Commonwealth Care Alliance Case Study

Commonwealth Care Alliance® (CCA) and Looker team to verify reliable, up to date data is used to ensure patients get the highest quality personalized care.

Google Cloud results

As COVID-19 presents new challenges in providing health services, the CCA DS team quickly creates custom concepts and metrics, such as “COVID high risk,” to accurately monitor member needs and risks
Clinician-centric dashboards, internally referred to as action boards, provide all CCA care managers with updated facts about their members and needs for intervention
Using BigQuery and Looker, CCA DS provides company-wide access to trusted data in a secure way without burdening internal resources

Commonwealth Care Alliance® (CCA) is committed to using the latest information and resources to ensure their members receive the highest quality of personalized care. When the COVID-19 pandemic started to spread through the world, CCA knew their members, who live with complex medical, behavioral health, and social needs, would be faced with new challenges and require an enhanced level of support and care. CCA also knew that the best way to understand, prepare, and protect their members was through the use of timely and reliable data.

Over the past year, the CCA Data Science (DS) team has utilized Looker and Google Cloud to build an analytics data architecture and deliver valuable information and predictive insights to CCA’s clinicians. When the COVID-19 outbreak began, the DS team built their first COVID-19 monitoring dashboards within a day and infused COVID-19 information throughout other existing clinical dashboards. This allowed CCA’s clinical workforce to understand how their members were being impacted, their potential risk, and the best next steps to help each member. The CCA DS team has also been rapidly iterating on these dashboards to keep pace with the COVID-19 pandemic, ensuring their clinicians have access to the latest information.

“In under a minute, I was able to define a dimension that says ‘is COVID-19 high risk.’ I added the logic and then instantly, across our entire member population, we had this ‘is COVID-19 high risk’ flag.”

—Dr. Valmeek Kudesia, CCA Vice President of Clinical Informatics & Advanced Analytics

Identifying the right information

As the pandemic evolves, CCA continually uses the latest available information to update and guide its data and care strategies. The CCA DS team quickly creates and updates custom concepts in Looker’s flexible modeling layer, LookML. Those important concepts are instantly pushed to CCA’s clinicians via Looker dashboards to allow decision-making on up-to-date key variables, such as factors that are known to increase member risk. “We must have the flexibility to capture and move with what we know at the time,” says Dr. Valmeek Kudesia, CCA Vice President of Clinical Informatics and Advanced Analytics.

At the beginning, CCA focused on identifying early hotspots in relation to where their members lived, then used known risk factors such as age, pre-existing conditions, and living conditions to identify which patients were most susceptible to complications if they were infected by COVID-19.

“In under a minute, I was able to define a dimension that says ‘is COVID-19 high risk.’ I added the logic and then instantly, across our entire member population, we had this ‘is COVID-19 high risk’ flag. We also know who is in a group housing facility, so we created another flag that is for ‘congregate living’ so the CCA team knows which members are vulnerable to a poor outcome or living in a high-risk situation,” explains Kudesia.

Over time, CCA has expanded the variables they monitor, such as members whose elective surgeries were cancelled and the changing availability of services that provide seniors with food, socialization, and medical assessments.

Reacting to changing challenges

In addition to identifying risk, the CCA DS team leverages Looker’s user attributes and permissions and integrates data that describes service disruption, allowing CCA clinicians to quickly understand how specific service closures impact their members and anticipate and react to the constantly evolving situation. This enables the CCA DS team to help CCA clinicians take quick action on the most relevant data.

For example, as CCA learned which Adult Day Health centers were temporarily closing, the tailored dashboard for each CCA clinician was instantly updated with which of their senior members would experience service disruptions. With this updated information, the care managers could help those members get the services they need. “As fast as we were learning about service impacts, it was a matter of two minutes to model that in LookML and deliver those updates to the care managers who need to take action,” adds Kudesia. As CCA care managers understand which closures impact which members, they prioritize their outreach to assist with additional meal plans, help them continue to receive critical medication, and proactively call to provide critical social connection over the phone.

“We provide and coordinate care for a vulnerable population of people, and through this data we have been able to address risk factors for the members who need us most, in a timely manner,” explains Kudesia.

“There’s a very tight relationship between the data and the expectation that something needs to be done with it.”

—Dr. Valmeek Kudesia, CCA Vice President of Clinical Informatics & Advanced Analytics

Helping members by taking action

Custom dashboards with updated member and regional information tailored to each care manager are internally referred to as action boards, because each tile shows data and points to a particular action that needs to be taken.

“There’s a very tight relationship between the data and the expectation that something needs to be done with it,” explains Kudesia.

The action boards allow the CCA DS team to insert tiles with links to other information, such as overdue assessments or predictive insights. BigQuery and Looker are updated over the day with data from core tools, such as CCA’s care management platform and electronic health record. Kudesia shares that “it allows our care managers to see the effects of their actions, and it updates throughout the day to provide a real-time updated priority list. The action boards are data and role designed, allowing us to make a logical connection; if a care partner sees X, they know they should take Y next step,” explains Kudesia.

A flexible foundation

The CCA DS team has been able to react and iterate quickly because of the solutions, processes, and problem-solving methods they already had in place. Their goals focus on delivering actionable insights to CCA clinicians to achieve better outcomes for CCA members. CCA DS requires secure, lean, and broad flexibility to ingest and compute varied types of data—the selected data architecture platforms Google Cloud, BigQuery, and Looker—all to provide insights to CCA clinicians.

“The DS team is focused on supporting every CCA clinician by providing them with the insights they need to continue their amazing work during these unprecedented COVID-19 related challenges. Our use of BigQuery and Looker allows us to focus our resources on what matters most—finding the appropriate interventions and providing the highest quality of care that improves outcomes for our members,” shares Kudesia.

With BigQuery, the CCA DS team is able to incorporate a wide range of data. With Looker’s modeling layer, LookML, CCA is able to define custom metrics once and then give their care providers reliable and consistent metrics.

Some of the early custom metrics they developed and monitored in Looker describe hospitalizations and a patient’s predicted risk of unplanned hospitalization. To do this, the CCA DS team produced a predictive model and made that metric available. CCA clinicians quickly saw the predictive score for members without needing to ask the DS team to run a query or use SQL themselves. This ability to define and add metric definitions at scale has become especially useful since the CCA team began to track, plan, and react to COVID-19.

“A lot of what we’ve been able to do so quickly with COVID-19 was because we already had the pipes and processes connected; we were just able to pull new components into an existing structure,” adds Kudesia.

Preparing for the future “new normal”

Within a constantly changing situation, CCA is able to make use of the latest facts and guidance to plan and proactively help their members. “In the beginning of COVID-19, we were racing to understand what challenges were in front of us. Now we are thinking about the go-forward plan and how we take care of people and anticipate their needs in this new normal,” says Kudesia.

While the DS team had long been looking into ways to learn and provide help to CCA clinicians with data, the COVID-19 situation raised additional challenges. Using the flexible solution, agile processes, and problem-solving mentality they already had in place, CCA has been able to react in a way that is providing internal visibility and data-informed next steps, improving member care and developing scalable practices as we prepare for the new normal.

Commonwealth Care Alliance: Using data to monitor, react, and help patients with COVID-19

About Commonwealth Care Alliance

Tell us your challenge. We're here to help.