Population Health and the Role of Health Analytics
Analytics is poised to transform multiple areas of the healthcare sector, and we have already explored the role of health analytics in value-based care. For this article, however, wanted to look at another area where analytics can be the driving force behind higher quality care and more efficient operations: population health.
Although the term population health has not always been well-defined, the concepts it encourages us to think about have been explored throughout history. During World War II, researchers were able to link medical conditions, including rheumatic heart disease and peptic ulcer, with changes in social conditions (e.g., unemployment). Population health approaches are increasingly important for ensuring good outcomes for individuals and reducing the overall societal cost of providing care because many medical conditions today are chronic, and the rate of these conditions is increasing.
Population health also presents a good example of how healthcare informatics and analytics come together for achieving success. Because the challenges associated with implementing population health initiatives involve complexities from a data governance standpoint as well as an organizational leadership standpoint, understanding population health is highly valuable for professionals in both fields.
Population Health vs. Population Health Management vs. Population Health Analytics
|Population Health||Population Health Management||Population Health Analytics|
|An approach to healthcare that considers the health outcomes of a group.||The strategies, processes and tools related to managing and compiling patient data for the purposes of implementing population health practices.||The application of quantitative methods to population health management to reach more sophisticated insight about groups or determine the most optimal strategies for treating health issues within that group.|
What is Population Health?
Population health as a purposeful and strategic approach to healthcare has come more into focus in recent history. When researchers in 2003 noted that population health had not yet been fully defined, they proposed the following: the health outcomes of a group of individuals, including the distribution of such outcomes within the group. Organizations such as the Healthcare Information and Management Systems Society have continued using similar definitions, and the core idea behind the term is to look at health at the group level by compiling data from different disciplines within and outside the health sector— including public health organizations, academic institutions, government and other entities.
What is Population Health Analytics?
While definitions may vary depending on who is describing the term, we use population health analytics to describe the act of applying quantitative methods and technology to reach advanced insight about a group. Whereas population health management covers the actions and tools associated with compiling relevant data, population health analytics involves taking the next step, which is to apply analysis to that data in order to reach actionable information. For example, analytics can be used to develop a more accurate model of the typical disease trajectory for a given group of individuals. Providers can then use this model to offer more proactive care that mitigates the progression of the condition.
Challenges and Solutions to Implementing Population Health Analytics
Although population health analytics presents both providers and payers with considerable opportunity, it can be challenging to implement the technology, processes and organizational change necessary to make these efforts successful. One of the first challenges that tends to emerge is the difficulties associated with bringing together data from many different sources.
Challenge: Data Integration and Governance
As reported by Health IT Analytics, comprehensive data about patients and populations can be difficult to integrate and the combination of volume and variety can be overwhelming for organizations attempting to compile this information. Other issues stem from a regulatory standpoint, as there is growing concern with how predictive models and advanced algorithms may be handling patient data.
Solution: Stronger Collaboration Between Healthcare Informatics and Analytics
With the challenges presented by growing data volume and variety, professionals within the discipline of healthcare informatics will continue to play a central role within their organizations. The major change to consider is that the field of healthcare informatics itself is likely to become more collaborative and more deeply entwined with other healthcare disciplines.
This is particularly likely with the relationship between healthcare informatics and analytics. In addition to the traditional data stewardship responsibilities informatics professionals have always had, they may also be challenged to provide strategic guidance on how best to capture and integrate data that their organizations may need for health analytics initiatives.
Challenge: Non-Medical Variables in Population Health Management
Related to the challenge of bringing together and using raw data is the fact that population health analytics put more pressure on healthcare professionals to consider information that isn’t entirely medical. Even when looking at the purely medical data, we can expect a massive influx of new information with trends such as wearable technology contributing to far more data availability than ever before.
However, not all of the information related to population health may be directly medical. For example, unemployment has been linked to higher risk of heart disease within specific patient groups. Developing a model for predicting high-risk patients based on this finding would require both medical and employment data for a large enough group to verify the effectiveness of the model. In addition, treating individual patients would require knowledge of both their medical history and employment status, as well as any other non-medical variables considered.
