RTI Health Advance uses cookies to offer you the best experience online. By and clicking "accept" on this website, you opt in and you agree to the use of cookies. If you would like to know more about how RTI Health Advance uses cookies and how to manage them please view our Privacy Policy here. Click "accept" to agree.

Accept
Stratifying Data Is Key to Population Health Management Success
Article

Stratifying Data Is Key to Population Health Management Success

Population health initiatives require specific, validated, and timely data. Organizations that efficiently carve through large swaths of information are best positioned for success.

Population health management involves identifying cohorts of patients, sometimes thousands of people, at risk for costly healthcare events that negatively affect outcomes and quality of life. Healthcare providers and payers then implement interventions for these cohorts of patients with the anticipation to curb healthcare spending and provide value-based care.

Who belongs in these cohorts? What interventions can improve their health? How will you know if your efforts are successful? The answers lie in a sea of data.

It only takes a subset of information to guide population health management efforts. Carving out a discrete dataset can minimize administrative burden, freeing up organizational capacity to focus on improvement. Find out how top organizations achieve population health success through thoughtful data stratification.

How stratifying data aids successful population health management efforts

Data should guide every milestone in a population health initiative, which is why data stratification is essential. Stratified data help healthcare organizations:

  • Identify patient cohorts: Cohorts may include people with a particular diagnosis, combination of diagnoses, or those who overutilize acute services. These individuals are likely to experience adverse outcomes based on other factors, like prescription medications that increase fall risks.
  • Assess the risk of adverse outcomes: Artificial intelligence and machine learning help determine risk in a cohort or individual members. These population health data management methods use technology and sophisticated algorithms to transform large data sets into meaningful insights.
  • Determine measures of success: In people with diabetes, a measure of success may be a reduction in blood sugar (A1c) levels. Capturing baseline A1c levels for cohorts of patients with diabetes and regularly monitoring levels to confirm patients are receiving the most effective care is critical to assessing the effectiveness of population health management initiatives.
  • Lay the groundwork for future efforts: Learnings from previous population health efforts should inform future strategies. Reliable data help you measure how far an organization has come and show where future opportunities lie.

When it comes to population health, good data matters

When moving the needle on care and outcomes for a large patient cohort, timely, actionable, and reliable data are essential. Not only is the well-being of your patient population on the line, but your organization likely has some financial skin in the game through a growing number of incentive payment programs.

There are many potential population health data sources, including

  • Prospective and retrospective claims data analysis from payers
  • Medical, surgical, pharmacy, and mental health encounter data
  • Facility-based and statewide patient registries of chronic disease
  • Quality metrics from CMS or local public reporting agencies
  • Qualitative inputs, such as clinician notes from your organization’s EHR system
  • Patient-reported data from risk scoring tools, such as the Patient Health Questionnaire (PHQ-9) for depression
  • Race, ethnicity and language data (REaL) and other information about social determinants of health (SDoH)

Aggregating population health data from multiple sources fosters comprehensive analysis

Stratifying data involves combing through large amounts of information. Aggregating data from various sources before stratifying it complicates efforts but provides a complete view of risk, care delivery, and outcomes.

For example, you can determine precisely how many patients in a heart disease cohort are receiving high-acuity services. Patients with poorly controlled symptoms are more likely to frequent the emergency department and need costly inpatient stays.

Data stratification, especially using multiple data sources, provides clues as to *why* patients are seeking high-acuity services, providing valuable insights that inform healthcare interventions at the individual patient level or a cohort level and drive quality improvement initiatives.

Stratified data makes it easier to get answers to important questions about patients, such as

  • When was the last time they saw their cardiologist?
  • Are they filling (and refilling) heart disease prescriptions?
  • Are they missing out on value-based services, like care coordination?
  • Are there comorbid conditions like depression or diabetes that providers might not be aware of but increase the risk of adverse outcomes?
  • Could social determinants of health be impacting care access or medication compliance?

Technology and many hands make light work of data stratification

When it comes to stratifying data for population health initiatives, you don’t have to go it alone. Strategic partnerships and specialized software can give you a leg up.

These three strategies can make light work of data stratification:

1. Use off-the-shelf tools

Many EHR systems offer plug-ins for population health. This solution is useful for organizations not ready to take the plunge with multiple data sources. Plug-ins use EHR data, including diagnostic and clinical inputs, to identify high-risk patients.

They also help your organization manage population health initiatives with smart features, like

  • Data dashboards, which make raw data easier to understand
  • Identification of patients who might benefit from care coordination
  • Quality metric tracking

2. Leverage accountable care organizations (ACOs) as partners

ACOs are partnerships among hospitals, physician groups, and other clinicians that streamline care delivery. They offer an opportunity to bolster population health efforts by pooling resources.

Various groups in an ACO may collaborate to implement data stratification efforts:

  • Health informatics professionals use technology and information systems to store and manage massive amounts of data securely.
  • Data analysts gather, cleanse, and analyze datasets to inform critical business decisions.
  • Quality improvement coordinators oversee initiatives to raise the level of quality an organization provides. Stratification of data is particularly important to closing gaps in quality of care and achieving health equity.

3. Participate in local health information exchanges

Health information exchanges offer multiple healthcare organizations a convenient, secure option to share vital data. They enable population health programs to utilize data beyond their organization’s firewall.

Participating in an exchange makes it possible to capture more of the services patients are receiving, even if they are from

  • More than one local health system
  • Ancillary providers, including anesthesiologists and radiologists
  • National pharmacy or lab chains that are not part of an integrated system
  • Providers in different parts of the country

Final thoughts on data stratification in population health management

Population health management and data stratification go hand in hand. For an organization to raise the level of health and well-being in at-risk populations, data must guide the way.

Stratifying data can be challenging, especially when it comes from multiple sources. Making smart use of available resources can ease this burden so that organizations can focus on optimizing care and outcomes for vulnerable populations.

Learn more about our population health solutions.

Stay informed on your favorite topics