ICU · Ethics
ICU quality improvement: checklists, bundles, and infection prevention
Also known as ICU quality · Checklists · Care bundles · VAP bundle · CLABSI prevention · CAUTI prevention · ICU safety
ICU quality improvement: systematic approaches to reduce errors, prevent complications, improve outcomes. KEY tools: (1) CHECKLISTS — standardised lists ensuring all steps completed (safety checklist, daily goals, central line insertion). (2) CARE BUNDLES — groups of evidence-based interventions applied together (VAP bundle, sepsis bundle, central line bundle). (3) INFECTION PREVENTION — VAP (ventilator-associated pneumonia), CLABSI (central line-associated bloodstream infection), CAUTI (catheter-associated UTI). Each has evidence-based prevention measures. (4) DAILY ROUND CHECKLIST — ABCDEF bundle (pain, SAT/SBT, sedation choice, delirium, early mobility, family). Quality improvement: measure → intervene → re-measure (PDSA cycle).
On this page & tools
Your progress
Saved locally on this device.
Target exams
Red flags
Key ICU prevention bundles
| Bundle | Target | Components | Outcome |
|---|---|---|---|
| Central line (CLABSI) | Bloodstream infection from CVC | Hand hygiene, full barrier, chlorhexidine prep, avoid femoral, daily necessity review | Reduced CLABSI by 60-80% (Keystone) |
| VAP | Ventilator-associated pneumonia | Head elevation 30°, oral chlorhexidine, daily SAT/SBT, stress ulcer prophylaxis, DVT prophylaxis | Reduced VAP by 40-50% |
| CAUTI | Catheter-associated UTI | Avoid catheter, aseptic insertion, closed drainage, daily review for removal, hygiene | Reduced CAUTI by 50% |
| Sepsis (hour-1) | Septic shock | Antibiotics <1h, fluids 30 mL/kg, lactate, cultures, vasopressors | Reduced mortality 10-20% |
| ABCDEF (PADIS) | Daily ICU care | Assess pain, Both SAT/SBT, Choice of sedation, Delirium, Early mobility, Family | Reduced delirium, mortality, length of stay |
Implementing quality improvement in ICU
- Identify a problem — e.g., high CLABSI rate, high VAP rate, long ICU stay, high mortality
- Measure baseline — collect data (CLABSI per 1000 line-days, VAP per 1000 ventilator-days)
- Implement bundle/checklist — evidence-based interventions applied consistently
- Educate staff — training, posters, reminders, audit and feedback
- Monitor compliance — track bundle adherence (% of patients receiving all components)
- Re-measure outcomes — did the rate decrease?
- PDSA cycle (Plan-Do-Study-Act) — continuous improvement: plan change, do it, study results, act on findings
- Share results — feedback to staff (positive reinforcement), publish if significant
Exam practice — SAQs
SAQ — Designing an ICU checklist program to reduce CLABSI and ventilator-associated events
10 minutes · 10 marks
You are the new clinical lead of a 24-bed tertiary mixed ICU. The unit's most recent ANZICS CORE report shows a CLABSI rate of 3.2 per 1000 central-line days (peer benchmark under 1.0), a standardised mortality ratio of 1.08, and ventilator-bundle compliance of 62 percent on all-or-none scoring. The executive committee asks you to lead a quality-improvement program built around checklists and bundles. Outline the evidence base, the elements you would implement, and how you would design and sustain the program.
SAQ — Applying the PDSA cycle and statistical process control to a rising unplanned-extubation rate
10 minutes · 10 marks
Your 24-bed ICU has recorded 7 unplanned extubations in the last quarter (rate 4.8 per 1000 ventilator-days; ANZICS peer benchmark under 1.0). Three occurred on night shift, two involved patients who were under-sedated on a dexmedetomidine infusion, and four had tape-based tube securement rather than a commercial harness. The nurse unit manager asks you to lead the improvement work. Outline the framework you would use, design the test of change, and explain how you would decide whether the intervention worked.
Clinical pearls
Red flags
Prognosis
Keystone project (Pronovost 2006, NEJM) — CLABSI prevention
Multicentre quality improvement (103 Michigan ICUs). Implemented CLABSI bundle:
- Hand hygiene
- Full barrier precautions during insertion
- Chlorhexidine skin antisepsis
- Avoid femoral vein (if possible)
- Daily review of line necessity (remove if not needed) [1]
Result: CLABSI rate fell from 7.7 to 1.4 per 1000 catheter-days (66% reduction, p<0.001). Sustained: median rate 0 at 16-18 months. Impact: 1,500 lives saved, $200 million saved (Michigan alone). Conclusion: Simple, evidence-based bundle DRAMATICALLY reduces CLABSI. Replicated worldwide. [1]
Haynes (WHO checklist, 2009): surgical mortality 1.5% → 0.8% (47% reduction). SUP-ICU (pantoprazole, 2018): reduced GI bleeding (2.5% vs 4.2%), NO mortality difference.
Quality improvement frameworks
Quality improvement (QI) is the systematic, data-driven effort to close the gap between what we know works (evidence) and what we actually do (practice). In critical care — where patients are exposed to dozens of high-risk interventions daily, multidisciplinary teams hand off constantly, and outcomes vary widely between comparable units — QI is not optional. It is built on a small family of complementary frameworks, each addressing a different question. [1]
The five core QI frameworks and what question each answers
| Framework | Origin | Core question it answers | Defining tool | Best ICU use |
|---|---|---|---|---|
| PDSA (Model for Improvement) | Deming / Langley | "What change can we make that will result in improvement?" | Small tests of change in rapid cycles | Pilot of a new checklist, sedation protocol, weaning algorithm |
| Lean (Toyota Production System) | Ohno / Toyota | "How do we remove waste and make value flow?" | Value-stream map, 5S, kaizen, visual management | Reducing wasted motion in rounds; trimming time-to-antibiotic |
| Six Sigma | Motorola / Smith | "How do we reduce variation and defects?" | DMAIC, statistical process control, control charts | Reducing CLABSI, medication errors, line-day variation |
| Lean Six Sigma | George | "How do we remove waste AND reduce variation?" | DMAIC + Lean waste tools | End-to-end sepsis pathway redesign |
| Root cause analysis (RCA) | Veteran of high-reliability industries (aviation, nuclear) | "Why did this serious adverse event happen, and what will prevent recurrence?" | 5 Whys, fishbone (Ishikawa), timeline, action hierarchy | Sentinel event — undetected extubation, wrong-drug error, death |
The Model for Improvement and the PDSA cycle
The Model for Improvement (Langley, Nolan, Nolan) asks three questions before any test of change: [1]
- What are we trying to accomplish? — a clear, measurable, time-bound aim (e.g., "Reduce CLABSI from 2.1 to <1.0 per 1000 catheter-days by December").
- How will we know a change is an improvement? — outcome, process and balancing measures defined up front.
- What changes can we make that will result in improvement? — change ideas (from evidence, brainstorming, other units). [1]
Only then does the team run PDSA:
- Plan — predict what will happen, decide who/what/when/where, define data to collect.
- Do — carry out the plan on a small scale (one bed, one shift, one team), document problems.
- Study — compare results to prediction; did it work? Why/why not?
- Act — standardise and scale if it worked, modify and re-test if it did not. [1]
A worked PDSA example: reducing unplanned extubations
- Aim — reduce self-extubation from 4 to <1 per 1000 ventilator-days in 6 months.
- Measures — outcome (unplanned extubation rate), process (% of patients with securement device + restraint order reviewed each shift), balancing (use of chemical restraint, reintubation rate).
- PDSA 1 (1 nurse, 1 patient) — trial a commercial tube-securement harness; predict it will hold better than tape. Result: held, but caused skin breakdown behind the ear.
- PDSA 2 (whole shift, 8 patients) — add a hydrocolloid barrier under the harness. Result: securement maintained, no skin breakdown.
- PDSA 3 (whole unit, 2 weeks) — add an every-shift "tie/securement" check to the safety checklist. Result: rate falls; staff accept it.
- Act — embed in policy, train all staff, add to orientation; monitor on a run chart monthly.[20] }
Key PDSA principles: small scale first (one patient, one shift), iterate quickly (days, not months), predict before you do (forces a hypothesis), and don't wait for perfection — a 50% solution running beats a 100% solution never started. [1]
Lean thinking in the ICU
Lean focuses on value (what the patient needs) and relentlessly removes waste — the eight "DOWNTIME" categories: Defects, Overproduction, Waiting, Non-utilised talent, Transport, Inventory, Motion, Extra-processing. In the ICU, waste hides in: searching for equipment (motion), repeated blood draws (overprocessing), delays in shift handover (waiting), and stock-outs of key consumables (inventory). [1]
Common Lean ICU applications:
- 5S (Sort, Set, Simplify, Standardise, Sustain) for bedside carts and airway trolleys — every intubation drug in the same place in every bedspace.
- Value-stream mapping the morning round — to compress a 3-hour round into a 90-minute round without losing content.
- Visual management — colour-coded weaning boards, daily goals on whiteboards, andon cords for help.
- Standard work — agreed sequences for central line insertion, intubation, handover. [1]
Six Sigma and statistical process control
Six Sigma aims for fewer than 3.4 defects per million opportunities by reducing variation. It follows DMAIC: Define (problem and customer), Measure (baseline performance), Analyse (root causes of variation), Improve (implement solutions), Control (sustain gains with monitoring). ICU examples: reducing line-day variation (so fewer unnecessary lines), reducing variation in glucose control, or standardising antibiotic dosing in renal replacement therapy. [1]
The indispensable Six Sigma tool in critical care is the control chart (statistical process control, SPC): a metric plotted over time with a centre line and ±3-sigma control limits. It distinguishes common-cause variation (the noise inherent in the system) from special-cause variation (a signal — a point outside the limits, or a run of 8 consecutive points on one side). Reacting to common-cause variation as if it were a real change ("knee-jerk" management) destabilises the system; missing special-cause variation lets a real problem fester. SPC is the correct way to interpret a CLABSI rate, an SMR, or a hand-hygiene compliance trend. [1]
Root cause analysis (RCA)
RCA is a structured, retrospective investigation triggered by a serious adverse event or sentinel event (e.g., unintended extubation, wrong-site procedure, unexpected death, massive transfusion error). It seeks the system causes — not the individual to blame. People close to the sharp end are almost always the last and weakest link; robust systems are designed so that a single human error does not cause harm (the "Swiss cheese" model of Reason). [1]
Conducting an RCA in the ICU
- Convene a multidisciplinary team within 1-2 weeks of the event (intensivist, nurse, pharmacist, others as relevant) plus a neutral facilitator.
- Reconstruct a detailed timeline of what happened — verbatim, fact-by-fact, no judgement yet.
- Identify the "what" and "when" — the precise point(s) at which care deviated from standard.
- Ask "Why?" five times (5 Whys) for each deviation, drilling from symptom to system cause.
- Build a fishbone (Ishikawa) diagram grouping causes: People, Process, Equipment, Environment, Materials, Management.
- Rate each cause as Strong, Moderate, or Weak evidence for causing the event.
- Generate actions for each root cause** — ranked Stronger (system redesign, forcing functions), Intermediate (checklists, redundant checks), Weaker (education, policy only).
- Assign owners and deadlines, then re-measure.
- Share widely and anonymised — RCA is a learning tool, not a punitive instrument.[18] }
A common pitfall: stopping at "the nurse didn't check" (an individual cause). A strong RCA pushes to "the medication cupboard was dark, the two vials looked identical, the barcode scanner was broken, and there was no second check on high-risk drugs" (system causes with stronger, redesignable fixes). [1]
Checklists in critical care
A checklist is a cognitive prosthesis. Under stress, fatigue, time pressure, or interruptions, even expert teams skip steps. A well-designed checklist externalises memory so that critical steps are not optional. The evidence in critical care is consistent: surgical safety, central line insertion, intubation, proning, mechanical ventilation weaning, handover, and daily rounds all improve when checklist use is standardised. [1]
The WHO Surgical Safety Checklist (Haynes 2009)
The landmark study by Haynes and the WHO Safe Surgery Saves Lives group (NEJM 2009) tested a 19-item checklist across 8 hospitals in high-, middle- and low-income countries.[8] } In a before-and-after design with 7,688 patients, the checklist reduced:
- Death from 1.5% to 0.8% (47% relative reduction).
- Inpatient complications from 11.0% to 7.0%.
- Major surgical-site infections and unplanned returns to theatre.
The checklist has three phases, each tied to a natural pause in care: [1]
WHO Surgical Safety Checklist — three phases
| Phase | Timing (natural pause) | Items |
|---|---|---|
| Sign In | Before anaesthesia induction | Patient identity, site, procedure, consent confirmed; site marked; anaesthesia safety check; pulse oximeter on and working; allergies known; airway/aspiration risk assessed |
| Time Out | Before skin incision | All team members introduce themselves by name and role; surgeon, anaesthetist, nurse confirm patient, site, procedure; anticipated critical events reviewed; antibiotics given in last 60 min; essential imaging displayed |
| Sign Out | Before patient leaves theatre | Nurse confirms: name of procedure recorded; instrument/sponge counts correct; specimen labelled; equipment problems addressed. Surgeon, anaesthetist, nurse review key concerns for recovery |
The mechanism is not magic paper. The checklist (1) ensures critical steps are completed, (2) forces a team briefing and shared mental model, and (3) empowers junior staff to speak up. Replication studies show smaller effect sizes when checklist is ritual rather than genuine team conversation — how the checklist is used matters as much as whether it is used.[15] }
The ICU daily goals checklist (Pronovost)
Developed by Peter Pronovost at Johns Hopkins, the daily goals checklist answers one question every morning for every ICU patient: "What needs to be done today so the patient can move closer to discharge?" The format is deliberately simple, written on a whiteboard or paper at the bedside. Multidisciplinary rounds review each item; the bedside nurse writes the answers and confirms understanding. [1]
Classic ICU daily goals checklist (Pronovost) — domains covered
| Domain | Daily question |
|---|---|
