Big data and artificial intelligence have dominated hospital technology executive concerns as IT leaders work to ensure they get the most investment return in an uncertain economic environment.
To learn more about the intersections between big data, artificial intelligence and healthcare, Becker's reached out to Kerri Webster, the chief analytics officer of Aurora-based Children's Hospital Colorado.
Question: How have hospitals incorporated big data into decision-making?
Kerri Webster: Big data and analytics have become part of our DNA. We use analytics to support hospital operations with dashboards, notifications and predictive models. We have forecasting models to support our census, staffing and supply chain decisions — ensuring we have the resources to care for our patients. At the bedside, we're using big data to augment the clinician experience to help with decision-making and quick action. Now more than ever, leveraging data to make decisions is key to success. The need to make data-driven decisions is no longer a nice to have — it's imperative to success.
Q: Did the COVID-19 pandemic change the way hospitals think about analytics?
KW: COVID-19 was a catalyst that provided the opportunity to leapfrog outdated practices and processes and replace them with advanced technologies. We've been able to use this innovation imperative to advance new technology initiatives in record time: scaling up our predictive analytics capability to support our clinicians with accurate census forecasting, supporting our supply chain by predicting supply needs and potential deficits, providing real-time alerting for air-flow monitoring, transforming manual process to automate the submission of data to the state. Ultimately, we were able to take advantage of the urgency of opportunity to deploy meaningful innovations to support our patients and caregivers.
Q: How can AI help health systems deal with the staffing crisis?
KW: There are several ways that AI can help with the staffing crisis: using predictive modeling to identify turnover risks, leveraging modeling to support creative staffing schedules, identifying patient acuity models to support patient assignments, leveraging predictive models to synthesize large amounts of clinical data to expedite decision-making, or identifying pipelines of the workforce to be ready to step into roles.
Q: What advice would you give to health systems working to integrate and use AI tools?
KW: It seems that AI in today's world is a solution in search of a problem. We need to flip that paradigm and truly understand the problem and insert the technology solutions to address specific issues — automate manual systems, augment expertise and take the guesswork out of processes. Also, start with the problem, not the technology. Work to deeply understand the problems that need to be addressed and leverage technology in support of the solution. Often people see AI/[machine learning] as a panacea to healthcare problems. If we view AI as another tool in our toolbox versus the answer, we can be successful with this technology.
Q: What health analytics innovations do you see coming in the next five years?
KW:
- Continued advancement of the use of modeling to predict and augment decision-making
- Enhanced workforce analytics — combined with patient data to address workforce challenges
- Analytics seamlessly embedded in workflows — people may not be aware they are using analytics
- Increased use of reliable natural language processing to support analytics insights
- Advancement of automated processes to support decision-making
- Large-scale analytics insights that use more than EHR data
- Addressing of equity and inclusion with analytics tools
- Precision medical tools becoming more accessible
- Enculturation of analytics-driven decision-making and processes