A complete clinical guide to CURB-65 — how to calculate it, what each score means for management, antibiotic selection, and when to consider ICU admission.
Pneumonia is one of the leading causes of death worldwide — and also one of the most common conditions a doctor will encounter in any clinical setting. Every year, over 450 million people develop pneumonia globally, and over 4 million die from it. In India, pneumonia remains a major killer across all age groups, from infants to the elderly.
When a patient presents with suspected pneumonia, one of the most important early decisions is deceptively simple: can this patient be safely treated at home, or do they need to be admitted to hospital? Send a high-risk patient home and they may deteriorate rapidly. Admit every patient unnecessarily and you overwhelm hospital resources, expose patients to nosocomial infections, and drive up costs.
The CURB-65 score — developed by the British Thoracic Society — is a quick, validated, five-point bedside tool designed to answer exactly that question. It predicts 30-day mortality in community-acquired pneumonia (CAP) and guides the appropriate level of care.
Community-acquired pneumonia (CAP) is a lung infection that develops in a person who has not recently been hospitalised and is not living in a healthcare facility. It is distinct from hospital-acquired (nosocomial) pneumonia, which involves different organisms and requires different management.
CAP is most commonly caused by bacteria, with Streptococcus pneumoniae (pneumococcus) being the single most common causative organism. Other common pathogens include Haemophilus influenzae, Mycoplasma pneumoniae, Klebsiella pneumoniae (particularly in diabetics and alcoholics in India), and respiratory viruses including influenza.
Clinically, patients typically present with fever, cough (productive or dry), breathlessness, pleuritic chest pain, and general malaise. Physical examination may reveal dullness to percussion, bronchial breathing, and increased vocal resonance over the affected area. A chest X-ray showing a new consolidation confirms the diagnosis.
CURB-65 assigns one point for each of five clinical parameters, giving a maximum score of 5:
New-onset confusion or disorientation. Defined as an AMT (Abbreviated Mental Test) score of ≤ 8, or new confusion as reported by patient or family.
Blood urea nitrogen (BUN) > 19 mg/dL. Reflects dehydration and early renal impairment — a marker of physiological stress.
Tachypnoea at rest. A key sign of respiratory compromise and the body's effort to maintain oxygenation.
Systolic BP < 90 mmHg or diastolic BP < 60 mmHg. Indicates haemodynamic compromise — a serious sign.
Advanced age is an independent predictor of poor outcomes in pneumonia. Elderly patients have reduced physiological reserve.
🫁 Use the RxMedCalc CURB-65 Calculator to score your patient instantly at the bedside.
| CURB-65 Score | 30-Day Mortality | Risk | Recommended Action |
|---|---|---|---|
| 0 | 0.6% | Low | Treat at home with oral antibiotics |
| 1 | 2.7% | Low | Home treatment generally safe; reassess if worsening |
| 2 | 6.8–9% | Moderate | Short-stay hospital admission or closely supervised outpatient care |
| 3 | 14–17% | High | Hospital admission; consider HDU if deteriorating |
| 4 | 27–36% | High | Urgent admission; assess for ICU need |
| 5 | 40–57% | Critical | ICU-level care; aggressive management |
⚠️ CURB-65 is a guide, not a rule. Clinical judgement must always accompany the score. A patient scoring 1 who lives alone, has no carer, lives far from a hospital, or is deteriorating rapidly may warrant admission regardless. Conversely, a score of 2 in a young patient with good social support and daily GP review may be safely managed at home.
Patient: 70-year-old man with 3-day history of fever, productive cough, and breathlessness. On examination: confused and disoriented, RR 34/min, BP 118/76 mmHg, temperature 38.9°C. Blood urea 9.2 mmol/L.
Scoring:
Total Score: 4
CURB-65 of 4 = high risk, predicted 30-day mortality ~27–36%. This patient requires urgent hospital admission. Assess for ICU suitability given confusion and significant tachypnoea. Start IV antibiotics immediately — do not delay for confirmatory investigations.
In primary care settings — a GP clinic, a rural health centre, or an emergency home visit — blood urea measurement may not be immediately available. The CRB-65 score removes the urea criterion and uses only the four clinical parameters:
| CRB-65 Score | Risk | Recommendation |
|---|---|---|
| 0 | Low | Home treatment appropriate |
| 1 – 2 | Moderate | Consider hospital assessment |
| 3 – 4 | High | Urgent hospital admission |
CRB-65 is particularly useful in Indian primary care settings where point-of-care blood tests may be unavailable, allowing rapid triage with information that can be gathered in under two minutes at the bedside.
The Pneumonia Severity Index (PSI) is an alternative scoring system developed by Fine et al. that uses 20 variables — including age, nursing home residence, comorbidities, physical examination findings, and laboratory results — to classify patients into 5 risk classes.
| Feature | CURB-65 | PSI / PORT Score |
|---|---|---|
| Variables | 5 | 20 |
| Calculation | Immediate, bedside | Requires lab results & calculation |
| Better at identifying | High-risk patients needing admission | Low-risk patients safe for discharge |
| Used in | UK (BTS), Europe, Asia | USA (IDSA/ATS guidelines) |
| Weakness | May overestimate risk in elderly with comorbidities | Unwieldy in acute settings; complex to calculate |
For most clinical settings — especially in India — CURB-65 is the preferred tool due to its simplicity and speed. PSI is better suited for research settings or where the primary goal is identifying patients safe for outpatient management.
Antibiotic choice depends on the severity of pneumonia (guided by CURB-65), likely pathogens based on clinical presentation, and local resistance patterns.
⚠️ Antibiotic resistance in India: Klebsiella pneumoniae producing extended-spectrum beta-lactamases (ESBLs) is increasingly common in India, including in community settings. In patients with risk factors (diabetes, prior hospitalisation, prior antibiotic use), consider broader empirical coverage and send cultures before starting antibiotics wherever possible.
The extent of investigation should match the severity:
CURB-65 identifies severity but is not specifically calibrated for ICU triage. The IDSA/ATS minor criteria for severe CAP and ICU admission include:
The presence of 1 major criterion (septic shock requiring vasopressors, or respiratory failure requiring mechanical ventilation) or 3 or more minor criteria strongly suggests ICU admission is needed.
CURB-65 was developed in a largely elderly population and performs well in this group. However, elderly patients may present atypically — without fever, with confusion as the only symptom, or with a relatively normal CXR early in the illness. A low CURB-65 score in an elderly patient with a general deterioration warrants careful reassessment before discharging.
CURB-65 was not validated for immunocompromised patients (HIV, transplant recipients, patients on chemotherapy or biologics). These patients may have atypical organisms (Pneumocystis jirovecii, fungi, CMV), require broader investigations, and should generally be assessed by a specialist. CURB-65 should be applied with caution in this group.
Pneumonia in pregnancy carries higher risks of preterm labour and foetal compromise. A lower threshold for admission is appropriate. Avoid fluoroquinolones and tetracyclines — use amoxicillin ± macrolide as first-line, and involve obstetrics early.
🫁 Calculate CURB-65 instantly at the bedside: RxMedCalc CURB-65 Calculator →
This article is written for educational purposes based on internationally recognised guidelines. It is not a substitute for professional medical advice. Antibiotic choices should always be guided by local resistance patterns and individual patient factors. Always consult a qualified physician for clinical decisions.
Built by an MBBS, AFIH Certified Physician in Punjab, India | RxMedCalc.com