The Role of High-Quality Data in Population Health Initiatives

Trushar Dungarani, MD

In this interview, Trushar Dungarani, MD, MDClone, discusses the importance of high-quality, accessible patient data for clinicians to enhance care delivery, address social determinants of health, and drive effective population health initiatives. He also shares insights on how health care stakeholders can overcome common data-related barriers.

Please introduce yourself by stating your name, title, organization, and relevant professional experience.

My name is Dr Trushar Dungarani, and I am the director of clinical and data science at MDClone. I’ve been a hospital medicine physician for over 15 years, and I’ve worked in 2 large health systems as a leader in hospital operations, quality, and clinical care. In my last role in the Johns Hopkins health system, I served as the co-lead physician for high-value care delivery. Along with clinical care, I’ve had roles in hospital medicine operations, insurance regulation as a physician advisor, and research in machine learning implementation. I have a passion for leveraging technology to improve patient outcomes, drive high-value patient care, and reduce burnout for health care workers. In addition, I currently serve as an elected member of the Quality Improvement and Patient Safety Committee for the Society of Hospital Medicine.

Why is high-quality patient data crucial for clinicians at the point of care? What are some of the risks involved when this data is lacking?

Health care is full of both standards and nuances in care. Having data that is meaningful and accessible is paramount. For too long, physicians and nurses have only had access to broad metrics such as readmission rates and length of stay (LOS). To act on these high-level metrics, health care teams need access and more data types to be able to find the solutions to these challenges. Behavior change is difficult if you don’t have the data you need. Giving our team members tools to come up with solutions will foster ownership of both problems and solutions.

Population health initiatives require extensive data on individual patients. That data is then analyzed by health systems to reveal opportunities to improve care, treatment, and outcomes for similar groups of patients. For example, a health system may seek to improve care for patients with chronic kidney disease by enabling early intervention through identification of patients at risk of developing the condition. To successfully perform a population health initiative that identifies these at-risk patients, the health system needs access to data that is structured, accurate, retrievable, and updated in real time.

This is necessary for decisions affecting a single patient, as well as for the planning and evaluation of broader population health initiatives and research. When the right data is lacking, quality of care suffers, opportunities to improve outcomes are missed, and preventable costs may escalate.

How can accurate analytics assist health care providers in understanding and addressing social determinants of health (SDoH)?

Accurate, comprehensive analytics help providers to identify social determinants of health (SDoH) that affect patients, enabling clinicians to optimize preventive care instead of waiting for patients to become ill. Say, for example, a patient lives in a food desert and lacks access to a consistent source of healthy and nutritious fresh food, which is exacerbating a chronic condition such as diabetes. When dealing with data, having tools that can reveal patterns in the data that can speak about correlation between factors is a huge start. SDoH data is just as important as traditional inputs— such as medications and vital signs—when trying to paint a trajectory for our patients. Data can really come alive when it’s in the right format for our team members to interpret, change, and track improvement over time.

What are some common data-related barriers that health care organizations face when implementing population health initiatives?

Health care has one of the toughest burdens for data trust in addition to having the lowest health literacy among its workers in all industries. Accurate, longitudinal data is critical to population health success, but we know interoperability is a challenge with the US health system. For example, many patients see several physicians and specialists for various health issues, sometimes from different health systems and different electronic medical records (EMRs), which often prevents their data from being centralized and easily accessible. One patient may have their data on 3 different EMRs, 2 insurance databases, and countless legacy sources.

Federal regulations on how data can be collected and shared, and patient control over their information can be another obstacle. Finally, hospitals often have differing technology and standards they are using to collect, structure, store, and transmit clinical information, making it difficult to aggregate and normalize that data in a format that is suitable for analysis by artificial intelligence and machine-learning algorithms.

What features and capabilities do hospitals and health systems need from a self-service data analytics platform to effectively support their population health efforts?

To roll out insights more quickly to the bedside, many leading health systems are embracing a “self-service” approach to data exploration. The self-service approach to data access enables health systems to empower clinicians and administrators by allowing them to have a dialogue with their own data. This drives greater volume and speed in quality improvement initiatives and performance improvement efforts. 

In terms of features and capabilities, providers need 3 things from their analytics platforms to help their population health initiatives succeed:

1. Data quality, trust, and maintenance because missing or faulty data make it impossible to perform accurate analytics
2. Fast, secure, and easy access to data, which enable users from across the organization to query data sets without needing any particular expertise in databases
3. The facilitation of open and high-quality communication between various groups in the hospital including clinical team members, IT/analytics, and various teams that input and code data

The self-service concept also delivers substantial cost advantages. By using self-service, health systems can drive existing clinicians and staff to become data-literate. Further, because clinicians throughout the organization can initiate their own projects, health systems gain the ability to dramatically accelerate the number of quality improvement initiatives that are conducted simultaneously.