The Opportunities of Technology-Enabled Risk-Sharing

Over 10% of the U.S. population is now covered by an accountable care organization (ACO), and the number of risk-adjusted lives is growing at roughly 15% to 20% each year. To date, risk adjustment has primarily consisted of payers retrospectively reconciling work done with payments owed.

With the Medicare Access and CHIP Reauthorization Act (MACRA) in 2015, the Merit-Based Incentive Payment System (MIPS) on its heels, and advanced alternative payment models (APMs) offering more incentives, thriving under value-based payment models is a complex but increasingly viable path for providers. With the right technological assistance, clinicians can better understand clinical and operational risk, how to intervene and manage it, and how to maximize appropriate reimbursements to ensure organizational stability and growth.

The Gulf Between Risk and Reward

Managing risk is complicated, especially when it moves from retrospective to prospective in clinical settings. Common risk-sharing contracts need forecasting and monitoring to provide effective benchmarking prior to that first contract year—influencing payments going forward. Access to risk expertise becomes essential. A cultural shift in providing care may also be needed to align new care initiatives around risk and avoid administrative burnout of clinicians.

Technologically speaking, the data itself must be harnessed, understood, and delivered where and when it’s needed. Under risk-sharing contracts, providers need to rely on data and analytics to address care and quality gaps and produce positive outcomes. Without the right data and analytics, providers can’t predict—or even notice—when an undiagnosed condition presents a material risk for a more costly care episode downstream. Accessing, leveraging, and surfacing the data required to successfully participate in value-based models at the right time, to the right user, can prove difficult.

Risk-Sharing Contracts White Paper

Achieving optimal outcomes under risk also relies on integrating care management into the access to this data. While there is significant opportunity to address risk at the encounter level, an even greater opportunity comes from patient outreach.

Any results from new, intelligence-driven outreach efforts on risk-related conditions are a boost to clinical and financial outcomes. Providers can know when and why to check on patients at a much deeper level, especially when socioeconomic factors are contributing to overall patient population risk. Without both the intelligence driven efforts and the outreach, though, there’s nothing.

The final piece is clinical documentation gaps. The Centers for Medicare and Medicaid Services (CMS) assigns risk scores to patients based on approximately 10,000 acute and chronic diagnoses. CMS combines this data with patient characteristics, including social determinants of health, demographics, and interaction factors (the combination of certain chronic diseases) to arrive at a risk adjustment factor (RAF). This RAF dictates the reimbursement for that patient’s total care. The care team documents diagnoses data during (or immediately after) the exam so coders can submit accurate, timely claims.

Many provider organizations attempt to capture this information manually increasing administrative strain on physicians already facing endemic burnout. Any documentation solution needs to account for, and possibly relieve, that burden.

Thoughtful Deployment for Impactful, Lasting Results

Successful risk adjustment enables a more complete understanding of the population’s underlying risks, which can fuel better care and cost management. Whether participating in government programs (Medicare ACOs and shared savings programs) or taking on private risk-sharing contracts with payer partners, organizations should obtain a complete picture of their patients’ acuity. This is critical to ensure proper reimbursements, effectively manage costs of high-risk members, and deliver quality care.

As provider organizations expand their risk-share footprint, the challenge in most of these organizations is that they have been set up to operate within fee-for-service. Some providers will immediately begin to attempt to capture risk gaps at the point of care. However, point-of-care risk capture is neither easy to implement nor a silver bullet, and there are no off-the-shelf solutions.

Even under the well-defined models of value-based care, every provider organization is different. Effective point-of-care workflows need pre-encounter and coding team support with the appropriate technology and processes.

Consider the following:


Prior to presenting information at the point of care, a sufficient amount of diligence is needed, especially when accounting for quality and risk. Analysis of historical data can identify gaps in risk capture for each member and can be used to create a stratified list of members to be scheduled for a visit.

However, the massive volumes of patient data available can make this untenable for a team—let alone in situations where one physician is operating alone. This is where artificial intelligence comes in: A clinically trained natural language processor (NLP) can process the full medical record, identify quality and risk gaps, and present these gaps so the care team can review them prior to the encounter.

With an appropriate team and technology, physician administrative work can be limited to the cases with the highest machine-identified confidence for a final decision. Meanwhile, staff can verify lower confidence suggestions of risk and quality gaps, and then conduct targeted patient outreach and schedule appointments.



A successful point-of-care solution should deliver gaps to the physician and allow the gaps to be closed—without leaving an electronic health record (EHR)-centered workflow or inhibiting a physician’s focus on the patient.

But a successful point-of-care workflow cannot stop there. It must also facilitate the capture of outstanding risk conditions through accurate documentation in the EHR. This is where re-centering risk on the patient and patient experience—without additional physician administrative work—pays off.

Through this approach, physicians can refocus attention on patients, with the confidence of insights from the full patient record to proactively ask about relevant chronic conditions. Patients no longer have to aggressively re-emphasize their conditions when visiting for ostensibly unrelated cases. The morale lift of this from a patient perspective cannot be understated.



Historically, the challenge with risk capture is that documenting risk conditions is insufficient and performed by payers, far from the point-of-care. Clinically, a patient record is similarly limited by this insufficiency. If conditions are not reliably documented, they cannot be used to address the patient’s needs on subsequent visits. At the same time, if data is thoroughly documented but inaccessible, it’s effectively moot. Any circumstances where providers could have intervened are continually missed.

To have an effective clinical and financial impact, providers must document and code the proper International Classification of Diseases (ICD) diagnoses on outbound claims. In some Medicare Advantage (MA) contracts, there is a retrospective timeframe during which coding “catch-up” can be conducted. However, in many non-MA arrangements, the code must be captured and put on the claim before initial submission. If the code fails to make it onto the claim, the work conducted by clinicians will not yield results.

Identifying uncoded but documented work can be a burdensome process, requiring coders to sift through the entire EHR. Thankfully, by leveraging technologies such as natural language processing (NLP), the coder only needs to verify identified code opportunities—prioritizing the capture of risk gaps and ensuring that outbound claims are accurately coded in a timely manner.

Finally, ongoing updates to analytics, based on new patient encounter information, can help identify additional gaps and promote a continual improvement process. The clinical opportunity to use outreach to address conditions early and improve care is not only possible—it’s an integrated element of any risk adjustment program at a provider organization.

Note that while each phase is an important component to establishing a successful risk capture program, providers cannot expect to reach this level of maturity overnight. Developing a plan must account for organizational readiness, long-term goals, and population needs. However, through thoughtful selection of an “entry” point into addressing risk, provider organizations can lead with easier early adoption of workflows. This allows for an early return on investment that can help organizations subsidize subsequent steps in the full-risk adjustment workflow adoption process.


Moving Forward

Risk adjustment isn’t the exclusive purview of payers anymore. Going forward, as a rapidly growing share of the U.S. population is covered under value-based care arrangements, provider organizations will need to stop seeing these models as “maybe one day” and start recognizing them as the future of healthcare.

With the right tools and transition planning, the adoption of value-based care doesn’t need to be a burdensome process. Successful examples of risk-based provider contracting and risk-adjustment technology solutions are a reality today—built on two decades of healthcare IT innovation and lessons learned in clinical environments.

When you reduce nonclinical work, center organizational initiatives on the patient, integrate effectively with workflows and existing technology, the health of populations and the financial stability of an organization will follow.


To learn more about risk-sharing contracts and how providers are successfully managing and adjusting risk, download this white paper.