Clinical Decision-Making

Pneumonia Severity: CURB-65 vs PSI/PORT — Choosing the Right Tool

CURB-65 is faster and simpler — 5 bedside variables, ideal for ED triage. PSI/PORT uses 20 variables and a two-step process, but more accurately identifies truly low-risk patients safe for outpatient treatment. Neither tool alone should drive admission decisions — use them with clinical judgment and IDSA/ATS minor criteria for ICU thresholds. In elderly patients, PSI has better discrimination than CURB-65.

Quick Medical Calculator Team · April 27, 2026 · 8 min read

TL;DR: CURB-65 is faster and simpler — 5 bedside variables, ideal for ED triage. PSI/PORT uses 20 variables and a two-step process, but more accurately identifies truly low-risk patients safe for outpatient treatment. Neither tool alone should drive admission decisions — use them with clinical judgment and IDSA/ATS minor criteria for ICU thresholds. In elderly patients, PSI has better discrimination than CURB-65.

Table of Contents

• The Clinical Problem
• CURB-65 — Overview and Scoring
• PSI/PORT Score — Overview and Scoring
• Head-to-Head Comparison
• Which Tool Should You Use?
• Evidence Summary
• Frequently Asked Questions
• References

The Clinical Problem

Community-acquired pneumonia (CAP) is the most common infectious cause of hospitalization in adults worldwide and carries a 30-day mortality of approximately 1–5% in outpatients and up to 25% in ICU patients. The central decision point — admit or discharge? floor or ICU? — has enormous consequences for both patient outcomes and healthcare resource use.

Two clinical prediction tools dominate this decision: CURB-65 and the Pneumonia Severity Index (PSI), also known as the PORT score. They were designed with different priorities, perform differently in different patient populations, and remain complementary rather than competing tools in 2025.

CURB-65 — Overview and Scoring

CURB-65 (Lim et al., 2003) prioritizes simplicity. It scores five variables, each worth one point:

Score range: 0–5

A no-lab version — CRB-65 — omits urea and is valid when labs aren't immediately available.

Strengths of CURB-65

• Computable in under 60 seconds from history, vitals, and one lab value
• CRB-65 requires zero lab work
• Strongly identifies high-severity patients
• Endorsed by BTS, IDSA/ATS, and WHO for resource-limited settings

PSI/PORT Score — Overview and Scoring

The Pneumonia Severity Index (Fine et al., 1997) was developed from the Pneumonia Patient Outcomes Research Team (PORT) cohort study. It stratifies patients into five risk classes using a two-step process.

Step 1: Class I Assignment (no points needed)

Patients <50 years with none of the following are automatically Class I (outpatient):

• Neoplastic disease
• Liver disease, CHF, cerebrovascular disease, renal disease
• Altered mental status
• Pulse ≥125 bpm
• RR ≥30/min
• SBP <90 mmHg
• Temperature <35°C or ≥40°C

Step 2: Point-Based Scoring (Classes II–V)

Patients who don't qualify for Class I enter the scoring system:

Demographic factors:

• Age (men: age in years; women: age – 10)
• Nursing home resident: +10

Comorbidities (+10 each): Neoplastic disease, liver disease, CHF, cerebrovascular disease, renal disease

Exam findings:

• Altered mental status: +20
• RR ≥30/min: +20
• SBP <90 mmHg: +20
• Temperature <35°C or ≥40°C: +15
• Pulse ≥125 bpm: +10

Labs/Imaging:

• pH <7.35: +30
• BUN ≥30 mg/dL: +20
• Sodium <130 mmol/L: +20
• Glucose ≥250 mg/dL: +10
• Hematocrit <30%: +10
• PaO₂ <60 mmHg or SaO₂ <90%: +10
• Pleural effusion: +10

PSI/PORT Risk Classification

Head-to-Head Comparison

Sensitivity for 30-day mortality (pooled estimates):

• CURB-65 ≥3: ~75% sensitive, ~78% specific
• PSI Class IV–V: ~85% sensitive, ~75% specific for high-risk identification

For identifying truly low-risk patients (safe for outpatient):

• PSI Class I–II: 30-day mortality <1% — the strongest evidence base for safe discharge
• CURB-65 0–1: 30-day mortality <3% — good, but slightly higher residual risk

Which Tool Should You Use?

