AskMed

CURB-65 score interpretation: what each score from 0 to 5 means

A CURB-65 score runs from 0 to 5. Here is the 30-day mortality the original study found at every score, the place of care NICE links to each band, how CRB-65 works without a blood test, and what the score leaves out.

Open the CURB-65 calculator

What a CURB-65 score means

A CURB-65 score is a count of five features in an adult with community-acquired pneumonia, one point each, that estimates the risk of death within 30 days. The features are confusion (an abbreviated Mental Test Score of 8 or less, or new disorientation in person, place or time), blood urea above 7 mmol/L, a respiratory rate of 30 or more a minute, low blood pressure (systolic below 90 mmHg or diastolic 60 mmHg or less) and age 65 or older. Lim and colleagues derived the score in Thorax in 2003 from 1,068 adults admitted to hospital in the UK, New Zealand and the Netherlands, of whom 9% died within 30 days. The paper sorted the totals into three groups: 0 or 1 is low risk, 2 is intermediate and 3 to 5 is high. The British Thoracic Society built its 2009 severity assessment on CURB-65, and NICE guideline NG250 (2025) uses them to guide place of care.

ScoreDied in 30 days, Lim 2003NICE NG250 bandPlace of care NICE suggests
0 or 1 · Low1.5% (5 of 324)Under 3%Home, with primary care follow-up and safety netting
2 · Intermediate9.2% (17 of 184)3% to 15%Virtual ward, SDEC unit, hospital at home or inpatient care
3 to 5 · High22% (47 of 210)Over 15%Inpatient care; critical care referral if appropriate

Mortality figures are from the derivation cohort of 718 patients. SDEC is a same-day emergency care unit. To score a patient, tick the features in the CURB-65 calculator; it shows the score, its group and the mortality below.

30-day mortality for each CURB-65 score

The 30-day mortality for each CURB-65 score comes from Table 4 of Lim's 2003 paper, which split its patients at random into a derivation cohort of 718 and a validation cohort of 214. In the derivation cohort, 1 of 137 patients with a score of 0 died (0.7%), 4 of 187 with a score of 1 (2.1%), 17 of 184 with a score of 2 (9.2%), 20 of 138 with a score of 3 (14.5%) and 26 of 65 with a score of 4 (40%). Only 7 patients scored 5: one died (14%), and three more needed mechanical ventilation in intensive care and survived. The validation cohort followed the same rising pattern, from no deaths at scores 0 and 1 to 26.3% at a score of 4. The paper's abstract quotes different per-score figures, including 3.2% at a score of 1 and 57% at a score of 5, that do not match its own Table 4. The CURB-65 calculator and this guide use Table 4.

CURB-65 scoreDerivation cohort diedValidation cohort died
01 of 137 (0.7%)0 of 36 (0%)
14 of 187 (2.1%)0 of 54 (0%)
217 of 184 (9.2%)5 of 60 (8.3%)
320 of 138 (14.5%)9 of 42 (21.4%)
426 of 65 (40%)5 of 19 (26.3%)
51 of 7 (14%)1 of 3 (33.3%)
30-day mortality by CURB-65 score in the derivation cohort of Lim 2003 (Table 4). Bars are scaled to 40%. Grey: only 7 patients scored 5, so that figure is unstable; read scores 3 to 5 together, 22%.

CURB-65 score of 0 or 1: low risk

A CURB-65 score of 0 or 1 places an adult with community-acquired pneumonia in the low-risk group. In the 2003 derivation cohort, 5 of 324 patients with these scores died within 30 days, a mortality of 1.5%, and the authors judged them likely suitable for home treatment, as hospital outpatients or by their general practitioner. NICE NG250 labels the same band as less than 3% mortality risk. Its 2025 place-of-care recommendation is to consider discharge home, with referral to primary care-led services and safety netting advice, decided with the patient and using clinical judgement. A score of 1 is often age alone: any patient aged 65 or older starts on 1 point before a single sign of severe illness is counted. A low score does not override other findings. NICE names low oxygen saturation, pleural complications and multiple comorbidities as reasons a patient with a low score may still have moderate- or high-severity disease.

