UK BTS guidanceUrea in mmol/L (UK standard). Score ≥3 = severe CAP, consider ICU. CRB-65 (no urea) for community use.

UK-developed scoreDECAF (Dyspnoea, Eosinopenia, Consolidation, Acidaemia, Atrial Fibrillation) outperforms APACHE II and BAP-65 in UK COPD populations (Steer et al., Thorax 2012).

⚠️ UK Validation NoteYEARS was developed and validated in Dutch populations. It is not formally endorsed by NICE NG158 or RCEM for routine UK ED use. NICE recommends Wells + D-dimer as the standard pathway. YEARS may be used in some UK centres but always follow local trust protocol. If in doubt, use Wells score.

Revised Geneva ScoreFully objective (no clinical judgement criterion). Validated in ED populations. NICE NG158 recommends Wells or Geneva for PE pretest probability. Both are acceptable UK practice.

PESI Class I–II (score ≤85)Supports outpatient treatment when combined with Hestia rule and clinical assessment. PESI III–V requires hospitalisation. ESC recommends PESI or sPESI for post-confirmation risk stratification of haemodynamically stable PE.

sPESI vs PESIsPESI is faster at the bedside. Score 0 predicts 1.0% 30-day mortality (vs PESI Class I–II ≤3.5%). Validated for outpatient PE management when combined with clinical assessment and Hestia rule.

Hestia RuleTick each criterion that applies. If ANY box is ticked, the patient is NOT suitable for outpatient treatment. Zero criteria = suitable for home treatment (if social situation permits). Use alongside clinical judgement and PESI/sPESI.

Prerequisites for Ottawa SAH RuleAge ≥15, GCS 15, non-traumatic headache reaching max intensity within 1 hour. Do NOT apply if patient has neurological deficit, known aneurysm/SAH/brain tumour, or headache caused by trauma, LP, or medication overuse.

NICE NG128 (2019) — do NOT use ABCD2 for triageNG128 recommends against using scoring systems such as ABCD2 to assess subsequent stroke risk or to determine urgency of referral. ALL suspected TIA: aspirin 300mg immediately (unless contraindicated or on anticoagulation) and specialist assessment within 24h of symptom onset. The score is retained here for education and communication of risk, not for decision-making. Admit or same-day review regardless of score if AF, on anticoagulation, or ≥2 TIAs in a week (crescendo).

Advantage over GCSFOUR can detect locked-in syndrome, vegetative state, and herniation patterns. Useful when verbal response is impossible (e.g. intubated patients).

UK Context — NICE NG253NICE recommends NEWS2 ≥5 as the primary trigger for sepsis assessment in UK hospitals, not qSOFA. qSOFA (≥2) is useful for rapid bedside identification in community/pre-hospital settings.

Septic Shock Definition (Sepsis-3)Sepsis + vasopressors required to maintain MAP ≥65 mmHg + lactate ≥2 mmol/L despite adequate fluid resuscitation.

NoteSIRS has been largely superseded by Sepsis-3 criteria (2016). Many UK trusts now use NEWS2 ≥5 as the primary sepsis trigger per NICE NG253. SIRS may still appear in local protocols.

FeverPAIN — 5 criteria (max 5)0–1: 13–18% strep — antibiotic not recommended. 2–3: 34–40% — no antibiotic or a back-up (delayed) prescription. 4–5: 62–65% — consider an immediate or back-up antibiotic. Apply clinical judgement (NICE NG84). First-line: phenoxymethylpenicillin; clarithromycin/erythromycin if penicillin-allergic.

Westley Croup Score (max 17)≤2 mild · 3–5 moderate · 6–11 severe · ≥12 impending respiratory failure. Oral dexamethasone 0.15 mg/kg for all severities; add nebulised adrenaline for severe. Keep the child calm — do not examine the throat or distress them.

UK-developed & validatedValidated in UK ED populations. Score >0 suggests stroke — activate stroke pathway. Always exclude hypoglycaemia first (BM check mandatory).

Thrombectomy threshold (UK/NICE)Consider thrombectomy if NIHSS ≥6, confirmed LVO on CTA, onset <24h (selected patients). Thrombolysis window: 4.5h from onset if no contraindications.

UK guidanceWCC in ×10⁹/L (UK standard). Score ≥7 → surgical referral. Score 4-6 → observe / CT abdomen. Score ≤3 → low risk, consider discharge with safety-netting.

