Moving the Needle for Population Health Management

While population health management has been a priority for United States healthcare for a decade now, many populations still face care access barriers, high cost of care, and environmental factors (behavioral and social determinants of health) that affect their health outcomes.

In a 2021 Commonwealth Fund report comparing the performance of eleven high-income countries, the U.S. spent far more of its gross domestic product (GDP) on healthcare but ranked last in terms of access to care, equity, administrative efficiency, and health outcomes. The nation continues to struggle with affordability, spending almost 18.3% of its gross domestic product on healthcare in 2021 without seeing improved patient health. In this article, we will explore the underlying problem that contributes to higher costs and poor outcomes and paint a picture of what population health management can look like when these problems are addressed.

In this article, we discuss:

• Healthcare’s Data Problem and Impact on Population Health Management
• Becoming an Agent of Change

What is Population Health Management?

Population Health Management (PHM) is a comprehensive approach to managing and improving the health of a group of individuals. It focuses on improving health outcomes and reducing healthcare costs by looking at the overall health of a population rather than one person at a time. PHM leverages data, technology, and clinical best practices to proactively identify risk factors, prevent illness, and help members access quality care when needed.

Why is Population Health Management Important?

>Population health management is a critical aspect of healthcare that has become increasingly important in recent years. This is due in large part to the changing landscape of healthcare, which includes a shift from a fee-for-service model of care to a value-based care model that focuses on improving health outcomes for entire populations rather than just treating individual patients.

There are several reasons why population health management is so important. One of the most significant is that it enables healthcare providers to identify and address population health issues proactively, rather than waiting until individuals become sick and need treatment. This can help prevent the development of chronic diseases and other health problems, which can be costly to treat and can have a significant impact on the quality of life of affected individuals.

Population health management is also important because it can help healthcare providers to better understand the needs of their patients and develop more effective treatment plans. By collecting and analyzing data on health outcomes and other factors, providers can identify patterns and trends that can inform their approach to care. This can lead to more personalized, effective treatment that is tailored to the unique needs of individual patients.

Healthcare’s Data problem

Many factors contribute to the United States’ combination of high spending and poor outcomes, such as interoperability, health equity, and lack of price transparency, but one factor that complicates all other population health management challenges is data, and more importantly, how we use the data we have. We certainly do not have a lack of data considering healthcare organizations collect 878% more data than in 2016. However, while we are data-rich, we are information-poor. Much of the data collected is not in an actionable form, leaving care team members to manually aggregate and interpret data and then make decisions that result in care plans that may not be consistently applied. Medication management, in particular, currently requires heavily manual processes to find patients in need of care, reconcile medications, create evidence-based care plans, and communicate care plan changes to patients and prescribers.

To make population health management truly successful, health plans, providers, and patients need actionable data that fits the following criteria:

• Usable: data that can be easily aggregated and processed.
• Trusted: data that care teams have confidence is healthy and ready to be acted on.
• Timely: care teams have the right data insights translated into actionable insights at the right time to reduce data overload.
• Personalized: the scope of the data includes insights that personalize care for each individual, and takes into account their unique and evolving circumstances and needs, including, but not limited to, social, financial, behavioral, and clinical.

Once care teams have usable, trusted, timely, and personalized patient data accompanied by contextualized evidence-based recommendations, they can improve patient outcomes and cost of care across large or small populations. Here are three ways this kind of data generates improved outcomes and reduced costs in population health management:

1. Proactive instead of reactive population health management

In the past, under fee-for-service models still predominant today, healthcare professionals used sporadic, episodic office visits to manage chronic diseases without coordination of care with other members of the care team. Optimal clinical outcomes are harder to achieve with this care model, as it becomes harder to proactively treat chronic conditions. Although health data is readily available, the data is often in disparate formats and multiple healthcare systems, making interpretation difficult and time-intensive. Care teams need a scalable way to translate that data into actionable insights that allow them to predict outcomes and address gaps in care.

Artificial intelligence is one approach care teams use to scale the analysis of patient data, derive insights, and improve care in a manner spanning individual patients and cohorts of patients, care team members, and populations. For instance, one health plan invested in Arine’s AI-driven medication intelligence solutions for its members and found that they were able to identify and resolve issues with chronic disease management that had previously resulted in lower Part D star ratings. Arine worked with the health plan to execute a quality improvement plan that incorporated predictive analytics to suggest outreach activities, automate comprehensive medication review (CMR) workflows, increase preventive screenings, and improve medication adherence rates. By the end of the year, the health plan surpassed the health outcome goals for its population, increasing its star ratings from two stars to five stars. Ultimately, they also expanded Arine’s technology to Part C chronic disease management initiatives, such as A1c control, blood pressure care, and breast cancer screenings. Arine achieved 5 Stars across all of the health plan’s Part C and D measures the following year.

2. Personalized care for diverse populations

One of the most significant challenges to population health is the widening health disparities among subpopulations. For healthcare professionals to address health equity challenges, they must look beyond the clinical setting and leverage social, economic, and behavioral data to personalize care for each individual.

During the pandemic, one health plan worked with Arine to address lower medication adherence rates among its Black and Latinx communities. The health plan utilized Arine’s machine learning technology to uniquely engage its members, such as sending culturally appropriate messages and reminders. By utilizing Arine’s platform, the health plan was able to achieve its goal of over 90% adherence to statins, antihypertensives, and anti-diabetes medications. Most importantly, they ensured patients who needed these medications were able to receive them — improving quality of life, decreasing the overall cost of care, and preventing medication-related adverse outcomes.

3. Improved efficiency to deliver outcomes at scale

While data has enabled healthcare providers and plans to personalize care and be more proactive, often this kind of care was difficult to deliver at scale, requiring large, expensive care teams to intervene and follow up with patients. Technology now has the capability, however, to improve efficiency for care teams, enabling smaller, more agile teams to work with large populations. Machine learning and AI-driven recommendations and workflow engines assign patients to the appropriate care team member and generate automatic and personalized care plans for each patient, allowing care teams to spend their time focusing on each patient instead of administrative tasks. One population health administrator noted the following about Arine’s AI-driven Medication Intelligence Platform, “The Arine Platform allows my team to serve more patients, better and faster. The Arine Platform is like a highly capable, senior member of our patient care team. It's as if we've added a new colleague whose sole purpose is to deliver tailored, actionable recommendations to our patients."

Becoming an agent of change

Population health is more than planning – it's taking the right action for each patient, and that starts with usable, trusted, timely, and personalized data.

To find out more about how you can optimize your population’s health, download Arine’s white paper: