Last week, I had the pleasure of moderating an analytics roundtable at the Caradigm Customer Summit (CCS), which took place on the Seattle waterfront. CCS was a fantastic event where provider organizations from around the country came to learn and share best practices as they lead the way with population health. You can read more about CCS in these posts – Day 1 recap and Day 2 recap. As this week is National Health IT Week, it’s an ideal time to share some of the discussion points from the roundtable to help raise awareness on the impact of data and analytics in transforming healthcare.
There were four key recommendations that came up during the discussion:
1. Build a Data Foundation
A population health analytics strategy starts with a data foundation. We heard in the roundtable that population health forces organizations to bring together a variety of data types including clinical, claims data from multiple payers (CMS and commercial payers), internal billing data, lab, pharmacy, etc. Aggregating clinical data continues to be a challenge as large health systems often are using dozens of EMRs and or are part of a clinically integrated network (CIN) that will likely always utilize many EMRs. Attendees said that while all EMR vendors publicly claim neutrality, it continues to be a challenge to get all the data that they need. Using a vendor neutral solution to aggregate data from disparate systems is one way to help overcome interoperability issues.
2. Increase Use of Predictive Analytics
The group also talked about how they are taking steps to evolve analytics efforts beyond retrospective reporting to predictive analytics that can have greater impact of patient outcomes. Most providers are still in the early stages as they have been focused on being able to report on and attain required quality metrics. Some are now starting to leverage predictive analytics in order to proactively impact patient care. For example, one attendee explained that they are giving care managers a patient readmission risk score along with the reasons for the score, which helps them take action with high-risk patients before they are discharged.
3. Surface Analytics in Workflows to Make Them Actionable
One of the most important considerations in a population health analytics strategy is to embed analytics in clinician workflows at the point-of-care. We heard repeatedly at CCS that clinicians are already faced with too much information, and that it’s not effective to present them with yet another report. What clinicians want is additional information presented in their existing workflow and tool that can help support clinical decisions. A great example of this that we heard about was surfacing a predictive Sepsis risk score for clinicians to see at the point-of-care. The Sepsis algorithm calculates a risk score based on real-time data (vital signs, lab results, medications and dates/times) and then stratifies patients as “At Risk”, “High Risk” and “Very High Risk”. Clinicians see the risk score including aggregated clinical data in their customary EMR and then examine at-risk patients to determine if they meet the criteria for severe sepsis treatment according to established protocols.
4. Establish Data Governance
Lastly, the group discussed how important it is to have a data governance structure with executive support representing all modalities of data such as quality measures, HCAHPS scores, fall risk, readmissions, hospital acquired conditions, etc. It’s critical to get different parts of the organization on the same page because there are inevitably many initiatives in motion at the same time. One attendee shared that they have built a dashboard for each modality, which helps align the different stakeholders.
CCS was an engaging event for industry leaders who are all striving to get to the same goal of population health. If you’d like to have a discussion about how to augment your population health analytics strategies or see the list of best practices we developed in conjunction with the analytics roundtable, then drop us a note here.