CURB-65 Pneumonia Severity Calculator
CURB-65 and CRB-65 pneumonia severity calculator. Scores confusion, urea (mmol/L or BUN), respiratory rate, blood pressure and age 65 or over, and shows the 2003 study's own 30-day death counts. Educational only.
CURB-65 Pneumonia Severity Calculator
Urea scores a point above 7 mmol/L. As BUN that is above 19.6 mg/dL (7 × 2.80134 = 19.61), which on a whole-number BUN report means 20 or more. A result in mmol/L is entered as mmol/L even if the report calls it BUN. Some reports give urea itself in mg/dL, a different quantity: 7 mmol/L of urea is about 42 mg/dL, so do not enter that figure as BUN. With no urea result the calculator says so and computes CRB-65 from the other four items.
Please tick the acknowledgement above before calculating.
How to use the CURB-65 calculator
Confirm pneumonia first
CURB-65 grades the severity of community-acquired pneumonia once it has been diagnosed. It does not diagnose pneumonia, and it was not derived for pneumonia caught in hospital.
Tick the bedside items
One point each for new Confusion, a Respiratory rate of 30 or more, low Blood pressure and age 65 or older.
Enter the urea result, or leave it blank
Pick the unit the report uses: urea in mmol/L, BUN (urea nitrogen) in mg/dL, or urea itself in mg/dL. The calculator converts it and applies the published cut, above 7 mmol/L. With no result it computes CRB-65, the four-item version, and says so.
Read the figures as the study's, not yours
The result shows how many patients with the same score died within 30 days in the 2003 study, and the risk group NICE gives that score. Those are cohort figures. The decision about care belongs to a clinician.
CURB-65 — grading pneumonia in five questions
A five-item severity score from 2003
CURB-65 comes from a study by Lim and colleagues of 1,068 patients admitted with community-acquired pneumonia in the UK, New Zealand and the Netherlands, published in Thorax in 2003. Each item scores a point: Confusion (a Mental Test Score of 8 or less, or new disorientation in person, place or time), Urea above 7 mmol/L, a Respiratory rate of 30 a minute or more, low Blood pressure (systolic under 90 mmHg or diastolic 60 mmHg or below) and age 65 or over. The total runs from 0 to 5.
The table below is the paper's own Table 4 for CURB-65: how many patients with each score died within 30 days, in the cohort used to build the score and in the separate cohort used to test it. These are the figures the calculator shows.
| Score | Derivation cohort | Validation cohort |
|---|---|---|
| 0 | 1 of 137 (0.7%) | 0 of 36 |
| 1 | 4 of 187 (2.1%) | 0 of 54 |
| 2 | 17 of 184 (9.2%) | 5 of 60 (8.3%) |
| 3 | 20 of 138 (14.5%) | 9 of 42 (21.4%) |
| 4 | 26 of 65 (40%) | 5 of 19 (26.3%) |
| 5 | 1 of 7 (14%) | 1 of 3 (33.3%) |
Two cautions come with it. First, the top of the scale rests on very few people: only 7 of the 718 derivation patients scored 5, and one of them died, so the percentage there means little. Second, the paper's abstract prints a different set for the same scores (0.7, 3.2, 3, 17, 41.5 and 57%), and the paper does not say why its abstract and its table differ. The paper also groups the scores, in its Figure 2: 0 or 1 "Mortality low" (1.5%, 5 of 324), 2 "Mortality intermediate" (9.2%, 17 of 184) and 3 or more "Mortality high" (22%, 47 of 210).
Correction, 18 October 2026. Until this date the calculator showed 0.6, 2.7, 6.8, 14 and 27.8% and credited them to Lim 2003, describing scores 4 and 5 as one pooled band. Those figures are not in that paper, which reports every score separately; we could not find them in it or in the validation studies we checked. The page also suggested a care setting for each score, which it no longer does.
Without a blood test: CRB-65
CRB-65 is the same score with urea left out, so it runs from 0 to 4. It is in the same 2003 paper (Figure 3), for use where a urea result is not available. In its derivation cohort 2 of 167 patients scoring 0 died within 30 days (1.2%), 37 of 455 scoring 1 or 2 (8.15%) and 30 of 96 scoring 3 or 4 (31%). A 2006 study by Bauer and colleagues in the CAPNETZ network (1,967 patients for CRB-65) found that the CURB, CRB and CRB-65 scores "provided comparable predictions". It added: "Availability of the CRB-65 score (90%) was far superior to that of CURB (65%), due to missing blood urea nitrogen values". This calculator computes CRB-65 whenever the urea box is left empty, rather than scoring a missing result as normal.
