CAUTI Doubles in a Neuro ICU: Finding the System, Not the Nurse
The neuro ICU in a 4-hospital system has 32 beds. Over 6 months, catheter-associated urinary tract infection (CAUTI) rose from 0.8 to 2.4 per 1,000 catheter-days; 7 cases occurred, 5 on night shift. Chart audits show a documented daily necessity review in 68% of catheter-days. The bladder scanner has been out for repair for 3 weeks. Night staffing is 3.8 HPPD versus 4.6 on days; turnover is 24%. The electronic order set defaults to 'Foley to gravity' and does not require a reason. A nurse-driven removal protocol exists but requires a provider co-sign. The CNS cochairs the infection prevention committee.
1. Map the contributors across system levels. Which factors are most likely to explain the night-shift clustering?
Read the worked answer
System levels:
- Policy/regulation: nurse-driven removal requires a provider co-sign, which is a built-in delay, especially at night.
- Technology/equipment: bladder scanner unavailable for 3 weeks removes the safe alternative to catheterization.
- Design/defaults: order set defaults to Foley without an indication, making insertion the path of least resistance.
- Staffing/workload: night HPPD 3.8 versus 4.6, turnover 24%, fewer resources for toileting assistance.
- Culture/education: daily necessity review documented only 68%; staff may believe the catheter is 'needed' because it has always been there.
Night clustering is not a night-nurse character flaw. At night there is less provider availability for co-sign, fewer assistive staff for toileting, and less oversight. The scanner outage disproportionately affects the shift trying to avoid catheterization. The system has made removal harder than insertion.
Avoid blaming the night shift; that drives underreporting. Frame the question as: what makes the safe behavior easy and the unsafe behavior hard?
2. Choose the two highest-leverage changes and predict two unintended consequences of each.
Read the worked answer
Change 1: Remove the provider co-sign from the nurse-driven removal protocol and embed objective removal criteria.
- Unintended consequence: premature removal in a patient who still needs strict output monitoring (e.g., neurosurgical patient on mannitol).
- Mitigation: explicit inclusion/exclusion criteria, charge-nurse review, and a bladder scanner.
- Second consequence: increased incontinence care and skin breakdown.
- Mitigation: toileting schedules and incontinence products.
Change 2: Redesign the order set to require an indication and auto-expire at 48 hours unless renewed.
- Unintended consequence: alert fatigue and click-through workarounds.
- Mitigation: one indication field, not a pop-up cascade.
- Second consequence: patients who genuinely need long-term catheters get removed and reinserted, increasing trauma and infection risk.
- Mitigation: clear approved indications and a daily review huddle.
Also fix the scanner as a prerequisite; it is a quick win that makes the other changes credible. High leverage means changing defaults and removing barriers, not adding another education module.
3. Design a measurement plan: process, outcome, and balancing measures, plus a 2-week PDSA.
Read the worked answer
Process measures: daily necessity documented (target >95%), catheter-days per 100 patient-days, appropriate indication at insertion, nurse-driven removals completed without co-sign, scanner availability.
Outcome measure: CAUTI per 1,000 catheter-days, with a run chart rather than a single rate; track by shift and unit.
Balancing measures: urinary retention requiring straight catheterization, falls, incontinence-associated dermatitis, nurse workload and satisfaction, and unreimbursed costs.
2-week PDSA on one night-shift pod:
- Plan: test nurse-driven removal on 10 patients using a one-page checklist and charge-nurse review; ensure scanner is available.
- Do: run for 14 nights; collect checklist completion and any retention events.
- Study: weekly huddle to review failures and near misses; ask nurses what got in the way.
- Act: adopt, adapt, or abandon; then spread to days.
Systems thinking test: if the intervention only works when the CNS is present, it is not a system change.