UK unitsBilirubin in µmol/L, albumin in g/L (UK standard). INR used rather than PT seconds.

Which method? (NICE NG203 2023)CKD-EPI 2021 is the UK standard (recommended by NICE, UKKA, RCPath). CKD-EPI 2021 removes race adjustment. MDRD underestimates at higher GFR. Cockcroft-Gault used for drug dosing (e.g. vancomycin, gentamicin, DOAC prescribing).

NICE AKI DefinitionRise in creatinine ≥26 µmol/L within 48h, OR ≥1.5× baseline within 7 days, OR urine output <0.5 mL/kg/h for ≥6h. Always check for nephrotoxics, obstruction, and volume status.

NICE NG196 (AF 2021)Men ≥2, Women ≥3 → offer anticoagulation. Men = 1, Women = 2 → consider anticoagulation (assess bleeding risk with ORBIT score). Do not offer anticoagulation for score 0 in men or 1 in women.

NICE NG196HAS-BLED is not a reason to withhold anticoagulation — it identifies modifiable bleeding risk factors to address. Score ≥3 = high risk; review and correct modifiable factors (uncontrolled BP, INR lability, alcohol excess, NSAIDs).

ORBIT vs HAS-BLEDORBIT was developed in a US/UK population of 7,411 AF patients and outperforms HAS-BLED in predicting major bleeding on anticoagulation. Uses objective lab values (Hb, eGFR) rather than subjective assessments. Haemoglobin in g/dL (divide g/L by 10). NICE NG196 recommends assessing bleeding risk before anticoagulation — ORBIT is the preferred tool in many UK centres. Score ≥3 = high bleeding risk.

6-hour salvage windowTesticular torsion requires surgical exploration within 6h for >95% salvage rate. Do NOT delay exploration to await USS if clinical suspicion is high. TWIST ≥5 → immediate urology referral and theatre.

UK Treatment Line (MHRA 2012)Single treatment line at 100mg/L at 4h (previously 200mg/L for low-risk). Staggered/unknown time overdose → treat. Always contact NPIS (0344 892 0111) for complex cases.

CIWA-Ar Scoring Guide (NICE CG100)Score each item 0–7 (except orientation 0–4). Total max 67. Score <8 = mild — monitoring. 8–15 = moderate — consider lorazepam/chlordiazepoxide. ≥16 = severe — IV/IM benzodiazepine, HDU, senior review. Re-score every 1–4h. Seizure risk peaks 24–48h after last drink. Wernicke's prophylaxis: thiamine IV (Pabrinex) in all at-risk patients.

COWS — 11 items, max 485–12 mild · 13–24 moderate · 25–36 moderately severe · >36 severe. Buprenorphine induction usually needs COWS ≥12 to avoid precipitated withdrawal — involve the substance-misuse / OST team. Symptomatic options: lofexidine (NICE TA), clonidine, antiemetics, loperamide, simple analgesia. Re-score regularly.

NICE CG141 / BSG — Blatchford et al., Lancet 20009-variable score (max 23). GBS = 0 → very low risk, consider early discharge with outpatient endoscopy within 24h. GBS ≥1 → admit. GBS ≥6 → high risk, urgent endoscopy within 24h. GBS ≥12 → very high risk, endoscopy within 12h. Urea in mmol/L, Hb in g/L (UK units). Pulse ≥100 bpm scores +1.

BSG Lower GI Bleeding Guidelines 2019 · Oakland et al., Gut 2017⚠️ Only use in stable LGIB — check shock index (HR÷SBP) first: if ≥1 the patient is unstable → skip Oakland, go direct to CT angiography. Score ≤8: minor bleed, 95% probability of safe discharge (defined as: no rebleeding, no RBC transfusion, no therapeutic intervention, no in-hospital death, no readmission with LGIB within 28 days). Discharge with urgent outpatient colonoscopy — within 2 weeks if high-risk features. Score >8: major bleed — admit for inpatient colonoscopy on next available list. Haemoglobin is the dominant variable (max 22 pts). Max score 35. For upper GI bleed, use Glasgow-Blatchford Score instead.

Important — NICE NG225SAD PERSONS should be used as a structured checklist to ensure all risk factors are considered — NOT as a sole decision-making tool. NICE does not recommend any single risk assessment tool for determining level of care. Always involve senior psychiatric input for all deliberate self-harm presentations. Mandatory safeguarding consideration for under-18s.

