Reduce infections in your hospital with artificial intelligence

Surgical site infections are a serious problem at any hospital, but new tools are being used to predict the likelihood of them occurring – and saving lives in the process. One health system’s investment in artificial intelligence (AI) has paid off in both infection reductions and cost savings, to the tune of $1.2 million. 

The University of Iowa Hospitals & Clinics have used machine learning to reduce surgical site infections by 74% over the past three years, according to Healthcare IT News.

Integrating artificial intelligence

Working with vendor DASH Analytics, the organization is using the DASH Analytics High-Definition Care Platform (HDCP). The system integrates with the hospital’s electronic health records (EHR) system to measure the risk for individual patients.

It then decides on best practices for each patient and offers advice to the provider at specific points in the care process via the EHR.

When dealing specifically with surgical site infections, the machine uses the World Health Organization’s Surgical Safety Checklist to make sure the correct care protocol is being followed.

It also collects data from the EHR, such as which surgeon performed the procedure, how long the surgery lasted and blood loss estimates, and combines that info with the patient’s history.

Once this data has been collected, the machine inputs everything into its prediction model and alerts the provider when the risk is particularly high.

Selective treatment

The predictive model allows providers to use expensive wound closure treatments only on patients who need them, rather than spending the money to give them to everyone.

When patients were flagged by the system as particularly high risk, providers used negative pressure wound therapy to reduce the possibility of infection.

Being able to take all of the necessary information and synthesize it quickly is the main benefit of AI, said Dr. John Cromwell, associate chief medical officer and director of surgical quality and safety at the University of Iowa Hospitals & Clinics.

“No matter how much experience one has, the exponential increase in medical knowledge makes it impossible for a caregiver to assimilate all of the data necessary to consistently apply best practices in every situation,” Dr. Cromwell said.

Plus, this risk assessment is objective, eliminating the possibility for discrepancies based on which provider is treating the patient.

Although AI systems are expensive and often tricky to implement, the benefits of predictive analytics can save lives and cut down on readmissions. Plus, they can eventually result in cost savings down the line.

Talk to your IT department and finance pros to see if machine learning technology would be a valuable and feasible addition to your facility.