By Peter Calderone and Deanna Rothwell

For many health systems and hospitals, the process and methodology of using real-world data to inform care and operational quality improvements is a long, onerous slog.

Specifically, gaining access to data in a timely manner while having the capacity and resources for complex analysis represents a significant challenge for most hospitals.

Health system leaders, of course, understand the importance of leveraging data to discover new ways of driving higher-quality patient care and better financial and operational performance, but they typically run into two major issues when attempting to implement these projects: cost and manpower.

U.S. healthcare providers have long faced tight budgets that leave limited funds for data analytics initiatives – and such challenges are not unique to this country. In Canada, for example, the healthcare data universe and use cases match the US experience with the added constraint of even lower IT and workforce budgets. To overcome their significant financial and manpower barriers, several healthcare organizations across Canada, the U.S., and Israel have deployed nimble organizational structures and self-service analytics to drive innovative performance-improvement initiatives.

The old way: Expensive data analysts and restricted data access for clinicians
Traditionally hospitals and health systems have derived insights from their own data by employing large teams of data analysts. For the clinicians and health system leaders developing hypotheses and questions, they send their queries to the data analytics team, who then must comb the health system’s data to assemble the right data set. For the data analysts, they spend their days re-pulling and refining data sets in administrative cycles that can take weeks and months. This can be a lengthy, convoluted process that forces curious clinicians and leaders to wait in line while high-value data analysts are wrangling menial data requests. Both groups involved in this equation are restricted from focusing on the high value, strategic initiatives and the health system suffers when this limited flow of research projects results in very little return on investment over time.

For clinicians, researchers, and administrators who are looking for answers to enhance care quality, the process is discouraging. For health organizations, the process is inefficient and expensive. In contrast, a self-service approach allows health systems to eliminate these bottlenecks.

The better way: Self-service access with centralized expertise and support on demand to enable dynamic 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 of quality- and performance-improvement initiatives. Further, when operational improvements require reporting automation and dashboard visualizations, analyst groups are freed from the need to perform initial data exploration, driving savings in time and resources.

This approach also provides significant cost advantages. By using self-service, hospitals empower their 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 concurrently. Hospital leaders are better able, using timely data, to select which projects to support, or where in the organization to make changes, when presented with data on the potential outcome or cost benefits. Because interventions cost time and money to implement and evaluate, using a data-driven approach prior to "acting" will lead to cost savings.

A look at self-service in action: The Ottawa Hospital (TOH)
One Canadian hospital that is benefiting from this self-service approach to data exploration is TOH, one of the largest academic and research hospitals in Canada, consisting of three hospital sites, 1,420 beds, and more than 12,000 employees and support staff.

Prior to adopting a self-service approach, the hospital employed a centralized data request process to support data-driven improvement projects. In the centralized process, a researcher, clinician, or healthcare professional would first submit a data request to the analytics team. The analytics team would then meet to determine the potential project’s needs, data plan, and cost estimate. The requestor would then need to obtain the necessary approvals from leadership and regulatory teams, and, if approved, the analytics teams would produce the requested data.

For complex inquiries, the process could take months, frustrating curious clinicians who must endure long waits before obtaining answers to their important questions. Determined to find a better way, the hospital decided to implement a self-service data platform that enables users to gather, use, and share data-driven insights while fully protecting patient privacy. Now the users on the platform and the IT teams supporting the technology are partners with a higher-level focus on quality improvement instead of managing data requests and prioritization.

In one project Marcel Miron-Celis, a Methodologist with the Ottawa Hospital Research Institute, worked on a study evaluating the safety of treating cancer-associated pulmonary embolism in an outpatient setting. Marcel stated the self-service approach "allowed us to narrow down the cohort to 739 patient encounters, which likely saved the clinician stakeholders hundreds of hours that they would have spent performing chart reviews on hundreds of thousands of patient encounters to identify potential patients that met those criteria.” Marcel compared this effort to working in traditional query tools and estimated the traditional approach could have taken up to four to six business days and required the expertise of analyst resources, whereas with the self-service technology, a researcher can easily do this work in as little as two days.

Among other projects that TOH clinicians launched was a study of risk factors associated with ischemic stroke in cancer patients. About 10% of patients with ischemic stroke, a condition in which the brain’s blood supply is reduced, also have cancer. Ischemic stroke is a leading cause of disability and the second-leading cause of death worldwide, so the ability to identify treatment strategies to prevent its occurrence in cancer patients is critical.

Using data from TOH Data Warehouse of 10,875 patients from 2000 to 2019, the research team compared two patient groups: First, all patients who had both a cancer diagnosis and an ischemic stroke within a 2-year period after their cancer diagnosis, and second, all ischemic stroke patients without cancer.

At the study’s conclusion, researchers found that cancer patients with ischemic stroke have a higher prevalence of chronic obstructive pulmonary disorder, previous ischemic stroke, and venous thromboembolism, and that previous ischemic stroke is an important predictor of recurrent stroke in cancer patients. The results highlight the importance of identifying optimal secondary prevention treatments in cancer patients, according to the researchers.

By coupling self-service data access with centralized expertise, TOH was able to leverage its centralized probability and statistical resources to generate a study at an accelerated pace.

Tackling budget and manpower challenges with self-service data analytics
The need for data-driven performance improvements will continue to grow as hospitals and health systems strive to optimize patient care and financial outcomes. Despite ever-present budget and manpower challenges, forward-thinking hospitals that embrace self-service data analytics have the opportunity to advance innovation and accelerate their data-driven initiatives.

About the authors: Peter Calderone is VP of customer success at MDClone, and Deanna Rothwell is director of analytics at The Ottawa Hospital.