Why Population Health Management Strategies Require Both Clinical and Claims Data (Executive Report)

disease management cohorts.

To illustrate the issues, Medicare has different requirements for how skilled nursing facilities (SNFs), home health agencies, dialysis centers, and other ambulatory care venues can bill them. For example, a skilled nursing facility may bundle all of their services into a single bill or bill for certain services separately depending upon the patient’s Medicare status, but a dialysis center  is limited to submitting a single charge per visit. To add further variation, hospital outpatient services bill according to ambulatory payment classified (APCs).

Professional claims must also be included in any comprehensive analysis of a population of patients across the continuum of care.

As claims data from across the outpatient continuum builds up — all with differing requirements for how Medicare will reimburse and many lacking clinical detail – using this data to define and then manage a disease cohort becomes a very complex prospect.

Additional Challenges

Figure 3 - Health Catalyst's Source Mart Designer analyzes the structure of data sources and recommends pragmatic data types and column names with well-structured, consistent data fields.

Figure 3 – Health Catalyst’s Source Mart Designer analyzes the structure of data sources and recommends pragmatic data types and column names with well-structured, consistent data fields. (Click to Enlarge Image)

In addition to the previously mentioned difficulties that must be overcome, the following challenges also limit the sharing and merging of claims and clinical data between payers and providers.

  • Both types of data — but particularly claims data — tend to be “dirty,” i.e., inaccurate, incomplete, or erroneous, requiring the data be scrubbed before using them for analytics.
  • Matching patient identifiers from each record to create a single, accurate, comprehensive patient record is a complex and difficult undertaking. The more diverse the sources of data, the more challenging it becomes to bring them under a single patient identifier.
  • Payers have to choose which claims data is the most accurate and which version of any particular claim to use. Each claim has multiple iterations, including (at the very least) the original claim the provider submits and the edited, adjudicated claim. While it may be easier for a provider to access its own claims data, there is a risk it may not reflect the final form the data takes.
  • Patient privacy legal requirements require the patient’s permission to share the data between the provider and the payer. Some patients may have reservations about giving their payers access their clinical data, or they may have concerns about the security of their personal health information.
  • Providers have different EMRs and relationships with multiple payers with specific data requirements, which makes abstracting and sharing data problematic.


“A superior population health management initiative is possible by selecting the right technology and mapping outpatient codes to clinical care process families.”

The task to create a systematized, consistent method to match codes with cohorts is complex, but not insurmountable. The following two strategies will help providers and payers address these complexities to make it easier to combine claims and clinical data and create a superior population health management initiative:

  • Map outpatient codes to clinical care process families. Analytics vendors and other organizations  can undertake a significant effort to map the various outpatient codes to clinical care process families (such as diabetes and heart failure). For example, the vendor would look at skilled nursing facilities and analyze the billing codes they commonly use. They would then map different groupings of these codes to the care process family. This type of sophisticated mapping would provide a consistent methodology for assigning patients to a disease cohort regardless of care setting. Health Catalyst is an example of a vendor that’s in the process of developing these types of mapping efforts.
  • Select Flexible and scalable technology. ACOs and other organizations must select a technology infrastructure to manage their populations that is flexible enough to integrate and map new data sources quickly, on an as-needed basis. Because population health management and accountable care are in their infancy, the types of data sources required and use cases for the data will often change. Many analytics solutions targeted at the population health market today lack this flexibility. Organizations should look for technology that is adaptable enough to handle rapidly evolving needs. They should also ensure it is sophisticated enough to effectively leverage both clinical and claims data to accurately define and manage disease cohorts.

Consider this example of the need for flexible technology. When a diabetic patient visits a doctor in the outpatient setting for a foot ulcer, the doctor may or may not document the diagnosis of “diabetes.” Instead, the diagnosis may simply be “foot ulcer.” By using a flexible analytics solution, however, analysts can easily categorize this patient in the diabetic population cohort even if the patient is seen for the primary complaint of a foot ulcer.

Correcting these issues may sound simple, but without a flexible solution, it isn’t. The coming years promise more changes to the healthcare industry (new research, new technologies, etc.), not fewer, so the more flexibility built into the technology, the better a job the organization can do to create population health management solutions that deliver the expected results.


Achieving population health management is a critical goal for the healthcare industry. But the traditional way of using either claims data by itself or clinical data by itself won’t yield the valuable insights necessary to understand the data. Instead, payers and providers need to work together to overcome the challenges to combine claims and clinical data. Once the challenges are overcome, a more complete view of patients from across all care settings will allow for more accurate definitions of disease cohorts, identification of variations in care and waste, and a better measurement of the effectiveness of treatment. When coupled with the flexibility to adjust to changes in our rapidly growing understanding of what population health actually entails, data aggregation and analysis become a powerful tool for managing population health.

Read More About This Topic

The Evolution from Payer-Driven Care Management to Provider-Driven Population Health Management
Brent Dover, President

New Payer / Provider Arrangements That Will Drive Better Process Improvement
Bobbi Brown, Vice President, Financial Engagement

The Best Way to Make Payer and Provider Healthcare Data Accessible
Bobbi Brown, Vice President, Financial Engagement

Population Health Management: Implementing a Strategy for Success
David Burton, MD, Senior Vice President and Former Executive Chairman

The Four Building Blocks of Population Health Management (webinar, slides, or transcript)
David Burton, MD, Senior Vice President and Former Executive Chairman


  1. Centers for Disease Control and Prevention. “Chronic Diseases and Health Promotion.” 2012. http://www.cdc.gov/chronicdisease/overview/index.htm?s_cid=ostltsdyk_govd_203
  2. Kung HC, Hoyert DL, Xu JQ, Murphy SL. “Deaths: Final Data for 2005.” National Vital Statistics Reports 2008; 56(10). https://www.cdc.gov/nchs/data/nvsr/nvsr56/nvsr56_10.pdf
  3. Kung HC, Hoyert DL, Xu JQ, Murphy SL. “Deaths: Final Data for 2005” National Vital Statistics Reports 2008; 56(10). https://www.cdc.gov/nchs/data/nvsr/nvsr56/nvsr56_10.pdf
  4. Centers for Disease Control and Prevention. “Prevalence of Doctor-diagnosed Arthritis and Arthritis-attributable Activity Limitation.” 2003–2005. http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6244a1.htm
  5. Centers for Disease Control and Prevention. “National Diabetes Fact Sheet, 2011.” www.cdc.gov/diabetes/pubs/pdf/ndfs_2011.pdf
  6. Dobson, Allen, DaVanzo, Joan, et al. “Clinically Appropriate and Cost Effective Placement (CACEP): Improving Health Care Quality and Efficiency.” 2012. http://www.ahhqi.org/
  7. Wellness.wikispaces.com. “Using Claims Data.” http://wellness.wikispaces.com/Using+Claims+Data



Page 3 of 3
1 2 3
Next Page