Health Data

Reduce Costs and Improve Outcomes Using Data Analytics in Population Health

Population health has been around for some time, but since the COVID-19 pandemic, population health efforts have soared. Population health has been defined in many ways, but the common theme of all the definitions is to achieve positive health outcomes across all populations.1 To do this, data is needed to bridge the gap. Population health relies on information technology to manage and analyze enormous amounts of data to provide actionable reports.2 With the endeavors of data analytics and population health, populations across all spectrums have been positively impacted.

Gathering and Using Data

The use of data analytics is becoming more common in healthcare for many reasons. Data analytics methods include analyzing raw data and making conclusions about that information.3 These methods are very beneficial in population health. Data can help identify populations in need of care, measure the care that is provided, and deliver care to the correct people.4

Data is becoming more available and is collected on a regular basis. Data comes from a variety of sources, such as patient demographic data like age and gender, financial data such as insurance information, and clinical data such as medical history and lab results. Collecting data and analyzing it can determine the big picture for the patient and the population. Big data is the process of creating value from data collected,5  and it is gaining popularity in population health.

Big data can be gathered from various sources and various properties that all contain information about a population. When the information is gathered from the various properties, it is considered a big data chain. In order to get usable information from the big data chain, the organization must manage, process, and analyze big data.6 For example, big data from the healthcare industry would include financial data, clinical data, biometric data, social media data, data from research activities, and patient satisfaction data. Population health would use this data to determine many factors, including:

  • Understanding environmental factors that influence a person’s overall health
  • Considering physical and social determinants of health
  • Shifting thinking from one-size-fits-all to value-based care method.
  • Increasing quality and accessibility of care for all populations
  • Reducing rising healthcare costs

These factors combined with a team of professionals that focuses on increasing quality of care and population health management are needed to help with the evolving healthcare landscape.

The Population Health Value Framework

In 2016, a team of health professionals at the University of California Los Angeles Health System implemented a system-wide population health approach to promote value-based care. Preliminary analysis deemed the framework successful.

The purpose of the health value framework was to identify patient populations with high expenses and promote value-based care that guided value creation.7 The framework was formed by four priorities:

  1. Identify patient populations with high expenses and reasons for spending.
  2. Create design teams to understand the patient story.
  3. Create custom analytics and spending-based risk stratification.
  4. Develop care pathways based on spending tiers.8

The subpopulations in the study were cancer patients, chronic kidney disease patients, and patients with dementia. For each patient subpopulation, teams were formed to address patient needs and to help identify the reasons for high spending.

Every patient had their own care pathway.9 The care pathways were tailored to individual patients to address their needs. All patients received the necessary interventions, but the level of interventions varied depending on their illness and treatment. Custom analytics were created to use clinical and administrative data for each subpopulation. Risk-stratification measures were added to subpopulations depending on early data analysis. Targeted care pathways were developed based on spending tiers. The analyzed data determined tiers for each subpopulation and the care pathway best suited for each patient.10

Early analysis showed a monthly reduction in inpatient bed admissions among dementia patients and a 2 percent decrease in hospital admissions for chronic kidney disease patients.11 Tailored care pathways for individual patients can significantly reduce costs driven by admissions and readmissions.

The health value framework showed how redesigning care could reduce costs, and improve outcomes and patient experiences. UCLA Health has moved forward to include all high-cost subpopulations with chronic conditions in its health value framework.12 The framework  is designed organize care across all specialties and grow a culture for value-based care.

Application of Data Analytics in Population Health

There are four different types of data analytics:

  1. Descriptive analytics examine and describe events that have already happened.
  2. Diagnostic analytics seek to understand the cause of an event.
  3. Predictive analytics examine historical data and past trends in order to predict future events.
  4. Prescriptive analytics identify particular actions individuals should take to reach  particular outcomes.

Data analytics serve different functions. For example, descriptive analytics can be used in population health to determine how contagious a particular virus is by examining the rates of positive tests for that virus in a population over a specific time period. Predictive analytics can be used in population health to forecast the spread of diseases by examining case-specific data for that disease from previous years.13

Improving Overall Health

Data analytics tools are particularly useful for population health. As healthcare has moved to value-based models of care, organizations have placed priority on identifying interventions that can improve the overall health of the population. Data is vital for measuring successful interventions that can improve health outcomes. Data analytic tools and techniques are particularly useful for population health in measuring and predicting disease outbreaks.

Notes

1. Appold, K. 2020. “Confused about population health? You’ve come to the right place.” Managed Healthcare Executive, (30)10. https://www.managedhealthcareexecutive.com/view/confused-about-population-health-you-ve-come-to-the-right-place

2. Ibid.

3. Batko, K., & Slezak, A. 2002. “The use of big analytics in healthcare.” Journal of Big Data, (9)1. https://doi.org/10.1186/s40537-021-00553

4. McNemar, E. 2021. “What is the role of data analytics in population health management?” Health IT Analytics. https://healthitanalytics.com/features/whatistherole-ofdataanalytics-inpopulationhealthmanagement

5. Ibid.

6. Ibid.

7. Reshma, G., Roh, L., Lee, C., Reuben, D., Naeim, A., Wilson, J., Skootsky, S. 2019. “The population health value framework: Creating value by reducing costs of care for patient subpopulations with chronic conditions.” Academic Medicine, (94)9, 1337-1342. https://doi: 10.1097/ACM.0000000000002739

8. Ibid.

9. Ibid.

10. Ibid.

11. Ibid.

12. Ibid.

13. Cote, C. 2021. “3 applications of data analytics in health care.” Harvard Business School Online. https://online.hbs.edu/blog/post/data-analytics-in-healthcare

Rachel Ellison (rachel.ellison@louisiana.edu) is an associate professor and health services administration program director at the University of Louisiana at Lafayette.

Lesley Clack (lclack@fgcu.edu) is an associate professor and department chair of health sciences at Florida Gulf Coast University.

Discover AHIMA education related to data analytics.

Tags: value-based care , population health , COVID-19 , data analytics , Big Data