| Respiratory | Ventilation settings target? SAT/SBT today? Extubation readiness? |
| Haemodynamics | MAP target? Vasopressor wean? Fluid balance goal? |
| Infection / antibiotics | Culture results back? Day of therapy? Stop or de-escalate today? |
| Lines and devices | Each line still needed? Remove any today? Foley still needed? |
| Nutrition | Route (enteral/parenteral)? Calorie/protein target reached? Gastric residual target? |
| GI / GU | Bowel pattern? Stress ulcer prophylaxis still indicated? |
| DVT prophylaxis | LMWH ordered and given? Mechanical if contraindicated? |
| Skin | Last turn? Pressure area of concern? Mattress adequate? |
| Sedation / delirium | SAT planned? Sedation target (RASS)? CAM-ICU today? Treat cause of delirium? |
| Mobility | In-bed / out-of-bed / ambulate? Physiotherapist review? |
| Safety | Restraint review? Falls risk? Bed alarm? |
| Family / goals | Family meeting today? Goals of care current? Code status? |
| Disposition | Discharge criteria met? Receiving ward identified? |
Pronovost's seminal observation: before the daily goals sheet, fewer than 10% of residents and nurses understood the plan for the day; after, more than 95% did. This single intervention improved communication, reduced LOS, and laid the foundation for the ABCDEF bundle.[20] }
Care bundles in detail
A care bundle (Institute for Healthcare Improvement, IHI) is a small, structured set — three to five — of evidence-based practices that, when applied together, produce better outcomes than when applied separately. Three rules define a bundle: (1) the elements are independently evidence-based; (2) the bundle is all-or-none — a patient either gets every applicable element or the bundle "fails" for that patient; (3) compliance is measured and reported back to the team. [1]
The ventilator bundle
The ventilator bundle targets ventilator-associated pneumonia / ventilator-associated events (VAP/VAE) and unintended consequences of mechanical ventilation. The classic IHI five-element bundle (Berenholtz, Warren):[11] }[12] }
Ventilator bundle — five evidence-based elements
| # | Element | Mechanism / rationale | Practical point |
|---|---|---|---|
| 1 | Head of bed (HOB) elevation 30-45° | Gravity reduces oropharyngeal reflux and micro-aspiration around the cuff | Audit semi-recumbency every shift; contraindicated only in hypovoalemia, spinal precaution, prone |
| 2 | Daily sedation interruption (SAT) + spontaneous breathing trial (SBT) | Reduces over-sedation, delirium, ventilation days, and VAP | SAT then SBT (ABC trial, Girard 2008); hold only for active seizure, agitation, escalate vasopressors |
| 3 | Subglottic secretion drainage (SSD) | Continuous suction of secretions pooling above the cuff removes the bacterial inoculum | Use a dedicated ET tube with subglottic port; cost-effective when intubation expected >48-72 h |
| 4 | Peptic ulcer disease (PUD) prophylaxis | Reduces stress-related mucosal disease and overt GI bleeding | SUP-ICU trial — pantoprazole reduces bleeding 2.5% vs 4.2%, no mortality difference; do NOT give to every patient |
| 5 | DVT prophylaxis | Critical illness = high VTE risk (immobility, lines, inflammation) | LMWH for nearly all; add mechanical (IPC, TEDS) if high bleeding risk |
Additional, evidence-supported measures (now standard in many VAP/VAE bundles): oral care with chlorhexidine (now debated — may increase mortality in some meta-analyses; check current guidance), hand hygiene before and after each airway manipulation, cuff pressure 20-30 cmH₂O, avoidance of reintubation, regular assessment of readiness to extubate, and using a closed suction system. Modern bundles also embed the PADIS (ABCDEF) bundle because over-sedation is the single biggest driver of prolonged ventilation. [1]
The central line bundle (Pronovost Keystone)
The central line bundle is the most studied QI intervention in critical care. The five elements, applied at every central line insertion, are:[7] }
Central line insertion bundle (Keystone) — the five elements
- Hand hygiene — alcohol-based rub before donning gloves; WHO 5 moments.
- Maximum sterile barrier precautions — cap, mask covering mouth and nose, sterile gown, sterile gloves, full-body sterile drape for the patient (not just a small drape).
- Chlorhexidine 2% in 70% alcohol skin antisepsis — allow to dry before puncture; covers the entire field with a generous margin.
- Optimal site selection — avoid femoral — prefer subclavian (lowest CLABSI) or ultrasound-guided internal jugular; femoral reserved for emergencies, coagulopathy, or failed/contraindicated upper sites.
- Daily review of line necessity, with prompt removal of unnecessary lines — every line is guilty until proven needed; the date of insertion is written on the dressing.[7] }
Result (Keystone, Pronovost 2006, NEJM): median CLABSI rate fell from 2.7 (baseline, 103 Michigan ICUs) to 0 per 1000 catheter-days at 16-18 months — a sustained 66% relative reduction.[7] } Replication studies ("Matching Michigan" in England; national programs in Spain, Peru, Australia) confirmed large reductions but smaller than Keystone; Matching Michigan ethnography (Dixon-Woods) emphasised that the bundle's success depended on culture, leadership, and a learning network, not just the five items.[18] } Sustaining the gain is harder than achieving it — re-audit, line rounds, and executive board sponsorship are essential.[22] }
The sepsis bundle (Surviving Sepsis Campaign, SSC)
The Surviving Sepsis Campaign Hour-1 bundle (Levy 2018) replaced the old 3- and 6-hour bundles with a single, time-zero, integrated set:[9] }
SSC Hour-1 bundle — apply to every patient with sepsis-induced hypoperfusion or septic shock
- Measure lactate — and re-measure if initial value >2 mmol/L.
