How Davies Winner St. Clair Supports Clinicians

St. Clair Hospital, Pittsburgh, derives benefits from its EHR by focusing on providing the information that caregivers really need.


Federal and state documentation mandates, including meaningful use, are ratcheting up the pressure on caregivers to make care decisions based on the best clinical and financial methodologies.

Provider organizations have responded with a focus on providing point-of-care decision support. But Rich Schaeffer, vice president and CIO at St. Clair Hospital, Pittsburgh, Pa., believes the efforts at some organizations have lost focus on what caregivers really need.

“If you ask physicians if the clinical decision support they’re using it making them more efficient, most of them will tell you no—there’s still a lot of frustration out there with how they have to interact with an EHR to document care and make clinically sound decisions,” Schaeffer says.

“Vendors have been occupied with building meaningful use and ICD-10 infrastructure into their systems, and there hasn’t been enough emphasis on creating efficiencies for caregivers,” Schaeffer adds. “We’ve focused our efforts on taking the complexity, and data entry, out of the process by putting automated decision support in the pathways clinicians are using.”

St. Clair’s relentless focus on efficiency is the reason the hospital was named one of four recipients of the 2015 HIMSS Enterprise Nicolas E. Davies Award of Excellence, which recognizes outstanding achievement of organizations that have utilized health information technology to substantially improve patient outcomes while achieving return on investment.

An example of St. Clair’s efforts to embed efficiencies into the EHR is the hospital’s initiative to reduce the amount of time it takes to begin antibiotic IVs. “Time is critical when a patient has an infection, and we had a workflow where physicians make the correct order, but for various reasons the start of an IV was delayed,” Schaeffer says.

Before June 2014, the daily rate of missing start times was 4.5 per day at the hospital. To address that problem, the EHR now auto-orders some medications to ensure that patients receive antibiotics immediately. As a result, the number of reduced antibiotic start times has been brought down to 0.088 per day.

Schaeffer says that building seamless logic into the workflow is the work of which he and his team are most proud, because it makes decision-making less burdensome for clinicians and eliminates the potential for alert fatigue among clinical staff.

When the hospital started devising a response to Clostridium difficile [C. Diff] infection rates, the pharmacy team recommended a probiotic regimen for patients receiving antibiotics that could lead to C. Diff infections. The decision to start probiotics is based on complex, behind-the-scenes logic that determines if the medication is an antibiotic, and if so, analyzes whether it is antiviral, antifungal, antimalarial, antituberculosis or anthelmintic. It then compares the medication an exclusion list and the patient type, age, duration of the medication therapy, lab results and probiotic duplication.

The pharmacy group initially wanted to embed alerts into the EHR when physicians ordered certain antibiotics, but physicians objected to the idea of having to go through additional screens and make extra clicks every time an antibiotic was ordered. So instead, once the antibiotic is ordered, the EHR automatically enters the order for the probiotic based on that embedded logic.

“The C. Diff effort required us to consider numerous data points along with physician workflow and build true decision support at the moment physicians needed it,” Schaeffer says. “We keep looking for these types of opportunities to get rid of nuisance alerts and make the EHR more usable in a clinically significant way.”

That strategy has spurred St. Clair to redesign its hypoglycemia protocol to decrease hyperglycemic episodes as well as identify undiagnosed diabetic patients. Like the C. Diff effort, the embedded EHR logic created automatic orders and standard order sets for patients based on test results, and created an automatic order set that triggers an HgbA1c test for any patient that doesn't have a record of an HgbA1C test within the last 90 days.

The effort has reduced the incidence of hyperglycemia from 659 per 1,000 patient stays to 416 per 1,000 patient stays over the past three years.

“Once you get an EHR in place, this is where you really have to get to work—finding those opportunities to assist physicians at each point in their decision-making process,” Schaeffer says. “It’s time-consuming and complex, but that’s what makes the system more usable and efficient, which are the most important goals.”

More for you

Loading data for hdm_tax_topic #better-outcomes...