Rahul Sharma

home

authors

rahul-sharma

There’s good news and bad news when it comes to the evolution of value-based care in the U.S. Let’s start with the good news. A 2021 study from the Health Care Payment Learning and Action Network (HCPLAN) shows a continued move from fee-for-service models to value-based care (VBC). According to the study, 39.3% of healthcare dollars spent are tied to traditional fee-for-service payments, 19.8% are tied to pay-for-performance, and 40.9% are part of some type of value-based contract (e.g., shared savings, shared risk, bundled payment, population-based payments, etc.).

Now for the bad news. While the trend toward VBC is promising, research also shows it can take a long time for healthcare payers to effectively implement these new payment models. According to a 2018 study, only 21% of payers say they can roll out a new episode of care in three to six months, more than 33% say they need a year, 21% need 18 months, and 13% require two years or more.

This inconsistent implementation pace can impact the overall adoption and success of VBC programs nationwide. Addressing this issue requires resolving several universal challenges that currently hinder the ability for many healthcare payers to quickly implement new VBC payment models.

A common implementation barrier is legacy claims and workflow technology investments that don’t support the management of the complex multi-stakeholder care networks involved in VBC or the episodic requirements of payment models that aren’t claim centric. Another barrier is the timeliness of data reporting and the inability to understand contract performance more prospectively as opposed to after-the-fact.

While technology and data exchange challenges impact both payers and providers, payers will likely need to take the lead in this area to make progress. This is primarily because payers are often responsible for not only educating providers on VBC models, but also bridging the gap between fee-for-service and VBC. Serving as a catalyst for this change and positioning the industry for VBC nimbleness requires payers to rethink the existing infrastructure paradigm. 

Successful administration of value-based programs requires an infrastructure that enables the many-to-many relationships between VBC stakeholders and their counterparts. This includes health insurance carriers as well as risk-bearing entities such as accountable care urganizations (ACOs), clinically integrated networks, carve-out programs for chronic disease management, primary care, care management programming, social service networks, and community-based organizations (CBOs). The administration of funding pools, including downstream distribution of funds and data exchange to participating partners, is one of the most critical functions of successful value-based execution.

For both payers and providers, traditional approaches and legacy systems do not support the hierarchical relationship structures necessary for onboarding stakeholders in value-based contracts. Likewise, orchestration of cascading payment models, where payer/provider collaborations incorporate risk-bearing entities and downstream participating providers, are not easily scaled, reducing the ability to accelerate adoption of varied alternative payment models. This hierarchical approach to partner onboarding, scaling of contract operationalization, and permissioned data sharing is a necessity for alignment of medical, social, behavioral, and environmental components of successful value-based program administration and high-performance networks that confidently deliver on whole-health patient outcomes.

An infrastructure for VBC not only requires hierarchy support, but also the ready exchange of a more comprehensive set of health data. Traditional electronic health record (EHR) systems were built to solve the challenges of transactional data related to claims submission, billing, and processing of adjudication, clinical data, and pharmacy data. VBC requires more insight into patient health and outcomes. It requires the creation of a longitudinal health record (LHR) that not only manages and shares traditional structured EHR data (e.g., data that fits neatly into fixed fields), but also enables the proper digitization and exchange of clinically rich data that resides in unstructured form as charts, notes, images, audio and video files, as well as freeform text of character large objects (CLOBs).

Advanced technologies such as artificial intelligence (AI) and machine learning (ML) make this possible, providing a unified view of the patient that allows for a move from transactional processing to an outcome-driven, value-based model focused on patient-centered care. However, a data infrastructure truly enabled for VBC also needs to be based on ontology mapping rather than traditional relational or nosql databases.

An ontology encompasses a representation, formal naming, and definition of the categories, properties, and relations between the concepts, data, and entities that substantiate one, many, or all domains of discourse. For example, changing a property from single-valued to multi-valued in a relational database requires the entire column for that property to be deleted and an entirely new table to be created that holds all the new property values. By contrast, all data modeling statements in ontological languages are incremental. Enhancing or modifying a data model after the fact can be accomplished simply by modifying the concept, making the synthetization of patient health data across entities much easier.

