In recent years, healthcare organizations (HCOs) have begun to appreciate the importance of incorporating new types of data, such as social determinants, into their analytics models.
However, many HCOs overlook a valuable source of insight that can help them capture greater revenues, optimize risk adjustment, boost quality scores and improve population health and care quality: laboratory data, which includes critical data such as cholesterol levels, red and white blood cell count, blood glucose levels, hemoglobin A1C, etc.
Lab data can deliver significant value to HCOs that are assuming greater risk in value-based care arrangements, particularly those that work with Medicare Advantage (MA) plans, by helping them discover unrecorded conditions that warrant risk adjustment intervention. While many electronic health records (EHR) systems include some lab data, the information is often fragmented and incomplete, reducing its efficacy for risk adjustment.
To be successful in MA and have the resources to deliver the right care, it’s essential to capture all of the risk adjusted revenues possible for the enrolled MA population. While CMS allows for adjustment of RAF score and payments, timing is critical for MA plans and health systems bearing downside risk. Relying on retrospective techniques is no longer sufficient.
HCOs’ clinical teams have long seen lab results as a valuable source of data for decisions on treatments and prescriptions. However, due in part to organizational siloes and fragmentation, few business and financial teams across HCOs have fully taken advantage of lab data’s value in predictive modeling to develop prospective risk profiles for their patient populations.
That is now starting to change, as healthcare leaders begin to realize that, by combining lab data directly from laboratory data providers with more traditional data sources continuously, HCOs can make the best decisions about affordable, quality care today, based on what is going to happen tomorrow.
How it works
Consider an example of a MA plan that enrolls new members at the beginning of each year. These plans generally have extremely limited visibility into the historical data of new members, in contrast with existing members, for some of whom the plans may have years of medical data on file. This gap in new members’ data can lead to unrealized revenue as a result of unrecorded conditions in these patients’ risk profiles.
Now, consider how the same MA plan could leverage lab data to avoid missed revenues. By developing a model based on its substantial collection of the plan population’s historical lab data and financial outcomes, the plan can identify conditions that new patients may be likely to have using prospective predictive models and the outstanding revenues on the line. Historical lab data also enriches EHR data for existing members enhancing the ability to predict new conditions that are likely to surface, which then translates to a more accurate capture and enhanced revenue to address these new conditions.
Better risk management, accelerated revenues
For MA plans, it pays to gather data about new members sooner rather than later. The ability to identify high-risk members, for new or existing, at the beginning of the plan year provides greater opportunity for interventional efforts to be successful and for full RAF realization. Early identification gives plans a significant advantage in managing the risk of new member populations and mitigates the likelihood of late or completely missed diagnoses.
Specifically, the early prediction of RAF scores allows an opportunity to service high-risk RAF members in the first half of the year so that their serviced claims can be submitted in the initial submission deadline during September. As a result of this, the underlying revenue can be received at an accelerated pace starting January instead of a lump-sum amount in August as an adjustment.
Challenges with analyzing lab data
While we expect the analysis of lab data to play an increasingly important role in risk-adjustment, there are several technical challenges with the process that have hindered its adoption. First, many – if not virtually all - patients in MA populations regularly take various medications, which can significantly affect lab measurements, making it imperative to have an intelligent and continuous analytics solution in place.
Next, members of the new enrollee population have different levels of lab data availability, which creates challenges in developing patient risk stratification groups. A proficient analytics system should be able to handle patients at different levels of data by appropriately handling missing data.
Different labs may have different reference ranges based on their test equipment and techniques used. Interpretation of data must be aligned with a lab’s reference ranges. This also makes handling missing data in labs a greater challenge.
Beyond revenue enhancement
In addition to unlocking additional revenue opportunities, the early identification of high-risk RAF candidates can also result in additional clinical benefits, in particular when a lab data analysis reveals risky measures for members. These advantages may include a reduction in future healthcare costs, improved outcomes and higher quality-of-life. Clinical teams have long appreciated the value lab data brings to patient care; operational and financial teams will soon realize the difference it can make in care management, quality improvement and risk adjustment.