There were times when I dreaded getting my report card in school. Not because I was a bad student, but because in some instances I didn’t know what letter grade my teacher was going to assign me. I never liked this guessing game. I preferred to receive assignment grades and progress reports throughout the semester, so I could track exactly how I was doing and what my final grade would be on my report card.
For some healthcare organizations, population health planning can be a lot like this. Just like it’s helpful to know what informed your grades on your report card, it’s critical for organizations to have access to all data that can inform their population health programs. This is where, in addition to EHR data, organizations are using claims data.
As more organizations are collecting and integrating claims data due to government programs and pay-for-performance contracts, they’re also using this data to inform the success of their population health initiatives. Claims data provides opportunities for organizations to improve patient health and satisfaction, implement appropriate outreach, and evaluate utilization. Below I’ve outlined some ways your organization can use claims data to get a full report on patient population health.
Using claims data to inform population health initiatives
Claims data can help your organization advance several financial and population health goals. Here are some key ways healthcare organizations are using claims data.
- Medication adherence tracking – This helps organizations ensure patients are getting prescriptions filled and re-filled on time.
- Patient history reporting – Claims data typically includes a breadth of information from across multiple healthcare organizations, allowing your providers to get a better picture of what's in a patient's history.
- Tracking of preventative services – The breadth of data also allows your organization to track the preventive services that patients have had in the past (for example, flu vaccines and colonoscopies) helping to target outreach to patients in need of these services.
- Aligning with payers’ view of quality – Quality measurement calculations based on claims data will better reflect what your payers see when they measure your performance on pay-for-performance contracts or reach out to you with quality initiatives.
- Utilization and resource reporting – When claims data includes payment information, you can track the expense and use of your healthcare services.
What to consider when analyzing claims data
Here are three important considerations to keep in mind when analyzing claims data:
1. Timing and frequency
Claims can often take up to three months (in some cases longer) to adjudicate after a service has been rendered. Here are some factors that may add to this three-month lag time:
- How often the payer is extracting the data
- When the payer sends and/or your organization collects data from the payer
- The frequency that data is loaded into your data warehouse or database
- How often a given report may be run
2. Data quality and interpretation
When receiving data from a payer, you'll want to make sure that the files you receive have clean data, are formatted correctly, and follow consistent standards. This will help streamline the extraction, transform, and load (ETL) process for your data warehouse or database. In addition, it's important that you and your payer are on the same page about what data you are receiving. For example, is the payer only sending adjudicated claims or are they including claims that are in process? You’ll also want to consider how the data should be interpreted. You may want to ask yourself, “What would reversed claims look like in this data?” or “How would an appealed claim look?”
3. Characteristics of the data
Finally, it's important to understand what is and isn't included in claims data. Claims data will (likely) not contain lab results or any clinical data that isn't tied to services billed to insurance (i.e., blood pressure screenings, BMIs, etc.). Claims data also does not encompass any data recorded in notes – again, it only includes what was billed to insurance. It will not contain any services paid for by the patient or another insurance company.
Claims data does not include a problem list either; it only comprises the diagnoses on the individual claim, each of which may be an acute condition or a chronic issue. Therefore, many quality measures based on claims data look for two claims with a given diagnosis before assigning a patient a chronic condition (one claim may simply be from a test to rule out a diagnosis or may have been billed in error).
What is in claims data, on the other hand, is a set of services, diagnoses, and medications that typically have been coded under a standard national system, allowing for national comparisons, benchmarking, and consistent analysis.
How Nordic can help
Whether you're new to evaluating claims data, need some help in analysis, or are looking for development, we can help. Our data & analytics team is experienced in claims analytics, reporting, and data warehousing.
If you're using Epic's Caboodle Data Warehouse, Nordic's analysts can help your organization with the development and implementation of SQL Server Integration Services packages, automate data acquisition from the payer, build custom Caboodle reporting tables, and use Epic's Caboodle APIs to notify administrators when work queue or error log issues arise. We can also show your organization how to use claims data to update registry metrics and implement and customize Epic's Cost & Utilization dashboard.
Please reach out if you’re interested in learning more or would like to speak with our data & analytics team.
Mo Mehta is a senior Data & Analytics consultant at Nordic. He has nine Epic certifications, and his broad experience in healthcare quality measures and analytics includes projects utilizing Clarity, Caboodle, Healthy Planet, and Constellation. Mo helps client partners with ETL development, analytics strategy, and quality measures implementation.