CURB-65 score of 2: intermediate risk

A CURB-65 score of 2 is the intermediate band, where the choice of place of care is least clear-cut. In Lim's derivation cohort, 17 of 184 patients with a score of 2 died within 30 days (9.2%), as did 5 of 60 in the validation cohort (8.3%). The paper suggested considering hospital-supervised treatment, such as a short inpatient stay or hospital-supervised outpatient care. NICE NG250 gives this band a 3% to 15% mortality risk and, since 2025, lists four options to consider: a virtual ward, a same-day emergency care (SDEC) unit, a hospital at home service, or inpatient care. NICE asks clinicians to decide with the patient, and their family or carers where appropriate, taking into account the person's preferences, any advance care plan or treatment escalation plan, comorbidities and frailty, the safety and suitability of the home, and the support available there. The same score of 2 can come from age plus urea or from confusion plus low blood pressure.

A worked example: a 72-year-old with a urea of 8.4 mmol/L, a respiratory rate of 24, a blood pressure of 128/76 mmHg and no confusion scores 2, one point for age and one for urea. This is the example the CURB-65 calculator opens with.

CURB-65 score of 3, 4 or 5: high risk

A CURB-65 score of 3 or more marks high-risk community-acquired pneumonia. In the derivation cohort, 47 of 210 patients scoring 3 to 5 died within 30 days, a mortality of 22%, and the authors advised managing these patients in hospital as severe pneumonia, with assessment for intensive care especially at a score of 4 or 5. NICE NG250 sets this band at more than 15% mortality risk and recommends considering inpatient care, with referral to critical care services if appropriate. The single-score figures rise steeply, from 14.5% at a score of 3 to 40% at a score of 4. CURB-65 was not designed to make the intensive care decision by itself. The 2019 American Thoracic Society and Infectious Diseases Society of America (ATS/IDSA) guideline uses separate severe pneumonia criteria for that, and recommends direct ICU admission for patients with hypotension needing vasopressors or respiratory failure needing mechanical ventilation, whatever their score.

CRB-65: the score without urea for primary care

CRB-65 is CURB-65 without the urea item, scored 0 to 4 from confusion, respiratory rate, blood pressure and age alone, so it can be used where no blood result is available, such as in general practice. The 2003 paper tested it alongside CURB-65: in the derivation cohort, mortality was 1.2% (2 of 167) at a score of 0, 8.15% (37 of 455) at a score of 1 or 2, and 31% (30 of 96) at a score of 3 or 4. The authors grouped these as likely suitable for home treatment, likely to need hospital referral and assessment, and urgent hospital admission. NICE NG250 uses CRB-65 for adults assessed in primary care, with risk bands of less than 1% at 0, 1% to 10% at 1 or 2, and more than 10% at 3 or 4. NICE advises considering hospital referral at a score of 2 or more, and primary care-led services with safety netting advice at a score of 0.

CRB-65Died, Lim 2003NICE NG250 bandPlace of care NICE suggests
02 of 167 (1.2%)Low, under 1%Primary care with safety netting
114 of 266 (5.3%)Intermediate, 1% to 10%Primary care, or a virtual ward, SDEC unit, hospital at home or hospital
223 of 189 (12.2%)Intermediate, 1% to 10%Referral to hospital
3 or 430 of 96 (31%)High, over 10%Referral to hospital

The study's CRB-65 figures come from patients already admitted to hospital, and its 12.2% at a score of 2 sits above the NICE band. NICE also advises hospital referral for anyone with signs of a more serious illness, such as cardiorespiratory failure or sepsis, whatever the score.

From CURB-65 score to severity and treatment

NICE NG250 separates the CURB-65 or CRB-65 mortality risk score from the clinician's assessment of disease severity, which it grades as low, moderate or high. The two usually match, but NICE says clinical judgement should always be used, because a patient with a low score who has an unusually low oxygen saturation, pleural complications or multiple comorbidities may have moderate- or high-severity disease. That severity grade then shapes the rest of care. NICE ties the choice of antibiotic to disease severity, alongside the risk of complications, local resistance data, recent antibiotic use and recent microbiology results, with separate first-line options for low-, moderate- and high-severity disease. It advises against routine microbiological tests in low-severity disease, and suggests considering blood cultures, sputum cultures and urinary antigen tests in moderate- or high-severity disease depending on the clinical picture. Antibiotics should start as soon as possible after diagnosis, within 4 hours of presentation to hospital. The CURB-65 score is a starting point, not the final severity grade.