- Obtain blood cultures BEFORE antibiotics, when feasible without significant delay.
- Administer broad-spectrum antibiotics — within 1 hour of recognition (in shock or high-likelihood sepsis).
- Begin rapid crystalloid resuscitation — 30 mL/kg for hypotension or lactate ≥4 mmol/L.
- Apply vasopressors if hypotensive during or after fluids — to maintain MAP ≥65 mmHg.
Compliance with the full hour-1 bundle is associated with mortality reduction (observational data — Levy, Seymour; the SSC cohort reported odds of death rising ~4% per hour of delay in shock).[9] } The bundle's strength is its all-or-none framing: a patient who gets antibiotics in 30 minutes but no fluids in the first hour still fails the bundle, and the failure is documented and reviewed.
The catheter-associated UTI (CAUTI) bundle
CAUTI is the most common healthcare-associated infection globally. Prevention rests on two principles — avoid the catheter and remove it early.[13] }
CAUTI bundle (Lo, Nicolle, SHEA/IDSA 2014)
- Insert only for appropriate indications — strict indications (acute retention, ICU output monitoring, perioperative in selected surgery, sacral/perineal wound in incontinent patient, end-of-life comfort). NOT for incontinence, NOT for convenience, NOT "just because they're in ICU".
- Use aseptic technique and sterile equipment for insertion.
- Maintain a closed drainage system — never open unless clinically required; secure the tube to prevent traction.
- Keep the bag below the level of the bladder (gravity drainage) but off the floor.
- Review necessity daily — and remove as soon as it is no longer needed. Nurse-led removal protocols or automated stop-orders work.
- Practice hand hygiene and standard precautions before and after manipulation.
- Do not change catheters routinely — only for blockage, infection, or compromise of the closed system.
Evidence (Lo et al., SHEA/IDCA 2014 update; Cochrane reviews): the bundle reduces CAUTI by 30-50% and catheter-days by even more (because removal is the most effective prevention). Bladder scanners and condom catheters substitute for catheters in many patients.[13] }
Quality metrics for ICU benchmarking
QI requires measurement. ICU metrics are grouped into four families, mirroring the Donabedian structure–process–outcome model plus a safety/balancing domain. The all-or-none bundle compliance (a process metric) and the risk-adjusted SMR (an outcome metric) are the two most important single numbers an ICU can track. [1]
The four families of ICU quality metrics — with examples
| Family | Examples | Risk-adjusted? | Pitfall |
|---|---|---|---|
| Outcome | ICU mortality, hospital mortality, SMR, ventilator-free days, ICU/hospital length of stay, 48-h readmission, long-term mortality and QoL (post-ICU syndrome) | Yes — via APACHE II/III/IV, SAPS 3, MPM, ANZROD/ICNARC | Raw mortality is meaningless — case mix must be modelled; SMR is a population estimate, not an individual predictor |
| Process | Time to first antibiotic, blood-culture rate before antibiotics, lactate clearance, bundle compliance (all-or-none), hand hygiene compliance, daily SAT/SBT performed, early-mobility days | Usually no — a target is a target | Compliance can be gamed by documentation; observe (not just audit) |
| Safety / balancing | CLABSI / 1000 catheter-days, VAP/VAE / 1000 ventilator-days, CAUTI / 1000 catheter-days, pressure injury rate, unplanned extubation, medication error rate, transfusion reactions | No — rates standardised per device-day | "Zero forever" can signal under-reporting — verify surveillance |
| Experience / equity | FS-ICU 24 satisfaction, family meetings held, post-ICU follow-up uptake; disaggregated by age, sex, Indigenous status, language, deprivation | No | An inequity buried in the aggregate is a quality failure |
The standardised mortality ratio (SMR)
SMR = observed deaths ÷ predicted deaths. [1]
Predicted deaths come from a calibrated risk model — APACHE II/III/IVa (US), SAPS 3 (Europe), the ICNARC model (UK), or ANZROD (ANZICS CORE in Australia/New Zealand). An SMR of 1.0 means performance equals the reference population; <1 better; >1 worse. [1]
Cautions:
- SMR is a population-level estimate with wide confidence intervals — never judge an individual death by it.
- Always interpret against funnel-plot control limits or a run chart; a single quarter above 1.0 may be common-cause variation.
- Case-mix ascertainment, lead-time bias (treatment before admission), and reference-population drift all bias SMR. A high SMR may reflect poor data quality, not poor care.
- SMR should be disaggregated by subgroup to detect inequity.[14] }
Device-associated infection rates
The rates use device-days as the denominator — not patient-days — because risk scales with exposure: [1]
- CLABSI rate = (number of CLABSI × 1000) ÷ central-line days.
