A new kind of health insurance network | Oscar Rx | hioscar Blog | Oscar
A new kind of health insurance network
The story of Oscar’s network philosophy, construction approach, and vision for the future.
Behind the ScenesOur Culture
8 min read
Nick Reber
Share article
Once upon a time, navigating the health care system was straightforward. People went to their local doctor for basic issues, and they went to the hospital for anything major. They found these providers through word-of-mouth or sheer proximity.
While today’s modern health care system is much more complex than it was in the early 1900s, many insurers have retained an antiquated view of how consumers find and access care. They assume that, given a long list of in-network doctors and medical systems in their area, consumers will figure out how to navigate the system.
There are several problems with this approach. First, it assumes that people know where to get care and what type of doctor to see for a specific issue. Second, it equates more options with a better healthcare experience. And third, it allows for any random number of pathways between doctors, clinics, hospitals, and labs.
All of these factors contribute to a confusing, expensive, disconnected experience for consumers who engage with the American health care system. That’s why Oscar is stepping in to help members coordinate care from the right providers at the right time – in large part through its networks.
Oscar’s network philosophy
From the beginning, Oscar wanted to build its own provider network to ensure the best possible member experience. Nick Reber, Vice President Provider Strategy & Operations at Oscar, describes the team’s philosophy this way:
“There’s real opportunity to add value to customer’s lives through the network, and not just be a payment passthrough. One of the big problems in health care is that consumers are left alone without the information or incentives they need to get high-quality care. They don’t know which doctors are good or bad, how to get the care they need, or what’s reasonable in terms of cost. Our challenge is to deliver the right incentives and provide good information to members. If we make smart decisions about who should and shouldn’t be in our network, and help members navigate the system, we can sell plans that are lower cost and higher quality than other offerings.”
Guided by this philosophy, Oscar has constructed networks in New York, San Antonio, Los Angeles / Orange County, and San Francisco that don’t fit comfortably into any existing category. They’re not the broad, catch-all networks like those built by traditional insurance companies, which provide lots of options but tend to be more expensive, of varying quality, and hard to navigate. They’re also not narrow networks built around a single provider group.
Instead, Oscar’s networks are curated. They provide a complete suite of care options like a broad network, while driving the cost savings, clinical integrations, and quality outcomes of a narrow network.
Building a new type of network requires a novel approach. Dr. Vinod Mitta, VP of Clinical Operations, and his team worked closely with the network team early on to divvy up the medical world into 500 different types of care members might need – down to the level of congenital cardiothoracic surgery and neurotology. (For context, most health insurers define only 40-50 types of care, which can lead to coverage gaps for less common medical specialities.) They also defined common provider pairings, like anesthesiologists and surgeons, to ensure the network contracted all of the key players for specific types of care.
The search for great providers
Once Dr. Mitta and team defined this core set of care types and pairings, Oscar’s Research and Technology teams used a variety of information sources to identify the best doctors in Oscar’s service areas across all medical specialities.
There are four key criteria Oscar uses to select potential network providers:
Network completeness: Contracting with providers who cover every type of medical specialty in every service area.
Provider quality: Finding providers who have a great member experience and strong clinical outcomes.
Cost: Evaluating unit costs and utilization rates, and comparing those guidelines for a particular type of provider in a particular region.
Frictions: Finding providers and medical groups who already work together via admitting privileges, discharge preferences for post-care facilities, and referral patterns.
Building a quality network
Getting thousands of providers onboard over the course of a few months was a huge undertaking. But finding the right doctors and medical groups was just the first step in a lengthy network construction process. Then came the operational hurdle of calling offices, negotiating rates, developing risk-sharing models with key hospitals and medical groups (opens in new tab), sending contracts, getting contracts signed, uploading signed contracts into Oscar’s provider database and claims processing systems (opens in new tab), and checking to make sure that provider information is accurate.
Neha Kumar, Senior Product Manager, works with a team of engineers on the infrastructure that powers provider data. This aspect of network construction is particularly complex to manage.
“Ingesting data from providers is easy to scale, but ensuring that data is accurate – and keeping that data up-to-date over time – is an ongoing challenge. We’ve invested in building an entire technical and operational infrastructure to support a clean, accurate database of provider information.”
Network construction - not just one and done
Building a network from scratch is a big project in and of itself, but the process doesn’t end when providers sign on. Oscar’s Research team is responsible for defining and evaluating the quality of the network on an ongoing basis.