Limits of the CURB-65 score

The CURB-65 score leaves out several things that change a pneumonia patient's risk. Hypoxia is not in the score: a PaO₂ below 8 kPa was linked to 30-day mortality in Lim's own univariate analysis, yet it did not enter the final rule, and a patient with an oxygen saturation of 85% can still score 0. Comorbidity is not counted either, so heart failure, chronic lung disease, frailty or immunosuppression add nothing. Low serum albumin was an independent predictor in the paper but was left out because it is not routinely available at admission. Age cuts both ways. Every patient aged 65 or older starts with a point, while a patient under 65 can score at most 4 and needs three acute abnormalities to reach the high-risk band, so a young patient with severe illness can score low. The study also excluded nursing home residents, immunocompromised patients, people in hospital in the previous 14 days, and people with tuberculosis, bronchiectasis, HIV infection or cancer.

NICE makes the same point in every place-of-care recommendation: use clinical judgement with the score, bearing in mind it can be affected by comorbidities or pregnancy. CURB-65 is also for adults with community-acquired pneumonia only; NG250 recommends no prediction tool for hospital-acquired pneumonia or for under 18s in primary care. For a score that does weigh age and chronic illness, see CURB-65 vs PSI.

Questions

What does a CURB-65 score of 2 mean?

Intermediate risk: 9.2% 30-day mortality in the 2003 derivation cohort, and a 3% to 15% band in NICE NG250. NICE suggests considering a virtual ward, SDEC unit, hospital at home service or inpatient care, decided with the patient.

Why do some sources say 57% mortality at a score of 5?

That figure is in the abstract of the 2003 paper, but the paper's Table 4 shows 1 death among 7 patients with a score of 5 (14%), with three more needing mechanical ventilation. The groups in the paper's own figure, 1.5%, 9.2% and 22%, match Table 4.

Is a CURB-65 of 0 always safe for home treatment?

No. Mortality was 0.7% in the derivation cohort, but NICE says a low score can still mean moderate- or high-severity disease when oxygen saturation is unusually low, there are pleural complications or there are multiple comorbidities.

What is the difference between CURB-65 and CRB-65?

CRB-65 drops urea and scores 0 to 4 from clinical signs only. NICE uses CRB-65 in primary care and CURB-65 in hospital, with different mortality bands for each.

Can AskMed explain a CURB-65 score with sources?

Yes. AskMed answers from journals and guidelines such as NICE and IDSA, citing each claim so you can open the source. It is on the App Store and Google Play, and you can use it on the web.

Sources

  1. Lim WS, van der Eerden MM, Laing R, et al. Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax. 2003;58(5):377–382. doi:10.1136/thorax.58.5.377
  2. National Institute for Health and Care Excellence. Pneumonia: diagnosis and management. NICE guideline NG250, published 2 September 2025 (replaces CG191 and NG138), recommendations 1.2.1 to 1.2.11, 1.4.5, 1.4.6, 1.5.1 and 1.6.1. nice.org.uk/guidance/ng250
  3. Lim WS, Baudouin SV, George RC, et al. BTS guidelines for the management of community acquired pneumonia in adults: update 2009. Thorax. 2009;64(Suppl 3):iii1–iii55. doi:10.1136/thx.2009.121434
  4. Metlay JP, Waterer GW, Long AC, et al. Diagnosis and treatment of adults with community-acquired pneumonia: an official clinical practice guideline of the ATS and IDSA. Am J Respir Crit Care Med. 2019;200(7):e45–e67. doi:10.1164/rccm.201908-1581ST

Cited medical answers
you can trust.

AskMed on iPhone, iPad, Android and the web.