Critical care outreach and early warning systems

Contents

Failure to Rescue

Failure to rescue describes the preventable death of a hospitalised patient who develops a recognised complication — not from the complication itself, but from failure to detect it and intervene in time. It is a key concept in patient safety.

Evidence Base

  • Goldhill et al (1999) and subsequent UK studies identified that the majority of ICU admissions are preceded by recognisable clinical deterioration occurring hours before the patient becomes critically unwell
  • Ward-based physiological deterioration (↑ HR, ↑ RR, ↓ SBP, altered consciousness) is detectable and actionable if observed and escalated
  • The failure is often not in detection itself but in escalation and response: observations recorded but not acted upon; nursing staff unable to escalate; doctors not responding appropriately

Antecedents to In-Hospital Cardiac Arrest

Schein (1990) described physiological disturbances in the 6–8 hours preceding in-hospital cardiac arrests: most had documented abnormal vital signs with inadequate escalation. The majority of in-hospital arrests are therefore potentially preventable.

Early Warning Scores — NEWS2

National Early Warning Score 2 (RCP, 2017) is the UK-wide standardised early warning system for detecting patient deterioration on general wards.

Scoring Parameters

Each parameter scored 0–3 (except O₂ — scored 0 or 2):

Parameter Score 3 Score 2 Score 1 Score 0 Score 1 Score 2 Score 3
Respiratory rate (/min) ≤8 — 9–11 12–20 — 21–24 ≥25
SpO₂ (%) ≤91 92–93 94–95 ≥96 — — —
Supplemental oxygen — Yes +2 — No — — —
SBP (mmHg) ≤90 91–100 101–110 111–219 — — ≥220
Pulse rate (/min) ≤40 — 41–50 51–90 91–110 111–130 ≥131
Consciousness Alert — — — Confused — V/P/U
Temperature (°C) ≤35.0 — 35.1–36.0 36.1–38.0 38.1–39.0 ≥39.1 —

Supplemental oxygen: if patient requires O₂ to maintain saturation, add +2 to aggregate score.

Two SpO₂ scales: Scale 1 (standard: target ≥96%); Scale 2 (hypercapnic COPD risk: target 88–92%) — use Scale 2 for patients with confirmed or risk of type II respiratory failure.

Response Thresholds

Score Action
0 Routine observations per ward schedule
1–4 Increase observation frequency; nurse assessment
5–6 (or single score of 3) Urgent clinical review by ward team; consider referral to CCO or critical care team
≥7 Emergency response; immediate critical care team involvement; consider HDU/ICU

NEWS2 Limitations

  • Validated for detecting acute illness — not validated for ICU or post-operative patients (where physiological parameters differ)
  • Does not incorporate drug or treatment context (e.g., a patient on a beta-blocker will not have tachycardia despite shock)
  • Confounded by chronic disease (a COPD patient with SpO₂ 92% on 2L O₂ at baseline may score high inappropriately)
  • Point-in-time score — single readings less informative than trends over time
  • Not validated in pregnancy (different normal ranges)
  • Automated/continuous vital sign monitoring platforms are being integrated with NEWS2 to detect trends earlier

Critical Care Outreach (CCO)

CCO teams are nurse-led (usually) specialist teams that respond to deteriorating patients outside the ICU, provide ICU follow-up on the ward, and promote early ICU discharge by supporting ward management of complex patients.

Functions

  1. Respond to deteriorating patients: respond to escalation by ward staff (high NEWS2, SBAR referral); assess, stabilise, determine appropriate level of care (ward / HDU / ICU)
  2. Facilitate ICU discharge: follow up recently discharged ICU patients on wards; manage residual critical illness issues; reduce readmission
  3. Education: train ward nursing and medical staff in recognition and management of deterioration; NEWS2 education; simulation
  4. Audit and quality improvement: track outcomes of patients referred to CCO; identify patterns in failure to escalate; service development

Evidence for Outreach

Mixed evidence — RCT evidence (MERIT trial, Australia, 2006) did not demonstrate statistically significant reduction in mortality from introducing a medical emergency team (MET); however, observational evidence and before-and-after studies in UK and international contexts suggest:

  • Reductions in unanticipated ICU admissions and in-hospital cardiac arrests
  • Improved patient safety culture
  • More timely ICU admissions (earlier → better outcomes)
  • The evidence is confounded by implementation variation and context

Rapid Response Systems

A rapid response system (RRS) consists of two linked components:

1. Afferent Limb (Detection)

  • Routine vital sign monitoring and NEWS2 scoring
  • Patient/family activation systems ("patient-activated rapid response")
  • Automated monitoring and alerts
  • Safety reporting and escalation culture

2. Efferent Limb (Response)

  • Medical Emergency Team (MET) or Critical Care Outreach responding to escalation
  • Predefined criteria trigger response: NEWS2 ≥7, single parameter of 3, nursing concern

The "Calling Criteria" Debate

Fixed calling criteria (e.g., HR >130, RR >25) trigger a response regardless of NEWS2 aggregate. The advantage is objectivity; the disadvantage is alert fatigue from over-sensitive thresholds. Structured clinical judgment tools (SBAR — Situation, Background, Assessment, Response) allow any nurse to escalate on nursing concern alone, supplementing score-based criteria.

