CURB-65 vs PSI: Pneumonia Severity Scoring Compared

Compare CURB-65 (simple, 5 criteria) and PSI/PORT Score (complex, 20 criteria) for community-acquired pneumonia. Learn when to use each for admission decisions and outpatient treatment.

By Online Medical Tools Editorial TeamPublished Invalid Date

CURB-65 vs PSI: Pneumonia Severity Scoring Compared

Quick Answer: CURB-65 is a simple 5-criteria tool (Confusion, Urea, Respiratory rate, Blood pressure, Age ≥65) designed to identify high-risk pneumonia patients who need hospital admission or ICU care—best for quick emergency department decisions. PSI (Pneumonia Severity Index, also called PORT Score) is a complex 20-variable tool designed to identify low-risk pneumonia patients who can safely be treated as outpatients—best for safely discharging patients home. Use CURB-65 when you need to decide if someone needs admission (simple, fast, identifies severe cases). Use PSI when you want to confirm someone is safe to send home (more comprehensive, rules out hidden risk factors).

Side-by-Side Comparison

| Feature | CURB-65 | PSI (PORT Score) | |---------|---------|------------------| | Full name | Confusion, Urea, Respiratory rate, Blood pressure, 65+ years | Pneumonia Severity Index (Patient Outcomes Research Team Score) | | Primary purpose | Identify HIGH-risk patients needing admission/ICU | Identify LOW-risk patients safe for outpatient treatment | | Number of criteria | 5 simple criteria | 20 variables (3 demographic, 5 comorbidities, 5 physical exam, 7 lab values) | | Score range | 0-5 | 0-395 (Risk Class I-V) | | Risk categories | 4 levels (0, 1, 2, 3-5) | 5 Risk Classes (I-V) | | Time to calculate | 1-2 minutes (bedside) | 5-10 minutes (requires labs) | | Complexity | Very simple (5 yes/no questions) | Complex (20 weighted variables) | | Lab tests required | Urea/BUN only | 7 labs required (pH, BUN, sodium, glucose, hematocrit, PaO₂, pleural effusion) | | Best for identifying | Severe cases (high mortality risk) | Mild cases (low mortality risk) | | Sensitivity | Lower for low-risk detection | Higher for low-risk detection | | Specificity | 74.6% (better at ruling in high risk) | 52.2% (more false positives) | | ICU admission prediction | Better (higher sensitivity) | Lower sensitivity for ICU need | | Outpatient treatment | CURB-65 = 0-1: consider outpatient | PSI Class I-II: outpatient recommended | | Hospital admission | CURB-65 = 2: consider admission | PSI Class III: consider brief admission or close outpatient follow-up | | ICU consideration | CURB-65 ≥3: consider ICU | PSI Class IV-V: inpatient, consider ICU | | Young patient bias | No age bias (65+ is only 1 point) | ✅ Age bias (young patients get lower scores even if severely ill) | | Elderly patient | Less age weighting (1 point for ≥65) | Heavy age weighting (age in years = points) | | Clinical use setting | Emergency department (quick triage) | Primary care, ED (confirm low risk for discharge) | | Evidence base | Widely validated in ED populations | Original PNEUMONIA study (14,000+ patients) | | Guideline recommendations | British Thoracic Society, IDSA/ATS | IDSA/ATS guidelines |

CURB-65: Simple High-Risk Identification

What It Measures

CURB-65 is a simple 5-point scoring system that identifies community-acquired pneumonia (CAP) patients at high risk of death who require hospital admission or ICU care. It's designed for quick bedside assessment in emergency departments.

Developed in 2003 by the British Thoracic Society, CURB-65 simplified the earlier CURB score by adding age ≥65 as a criterion.

The 5 CURB-65 Criteria (1 Point Each)

| Criteria | Definition | Points | |----------|-----------|--------| | C - Confusion | New onset disorientation (person, place, time) or altered mental status | 1 | | U - Urea | Blood urea nitrogen (BUN) >19 mg/dL (or >7 mmol/L) | 1 | | R - Respiratory rate | Respiratory rate ≥30 breaths per minute | 1 | | B - Blood pressure | Systolic BP <90 mmHg OR diastolic BP ≤60 mmHg | 1 | | 65 - Age | Age ≥65 years | 1 |

Total Score: 0-5 points

CURB-65 Risk Categories and Recommendations

| Score | Risk | 30-Day Mortality | Recommended Management | |-------|------|-----------------|------------------------| | 0 | Very low | <1% | Outpatient treatment (oral antibiotics) | | 1 | Low | 2-3% | Consider outpatient vs brief observation | | 2 | Moderate | 9% | Hospital admission recommended | | 3 | High | 15-17% | Hospital admission, consider ICU | | 4-5 | Very high | 30-40% | Urgent hospital admission, strongly consider ICU |

