Skip to main content
Skip to footer

Data analysis helps MU Health Care reduce IV waste and improve efficiency

by MU Health Care

Published on 9/5/2019

Staff at University of Missouri Health Care in Columbia, Missouri, knew their IV batch fill and distribution process had room for improvement.

“We knew we were not operating efficiently,” said Drew Jett, acute care pharmacy manager at University Hospital. “Too often, we prepared an IV order and it was changed or canceled prior to administration. We wasted time and resources to fill, deliver and return orders that went unused.”

The situation led pharmacy leaders to work with Cerner’s Value Management Advisory Services team and the on-site Tiger Institute team to quantify the organization’s IV obsolescence rate — the percentage of batch-filled IVs changed or canceled after the label printed but before the scheduled administration time.

The teams analyzed order data to refine pharmacy batch fill processes and adjust staff schedules, ultimately reducing the organization’s IV obsolescence rate by more than 43% over three months.1

The Cerner ITWorksSM client’s University Hospital traditionally filled each day’s routine IV orders in a single batch. But research links more frequent batch fills to efficiency improvements and waste reduction, so pharmacy leaders considered adding a second daily batch fill.2

“The Cerner team was able to pull data that allowed us to make data-driven decisions about how to set up the new IV fills and accurately predict the reduction in obsolescence,” explained Jett.

In addition to tracking IV obsolescence rate, the Cerner team analyzed MU Health Care’s clinical data to reveal patterns in IV order and administration times. The team uncovered behavioral insights that influence ordering patterns, like the fact that IV orders often change right before lunch, after providers complete their morning rounds.

After using data analysis to determine optimal IV batch fill times, the Value Management Advisory Services team simulated outcomes for newly proposed scenarios using MU Health Care’s past IV order data. Pharmacy leaders adjusted staffing models to accommodate new batch fill times.

Even the organization’s Women’s and Children’s Hospital, which was already processing two IV batches per day, benefitted from the opportunity to holistically reexamine pharmacy operations through a new, data-informed lens.

“With the obsolescence data, we were able to optimize fill times for our batches,” said Jodie Wehrman, inpatient pharmacy manager at Women’s and Children’s Hospital. “This was an insignificant change in process for our staff but led to a significant reduction in wasted doses.”

Following the changes, IV obsolescence rates aligned closely with pre-implementation projections and average lag time between IV batch fill and administration decreased 48%.3 Reducing lag time helps decrease waste by shortening the timeframe for order changes, patient transfers or discharges after an IV has been prepared and sent to the patient’s floor for administration.

“As an academic health system, we apply a research-driven approach to everything we do — from clinical care to operations,” explained Brad Myers, executive director of pharmacy and lab services. “This project’s value came from tailoring a research-backed concept to our unique needs and environment. We’re using data analysis to operate more efficiently.”

1 Average batch fill IV obsolescence rate decreased from 13.4% between Oct. 5 - Dec. 18, 2018 to 7.6% between Feb. 11 - April 30, 2019.

2 Abbasi G, Gay E. Impact of Sterile Compounding Batch Frequency on Pharmaceutical Waste. Hosp Pharm. 2017;52(1):60–64. doi:10.1310/hpj5201-60

3 Average difference between IV fill time and administration time decreased from 17.5 hours between Oct. 5 - Dec. 18, 2018 to 9.1 hours between Feb. 11 - April 30, 2019.

Client outcomes were achieved in respective settings and are not representative of benefits realized by all clients due to many variables, including solution scope, client capabilities and business and implementation models.