Resuscitation Plans and RESPECT

Rapid response should be preceded by a clearly documented DNACPR/RESPECT plan where appropriate:

  • Patients escalated to CCO but with a DNACPR order: CCO response is appropriate (CCO is not resuscitation — it is prevention)
  • RESPECT document guides what interventions are appropriate on deterioration (e.g., IV antibiotics and fluids — yes; escalation to ICU — no; CPR — no)

ICU Follow-Up and Rehabilitation

ICU survivorship carries a high burden of physical, cognitive, and psychological sequelae — collectively termed Post-Intensive Care Syndrome (PICS):

PICS Components

  • Physical: ICU-acquired weakness (ICUAW), neuropathy, myopathy; functional decline; dysphagia; breathlessness
  • Cognitive: impaired memory, attention, executive function; post-intensive care cognitive impairment (PICIC)
  • Psychological: PTSD (~25%), depression, anxiety; ICU diary-based interventions may reduce PTSD

ICU Follow-Up Services (FICM/GPICS Recommendations)

GPICS (Guidelines for the Provision of Intensive Care Services) recommends:

  • All ICU patients with ≥3 days stay or mechanical ventilation should receive a structured ICU follow-up appointment at 2–3 months post-discharge
  • ICU follow-up clinic: addresses PICS; functional assessment; psychological screening; referral to rehabilitation, psychology, physiotherapy, speech and language
  • ICU diary: maintained during ICU stay by nurses/family; given to patient at discharge; reduces incidence of PTSD

Early Rehabilitation in ICU

Early active physiotherapy and mobilisation in mechanically ventilated patients:

  • Associated with reduced ICUAW, shorter mechanical ventilation duration, and better functional outcomes at discharge
  • Daily sedation holds + early mobilisation form the "A-RACE" bundle (Awakening, Breathing, Coordination, Delirium, Early mobility, Family engagement)
  • Coordination between ICU, physiotherapy, occupational therapy, and speech and language is essential

Viva Questions

1. What is NEWS2 and what are its limitations in detecting deteriorating patients?

NEWS2 (National Early Warning Score 2) is the UK-standard physiological aggregate scoring system for detecting clinical deterioration on general wards. It scores seven parameters — respiratory rate, oxygen saturation, oxygen requirement, systolic BP, heart rate, consciousness (ACVPU), and temperature — on a 0–3 scale, with two saturation scales (standard and COPD target). An aggregate score ≥7, or any single parameter scoring 3, triggers an emergency response. NEWS2 has demonstrated good validation in emergency department and general ward populations for predicting ICU admission, cardiac arrest, and mortality. However, its limitations are important: it is a point-in-time measurement — trends over time are more informative than a single value; it is not validated in the ICU, post-operative patients, or pregnancy (all have different normal physiological ranges); it does not account for medications (beta-blockers, opioids blunt physiological responses); the two-scale oxygen system adds complexity and can lead to scoring errors; and single readings may trigger alerts in patients with known chronic baseline abnormalities (SpO₂ in severe COPD, tachycardia in anxiety). The score should always be interpreted in the clinical context — nursing concern (regardless of score) is a valid and important escalation trigger.

2. What is "failure to rescue" and how does it relate to the design of ICU outreach services?

Failure to rescue describes preventable death from a recognised complication — not because the complication could not have been managed, but because it was not detected and acted upon in time. Evidence (Goldhill, Schein, and subsequent UK audit data) shows that the majority of in-hospital cardiac arrests and unplanned ICU admissions are preceded by hours of documented physiological deterioration that was recognised but not adequately escalated. The failure is often systemic: vital signs recorded but not actioned; ward staff unable to reach the right person; medical staff underestimating severity. ICU outreach services directly address failure to rescue by providing an expert response pathway for deteriorating ward patients — a team that can assess, intervene, and determine appropriate escalation within minutes of a NEWS2 alert or nursing concern. CCO also works upstream — training ward staff in recognition, embedding escalation culture, ensuring RESPECT documentation is in place, and following up post-ICU patients who may deteriorate on the ward. The GPICS standards now mandate minimum CCO provision (24/7 in Level 3 ICUs).

3. What is Post-Intensive Care Syndrome (PICS), and how can the ICU team reduce its burden?

PICS is the cluster of new or worsened physical, cognitive, and psychological impairments persisting beyond acute critical illness. Physical impairments include ICU-acquired weakness (ICUAW) — a consequence of prolonged immobility, sedation, neuromuscular blocking agents, inflammation, and malnutrition — as well as dysphagia, breathlessness, and functional decline. Cognitive impairments include memory, attention, and executive function deficits (PICIC) — particularly prevalent after delirium, prolonged sedation, or hypoxia. Psychological sequelae include PTSD (~25%), depression, and anxiety. Interventions to reduce PICS: (1) Early physiotherapy and mobilisation — even in ventilated patients; reduces ICUAW and duration of ventilation; (2) Sedation minimisation — daily sedation holds, analgesia-first strategy; (3) Delirium prevention — ABCDEF bundle; cognitive stimulation; day-night cycle; family involvement; (4) ICU diary — maintained by nurses and family; given to patient at discharge; evidence suggests reduction in PTSD incidence; (5) ICU follow-up clinic — structured assessment at 2–3 months; identifies and refers for rehabilitation, psychology, speech therapy; (6) Family support — PICS affects families too (PICS-F); communication, family meetings, family engagement in care.