How Two Health Systems Solved Clinical Problems With Self-Service Data

Daniel Blumenthal, VP of Strategy, MDClone

Navigating clinical data is often challenging, time-consuming, and expensive for hospitals and health systems.

For clinicians and staff who are eager for data-driven answers to simple questions, the process may take months or longer due to siloed systems, complex data models, unstructured data, privacy regulations, and limited availability from overburdened IT and data teams.

In these scenarios, the layers between clinicians, staff, and other operators and their health systems’ own data create friction between departments. Often, the results of this misalignment include frustrated clinicians, overworked IT staff, wasteful bureaucracy, inability to scale insight generation, and missed opportunities to gain a competitive edge.

Clearly, this top-down approach to data exploration is not serving health systems well. In contrast, forward-thinking health systems are increasingly adopting solutions that enable self-service data discovery and speed-to-data at large volumes.

The Benefits of Self-Service Data Exploration

Self-service data platforms enable users to engage in independent data discovery, interacting with all patient data from any source, leveraging unstructured data, and collaborating freely using synthetic data. One major advantage of the self-service approach is that it allows leadership to nurture innovative ideas from the front lines, increasing the volumes and speed-to-value of quality and research projects.

Effectively, 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. It also offers the potential for significant savings in time and resources.

For example, because clinicians and staff throughout the health system can initiate their own data-exploration projects, health systems gain the ability to dramatically accelerate the number of quality-improvement initiatives that are conducted concurrently. Armed with timely data on the potential outcome or cost benefits, hospital leaders can make more informed decisions about which projects to support and where within the organization to make changes.

Additionally, because potential changes and interventions require the investment of time and money to implement and evaluate, using a data-driven approach prior to initiating these projects leads to cost savings.

Real-World Examples of Data-Driven Clinical Improvements

Following are two examples in which health systems used self-service data analytics to improve patient care and clinical outcomes:

• Intermountain Health: Radiation cystitis is a debilitating complication arising from pelvic radiation therapy for cancers that poses significant treatment challenges and can lead to symptoms such as hematuria (urinary bleeding), urinary pain, and incontinence, impacting patients’ quality of life.
  • Utah-based Intermountain Health sought to investigate the impact of hyperbaric oxygen (HB02) on mortality and blood transfusion requirements in radiation cystitis patients. Using a self-service data platform, Intermountain searched its database for patients diagnosed with radiation cystitis over the prior 15 years, revealing 572 patients.
  • The study found that those who completed 20, 30, or 40 HB02 sessions experienced a survival benefit. Multivariable regression analysis revealed that increased age, lower hematocrit, and decreased glomerular filtration rate correlated with a higher risk of death. However, as the number of HB02 sessions increased, the risk of death decreased. Additionally, patients starting HB02 more than 30 days after diagnosis were more likely to need blood transfusion with approximately 30% receiving blood transfusion in the six months before HB02 and 21% receiving transfusion in the first six months after starting HB02.
• The Ottawa Hospital: Approximately 40 000 hysterectomies are performed annually in Canada, and 600 000 in the US, with 30% of females undergoing this procedure by age 60 in North America. Anemia is prevalent among women scheduled for elective hysterectomy, significantly increasing the risk of complications, blood transfusions, prolonged hospital stays, and mortality. Heavy menstrual bleeding (HMB) contributes to regular blood loss, necessitating more than just iron supplementation for anemia correction.
  • Prior to gynecologic surgery, multidisciplinary coordination is crucial, involving gynecology (to reduce menstrual blood loss), hematology (for iron supplementation), and anesthesia (to assess overall fitness for surgery). However, there is no standardized pathway available in the literature specifically for hemoglobin optimization in gynecologic surgery.
  • Gynecologic surgeon Dr Innie Chen and hematology specialist Dr Karima Khamisa at The Ottawa Hospital sought to test the hypothesis that a gynecology-specific pathway would reduce transfusions, complications, and costs. The collaborative multidisciplinary team used a self-service platform to identify a cohort of all patients who underwent hysterectomy for non-cancerous indications over a ten-year period.
  • The results showed that 1305 patients underwent hysterectomy in the study period. The peri-operative transfusion rate was 3.6%, and peri-operative transfusion was associated with pre-operative anemia and abdominal hysterectomy.
  • Following the study intervention period, there was a substantial decrease in rates of pre-operative anemia from 19.4% to 16.5%, and rates of iron deficiency from 31.8% to 28.5%. This corresponded to a significant improvement in mean pre-operative Hb level (g/L) for all hysterectomies and for abdominal hysterectomy. There was also a significant reduction in peri-operative transfusion rate for all hysterectomies (4.4% to 1.7%) and for abdominal hysterectomy (13.3% to 5.1%).
  • The results demonstrated improved pre-operative anemia and reduced transfusion following implementation of the intervention bundle.

Conclusion

The promise of extracting actionable value from data doesn’t come from fancy programs but from enabling an entire organization full of talent to explore and discover opportunities within that data. With self-service data platforms, health systems gain the ability to leverage their own workforce to find data-driven answers to the clinical questions that interest them most.