Claims 2.0: Developing a Clinical Data Strategy in the Journey to Value Based Payment
I am the first to admit the value of claims for analytics and reporting. Many years ago when I managed the analytics department for two different managed Medicaid plans, claims data was our bread and butter. We had best of breed systems, ETL1, a data warehouse, data marts and a team of well-versed analysts. We were able to create cost and utilization reports, identify our high flyers, look at the performance of our provider network through Healthcare Effectiveness Data and Information Set (HEDIS) measures, and even do risk prediction with a set of claims-based tools.
However, that was back when we paid fee for service, and lived in a largely paper-based clinical world. Even though our internal team of analysts generated our HEDIS measures, we still needed to contract nurses to do the chart reviews for the hybrid measures.
We had monthly meetings with our chief actuary to pull data to figure out our IBNR (incurred but not reported), otherwise known as the claims lag. Unfortunately, a lot of our reporting was skewed by the claims lag, and although the data was helpful, it was like driving the car while looking in the rear view mirror.
We are now in the era of electronic health records (EHR), MACRA2 and a plethora of digital clinical data, which means we can take advantage of what happens today, sometimes in real-time. With the advent of value-based payment models, where healthcare organizations work to improve outcomes at the same or lower cost, the need to know what happens today becomes an imperative.
With the advent of value-based payment models, where healthcare organizations work to improve outcomes at the same or lower cost, the need to know what happens today becomes an imperative.
As we meet with payers across the country, the forward leaning organizations have started to think about how to integrate clinical data into their overall strategy. Many of the health plans realize that sharing clinical and claims data makes business sense in order to help provider networks succeed in taking on risk. Delivery reform accompanied by payment reform is probably one of the industry’s greatest challenges to date, and going at it alone for any organization will be extremely labor-intensive.
Augmenting existing risk prediction models
This allows payers to classify patients into risk categories to better identify candidates for care management programs and predict future costs. There are many good risk prediction models based on claims, but the best ones use clinical data (and even patient generated data) to know which patients to put in a care management or disease management program, or who might be at risk for a heart attack next year.
Enhance and streamline quality measurement
HEDIS requires chart reviews for the hybrid measures. Because of this, resources often must spend time going through electronic or paper charts to document the presence or absence of certain findings. Not to mention that HEDIS measures are largely about process; did a woman have a mammogram or not? What we really want to know is the outcome; did the intervention improve the health and wellness of our members?
Engage and support the provider network
As a health plan, your greatest asset is your provider network. Through your contracted network, you provide valuable care and services to your members. Their success is your success. Sharing clinical and claims data with your partners only strengthens the relationship. This helps them understand and manage cost and quality as part of risk based contracting strategies.
The ultimate goal is for patient care to be appropriately coordinated between all the various caretakers – both clinical and non-clinical. That doesn’t always happen with both a payer and provider reaching out to a patient for the same episode. One of the top uses of clinical data involves both the health plan and the clinical team receiving an alert that the patient was just admitted to an ED or hospital. When this happens, a prior agreement should exist about who will take the lead. Once the engagement has been initiated, care plans can be shared and viewed by all those involved in the patient’s care, including the patient themselves.
Streamline Utilization Management (UM) and Prior Authorization Processes
UM and prior authorization processes should be informed by clinical data and require collaboration and coordination between plan and provider. Determining medical necessity based on clinical guidelines and appropriateness of care should also be a streamlined exchange of information. A study of 12 primary care practices published in 2013 by the Journal of the American Board of Family Medicine put the mean annual projected cost per full-time equivalent physician for prior authorization activities between $2,161 and $3,430.
Once you have outlined the use cases that will support your key strategies, the next step involves determining what clinical data you need and how you will get access to it. This often is the hardest part of the strategy. Although some clinical data may be easily accessible (like lab results or medication fills), getting data from EHRs requires both governance and technical acumen. The good news is that provider-payer collaboration continues to accelerate. We have recently been involved in a number of these conversations, helping our clients and prospects think through their strategy and develop a plan to bring in the key clinical data they need. For payers that don’t have a clinical data strategy yet, we suggest they start thinking about it soon.
1 Extract, Transform and Load (ETL)
2 Medicare Access and CHIP Reauthorization Act of 2015 (MACRA)