Clinical decision thresholds:

Score 0-1: Outpatient treatment safe
Score 2: Hospitalization usually indicated
Score ≥3: Severe pneumonia - admit to hospital, consider ICU care

Strengths of CURB-65

Very simple - 5 yes/no questions, anyone can calculate
Fast - takes 1-2 minutes at bedside
Minimal labs - only requires BUN (or can skip and use CRB-65)
Good specificity (74.6%) - reliably identifies high-risk patients
Predicts ICU need - better than PSI at identifying patients who will need ICU
Widely adopted - recommended by British Thoracic Society, used internationally
CRB-65 variant - can drop "U" (urea) if labs unavailable, still useful

Limitations of CURB-65

Lower sensitivity for low-risk - may miss some patients who could go home
Binary criteria - doesn't account for degree of abnormality (e.g., BP 89 vs 60 both score 1)
Limited scope - only 5 variables, doesn't include comorbidities, oxygenation, or many lab values
Less granular - 5-point scale vs PSI's 5 risk classes
Not validated for all populations - less studied in nursing home, immunocompromised patients

PSI (Pneumonia Severity Index): Comprehensive Low-Risk Identification

What It Measures

PSI (also called the PORT Score - Patient Outcomes Research Team) is a comprehensive 20-variable scoring system that predicts 30-day mortality in community-acquired pneumonia. It's designed to identify low-risk patients who can be safely treated as outpatients, avoiding unnecessary hospitalizations.

Developed in 1997 from a study of 14,199 hospitalized pneumonia patients and validated in 38,039 additional patients.

PSI Risk Classes

PSI assigns patients to 5 Risk Classes (I-V) based on points:

| Risk Class | Score Range | 30-Day Mortality | Recommended Management | |-----------|-------------|------------------|------------------------| | Class I | Age <50 + no comorbidities + normal vitals + normal mental status | 0.1-0.4% | Outpatient treatment | | Class II | ≤70 points | 0.6-0.7% | Outpatient treatment | | Class III | 71-90 points | 0.9-2.8% | Brief inpatient vs close outpatient monitoring | | Class IV | 91-130 points | 8.2-9.3% | Inpatient treatment | | Class V | >130 points | 27-31% | Inpatient treatment, consider ICU |

The 20 PSI Variables (Complex Scoring)

Step 1: Is patient Risk Class I?

If all criteria met → Risk Class I (outpatient treatment):
- Age <50 years
- No cancer, liver disease, CHF, cerebrovascular disease, or kidney disease
- Normal mental status
- Pulse <125 bpm
- Respiratory rate <30/min
- Systolic BP ≥90 mmHg
- Temperature 35-40°C

Step 2: If NOT Risk Class I, calculate points:

Demographic factors:

Age in years (males) or age - 10 (females)
Nursing home resident: +10 points

Comorbidities:

Cancer: +30
Liver disease: +20
Congestive heart failure: +10
Cerebrovascular disease: +10
Kidney disease: +10

Physical exam findings:

Altered mental status: +20
Respiratory rate ≥30: +20
Systolic BP <90: +20
Temperature <35°C or ≥40°C: +15
Pulse ≥125: +10

Laboratory/Radiographic findings:

Arterial pH <7.35: +30
BUN ≥30 mg/dL: +20
Sodium <130 mEq/L: +20
Glucose ≥250 mg/dL: +10
Hematocrit <30%: +10
PaO₂ <60 mmHg or O₂ sat <90%: +10
Pleural effusion on X-ray: +10

Total points: 0-395

Risk Class assignment:

Class II: ≤70 points
Class III: 71-90 points
Class IV: 91-130 points
Class V: >130 points

Strengths of PSI

High sensitivity for low risk - excellent at identifying patients safe for outpatient treatment
Comprehensive - accounts for age, comorbidities, vitals, labs, radiology
Validated in large studies - 52,000+ patients in original and validation cohorts
Reduces unnecessary admissions - safely identifies ~30% of patients who can go home
Weighted scoring - more important variables get more points (e.g., pH = 30 points, pulse = 10)
Guideline-recommended - IDSA/ATS guidelines endorse PSI for risk stratification

Limitations of PSI

Complex - 20 variables, weighted scoring, difficult to calculate without computer
Time-consuming - requires complete labs and chest X-ray
Age bias - young patients automatically get low scores even if severely ill
- Example: 30-year-old with severe pneumonia (pH 7.2, respiratory rate 35, O₂ sat 85%) might still be Class II-III due to young age
Lower specificity (52.2%) - more false positives (predicts high risk when mortality is lower)
Less useful for ICU decisions - doesn't predict ICU need as well as CURB-65
Not bedside-friendly - requires calculator or electronic tool

