CURB-65 Score: A Complete Guide to Pneumonia Risk Stratification

When a patient arrives with fever, cough, and breathlessness, one of the first questions you ask yourself is: how sick is this person, and do they need to come in? The CURB-65 score is the most widely used tool to answer that question quickly and reliably in community-acquired pneumonia (CAP).

In this guide, we’ll break down each criterion, show you how to calculate and interpret the score, and explain how it should inform your management decisions — backed by the original British Thoracic Society (BTS) guidelines.

What Is the CURB-65 Score?

CURB-65 is a clinical prediction rule for assessing the severity of community-acquired pneumonia. It was developed by Lim et al. in 2003 and validated across thousands of patients in the UK, Europe, and beyond. Each letter stands for one clinical feature:

Maximum score: 5. Each criterion present scores 1 point.

The 5 Criteria Explained

C — Confusion

This refers to new-onset disorientation to person, place, or time — not pre-existing dementia. Use the Abbreviated Mental Test (AMT) score if needed: a score of ≤8 out of 10 indicates confusion. In practice, ask the patient their name, date of birth, where they are, and the current year.

U — Urea >7 mmol/L

Elevated blood urea indicates poor renal perfusion — a marker of systemic illness severity. Note: this requires a blood test, so in community settings where bloods aren’t available, some clinicians use the CRB-65 variant (which drops the U criterion).

R — Respiratory Rate ≥30/min

A respiratory rate of ≥30/min is a strong independent predictor of mortality in pneumonia. It reflects poor gas exchange and physiological compensation. Always count the respiratory rate yourself — nurses often estimate it.

B — Blood Pressure (Systolic <90 or Diastolic ≤60 mmHg)

Hypotension in this context suggests septic physiology. Combined with other features, it significantly raises 30-day mortality risk and should prompt urgent senior review and consideration of HDU/ICU care.

65 — Age ≥65 Years

Older patients tolerate pneumonia less well due to reduced physiological reserve, immune senescence, and higher burden of comorbidity. Age ≥65 automatically adds 1 point regardless of other features.

Interpreting the Score

Management Based on CURB-65

Score 0–1: Low Risk — Consider Home Treatment

• Oral amoxicillin 500mg TDS for 5 days (BTS/NICE guidance)
• If atypical features (dry cough, younger patient, no sputum): add or switch to doxycycline 200mg then 100mg OD or clarithromycin 500mg BD
• Safety-net advice: return if worsening, new breathlessness, or not improving in 48 hours
• Review in 48 hours

Score 2: Moderate Risk — Hospital Admission

• Consider short-stay admission or hospital-at-home with close follow-up
• IV co-amoxiclav + clarithromycin if severe features developing
• Bloods: FBC, U&E, LFTs, CRP, blood cultures before antibiotics
• CXR to confirm and assess extent
• Oxygen if SpO₂ <94%

Score 3–5: Severe — Urgent Admission + Senior Review

• IV co-amoxiclav 1.2g TDS + IV clarithromycin 500mg BD
• If penicillin allergic: IV levofloxacin 500mg BD
• Assess for HDU/ICU criteria: persisting hypoxia, haemodynamic instability, failure to respond to antibiotics in 24–48 hours
• Sepsis Six if septic: blood cultures, IV antibiotics, IV fluids, oxygen, catheter, lactate
• Consider early liaison with ITU team

CURB-65 vs CRB-65: Which Should You Use?

CRB-65 drops the Urea criterion (U), making it usable in community settings without blood tests. It scores out of 4.

Limitations of CURB-65

No scoring tool is perfect, and CURB-65 has well-documented limitations you should be aware of:

• Does not account for comorbidities: A patient with severe COPD, active malignancy, or immunosuppression may be far sicker than their score suggests
• Underestimates severity in younger patients: A fit 30-year-old with bilateral pneumonia and SpO₂ 88% may score 0
• Does not include oxygenation: SpO₂ and PaO₂/FiO₂ ratio are not included — always check oxygen saturation
• Social factors ignored: Lives alone, unable to self-manage, no carer — these matter clinically even if the score is low
• Radiology not included: Bilateral or multilobar consolidation carries a worse prognosis not captured by CURB-65

CURB-65 and the Sepsis Overlap

Severe pneumonia frequently meets sepsis criteria (infection + organ dysfunction). A CURB-65 ≥3 should prompt you to run through your hospital’s sepsis pathway concurrently — do not delay antibiotics while calculating scores.

Key sepsis markers to check alongside CURB-65:

• Lactate ≥2 mmol/L → possible tissue hypoperfusion
• Creatinine rise, oliguria → early AKI
• Platelet drop → DIC screening
• NEWS2 score ≥5 → escalate to senior

Documentation: Writing a Pneumonia Assessment

When documenting a CAP admission, include the CURB-65 score explicitly. A well-written assessment might read:

Need help writing discharge summaries or clinical notes quickly? Try our AI Medical Report Generator — free, no signup, powered by Claude AI.

Key Takeaways

• CURB-65 is a 5-point score: Confusion, Urea >7, RR ≥30, BP low, Age ≥65
• Score 0–1 = low risk (consider home); Score 2 = moderate (admit); Score 3–5 = severe (urgent admission)
• Always use clinical judgement alongside the score — oxygenation, comorbidities, and social context matter
• CRB-65 is the community version (no blood test needed)
• In severe pneumonia, initiate the sepsis pathway simultaneously
• Document the CURB-65 score and risk group explicitly in all CAP admissions

References

1. Lim WS, et al. Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax. 2003;58(5):377-382.
2. British Thoracic Society. BTS Guidelines for the Management of Community Acquired Pneumonia in Adults. 2009 Update.
3. NICE. Pneumonia (community-acquired): antimicrobial prescribing. NICE guideline NG138. 2019.
4. Mandell LA, et al. Infectious Diseases Society of America/American Thoracic Society Consensus Guidelines on the Management of Community-Acquired Pneumonia in Adults. Clin Infect Dis. 2007;44(Suppl 2):S27-72.