When it comes to ontologies for the patient data sets, there are 18 different sets that these would fall into:


Clinical data (e.g., notes, charts, X-Ray, MRI, radiographs etc.)

Financial Data (from banks for loans, AR reports, patient responsibility from claims)

Social Determinants of Health (SDOH) Data

External summary data sources for analysis (WHO, CDC, Johns Hopkins etc.)

All this data needs to have an underlying Enterprise Master Patient Index (EMPI) to tie the data sets to the right patient together and tie it to the unique identity enumeration done for that patient.

Microservices aid permission-based sharing

The underlying data infrastructure outlined in the previous sections can then be exposed via a secure and scalable Data as a Service (DaaS) layer on which different applications and integrations are built, creating a network of networks for permissioned-based data sharing that enables a move to VBC. This DaaS layer is important because it allows an adaptable VBC infrastructure to be realized without a rip-and-replace strategy. Data layers can instead be integrated and extended seamlessly, allowing stakeholders to use existing applications served up via microservices and extend or create microservices and business applications for their own needs.

The trend toward VBC is driven by a desire to improve patient outcomes while lowering healthcare costs. Payers play a central role in establishing the hierarchical infrastructure necessary to accelerate and scale the implementation of new VBC models throughout the continuum of care.

APM Measurement: Progress of Alternate Payment Models – HCPLAN – 2020-2021

Payers moving to value-based care faster than expected – Healthcare Dive – Len Masterson – June 18, 2018

is CEO of HSBlox, a healthcare technology in suburban Atlanta that supports value-based technology. 

is the chief operations officer of the company. 

Articles

While the trend toward value-based care is promising, research shows it can take a long time for healthcare payers to effectively implement the model. Value-based programs require an infrastructure that enables the many-to-many relationships between value-based care stakeholders and their counterparts.

  Transforming Payer Infrastructure to Accelerate Value-based Care

Social determinants of health (SDOH) account for up to 80% of health outcomes, while medical care accounts for the balance, according to 

Our experiences with COVID-19 are a stark reminder that unaddressed social determinants and health inequities play significant roles in the comorbidities that have led to countless deaths attributed to the coronavirus since early 2020.

Given the importance of SDOH in determining individual and population health outcomes, it’s clear that payers and providers can benefit from a comprehensive and secure medical and social longitudinal health record that captures SDOH data points to connect patients with community resources that address unmet social needs – reducing healthcare costs while improving lives. By having SDOH data available at the point of care, clinicians and other caregivers have a comprehensive view of the patient, enabling them to approach care holistically and take action.

Unfortunately, efforts to capture and integrate SDOH data across the healthcare continuum largely have been ineffective and plagued with interoperability issues, culture gaps, and lack of coordination. A 2019 Dartmouth 

 showed that only 24% of hospitals and 16% of physician practices in the U.S. reported screening for SDOH factors like food insecurity, housing instability, interpersonal violence, transportation needs, and utility needs.

Technology can make it possible to collect and synthesize data from different sources and in different forms (structured and unstructured). Screening tools can be augmented with external data sets, such as the 12 Dimensions of the Social Environment, for which the CDC has created a 

This data can be incorporated into longitudinal health records for individual patients. Digitization of the unstructured data sets using Artificial Intelligence (AI) algorithms can provide a wealth of information for enhancing the longitudinal health record of the patients.Machine learning (ML) can then be applied to build models for patient medical and social risk scores, cost prediction, and patient behavior.