Key Differences: CURB-65 vs PSI

1. Design Philosophy: High-Risk vs Low-Risk Identification

CURB-65: Designed to identify severe pneumonia (high mortality risk)

Focus: "Who needs to be admitted or go to ICU?"
Goal: Don't miss severe cases
Approach: If CURB-65 ≥2, admit

PSI: Designed to identify mild pneumonia (low mortality risk)

Focus: "Who is safe to send home?"
Goal: Avoid unnecessary hospitalizations
Approach: If PSI Class I-II, can treat outpatient

Clinical implication: Use CURB-65 to rule IN admission. Use PSI to rule OUT admission.

2. Simplicity vs Comprehensiveness

CURB-65:

5 simple criteria (yes/no)
Can calculate at bedside in 1-2 minutes
Only requires BUN lab (or none for CRB-65)
Easy to remember: C-U-R-B-65

PSI:

20 complex variables (weighted scoring)
Requires 5-10 minutes + complete labs
Needs calculator or electronic tool
Difficult to do manually

Clinical implication: CURB-65 is ideal for busy EDs. PSI is better when you have time and full lab results.

3. Age Weighting

CURB-65:

Age ≥65 = 1 point (out of 5 total)
Less age bias - young patients can still score high if severely ill

PSI:

Age in years = points (e.g., 75-year-old = 75 points before any other factors)
Heavy age bias - elderly patients automatically get high scores
Problem: Young, otherwise healthy patients with severe pneumonia get low scores
- Example: 25-year-old with hypoxia (O₂ sat 85%), tachypnea (RR 35), hypotension (BP 85/50) → might still be Class II-III due to age weighting

Clinical implication: PSI may underestimate severity in young patients. Always consider clinical judgment.

4. Sensitivity vs Specificity

CURB-65:

Higher specificity (74.6%) - when it says high risk, it usually is
Lower sensitivity for low risk - may not identify all safe-to-discharge patients
Better for ruling IN admission - if CURB-65 ≥2, patient likely needs admission

PSI:

Higher sensitivity - better at identifying truly low-risk patients
Lower specificity (52.2%) - more false positives (predicts high risk when patient does well)
Better for ruling OUT admission - if PSI Class I-II, patient likely safe to discharge

5. ICU Admission Prediction

CURB-65:

Better at predicting ICU need - CURB-65 ≥3 correlates with ICU admission
Higher sensitivity for severe pneumonia requiring intensive care

PSI:

Less accurate for ICU prediction - was designed for mortality, not ICU need
May classify patients as moderate risk who actually need ICU

Clinical implication: Use CURB-65 when considering ICU admission.

6. Guidelines and Adoption

CURB-65:

British Thoracic Society - primary recommendation
IDSA/ATS guidelines - recognized as alternative
More popular in Europe and UK

PSI:

IDSA/ATS guidelines - recommended for identifying low-risk patients
Original US-developed tool
More popular in United States

When to Use CURB-65

Use CURB-65 when:

1. Emergency Department Triage (Quick Decisions)

Patient presents with pneumonia symptoms
Need rapid assessment of severity
Decide: admit vs discharge?

Workflow:

Calculate CURB-65 at bedside (1-2 minutes)
Score 0-1: Consider outpatient treatment
Score 2: Likely needs admission
Score ≥3: Admit to hospital, consider ICU

2. Identifying High-Risk Patients

Patient appears sick, need objective severity assessment
CURB-65 ≥3 indicates severe pneumonia
Triggers ICU consultation

3. No Full Labs Available

Urgent care, clinic, or pre-hospital setting
Can use CRB-65 (drop the "U" for urea) if no BUN available
Still useful for risk stratification

4. Busy Clinical Settings

High patient volume ED
Need simple, fast tool
CURB-65 is memorizable and quick

5. Young Patients

CURB-65 doesn't heavily penalize youth
Better captures severe pneumonia in otherwise healthy young adults

When to Use PSI

Use PSI when:

1. Confirming Low-Risk Status for Discharge

Patient appears mild but want objective confirmation
PSI Class I-II strongly supports outpatient treatment
Reduces unnecessary admissions

Workflow:

Patient presents with mild-moderate symptoms
Calculate PSI (requires full labs + chest X-ray)
Class I-II: Outpatient treatment safe
Class III: Consider brief admission or close outpatient follow-up
Class IV-V: Admission required