Decision Algorithm

Step 1: Apply CURB-65 at the bedside

If score ≥3 → high severity. Admit. Consider ICU if ≥4. No need for PSI.

If score 0–1 and patient appears well → likely safe for outpatient. Confirm with PSI if any doubt.

If score = 2 → intermediate. Apply PSI to refine risk before making disposition decision.

Step 2: Apply PSI to refine low/intermediate risk

If PSI Class I–II → strong evidence for safe outpatient treatment (assuming no social barriers).

If PSI Class III → borderline; consider 24-hour observation or very close outpatient follow-up.

If PSI Class IV–V → admit; ICU workup for Class V.

Always apply IDSA/ATS minor criteria before discharging any patient with score ≥2 or Class III. Three or more minor criteria warrant ICU admission regardless of CURB-65 or PSI.

Scenario-Specific Guidance

Evidence Summary

Based on articles retrieved from PubMed:

Huang L et al. (2024) studied 822 elderly inpatients with CAP and compared PSI, CURB-65, CRB-65, A-DROP, and SMART-COP. PSI had the highest AUC for 30-day mortality (highest sensitivity and negative predictive value), confirming its superiority in older patients. CURB-65 had lower discrimination than PSI in this population but remained clinically practical. SMART-COP best predicted ICU admission and need for mechanical ventilation. Clin Microbiol Infect. 2024;30(11):1426–1432. PMID: 39002660.

Preti C et al. (2022) evaluated CURB-65, PSI, and MuLBSTA in 431 COVID-19 pneumonia patients. All three scores showed moderate discrimination for 28-day mortality (AUC 0.725, 0.776, and 0.743, respectively — no statistically significant differences). Crucially, all scores underestimated severity in low-risk classes, leading the authors to recommend against using them as sole triage tools for COVID-19. Monaldi Arch Chest Dis. 2022;92(4). PMID: 35225441.

Frequently Asked Questions

Q: Can I use CURB-65 and PSI together? Yes — this is often the optimal approach. CURB-65 flags high-risk patients rapidly; PSI refines risk in low-to-intermediate cases where safe discharge is being considered.

Q: What are the IDSA/ATS minor criteria for ICU admission? RR ≥30/min, PaO₂/FiO₂ ≤250, multilobar infiltrates, confusion, BUN ≥20 mg/dL, WBC <4,000/µL, platelets <100,000/µL, core temp <36°C, hypotension requiring aggressive fluids. Three or more = ICU regardless of CURB-65 or PSI score.

Q: Which tool should I use for immunocompromised patients? Neither was validated in immunocompromised populations. For HIV patients, transplant recipients, or those on chronic steroids, clinical judgment and specialist input outweigh any scoring tool.

Q: Is there a better tool for healthcare-associated pneumonia? HCAP as a distinct category has been removed from 2016 IDSA/ATS guidelines. CURB-65 and PSI apply to all CAP, but antibiotic coverage decisions should be based on individual risk factors for resistant organisms.

Q: What's the CRB-65 score used for? CRB-65 omits the urea criterion — it's validated for use when lab results aren't immediately available (primary care, remote settings, resource-limited environments). Score 0 = low risk; 1–2 = intermediate; ≥3 = high.

References

1. Huang L, Weng B, Gu X, et al. Clin Microbiol Infect. 2024;30(11):1426–1432. PMID: 39002660.
2. Preti C, Biza R, Novelli L, et al. Monaldi Arch Chest Dis. 2022;92(4). PMID: 35225441.
3. Fine MJ, Auble TE, Yealy DM, et al. N Engl J Med. 1997;336(4):243–250. PMID: 8995086.
4. Mandell LA, Wunderink RG, Anzueto A, et al. Clin Infect Dis. 2007;44 Suppl 2:S27–72. PMID: 17278083.