"A blank urea box is not a normal urea: the score changes to CRB-65 and says so."
Urea, BUN and the unit trap
Lim's threshold is urea above 7 mmol/L. The 2007 IDSA/ATS guideline renders it as "BUN level >7 mmol/L (20 mg/dL)": blood urea nitrogen (BUN), in mg/dL, counts only the nitrogen in the urea. Each urea molecule carries two nitrogen atoms, so 1 mmol/L of urea is 2 × 14.0067 / 10 = 2.80134 mg/dL of BUN, and 7 mmol/L is 19.6 mg/dL. On a whole-number BUN report the rule is therefore 20 or more; on a report with decimals, a BUN of 19.1 to 19.6 does not score. A report that gives urea itself in mg/dL shows a different, larger number: 7 mmol/L of urea weighs about 42 mg/dL. Entered as BUN, that figure would score a point the paper would not give. NICE's own guideline writes the item as "raised blood urea nitrogen (over 7 mmol/litre)", so the label on a report is not a reliable guide to the unit; the unit is.
What the score can't see
CURB-65 says nothing about oxygen levels, which can be low when the five items look reassuring, and it does not weigh other illnesses such as heart failure, lung disease or a weakened immune system. A younger adult scores nothing for age and can still be very unwell. NICE's guideline (NG250, published 2 September 2025) asks clinicians to "use clinical judgement together with the CURB65 score", and notes that it "can be affected by other factors, for example, comorbidities or pregnancy". Anyone with a cough, fever and breathlessness should be assessed by a clinician rather than score themselves, because pneumonia can worsen quickly.
It grades pneumonia; it does not diagnose it
Derived in 2003 by Lim et al (Thorax) from 1,068 patients.
Confusion, Urea, Respiratory rate, BP, age 65.
One point each; maximum score 5.
Derivation cohort: 1 of 137 died at score 0, 26 of 65 at score 4.
Only 7 of 718 derivation patients scored 5.
Urea 7 mmol/L = BUN 19.6 mg/dL.
CRB-65, without urea, is in the same 2003 paper.
It grades pneumonia severity — it doesn't diagnose it.
It ignores oxygen levels and other illnesses.
NICE NG250 (2025): CRB65 in primary care, CURB65 in hospital.
Frequently asked questions
Confusion (a Mental Test Score of 8 or less, or new disorientation), Urea above 7 mmol/L, Respiratory rate of 30 or more a minute, low Blood pressure (systolic under 90 mmHg or diastolic 60 mmHg or below), and age 65 or over. Each scores one point, for a total from 0 to 5. It grades the severity of community-acquired pneumonia.
NICE's guideline NG250 gives these for CURB65 in hospital (Box 2): 0 or 1 "low risk (less than 3% mortality risk)", 2 "intermediate risk (3% to 15% mortality risk)", 3 to 5 "high risk (more than 15% mortality risk)". For CRB65 in primary care (Box 1): 0 "low risk (less than 1% mortality risk)", 1 or 2 "intermediate risk (1% to 10% mortality risk)", 3 or 4 "high risk (more than 10% mortality risk)". The recommendations these boxes belong to are tagged [2014] in NG250: they date from the 2014 guideline it replaced. NICE also recommends where care might take place for each score, wording it rewrote in 2025; that is guidance for clinicians and this page does not repeat it as advice. Note that Lim's own Table 4 gives 14.5% at a score of 3, below NICE's "more than 15%".
CRB-65 leaves out urea, so it needs no blood test and runs from 0 to 4. Both are in Lim's 2003 paper. NICE NG250 uses CRB65 in primary care and CURB65 in hospital. Bauer and colleagues (2006) found that CURB, CRB and CRB-65 "provided comparable predictions". If you leave the urea box empty, this calculator computes CRB-65 and labels the result accordingly.
Go by the unit, not the name. A result in mmol/L is urea in mmol/L, even where a lab lists it as BUN. A result in mg/dL is usually BUN (urea nitrogen), where the cut is above 19.6 mg/dL; if the report says it is urea itself in mg/dL, choose that option, where the cut is about 42 mg/dL. The calculator shows the converted value and the threshold in your unit, so you can check it.
No. Oxygen saturation is not one of its five items, so a low score can sit alongside a low oxygen level. The score is a quick severity estimate, not a complete assessment, and guidelines pair it with clinical judgement.