2. Primary Care or Outpatient Settings

Have time to gather complete data (labs, X-ray)
Want comprehensive risk assessment
PSI's 20 variables capture more risk factors than CURB-65

3. Elderly Patients

PSI appropriately weighs age-related risk
Accounts for multiple comorbidities (common in elderly)
More nuanced than CURB-65's single "≥65" criterion

4. Avoiding Unnecessary Hospitalizations

Hospital capacity is limited
Want to safely discharge low-risk patients
PSI identifies ~30% of patients safe for outpatient treatment

5. Research and Quality Improvement

PSI is validated in large studies (52,000+ patients)
Standard tool for pneumonia research
Useful for comparing hospital outcomes

Clinical Scenarios: Which Score to Use

Scenario 1: 72-Year-Old Woman, Cough, Fever, No Confusion, BP 130/80, RR 24, BUN 25

Use CURB-65:

C = 0 (no confusion)
U = 1 (BUN 25 > 19)
R = 0 (RR 24 < 30)
B = 0 (BP normal)
65 = 1 (age 72)
Total = 2 → Hospital admission recommended

Use PSI (if full labs available):

Age 72 (female) = 62 points
BUN 25 = 0 (needs ≥30)
Add other variables...
Estimated Class III → Brief admission or close outpatient monitoring

Clinical decision: Both scores suggest admission or close monitoring. Admit for observation.

Scenario 2: 28-Year-Old Man, Severe Dyspnea, RR 38, O₂ Sat 82%, BP 85/55, No Comorbidities

Use CURB-65:

C = 0 (no confusion documented)
U = assume 0 (not elevated in young, healthy)
R = 1 (RR 38 ≥ 30)
B = 1 (SBP 85 < 90)
65 = 0 (age 28)
Total = 2 → Hospital admission recommended (possibly ICU due to hypoxia)

Use PSI:

Age 28 = 28 points
Hypoxia (O₂ sat 82%) = +10
Tachypnea (RR 38) = +20
Hypotension (SBP 85) = +20
Total ~78 points = Class III (only 0.9-2.8% mortality)

Problem with PSI: Despite severe respiratory distress and hypotension, PSI says Class III (low mortality). This is the age bias problem - young age keeps score low even with severe illness.

Clinical decision: Trust CURB-65 (or clinical judgment) over PSI here. Patient needs admission, likely ICU, despite PSI Class III. PSI underestimates severity in young patients.

Scenario 3: 55-Year-Old Woman, Mild Cough, Low-Grade Fever, All Vitals Normal, No Confusion, BUN 12

Use CURB-65:

C = 0
U = 0 (BUN 12 < 19)
R = 0
B = 0
65 = 0 (age 55)
Total = 0 → Outpatient treatment safe

Use PSI:

Age 55 (female) = 45 points
No comorbidities, normal vitals, normal labs
Total ~45 points = Class II (outpatient treatment)

Clinical decision: Both scores agree - outpatient treatment with oral antibiotics. No admission needed.

Scenario 4: 80-Year-Old Man, Nursing Home Resident, Diabetes, CHF, Mild Confusion, RR 22, BP 110/70

Use CURB-65:

C = 1 (mild confusion)
U = assume 0 (not mentioned)
R = 0 (RR 22 < 30)
B = 0 (BP normal)
65 = 1 (age 80)
Total = 2 → Hospital admission recommended

Use PSI:

Age 80 = 80 points
Nursing home = +10
CHF = +10
Diabetes (if glucose ≥250) = +10
Confusion = +20
Total ≥130 points = Class IV-V (high mortality risk)

Clinical decision: PSI provides more granular assessment - captures multiple comorbidities and nursing home status. Both scores indicate admission needed. PSI's comprehensiveness is valuable here - highlights very high baseline risk.

Scenario 5: Busy ED, Multiple Pneumonia Patients, Need Quick Triage

Use CURB-65 for all patients:

Takes 1-2 minutes per patient
Rapidly stratify: Score 0-1 (consider discharge), Score 2+ (admit)
Later, can calculate PSI for borderline cases if time permits

Clinical decision: CURB-65 is ideal for high-volume, time-pressured settings. PSI is too slow for initial triage.

Scenario 6: Primary Care Clinic, Patient Wants to Avoid Hospitalization, Has Mild Symptoms

Use PSI:

Have time to obtain full labs and chest X-ray
Calculate comprehensive PSI score
PSI Class I-II provides strong evidence for outpatient treatment
Reassures patient and physician that outpatient care is safe

Clinical decision: PSI is ideal here - comprehensive assessment confirms low risk, supports shared decision-making for outpatient treatment.

Can You Use Both Together?