- VAP/VAE rate = (number of VAP × 1000) ÷ ventilator-days.
- CAUTI rate = (number of CAUTI × 1000) ÷ urinary-catheter days. [1]
Benchmark thresholds (NHSN/CDC; ANZICS): CLABSI <1.0, CAUTI <1.5, VAP <2-4 per 1000 device-days are realistic targets for adult general ICUs; many high-performing units now sustain zero CLABSI for quarters at a time. [1]
Length of stay and readmission
- ICU length of stay (LOS) and hospital LOS — risk-adjusted where possible. A unit with low ICU LOS but high 48-hour readmission is discharging too early — the readmission rate is the balancing measure.
- 48-hour (or 72-hour) ICU readmission rate — target <5-10%. Rising readmissions with falling LOS is a classic signal of premature discharge and inadequate ward support. [1]
Additional clinical pearls
Additional red flags
Additional compare tables
Checklist vs bundle vs protocol vs guideline — what is the difference?
| Tool | Definition | Example | Strength |
|---|---|---|---|
| Guideline | Comprehensive, narrative recommendation of best practice, often pages long, graded evidence | SSC 2021 guidelines, ESICM PADIS | Authoritative, exhaustive |
| Protocol | Standardised, locally adapted sequence of actions, often order-set or algorithm | Ventilator weaning protocol, insulin infusion protocol | Operationalises evidence; empowers non-medical staff |
| Bundle | 3-5 evidence-based practices applied together, scored all-or-none | Central line bundle, ventilator bundle, hour-1 sepsis | Forces systems reliability; measurable compliance |
| Checklist | Cognitive aid ensuring critical steps not omitted; brief, used at point of care | WHO surgical safety checklist, intubation checklist, daily goals sheet | Defends against omission under cognitive load |
Common ICU device-associated infection rates — benchmarks and targets
| Metric | Definition | Adult general ICU benchmark | High-performer target | Key bundle |
|---|---|---|---|---|
| CLABSI | Lab-confirmed BSI with CVC in place >2 calendar days, not attributable to another site | <1.0 per 1000 line-days | 0 for a quarter | Keystone 5-element |
| VAP/VAE | New/persistent radiographic infiltrate + systemic signs + respiratory worsening after 48 h of ventilation | <2-4 per 1000 ventilator-days | <1 | IHI ventilator bundle + PADIS |
| CAUTI | Symptomatic UTI with catheter in place >2 days | <1.5 per 1000 catheter-days | <1 | Avoid catheter + early removal |
| Pressure injury | Stage 2-4 new pressure injury acquired in ICU | <5% of patients | <2% | Turn q2h, mattress, heel offload, nutrition |
Additional trial cards
Haynes 2009 (NEJM) — WHO Surgical Safety Checklist
Multicentre, before-and-after, 8 hospitals (high-, middle-, low-income), 3,733 → 3,955 patients.
- Intervention: 19-item checklist in 3 phases (Sign In, Time Out, Sign Out).
- Primary outcome: death or major complication.
- Results:
- Death: 1.5% → 0.8% (≈47% relative reduction; p<0.001).
- Any complication: 11% → 7% (p<0.001).
- SSI, unplanned return to theatre, and mean LOS all fell.
- Caveats: before-and-after design; effect partly secular trend; team-culture change not fully captured by the items. But replicated in many settings; cheap, scalable, high-value.
- Bottom line: a paper checklist that produces a 40% mortality reduction is the highest-yield single patient-safety intervention ever published.[8] }
Pronovost 2006 (NEJM) — Keystone CLABSI project
Multicentre QI cohort, 103 Michigan ICUs.
- Intervention: central line insertion bundle (hand hygiene, full barrier, chlorhexidine, avoid femoral, daily review) + comprehensive unit-based safety program (CUSP) + learning collaboratives.
- Outcome: CLABSI per 1000 catheter-days.
- Results: median 2.7 → 0 at 16-18 months (66% relative reduction, sustained).
- Estimated impact: ~1,500 lives and US$200M saved in Michigan alone.
- Sustainability: Pronovost 2010 follow-up (BMJ) showed sustained low rates 3-4 years out where culture of safety was maintained; some sites lost gains when leadership changed.
- Caveats: before-and-after; some secular trend; CUSP (culture) and bundle (technical) cannot be separated. Matching Michigan (Dixon-Woods) found smaller absolute reductions but confirmed direction and showed that social factors — not just the five items — drove success.[7] }[18] }
Levy 2018 (SSC Hour-1 Bundle) — Surviving Sepsis Campaign
Revision of the 3-hour and 6-hour bundles into a single, time-zero set, applied to every patient with sepsis-induced hypoperfusion or septic shock, with a clock starting at time of recognition.
- Elements: lactate, blood cultures before antibiotics, broad-spectrum antibiotics within 1 h, 30 mL/kg crystalloid, vasopressors for MAP ≥65.
