost networks will not be eliminated by the manual attestation methods of the past. Manual verification is labor-intensive, expensive, subject to human error and time and again, it’s proven ineffective. The fact is, “attested” data is frequently not “correct” or “accurate.”
Veda’s technology takes provider attestation out of the equation and, in doing so, can more accurately identify where ghost networks exist (by pointing out inaccuracies in the health plans’ provider directory data) and provide solutions to the ghost network problem (by filling in the gaps that are created when the bad data is removed).
It’s time to let technology like Veda’s solve the ghost network problem once and for all. Give your members the accurate information they need to make an appointment for care on their first try. Contact Veda for a free data assessment.
Supervised vs. Unsupervised Machine Learning: Using The Right Tool For The Job
It’s Veda’s philosophy that any new technology tools we utilize are not meant to wholly replace human engagement. We believe that technology should help elevate humanity. With a focus on performing meaningful work, people can achieve their highest value. Technology should help people help people.
As a Data Science as a Service (DSaaS) company, Veda leverages scientific principles within our unique AI and machine learning systems to perpetually clean, correct, and monitor evolving provider and facility data. A lot of companies claim to offer accurate provider data, but none are as committed to using science to solve deeply entrenched provider data problems.
Dr. Bob Lindner describes supervised vs. semi-supervised learning
At Veda, we use AI systems including natural language processing (NLP), supervised, and unsupervised learning components that can be leveraged to solve a wide array of payer data challenges depending on what tool is right for the job. No matter what, we start by understanding the problem—not by applying a method.
A lot of companies claim to offer accurate provider data, but none are as committed to using science to solve deeply entrenched provider data problems.
The Roles of Supervised and Unsupervised Learning
Supervised Learning
Veda utilizes supervised learning because it doesn’t require “perfect” sources of data— it can make use of the good parts of any data source and knows how to ignore errors. With supervised learning, a data scientist is watching and helping train a model with all the healthcare nuances and industry-specific language, etc. to make the model a good one.
We’ve measured individual sources of data for years—including attestation— and haven’t found any at 90% or above yet. Supervised learning is incredibly accurate with the data we have access to today. The benefit is the highest accuracy which is flexible and not dependent on a single data source.
Unsupervised Learning
We also use unsupervised learning for offline data exploration and research to learn more about a dataset, to help design better machine learning features for supervised learning systems. That’s because unsupervised learning separates big collections of data into groups on its own. The benefit of unsupervised learning is its ability to pick out patterns in the data.
Unsupervised learning separates big collections of data into groups on its own. Some will claim that unsupervised learning, alone, is superior to supervised learning because of the lack of human intervention. However, most algorithms require the user to specify upfront how many groups they want the data separated into. So no matter what data is being grouped, one would have to delineate exactly how many groups are wanted ahead of time and the exact number of groups the data is sorted into regardless of whether the groups match up well with the data. This requires the user to have upfront knowledge of exactly what labels and groups they need.
But the biggest pitfall of unsupervised learning is that there’s no labeled training data, which means there’s no actual measurement of how well it’s working and placing the items into the correct groups. And with no way of knowing how well it’s working, it’s impossible to depend on unsupervised learning as a primary method for accuracy.
The Right Tool For the Job
Supervised and unsupervised learning are tools, and just as you wouldn’t remodel your kitchen and only use a saw, you shouldn’t only use one kind of machine learning model.
Veda’s technology and approach to data challenges are fundamentally different from other provider data technology companies in that we focus on fully automating both the static information and the more challenging temporal information about a provider—data that changes at varied rates over time, like practice address, phone, and group affiliation. Our patented systems do not require manual outreach to providers, rather they rely on data created by providers throughout their established workflows. This increases data accuracy by reducing human error while also decreasing provider abrasion. Validating millions of temporal data elements in real-time requires Veda’s full automation system and could not be solved manually.
Above all, we believe that AI and machine learning are the best ways to solve the provider data quality problem because:
These techniques do not require us to know how accurate our sources of data are ahead of time—the machine figures out how to tell good data from bad
AI makes the most of imperfect and changing data
They do not require provider participation—we use data they already create in their day-to-day workflows, so no need to persuade providers to take additional action
It works—we have scientifically tested attestation along with “source of truth” modeling, and Veda’s approach has the highest measured accuracy of any approach in the industry.
Veda’s provider data solutions help healthcare organizations reduce manual work, meet compliance requirements, and improve member experience through accurate provider directories. Select your path to accurate data.
Velocity
ROSTER AUTOMATION
Standardize and verify unstructured data with unprecedented speed and accuracy.