Yes, and this can provide complementary information:

Combined Approach

Step 1: Calculate CURB-65 first (quick screening)

CURB-65 ≥3: Admit to hospital, consider ICU → skip PSI (already know patient is high risk)
CURB-65 = 0: Very low risk → consider outpatient, skip PSI (already know patient is low risk)
CURB-65 = 1-2: Borderline → proceed to PSI for more detailed assessment

Step 2: If CURB-65 = 1-2, calculate PSI

PSI Class I-II: Outpatient treatment
PSI Class III: Brief admission vs close outpatient follow-up (shared decision-making)
PSI Class IV-V: Hospital admission

When Scores Disagree

CURB-65 high, PSI low (rare):

Example: Young patient with severe pneumonia (CURB-65 = 3, PSI Class II)
Trust CURB-65 - PSI likely underestimating due to age bias
Admit patient

CURB-65 low, PSI high (common in elderly):

Example: 80-year-old with mild symptoms (CURB-65 = 1, PSI Class IV)
Use clinical judgment - PSI may be correct (elderly with comorbidities are higher risk)
Consider brief admission or very close outpatient monitoring

Best Practice

Use CURB-65 as initial screen, PSI for borderline cases:

Calculate CURB-65 at bedside (1-2 minutes)
If CURB-65 clear-cut (0 or ≥3): Make decision based on CURB-65
If CURB-65 borderline (1-2): Calculate PSI for more nuanced assessment
If scores conflict: Use clinical judgment + consider age bias in PSI

Evidence: Which is Better?

Mortality Prediction

Tie: Both predict 30-day mortality with similar accuracy

CURB-65 AUROC: 0.75-0.80
PSI AUROC: 0.75-0.85
Recent meta-analysis: no significant difference in mortality prediction

ICU Admission Prediction

Winner: CURB-65

Higher sensitivity for identifying patients who will need ICU
CURB-65 ≥3 strongly predicts ICU admission
PSI less accurate for ICU need

Identifying Low-Risk Patients (Safe for Discharge)

Winner: PSI

PSI Class I-II has very low mortality (0.1-0.7%)
Better at confidently ruling out high risk
CURB-65 = 0 also excellent (0.6% mortality), but less comprehensive

Simplicity and Speed

Winner: CURB-65

Takes 1-2 minutes vs 5-10 minutes for PSI
Can be done at bedside without calculator
Easier to remember and teach

Young Patients with Severe Pneumonia

Winner: CURB-65

PSI underestimates severity in young patients due to age weighting
CURB-65 captures severity regardless of age

Elderly Patients with Comorbidities

Winner: PSI

Accounts for multiple comorbidities (CHF, cancer, liver disease, kidney disease)
Age-weighted scoring appropriately reflects baseline risk
CURB-65's single "≥65" criterion is less nuanced

Clinical Guidelines

Both recommended:

IDSA/ATS guidelines: Endorse both CURB-65 and PSI
British Thoracic Society: Recommends CURB-65 as primary tool
Choice depends on clinical setting and patient population

Key Takeaways

✅ CURB-65 is simple (5 criteria), PSI is complex (20 variables)

✅ CURB-65 identifies high-risk patients needing admission, PSI identifies low-risk patients safe for discharge

✅ Use CURB-65 for quick ED triage and ICU decisions - takes 1-2 minutes, predicts ICU need well

✅ Use PSI when you have time and full labs - more comprehensive, confirms low-risk status for outpatient treatment

✅ PSI has age bias - underestimates severity in young patients, always use clinical judgment

✅ CURB-65 ≥3 = severe pneumonia - admit to hospital, consider ICU

✅ PSI Class I-II = low risk - outpatient treatment safe (0.1-0.7% mortality)

✅ Both predict 30-day mortality equally well (AUROC ~0.75-0.85)

✅ Can use both together - CURB-65 first for screening, PSI for borderline cases

✅ When scores disagree, trust clinical judgment - especially for young patients (trust CURB-65 over PSI)

Related Tools

CURB-65 Calculator - Simple 5-criteria pneumonia severity score
PSI Calculator - Comprehensive Pneumonia Severity Index (PORT Score)
Emergency Medicine Guide - Complete guide to CURB-65, PSI, and pneumonia management

Sources

Disclaimer: CURB-65 and PSI are clinical decision tools for healthcare professionals. They are not substitutes for clinical judgment and should be used as part of comprehensive patient assessment. If you have pneumonia symptoms (cough, fever, shortness of breath, chest pain), seek medical evaluation immediately. These tools are for educational and informational purposes only.

Disclaimer: This tool is for educational and informational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider with questions about your health.