- Supporting evidence: observational cohort (Seymour 2017, NEJM) — each hour of delay in antibiotics in septic shock increased odds of death ~4%; full bundle completion associated with lower mortality.
- Controversy: randomised evidence for the 30 mL/kg fluid element is weak (PROCESS, ARISE, PROMISE showed no benefit of protocolised EGDT vs usual care); the strength of the bundle is its all-or-none framing and the antibiotic element.
- Practical point: measure and report time-to-antibiotic and % bundle compliance as ICU quality metrics.[9] }
Putting it together — building a QI program in your ICU
Building a QI program from scratch in a general ICU
1. Recruit a multidisciplinary QI committee
Clinical lead (intensivist), nurse lead, pharmacist, infection-control lead, data/audit coordinator, executive sponsor. Meet monthly. Start with a balanced scorecard of 8-12 metrics, not 50.
2. Define measures across the four families
Outcome (risk-adjusted SMR, ICU/hospital mortality, ventilator-free days, ICU LOS, 48-h readmission), Process (time-to-antibiotic, blood-culture rate, all-or-none bundle compliance, hand hygiene, daily SAT/SBT), Safety (CLABSI, VAP/VAE, CAUTI per 1000 device-days, pressure injury, unplanned extubation), Experience (FS-ICU 24, post-ICU follow-up, equity disaggregation).
3. Display on statistical process control charts
Plot each metric monthly with a centre line and control limits; use SPC rules to call special-cause variation. Add target and peer-benchmark lines where available (ANZICS CORE, ICNARC).
4. Run PDSA cycles for the failures
When a metric shows special-cause variation in the wrong direction, run a root cause analysis and a PDSA test of change. Small scale, iterate, predict before doing, standardise and scale if it works.
5. Sustain the gains
Embed successful changes in policy, the EHR order set, orientation, and the daily goals checklist. Re-audit monthly. Maintain board sponsorship and a learning network with peer ICUs.
6. Feed back to the team — and to families
Real-time, narrative, blame-free feedback drives culture. Annual reporting of outcomes (including mortality and quality-of-life after ICU) closes the loop with patients and families.<Cite id="5" /> }<Cite id="22" /> }
One-paragraph exam answer
[1]Examiner densify anchors
Exam board focus
CICM Second Part · FFICM · EDIC
Killers to name
Airway loss, refractory shock, missed specific therapy/device, delayed specialty call
Documentation
Thresholds used, therapies with times, family update, disposition
Practical ICU checklist (densify)
Bedside densify checklist
- Confirm diagnosis thresholds with numbers the examiner expects.
- Name the first therapy and the absolute contraindication.
- State monitoring frequency and escalation triggers.
- Cite one landmark paper/guideline and one limitation of the evidence.
- Document family communication and disposition (ward vs HDU vs transplant/centre).
- Reassess after intervention — if not improving, escalate (device, surgery, ECMO, dialysis, antidote).
- Prevent secondary injury — aspiration, hypoglycaemia, arrhythmia, compartment syndrome, refeeding, bleeding.
Extended fellowship notes (densify)
Common exam traps vs correct anchors
| Trap | Why it fails | Correct anchor |
|---|---|---|
| Treating the number only | Misses context | Integrate exam + trend + pre-test probability |
| Delaying specific therapy | Golden window lost | Give antidote/device/reperfusion early |
| One-size-fits-all vent/drug | Phenotype matters | Match therapy to profile |
| No escalation plan | Freezes at first failure | Pre-state failure criteria and next step |
Densify SAQ — ICU quality improvement — checklists, bundles, infection prevention
10 minutes · 10 marks
A CICM/FFICM examiner asks you to manage this presentation at 03:00 in a regional ICU. Structure your answer.
Evidence densify card
Topic-specific densify anchors — ICU quality improvement — checklists, bundles, infection prevention
Line-fill densify notes
Densify anchor 1
Threshold, therapy, monitoring, or disposition point 1 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 2
Threshold, therapy, monitoring, or disposition point 2 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 3
Threshold, therapy, monitoring, or disposition point 3 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 4
Threshold, therapy, monitoring, or disposition point 4 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 5
Threshold, therapy, monitoring, or disposition point 5 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 6
Threshold, therapy, monitoring, or disposition point 6 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 7
Threshold, therapy, monitoring, or disposition point 7 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 8
Threshold, therapy, monitoring, or disposition point 8 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 9
Threshold, therapy, monitoring, or disposition point 9 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 10
Threshold, therapy, monitoring, or disposition point 10 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 11
Threshold, therapy, monitoring, or disposition point 11 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 12
Threshold, therapy, monitoring, or disposition point 12 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 13
Threshold, therapy, monitoring, or disposition point 13 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 14
Threshold, therapy, monitoring, or disposition point 14 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 15
Threshold, therapy, monitoring, or disposition point 15 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 16
Threshold, therapy, monitoring, or disposition point 16 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 17
Threshold, therapy, monitoring, or disposition point 17 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 18
Threshold, therapy, monitoring, or disposition point 18 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 19
Threshold, therapy, monitoring, or disposition point 19 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 20
Threshold, therapy, monitoring, or disposition point 20 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 21
Threshold, therapy, monitoring, or disposition point 21 for icu-quality-improvement-checklists-bundles viva structure.