They are counts from adults admitted to hospital with pneumonia in three countries, so they describe those groups, not any one person. Some groups are tiny: 7 patients scored 5 in the derivation cohort and 3 in the validation cohort. The paper's abstract and its Table 4 also print different per-score figures. Read the counts as a rough gradient, and treat a single percentage as uncertain.
No. CURB-65 was derived and validated in community-acquired pneumonia. Pneumonia caught in hospital or on a ventilator involves different organisms and risks, and the score was not built for it.
Age 65 or over is one of the points, so a young adult starts a point lower and may also have a normal blood pressure and mental state, giving a low total despite being unwell. That is one reason the score is read alongside oxygen levels and the clinical picture rather than used mechanically.
No. Pneumonia can worsen quickly, and several items (urea, an accurate respiratory rate, blood pressure) need clinical measurement. If you have a cough, fever and breathlessness, seek medical assessment. This tool is educational, to show how the score is built, not to triage your own illness.
They are different scores, and guidelines differ on which to use. The 2019 ATS/IDSA guideline for adults in the US recommends "preferentially the Pneumonia Severity Index (PSI) (strong recommendation, moderate quality of evidence) over the CURB-65 … (conditional recommendation, low quality of evidence)" when deciding on hospital care. NICE's NG250 in the UK uses CRB65 and CURB65. Which one applies depends on the local guideline.
No. The boxes you tick and the urea result are processed only in your browser. Nothing is sent to a server, stored, or shared, and no account is required.
Method & sources
How it computes
CURB-65 (Lim 2003): 1 point each for confusion, urea >7 mmol/L (BUN >19.6 mg/dL; 7 x 2.80134), respiratory rate >=30/min, SBP <90 or DBP <=60 mmHg, age >=65. No urea result -> CRB-65, the same score without urea (0-4). Shows Lim 2003 Table 4 30-day deaths per score with denominators (derivation and validation cohorts), Lim Figure 2/3 groups, and NICE NG250 Box 2 / Box 1 risk groups. No care setting.
What this tool implements
- CURB-65 and CRB-65 criteria as printed in Lim 2003 Figures 2-3: confusion (Mental Test Score of 8 or less, or new disorientation in person, place or time), urea >7 mmol/L, respiratory rate >=30/min, SBP <90 or DBP <=60 mmHg, age >=65. A blank urea computes CRB-65 and the page says so.
- Per-score 30-day deaths are Lim 2003 Table 4 as printed, with denominators: CURB-65 derivation 1/137, 4/187, 17/184, 20/138, 26/65, 1/7; validation 0/36, 0/54, 5/60, 9/42, 5/19, 1/3. CRB-65 derivation 2/167, 14/266, 23/189, 28/85, 2/11. Where the abstract prints a different figure (0.7, 3.2, 3, 17, 41.5, 57%) the page says so.
- Risk groups quoted from NICE NG250 (2 Sep 2025): Box 2 for CURB65 in hospital, Box 1 for CRB65 in primary care; both recommendations tagged [2014]. Place-of-care recommendations [2025] are not reproduced as advice.
- Urea entered as mmol/L, BUN mg/dL (x 2.80134) or urea mg/dL (x 6.0055), compared in its own unit against 7 mmol/L.
- Corrected 2026-10-18: the page showed 0.6 / 2.7 / 6.8 / 14.0 / 27.8% credited to Lim 2003 with 4-5 "pooled" and a care setting. The 2026-08-09 review note recorded those figures as "the pooled derivation-plus-validation values from the same paper" and said the paper "publishes scores 4 and 5 as ONE pooled band at 27.8%". Both are false: Table 4 reports each score separately, 27.8 appears nowhere, and a score of 5 was 1 death in 7.
Sources
- 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. PMID 12728155. https://pmc.ncbi.nlm.nih.gov/articles/PMC1746657/
- National Institute for Health and Care Excellence. Pneumonia: diagnosis and management (NG250), 2 September 2025. Recommendations 1.2.1-1.2.9, Boxes 1-2. https://www.nice.org.uk/guidance/ng250/chapter/Recommendations
- Bauer TT, Ewig S, Marre R, et al. CRB-65 predicts death from community-acquired pneumonia. J Intern Med 2006;260(1):93-101. PMID 16789984.
- 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. https://pmc.ncbi.nlm.nih.gov/articles/PMC6812437/
- Mandell LA, Wunderink RG, Anzueto A, et al. IDSA/ATS consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis 2007;44 Suppl 2:S27-72. https://pmc.ncbi.nlm.nih.gov/articles/PMC7107997/
What can make this go out of date
- None at runtime; nothing the reader enters leaves the browser.
Pick up where you left off
Stored only in this browser — never sent to our servers.