The number of coronavirus cases is growing, and Cloud4Y again switched to a remote work format. We see how our doctors are struggling to cope with the incredible workload, and we do not want to add to them the hassle. It is sad that even in such a situation, modern technologies are poorly used in our country. But in the United States, for example, a medical organization has built a cloud system that makes it possible to better take care of the country’s most vulnerable residents. We bring this story to your attention.
When the COVID-19 pandemic hit in early 2020, healthcare organizations had to change or accelerate their patient care plans. The American Commonwealth Care Alliance (CCA) used cloud-based data analytics to connect doctors and healthcare professionals with at-risk members. CCA Vice President Valmik Kudesia, in charge of Clinical Informatics and CCA Advanced Analytics, shares their story.
CCA Is a nationally recognized healthcare organization as a leader in providing and coordinating costly care for people in need who are eligible for Medicaid and Medicare. The CCA fulfills the responsibilities of the payer of health care, health care management, and directly the provider of care to patients. CCA “wards” live with health, behavioral and social problems. Many of them have difficult living conditions, they are vulnerable and marginalized. When news of COVID-19 hit the United States in the winter, organizations quickly realized that this category of people would need a lot of care and attention. CCA staff needed to continue to fulfill their responsibilities, given many new, ever-changing factors.
We needed reliable data that could be retrieved quickly and integrated into their workflows. CCA pre-built a cloud-based advanced analytics platform with BigQuery and Looker. Six months later, with confidence in the solution’s reliability and functionality, we continue to provide clinicians with a more holistic view of the needs of people in need. CCA is developing a human-centered use of data and analytics to tackle the upcoming seasonal COVID-19 / influenza combo.
The data you need to make decisions faster
The cloud platform was created for situations when users must move quickly and in different directions, as well as change the parameters of the requested information. This meant that when COVID-19 arrived, the CCA didn’t have to change much in its processes.
The data analysis team used Looker and BigQuery in conjunction with other technologies to develop and deploy data processing operations in conjunction with machine learning capabilities. Cloud services comply with HIPAA requirements (a kind of analogue of our FZ-152, but with a medical bias – approx. translator), and BigQuery was (and remains) flexible and available as a service. This allowed a small data analytics team to focus on working with the data and rapidly evolve the project while maintaining compatibility and delivering excellent platform performance.
Our analysts used a query abstraction and column-based data engine to provide the team with a number of benefits when working in a COVID-19 environment. We had many ways of describing individual values, combined with very fast cycles to determine “what matters” at a given moment of the day or week. The columnar format allowed most of the questions to be answered in advance, and through direct virtualization of queries, it was possible to determine what information doctors needed and quickly provide them with that data.
CCA staff could move with clinicians to generate data and predictions using shared and task-specific dashboards called Action boards. These boards don’t just keep users informed, they offer the information they need to decide what the doctor will do next.
The use of daily and even hourly data from different sources was necessary so that the vulnerable segments of the population could get what they need – food at home, medicine or other services. In some cases, we already had all the necessary data. For example, in less than 30 minutes, we were able to implement the CDC’s definition of high risk of COVID-19 complications in LookML (Looker Query Abstraction Layer) and connect the concept of “high risk COVID-19 complications” with our information model.
We also created our core COVID-19 dashboards throughout the business day and implemented relevant pandemic data into other clinical dashboards and Action boards. The flexibility of the solution, achieved through abstraction and fast delivery of data, allowed us to quickly identify each person at high risk of an adverse outcome from COVID-19 and provide this knowledge to CCA doctors.
Some of the data needed was not easy to obtain. For example, at the beginning of the outbreak, there were no COVID-19 data repositories or data transfer services. It was important to collect all possible data to serve the needy groups of people. And in many cases, we collected and used this data ourselves. For example, in the early stages of the COVID-19 pandemic, Massachusetts gradually began to close down adult day health centers (ADHs), community centers that provide essential services for seniors, and then suddenly it went widespread. But we got the knack to transfer this knowledge to every person who visited these institutions, literally a few minutes after we learned about another closed ADH. A little later, the Massachusetts Department of Public Health began to receive data from the Massachusetts Department of Public Health on positive tests, which provided an idea of u200b u200bthe concentration of at-risk people living in areas with high or growing COVID-19 infection.
Moving from “just data” to “essential element for treatment and support”
As the COVID-19 pandemic continues, we are using the latest information available to update and change our support and care strategies. CCA employees are now much more comfortable working with data thanks to the cloud. We usually have over 450 active users per week, and information is requested almost every second during the working day. Thanks to the large collected database, you can see how some of the data matches the overall picture or does not fit into it. Instead of using the data as separate pieces of the puzzle, we use it in an integrated manner. That is, thanks to a solution built on a cloud platform, we use data in the interests of patients, we have built new technology into our daily life.
With this handy tool in hand, the data science team has moved into deep function and causal engineering to make the information available to CCA staff and clinicians more complete and understandable. In turn, doctors and our other employees expect Big Data to help them better take care of the people who need help.
A retrospective of data-driven decisions
The journey to data-driven decision making takes time to build trust in the system. COVID-19 has helped build that trust, and our doctors now understand that the information the CCA receives and processes will help them do their jobs better. We also learned that you need to be able to iterate quickly to get data and a platform (not perfect, but good enough) to work with that data. And technology must provide the required iteration speed. We understand that when people do not see the big picture or do not have all the information, they can make not the best or radically wrong choice. Plus, they have habits or preferences that make the problem worse.
Now the person interacts with the machine, with the data coming to him. If everything is done correctly, the data can be included in the decision-making process without adding unnecessary steps that delay the journey from task to result. After all, information is a natural part of the decision-making process. In our case, concerns about a person. For example, studying chest pain often requires deciphering an electrocardiogram (ECG), which carries a wealth of useful information. And doctors understandably expect the ECG to help them take better care of the person, rather than focusing on the data itself that is generated when the ECG is taken.
The COVID-19 pandemic has shown us that the right use of data can empower people and become an ally in the fight for public health.
What else is interesting in the blog Cloud4Y
→ Found a long-lost manual for the world’s oldest computer
→ US Border Patrol plans to store data from travelers’ gadgets for 75 years
→ Definitely not Windows 95: which operating systems support space?
→ We tell you about state protected services and networks
→ Inside the Bell Labs data center, 1960s
Subscribe to our Telegram-channel so as not to miss another article. We write no more than twice a week and only on business