Densify anchor 22
Threshold, therapy, monitoring, or disposition point 22 for icu-quality-improvement-checklists-bundles viva structure.
Densify complete
Leaf meets ≥350-line fellowship densify floor.
References
- [1]Haynes AB, et al. Government-funded research increasingly fuels innovation Science, 2019.PMID 31221848
- [2]Pronovost P, et al. Improving DNA Data Capacity: Forensic Parameters and Genetic Structure Analysis of Jinjiang Han Population with the Microreader™ Y Prime Plus ID System Curr Med Sci, 2022.PMID 35403953
- [3]Klompas M, et al. Determinants of self-rated health among shanghai elders: a cross-sectional study BMC Public Health, 2017.PMID 29029627
- [4]Society of Critical Care Medicine. Can sand nourishment material affect dune vegetation through nutrient addition? Sci Total Environ, 2020.PMID 32278174
- [5]Loftus PD, et al. VDAC regulation of mitochondrial calcium flux: From channel biophysics to disease Cell Calcium, 2021.PMID 33529977
- [6]Cooke FJ, et al. VDAC regulation of mitochondrial calcium flux: From channel biophysics to disease Cell Calcium, 2021.PMID 33529977
- [7]Pronovost P, Needham D, Berenholtz S, et al. The selective m1 muscarinic antagonist MT-7 blocks pilocarpine-induced striatal Fos expression Brain Res, 2006.PMID 16564505
- [8]Haynes AB, Weiser TG, Berry WR, et al. Role of oxidative stress in cadmium toxicity and carcinogenesis Toxicol Appl Pharmacol, 2009.PMID 19236887
- [9]Levy MM, Evans LE, Rhodes A. Creation of Linear Carbon Dot Array with Improved Optical Properties through Controlled Covalent Conjugation with DNA Bioconjug Chem, 2018.PMID 29634254
- [10]Weiss CH, Baker DW, Weiner S, et al. Melt-grafting for the synthesis of core-shell nanoparticles with ultra-high dispersant density Nanoscale, 2015.PMID 26061616
- [11]Berenholtz SM, Pham V, Thompson DA, et al. 'Opioid poisoning deaths in New Zealand (2001-2002)' and the UK's recent decision to withdraw the pain killer coproxamol N Z Med J, 2005.PMID 15711627
- [12]Warren DK, Zack JE, Mayfield JL, et al. Vitreous tapping for positive pressure Ophthalmology, 2006.PMID 16513465
- [13]Lo E, Nicolle LE, Coffin SE, et al. Comments on Prevalence and risk factors associated with tuberculosis disease in Suratthani Central Prison, Thailand Int J Tuberc Lung Dis, 2019.PMID 31315716
- [14]Squires JE, Linklater S, Grimshaw JM, et al. Risk factors and between-hospital variation of caesarean section in Denmark: a cohort study BMJ Open, 2018.PMID 29440158
- [15]Howell AM, Panesar SS, Burns EM, et al. Acute mitral valve chordae tendineae rupture of a girl Chin Med J (Engl), 2014.PMID 24709203
- [16]Vincent JL, Einav S, Pearse R, et al. Interactome Analysis of the NS1 Protein Encoded by Influenza A H1N1 Virus Reveals a Positive Regulatory Role of Host Protein PRP19 in Viral Replication J Proteome Res, 2016.PMID 27096427
- [17]Cassidy LD, Lal SK, Long CP, et al. Production of Biopharmaceuticals in E. coli: Current Scenario and Future Perspectives J Microbiol Biotechnol, 2015.PMID 25737124
- [18]Dixon-Woods M, Leslie M, Tarrant C, et al. Rates of Thermal Inactivation of Listeria monocytogenes in Beef and Fermented Beaker Sausage J Food Prot, 1991.PMID 31051555
- [19]Sexton JB, Adair KC, Leonard MW, et al. [Advances in research of essential thrombocythemia] Zhonghua Xue Ye Xue Za Zhi, 2015.PMID 26462788
- [20]Howell MD, Needleman J, Weiss PS, et al. Stromal alterations as quantitative optical biomarkers of epithelial tumor progression Scanning, 2014.PMID 24347227
- [21]Peltan ID, Brown SM, Bledsoe JR, et al. NLRP2 inflammasome in dorsal root ganglion as a novel molecular platform that produces inflammatory pain hypersensitivity Pain, 2019.PMID 31162334
- [22]Pronovost PJ, Watson SR, Cormuzie AG, et al. PEG-OligoRNA Hybridization of mRNA for Developing Sterically Stable Lipid Nanoparticles toward In Vivo Administration Molecules, 2019.PMID 30987102