Rady Children's Hospital used predictive analytics to reduce length of stay in the emergency department and improve patient satisfaction.
Emergency departments (ED) are the front door to organizations and often the only door patients enter at a hospital. It's a critical opportunity to make a good first impression. Two years ago at Rady Children's Hospital San Diego, it was not unusual for there to be 40 to 60 families waiting in the ED during peak hours. "It wasn't fun to work in our ED," says Keri Carstairs, M.D., FACEP, Emergency Department division chief and medical director at Rady Children's. "But what we've done in our ED to make things better for patients, families and staff members can be done anywhere, whether it's a big or small hospital."
Rady Children's pediatric ED, a Level 1 pediatric trauma center, serves more than 90,000 patients a year. The organization's ED median length of stay and left without being seen rate exceeded industry benchmarks, and patient satisfaction scores were suboptimal during the 2013-14 fiscal year.
At the 2016 Annual Leadership Conference, Carstairs, along with Lisa McDonough, RN, Emergency Department nurse manager, discussed how the executive team, ED leadership, hospital flow committee, IT, and the quality management and finance departments used demand capacity analysis to improve ED flow and staffing to improve quality metrics. "We've had two years of sustainable results," Carstairs says. "We only have 40 to 50 beds in our ED. Based on demand, we really need more like 70 beds. But we've created efficiencies to improve capacity to meet demand."
Managing ED demand and capacity
To tackle this, you must ask the right questions. "Don't ask why a family is here—no parent plans an emergency," Carstairs says. "Instead ask who, when and what: who's coming, when are they coming, and what will they need when they arrive? Then look at capacity. How will we immediately meet patients' needs when they arrive? Where are we going to put them? Will we have the right resources at the right time to meet those needs?"
Innovative staffing plan
It's a common belief that the ED is an unpredictable environment. But using predictive analytics, an organization can get a clearer picture. "When we first looked at our data, we pulled 4 years of arrival data," Carstairs says. We knew by looking at what we weren't doing right, we had to start over. Since our census has grown, we've had to change."
The EMR has answers
Every ED has the same arrival pattern. Once an organization realizes this, it can figure out where the gaps are. The electronic medical record will help:
- Match hourly operations to patient demand
- Measure data so you can manage more effectively
- Examine expected arrivals and layer that data over length of stay
- Look at seasonal variations and days of the week
Carstairs says it's not just about arrivals but what you do when those arrivals show up. Use data as a framework and know what your medical team can do and where you can stretch them. Add operational insights and keep in mind the realities of shift work and fatigue. "We don't want to staff in real time, we want to staff based on what is going to happen down the road," she says. "We may have empty beds but we are fully staffed because we know what's coming down the road."
Generating revenue and patient satisfaction
The result is a less stressed staff who can more quickly accommodate patients and families. "To go in and know every night there were 50 patients waiting to see a bed—that was so defeating to the care team," McDonough says. "But now, physician retention isn't an issue anymore. They don't mind working nights because they know what to expect, and patients are getting what they deserve. They didn't come to sit in the waiting room."
In addition to happier patients and staff, the hospital has reduced length of stay by 40 minutes. That equals 167 hours of nursing time each day that can be reinvested into other patients. The ED has also generated $3 million additional revenue since 2014 because of these scheduling efforts.