Quality Care

Utilizing Data in Support of a Primary Care First Payment Model

Provider shifting to value-based care with Primary Care First model

The Centers for Medicare & Medicaid Services' (CMS) Primary Cares Initiative is a physician-focused, voluntary Medicare program consisting of five care delivery models to be implemented in 2020. Three of the models focus on direct contracting with physicians and physician groups, and two of the models focus on primary care providers, called Primary Care First—enabling primary care providers to shift from providing fee-for-service activity to providing more value-based care for both general and high-needs patient populations.

This approach uses a simplified payment structure, paying a fixed amount for each patient and offering bonuses for keeping patients out of the hospital and providing disincentives for extra spending incurred by admissions. Let’s talk about Primary Care First as it has unique data utilization needs in support of the primary care providers.

Primary Care First Payment Model

The Primary Care First payment model “tests whether delivery of advanced primary care can reduce total cost of care”, focusing on both primary care providers for standard populations and primary care providers focusing on hospice or palliative care services. Hospice and palliative care providers are managing seriously ill populations who are ready to assume greater financial risk in exchange for reduced administrative burden and performance-based payments.

Although primary care providers make up a small percentage of Medicare/Medicaid expenditures, 2.12% to 4.88% of total medical and prescription spending, they are on the front lines of patient care and can influence a patient’s healthcare trajectory, potentially preventing significant costs per patient down the road.

Primary care providers are frequently called on by insurers and health systems to control access to higher cost specialists, presumably providing a measure of cost control through the initial consultation and ongoing preventative care—they are frequently a patient’s first healthcare contact. Over time, the continuity of the primary care provider relationship provides one of the richest sources for clinical data in a longitudinal health record in quality, specificity, quantity and variety of data points. Primary care providers offer not only a source of information and care, but also a convenient data reception point for patients to record data from common chronic illnesses such as hypertension, diabetes and asthma. This becomes even more important as our population ages and requires more efficient healthcare services.

Participating in the Primary Care First payment model provides several benefits for healthcare providers, but one of the most useful benefits is the ability to send specific claims-related data to CMS and receive in return aggregated quarterly monthly claims data including expenditure and utilization data.

These CMS data files can be used with clinical data collected from the primary care providers, EHRs and available data exchanges to provide more in-depth views into their practices and patients. This data-based feedback is similar to the Comprehensive Primary Care Plus payment model, with CMS and payers providing claims and enrollment data to participating providers on Medicare Fee for Service beneficiaries. In addition, the files provide payer and practice application data along with care delivery and financial data reported by practices to CMS for both the participating practice and other practice participants.

Considerations for Using Data Sources

Primary care providers using CMS data files, or any other incoming data, either financial, clinical or other supporting data should consider the following important elements such as resource management, data governance, and business and clinical intelligence to get the most out of their data.

Resource Management

Many primary care practices are small, usually consisting of either one or several physicians or non-physician health professionals in either a system-owned or an independent structure. The system-owned practices frequently benefit from a standing healthcare IT department that provides support by creating, managing and integrating data from multiple sources including CMS data files directly into reports, EHRs, or other applications. In addition, these practices can augment their IT staff as needed with vendors to provide needed functionality.

Independently owned practices usually do not have access to a substantial healthcare IT department to facilitate data management, and so frequently resort to either using vendors or manual processes, sacrificing speed and innovation.

For both types of practices, healthcare vendors are often still developing and enhancing their own health IT functionalities and products to support comprehensive primary care. Resource management recommendations include selecting a health IT vendor that specializes in data management and integrations, primary care providers should look for or develop a platform modeled after their processes and workflows.

Data Governance

Even in a small primary care practice, data governance is needed to maintain data quality, using rigorously enforced procedures and processes to ensure success. A strong data governance practice focusing on topics such as policy, security, compliance, architecture with a clear strategy ensures confidence in a practice’s data. Data governance determines how data is to be handled, such as processes for how data is gathered, cleaned and stored. A large number of people are not needed for appropriate data governance, but clear and consistent processes are required to ensure continuity.

Business and Clinical Intelligence

Once the needed data is prepared, business and clinical intelligence—including analytics and modeling—is used to turn this data into actionable information. After obtaining buy-in and prioritization from providers, a thorough understanding of the practice’s business and workflow is necessary.

Analytics include identifying what questions the practice is seeking to answer with the data and identifying appropriate and flexible healthcare predictive models and supporting algorithms, ideally resulting in reports and dashboards, to identify patterns or clusters of identifiable information.

These data groupings can be used to not only provide descriptive information about what events have happened, which we customarily think of as reporting, but can also be used with appropriate models to provide predictive analytics. With this information, it is possible to identify what events are most likely to happen, such as which patients are likely to respond to a given event like changes in care management, giving primary care providers strategy support from the patient to the practice level.

Common applications of predictive analytics by practices include risk stratification of patients. This looks at severity and concomitancy of disease, utilization of services, social determinants of health and behavioral health data, and number of claims to identify patients needing greater levels of care as well as readmission patterns.

Reporting of CMS feedback care delivery data includes reviewing referral rates, providing insights into specific patients, groups, conditions and service levels. This use of data supports not only primary care provider activities, but provides performance monitoring of initiatives and reporting to outside entities, including CMS.

Next Steps

Resource management, data governance, and business and clinical intelligence are critical to create actionable information and insights from the CMS data feedback file and other data sources. Careful consideration of these factors can result in informed initiatives and actions to improve the primary care providers’ patient population and reap the most benefits from participating in the Primary Care First model.

The views and opinions expressed in this content or by commenters are those of the author and do not necessarily reflect the official policy or position of HIMSS or its affiliates.

Originally published June 26, 2019