Traditional electronic health records are designed around the care delivered during encounters with individual patients. However, incentives for improved outcomes, value, and expectations of those paying the bills, including patients and taxpayers, require more efficient and effective care processes and decision-making.

This is driving a shift in the healthcare delivery system and the way we think about care processes. Certainly, it includes the individual patient standing in front of a provider or virtually accessing a service.

But it also now includes analysis and understanding of specific populations, such as those managing chronic conditions (diabetes, hypertension, asthma, etc.) and patients discharged following cardiac surgery or from intensive care. Targeted populations could also include teens requesting pregnancy or STD testing, anyone with multiple ER visits or smokers of any age.

Providers with a vested interest in improving patient outcomes and the health status of their patient population are realizing the need for more robust information and data. This includes the ability to access additional sources of patient information for analysis, stratification and assignment into risk categories to facilitate prioritization and more directed management of care.

For example, a primary care provider might want to start with identifying patients who don’t have a recent A1C level noted in their chart. Perhaps a specialist ordered the test result, but it hasn’t made its way into the primary care provider’s record.

Maybe the patient has forgotten, has had difficulty arranging transportation, or is experiencing some other challenge that the care team can address. Claims data to identify patients receiving care from an endocrinologist, or lab results ordered by another provider would help with the prioritization of patients needing to be contacted.

Adding an analysis of socioeconomic and demographic data could help refine the results to identify those most at risk and in need of additional support. Or, perhaps the primary care provider wants to identify patients with an A1C level of eight or above and determine if they also have co-morbidities.

Again, an analysis of this population would produce a list of patients who could be stratified according to risk to create more manageable groups of individuals for more targeted strategies. This could include development of a short list of patients with co-morbidities where clinical practice guidelines might be insufficient for addressing their needs.

This is the power of data analytics.

A recent article in Health Management Technology outlined the five pillars of data analytics. They include:

  • Build a data architecture that leverages claims data to build a more complete picture of your assigned patients, and where else they may be receiving care.
  • Include predictive modeling and risk analytics tools that are tuned into the characteristics of patient populations to help to identify those who are at risk or adjust risk based upon patient characteristics. These include those with multiple chronic conditions, high utilizers of inpatient/ER/urgent care, age, homelessness, etc.
  • Stratify populations to identify those needing case/disease management or empanelment to the MD instead of the PA or NP. Break high-risk patients into smaller groups that can be more easily managed or flag them for care coordination teams.
  • Leverage other sources of data by integrating claims, a health risk assessment survey, biometric data sources, activation scores, registries, etc.
  • Monitor trends, utilization, quality, outcome, cost and other metrics to assess the impact of improvement efforts. Use lessons learned to improve processes and identify tools needed to better support patient needs, including opportunities for telehealth or the engagement of patients/family caregivers with technology.

Analyzing data to better understand patient needs from both the macro and micro levels will help providers and their healthcare organizations create more focused and effective strategies that result in wise use of limited resources. In the end, it will also help improve the health status of individual patients.