With more and more data available to healthcare providers, it’s difficult to know where to begin sourcing data for your population health management program—especially in the early days of program creation. However, choosing the right data to start with is critical to getting the fastest results from your program and demonstrating its value to stakeholders. Here are the data sources we suggest getting started with.
Keep in mind that the value of each data set is strengthened when paired with others to create a holistic picture.
Claims data.
This includes patient demographics, diagnosis codes, dates of service and cost of service. The vast majority of providers will only have access to in-network claims data, so the value here mainly lies in patient identification—linking specific patients with other data sets that provide clinical details.
Electronic Health Records (EHR) data.
This is the data that begins to include clinical details. These records usually contain information about the process of care, provider impressions, patient concerns (even those that don’t result in a diagnosis), vital signs, medications, allergies, imaging reports, lab data, and immunization dates. The challenge here is that EHRs may contain unstructured data like hard to read or incomplete text fields. There is no guarantee that the person filling out the EHR information is doing it correctly or accurately, and some file formats like PDFs may need additional steps to process before you can add them to the data sets and analyze them.
Social and community determinants of health data.
This includes average incomes, language proficiency, food availability, violence rates, transportation access, unemployment rates, education levels, etc. If you can access this data (public health organizations are a good source) it can help you understand what your patients face when they step outside your doors—a major factor in their health outcomes. Today, there aren’t many EHR systems that offer fields to input this information, however, it’s a critical piece for any population health program to reach its fullest potential. The good news is that the Centers for Medicare & Medicaid Services (CMS) implemented an Equity Plan for Improving Quality in Medicare which prioritizes the collection and analysis of this information, pushing the industry to develop strategies to do so.
Patient-generated health data.
This includes satisfaction surveys, patient-reported outcomes, and even data from fitness wearables. A lot of this data is available but it’s mostly unstructured, and integrating this information into population health workflows is a fairly new challenge for the industry. Many payers are offering benefits to patients who integrate smart health monitors, which will continue to encourage adoption and availability of personal health data.
Prescription and medication adherence data.
Medication is critical for disease management, so this information is extremely valuable in understanding who may benefit from targeted intervention. A patient may go to the emergency department and receive medication that conflicts with the medication instructions from their primary care physician. This issue of medication reconciliation is a major problem for doctors and patients today, so integrating this data set into your program can put you on the path to identifying where issues are prevalent and preventing them in the future.
Now that you have an idea of where to begin your data collection efforts, you probably want to know more. Our position as a data analytics provider in the Midwest has provided us with the opportunity to work with providers just like you, so we understand the challenges you face and the opportunities you have available. Contact us to find out how you can give your care delivery and population health program a major advantage with data.