Which method to use?Bazett (most common in clinical practice, tends to overcorrect at high HR). Fridericia (better at high/low HR extremes, preferred by cardiologists). Framingham (population-based regression, least used). Use Fridericia if HR >80 or <60 bpm for more accurate correction.

UK FormulaCorrected Ca = Measured Ca + 0.02 × (40 − albumin). Albumin in g/L, calcium in mmol/L. Normal corrected Ca: 2.2–2.6 mmol/L.

FormulaCorrected Na = Measured Na + 0.3 × (glucose − 5.5). For every 5.5 mmol/L rise in glucose above normal, add 1.6 mmol/L to sodium. Important in HHS and DKA.

UK FormulaCalculated Osm = 2×Na + Glucose + Urea (all in mmol/L). Osmol Gap = Measured − Calculated. Normal <10 mOsm/kg. Gap >10 → methanol, ethylene glycol, propylene glycol, ethanol.

Interpretation<0.4 = pure NAGMA. 0.4–1.0 = HAGMA + concurrent NAGMA. 1.0–2.0 = pure HAGMA. >2.0 = HAGMA + concurrent metabolic alkalosis.

MARSI-MEWS — MARSIPAN-derived EWS for Anorexia Nervosa InpatientsStandard EWS thresholds are unreliable in patients with anorexia nervosa due to chronically altered physiology. MARSI-MEWS uses adapted ranges across 5 parameters (total 0–15). Score 0: normal monitoring. Score 1–2: increase to 4-hourly observations. Score 3–4: hourly obs + 30-min nursing assessment. Score 5: 30-min obs + immediate medical assessment. Score ≥6: close continuous monitoring + immediate consultant involvement. Additional alerts: AVPU not Alert → respond immediately; blood glucose <4 mmol/L → respond immediately. ⚠️ Note: "MARSI-MEWS" as named/scored here could not be independently verified against a single citable published source — treat as a locally-adapted early warning tool consistent with MARSIPAN/MEED (RCPsych CR233, 2022) principles, and confirm thresholds against your trust's adopted eating-disorder EWS chart before relying on it.

NEWS2 Thresholds (RCP 2017 / NICE NG253)Score ≥5 → urgent clinical assessment + consider sepsis. Score ≥7 → continuous monitoring + emergency response. Single parameter extreme (score 3) → same urgency as aggregate ≥7. Select SpO2 Scale 2 for patients with hypercapnic respiratory failure (target 88–92%).

PEWS Escalation (Monaghan 2005 / NHS England)Score 0: routine care. Score 1–2: notify nurse in charge, increase observations, reassess within 1 hour. Score 3: immediate senior nurse review + call doctor — medical review within 30 min. Score ≥4: immediate senior medical review, consider PICU referral. Each "Plus 2" modifier adds directly to the total and represents a high-risk clinical feature. Always apply clinical judgement — a child who "looks wrong" should be escalated regardless of score.

MARSI-MEWS — MARSIPAN-derived EWS for Anorexia Nervosa InpatientsStandard EWS thresholds are unreliable in patients with anorexia nervosa due to chronically altered physiology. MARSI-MEWS uses adapted ranges across 5 parameters (total 0–15). Score 0: normal monitoring. Score 1–2: increase to 4-hourly observations. Score 3–4: hourly obs + 30-min nursing assessment. Score 5: 30-min obs + immediate medical assessment. Score ≥6: close continuous monitoring + immediate consultant involvement. Additional alerts: AVPU not Alert → respond immediately; blood glucose <4 mmol/L → respond immediately. ⚠️ Note: "MARSI-MEWS" as named/scored here could not be independently verified against a single citable published source — treat as a locally-adapted early warning tool consistent with MARSIPAN/MEED (RCPsych CR233, 2022) principles, and confirm thresholds against your trust's adopted eating-disorder EWS chart before relying on it.

STUMBL Score (Battle CE et al., Emerg Med J 2014)Weighted model predicting pulmonary complications after isolated blunt chest wall trauma (sensitivity 80%, specificity 96%). Thresholds: <11 low risk — consider discharge with safety-netting; ≥11 significant complication risk — admit; ≥26 high risk — consider critical care. Adequate analgesia is essential: rib-fracture pain causes splinting, atelectasis and pneumonia — regional anaesthesia (thoracic epidural, serratus anterior plane block, intercostal blocks) is superior to systemic opioids where available. Note: age and rib-fracture point weights here approximate the published model — verify against a validated calculator (e.g. MDCalc) for borderline critical decisions.