The COVID-19 pandemic highlighted the importance of having quality, verified data accessible to care providers and public officials to help combat the spread of a communicable disease. Technology can then make it possible to synthesize data sources of different kinds – supply chain, disease registries, Internet of Medical Things (IoMT) and contact tracing data, for example – to address their inconsistencies, help identify errors or misreporting, and seamlessly integrate credible new feeds. Proper analysis done in near real-time can guide mitigation approaches and forecasting to optimize health resources. Similarly, accessible and verifiable SDOH data enables payers and providers to discern and act upon specific social factors impacting patient health.

There are several obstacles the healthcare industry must overcome to fully leverage SDOH.

One is disparate and siloed data from multiple sources across industries, agencies, and organizations in the public and private sectors, each with its own unique structures.

Second, roughly 80% of healthcare data is unstructured. This data can take the form of clinician electronic notes, patient-reported information such as portal messages, IoMT data and transcripts from telehealth visits that often are hard to access. Healthcare organizations need applied technology capabilities such as Natural Language Processing (NLP), multilingual support and translation service capabilities, to digitize this unstructured data.

Unstructured data such as handwritten notes, documents, free form text, images and audio and video recordings must be digitized and structured for use in applied analytics to provide valuable decision-making insights to healthcare professionals and other authorized care team stakeholders.

One way to ensure authorized disclosure and use of SDOH and other healthcare data is to leverage distributed ledger technology (DLT) for secure and permissioned sharing in near real-time, and at a granular level. This approach ensures the reliability of the data and its source(s), providing the transparency necessary for decision-making.It also ensures that the concerned parties are seeing the same data without any need for reconciliation.

Any strategy for sharing patient data across stakeholders must make protected health and other personally identifiable information a top priority. Safeguards can include biometrics data for proper patient identity verification and matching. Data should be encrypted at rest and in transit.

Third, in the realm of social determinants, there are stakeholders not subject to HIPAA regulations, so the need arises to obtain and manage patient consent for sharing data with and between stakeholders. As a result, the underlying technology must incorporate a consent management capability to ensure patient-permissioned data sharing with an immutable audit trail of disclosures.It also must support multi-channel communication approaches for obtaining and managing consent, such as mobile applications, secure text, audio, and video. If consent is being obtained through a caregiver, then Electronic Visit Verification (EVV) capture is another key data point that should be recorded.

Just as different data systems need a way to communicate with each other, different SDOH stakeholders must be able to securely exchange information. A next-generation communications engine that provides end-to-end secure communication capabilities for B2B and B2C using protocols like Email, SMS, MMS with extensions to widely used instant messaging software can eliminate communication barriers among health plans, hospitals, clinics, social service agencies, and patients.

Furthermore, technology supporting SDOH networks must have a simple user interface and an easy to navigate workflow from which data can be accessed, imported, or exported. Crucial to the creation of longitudinal health records is support for key patient data categories and information sources, including clinical data, administrative claims data, prescriptions, demographics, disease registries, immunity tests, employment, sensor data, SDOH data, and many more. While some datasets are standardized, others must be translated from unstructured data to be accessible.

If COVID-19 exposed the problems in healthcare caused by social factors, it also underscored the importance of the physical supply chain, especially in times of crisis. An effective healthcare approach features near real-time integration of supply chain data both from different medical/healthcare vendors as well as delivery companies like FedEx and UPS.

Improving individual and population health while reducing healthcare costs is a win for patients, providers, communities, and payers. To accomplish this goal, healthcare stakeholders need to capture and leverage social determinants of health data, which research shows accounts for the vast majority of patient outcomes and population health trends. By ensuring access to relevant and reliable SDOH data from verified sources, healthcare professionals can identify and mitigate health risks that overwhelm our system. This can be done by deploying multi-stakeholder collaboration technology with a longitudinal health record to optimize care coordination and population health measures.

, which enables SDOH risk-stratification, care coordination and permissioned data sharing through its digital health platform.

Providers and payers need data to address social determinants of health effectively. But that data is often siloed, unstructured and difficult to use. 

Rahul Sharma

Articles

Transforming Payer Infrastructure to Accelerate Value-based Care

How Technology Can Unlock the Value of SDOH