Solution: Modernize IT Alongside Organization-Wide Objectives
One of the major challenges for organizations looking toward solutions like analytics is that they are often faced with a mixture of legacy systems and hardware, and newer platforms. The health sector has seen this with the adoption of electronic health records (EHRs)— in addition to migrating paper records, organizations were faced with changing processes to ensure new patient data was entered correctly and efficiently.
A 2018 report from McKinsey noted that this trend affects all organizations as they push toward “digital reinvention.” The healthcare sector has started to feel changes to core operations and processes as a result of investments in digital solutions like EHR adoption.
Health organizations can start by looking at their short- and long-term goals to see where population health analytics may best fit. For example, a report from the Association of American Medical Colleges highlights the advantage of leveraging predictive models for allocating resources within population health management programs, as well as for identifying potential high-risk patients.
Modernizing health IT will help to ensure that all relevant patient data is captured accurately and can be integrated alongside advanced analytics more easily. Digital transformation could also yield benefits for patient experiences, such as by providing more transparency into the care being offered, through patient record access portals or other platforms designed to improve information sharing between providers and patients.
Challenge: Transforming Data into Actionable Insight
Identifying the issues and traits of a population is only the first step. This information must then be used to create actionable insight. One part of this issue is that the majority of organizations struggle to successfully adopt analytics.
This challenge then persists, as one of the keys to success with analytics in healthcare is to adopt a culture of continuous improvement. Trends like greater adoption of value-based healthcare, increasing rates of chronic conditions and changing regulatory environments put considerable pressure on health organizations to make smarter investments and constantly improve the quality of services they provide.
Solution: Build Health Analytics Proficiency Among Leaders
While we often think of analytics as a technical challenge, there are both technology and organizational culture elements to implementing health analytics. In fact, some of the biggest challenges associated with these initiatives stem from a lack of understanding throughout management and a lack of organizational alignment to overall analytics strategy, according to a 2017 report from New Vantage Partners. More specifically, these challenges range from difficulties in creating a well-defined and consistent data governance strategy to difficulty in defining the roles associated with data and analytics leadership.
Challenges like these suggest that a proper understanding of analytics must be encouraged and embraced throughout leadership and middle management so that initiatives are guided by both business strategy and technology expertise.
Healthcare leaders can help to bring together the expertise needed to solve both the technical and IT infrastructure challenges that arise from analytics and ensure that analytics projects are guided by common goals among all stakeholders. It is helpful to start by looking at specific problems to solve, rather than integrating vast amounts of data without an objective in mind. By starting with a narrower scope, health leaders can avoid wasting unnecessary resources on initiatives that provide little value. Furthermore, when an approach does yield value for population health analytics, organizations can build and improve their initiatives with a purposeful, strategic approach.
Effective Population Health Analytics
In addition to standard analytics tools, population health presents a huge opportunity for innovative technology such as machine learning and artificial intelligence. These tools often require vast amounts of historical data which is used to train algorithms for identifying risk factors, conditions and for many other use cases throughout healthcare.
As an example, one study highlights how machine learning can be utilized to predict the likelihood of acute kidney failure in hospitalized patients, leading to proactive measures that reduce the possibility of long-term kidney damage.
Success stories like these suggest that organizations and healthcare professionals who can successfully navigate the challenges of health analytics will not only be able to better care for their patients but can contribute to overall societal good by reducing the costs associated with long-term and more severe conditions. As with other areas of technology, it is important to carefully evaluate solutions and take a measured approach so that health professionals can fundamentally transform their ability to provide quality care.
About the MSQM: Health Analytics at the Fuqua School of Business
The online health analytics master’s program at the Fuqua School of Business was designed to empower health professionals and leaders with the interdisciplinary knowledge they need to drive analytics forward in their organizations.
Our curriculum combines business expertise with health sector knowledge, giving graduates the technical proficiency and industry context needed to transform their organization’s data into valuable, actionable insight.
To learn more about the online MSQM: Health Analytics program and download a free brochure, fill out the fields below. If you have additional questions, please call +1.877.755.1342 to speak with an admission counselor.