Daniel Salas and Julian Freed-Brown, Researchers at Oscar, are leading the charge on network scoring and optimization. They’re responsible for gathering data from multiple sources (internal claims, Medicare claims, state-level databases, the provider database), analyzing that data, and developing algorithms that parse this data into micro- and macro-level trends. These aggregated trends are then plugged into Oscar’s Network Scorecard – a system they developed in order to show members’ access to care and quality of experience in each region.
Daniel says, “Up until now, we’ve focused mostly on measuring geographic access to care. As we build brand new networks, it’s critical to know where potential access gaps are so we can quickly get contracts in place to fill them. While we continue to add new providers in each of our service areas, we’re starting to shift our focus to evaluating quality indicators so we can optimize our existing provider base.”
Quality is an important, but often challenging, factor to define. “Our internal definition of ‘quality’ is just now finding a stable home with our team,” says Julian.
Part of the challenge is defining and weighing the factors that contribute to provider quality. Medical training, years of experience, complication rates, and number of services performed each year are all part of the equation, plus a number of other variables. Then there’s the complexity of determining value based on quality versus cost, which allows the team to reward high-quality providers with lower costs and work on improvement plans for lower-quality providers with higher costs.
As the Research team continues to refine their quality measurement process, they can tailor the Network Scorecard to provide a more sophisticated view of provider-member interactions in every market – an invaluable resource for many internal teams, and a great thing for members as well.
Scaling curated network construction
So far, Oscar has curated networks in Texas, California, and New York. As the company expands into new markets (opens in new tab), the critical question is: How will they scale this network approach?
Nick’s answer: A combination of increased manpower and smart automation.
“Some parts of network contracting, such as calling offices, emailing and faxing contracts, and processing paperwork, will remain manual, and will scale with manpower,” says Nick. “Other parts are automated. Still more will be automated in the future to reduce manual labor.”
Already, low-level automation such as electronic contract signatures and ingestion expedite the network construction process. The Research team is also developing algorithms that will reduce manual decision making and minimize the need for high-level experts on things like provider outreach and rate negotiations. Over time, technology will empower workers with broad skillsets to own the contracting process from end to end.
Building care pathways into the network
The traditional way of thinking about networks – contracting a full suite of providers across a range of specialities in a given service area – makes sense from the perspective of an insurance company or broker, who view health care on a macro level. But for individual consumers, this view of a network isn’t very helpful.
When a patient needs to get a colonoscopy, for instance, they don’t care how many gastroenterologists an insurance company has in their geographic region. They only care about finding a good gastroenterologist who performs procedures at a clinic near them – and who won’t charge an arm and a leg for the service.
Oscar’s Concierge teams already guide members through the network in a way that focuses on specific care pathways – that is, an optimized set of treatments, given in a set order, determined by evidence-based medicine. In the future, the team plans to align the network itself with a more consumer-oriented perspective.
Dr. Mitta puts it this way: “The expectation members have for their insurance companies is changing. Insurance companies are starting to guide members on where and when to get care, and under this new model, having a big network matters a lot less than having the right network of great providers. Instead of having to cast the widest possible net – contracting with everyone to support the often random pathways patients take through a network – the network itself can be built around specific pathways members are guided to follow. In turn, it becomes much easier to ensure a quality experience along every pathway for every member.”
The work Daniel and Julian’s team is doing now will pave the way toward this kind of network design.
“Using data from our networks, we’re able to evaluate the distribution of providers and see where members go for care. We can also look at cost and utilization rates in each of our service areas. Over time, we can use this information to inform how we guide members to get the right care for a specific medical issue from providers in their area,” says Julian.
Pre-defined care pathways could have a significant impact on member experience. Instead wading through a directory of providers and figuring out who to see, whether a physician referral is in network, and what prescriptions are on formulary at each step of a treatment plan, members would be guided through the process by the insurance network based on their type of condition (diabetes, congestive heart failure, etc.) or the procedure they need (knee replacement, cardiac valve repair, etc.). As a result, members gain access to a more intuitive, high-quality, affordable way to get care, and develop a better understanding of how each part of the system works together.
Given Oscar’s goal of providing a better health care experience, this evolution is the natural next step in an ongoing process of assessing, growing, and optimizing its networks.
