A staple in patient safety efforts, root cause analysis (RCA) has led to all kinds of lifesaving improvements at children’s hospitals.
But the outcomes of the tried-and-true analysis come only after patient harm occurs. In hopes to get ahead of harm, Johns Hopkins Children’s Center adapted an approach used in other industries called learning teams.
“Learning teams focus on not just harms but hazards, trying to identify things before they become a problem,” said Rahul Koka, MD, surgical director of quality and logistics for the hospital. “It ensures our teams are learning from everyday work rather than from accidents alone.”
Learning teams study work-as-done rather than work-as-imagined. By focusing on what staff are actually doing every day (rather than assuming they’re following policies), safety teams are more likely to spot potential risks.
With learning teams, the hospital’s outcomes improved substantially across departments in a variety of areas, including pressure injuries, hypothermia, and craniofacial surgeries.
The hospital uses three main types of learning teams:
- Event‑based: reactive like RCAs, but more collaborative
- Continuous: ongoing, proactive reviews
- High‑risk: focused on specific patient groups or procedures
Event-based learning teams
Learning teams review safety events on a regular cadence through curious conversations in a psychologically safe space. They focus on identifying patterns, seeking to understand, and examining the actual work rather than deviations from protocol.
“RCAs still have their use, and learning teams won’t work with everything,” Koka said. “But if the RCA doesn’t sit well with you, try a learning team in addition.”
| Root Cause Analysis | Learning Team |
|---|---|
| Investigates (not always blame-free) | Focuses on understanding |
| Looks for mistakes | Looks for patterns |
| Explores deviations from protocol | Explores work as it actually happens |
| Quick fixes | Learning translated into real actions |
| Minimal long-term follow up | Feedback loops |
Case: A patient undergoing a 16-hour surgery underwent multiple failed extubations because of an unintentionally retained gauze throat pack despite correct material counts following the operation.
An RCA pointed to protocol deviations, noted surgeons should avoid it next time, and moved on. Unsatisfied with this result, the learning team gathered frontline staff and quality improvement specialists to:
- Ask open questions like “What surprised you? What was different?”
- Identify the irreversible last step with a “critical steps analysis”
- Brainstorm real‑world safeguards, assuming counts are correct and a throat pack will be placed
This process enabled the team to come up with several solutions to prevent the event in the future (e.g., a tape “tail” affixed to the throat pack so it exits with the endotracheal tube or making a note of the pack’s presence on the patient’s forehead).
Continuous learning teams
These multidisciplinary groups meet regularly to examine common or high-risk procedures, emphasizing curiosity.
In addition to following typical quality improvement processes (process mapping, fishbone, checklist, etc.), the teams add the “Four Ds” to their discussions: what is dumb, dangerous, different, or difficult?
“This one really connects with people,” Koka said. “It prompts a lot of discussion, and it helped us gain a lot of traction.”
Case: A multidisciplinary team met monthly for 45 minutes to discuss newborns returning from the OR with hypothermia.
Evaluating the cases using the Four Ds led to small, frontline‑suggested fixes — like relocating checklists and securing transport equipment — that yielded a sustained drop in hypothermia from 17 % to 0.9% over six months.
High‑risk learning teams
These continuous learning teams focus on high-risk patients, meeting once a week for 20 minutes.
By meeting every week, we can really individualize and personalize the care we're providing,” said Marissa Leff, quality and safety coordinator.
'Ultimately, if you take frontline knowledge and add some quality improvement on a continuous basis, you can get a lot done.'
Frontline staff from all phases of care discuss the patient’s status from the last week and look forward to the next week(s). At the meetings, they ask questions like:
- What harms could happen to the patient?
- What critical steps produce safety threats?
- What key work activities cannot fail?
- What is different about the next patient?
“Frontline clinicians and people who are with the patient every single day weren't always sought for their feedback on these questions, and they have helped us tremendously to change our processes and make improvements,” Leff said.
The high-risk learning teams have led to reduced pressure injuries, shortened ICU stays through revised sedation policies, and improved pain control through pre‑op nerve patches and intra‑op Botox.
“We’ve been able to do all of this with just 20 minutes a week when people show up and are engaged,” she said.
Case: Length-of-stay for patients undergoing bladder exstrophy repairs was less than ideal.
Weekly 20‑minute conversations among surgeons, nurses, anesthesiologists, and quality improvement staff led to the realization that many pre-admissions were unnecessary, allowing them to cut pre‑admission days from 94% of patients to 20% — eliminating an entire day from patients’ length-of-stay.
Key lessons
- Narrow the scope: Smaller, achievable objectives help maintain focus.
- Engage frontline staff: Bottom‑up participation improves results.
- Keep teams small: Small teams result in better efficiency and collaboration.
- Emphasize qualitative insight: Qualitative assessments drive improvements that data alone can’t provide.
- Celebrate wins: Highlighting success leads to more interest and builds momentum.
Koka said, “Ultimately, if you take frontline knowledge and add some quality improvement on a continuous basis, you can get a lot done.”
This article is based on a presentation at the Children’s Hospital Association’s 2025 Transforming Quality Conference.
