Infection & Sepsis Risk score Strong evidence Clinical review pending v1.0.0

Pneumonia Severity Index (PSI/PORT)

What is the 30-day mortality risk in community-acquired pneumonia, and where should it be treated?

Add to favourites

Calculator

Fields marked * are required. Units can be changed per field.

Demographics

years
Sex Women score age − 10.

Coexisting illness

Physical examination

Laboratory and radiographic findings

Reset

Result

Your result will appear here

Complete the inputs and select Calculate. Results update instantly for point scores.

About this tool

Twenty-variable prediction rule that assigns adults with community-acquired pneumonia to five risk classes for 30-day mortality, supporting site-of-care decisions.

Intended use

Adults with community-acquired pneumonia, to identify low-risk patients suitable for outpatient care.

Outcome

30-day mortality

Time horizon

30 days

Derivation population

14,199 adult inpatients (derivation); validated in 38,039 inpatients and 2,287 PORT patients.

Do not use in

Derived in community-acquired pneumonia; not intended for hospital-acquired pneumonia.

Limitations

  • Age contributes heavily, so younger patients with severe physiology can be under-classified.
  • Requires laboratory and arterial blood gas data for full scoring.
  • Does not capture social factors, hypoxaemia severity beyond PaO₂ < 60 or inability to take oral therapy.

Clinical problems

Evidence

Evidence level

Strong evidence

Derived and externally validated in multiple cohorts (including prospective studies) and incorporated into major clinical guidelines. Methodology

Guideline status

Preferred over CURB-65 for site-of-care decisions in the 2019 ATS/IDSA CAP guideline.

Validation studies

Derivation and validation (Fine 1997)
54,525 patients across three cohortsProspectiveExternal

MedisGroups inpatient cohorts and Pneumonia PORT inpatients and outpatients

No significant difference in mortality by class across three cohorts: class I 0.1–0.4%, class II 0.6–0.7%, class III 0.9–2.8%.

Evidence summaries for this tool are awaiting independent clinical review. Statistics are shown only where they could be attributed to a specific publication.

References

  1. Original publicationFine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997;336:243-250. Find on PubMed (opens in a new tab)
  2. GuidelineMetlay 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:e45-e67. Find on PubMed (opens in a new tab)

Version history

VersionFormulaEffectiveReviewerChanges
v1.0.0Fine 1997 point allocation and class rules2026-10-03—Initial implementation with reference test vectors.

Related tools

Clinical decision support only. Interpret results in the context of the individual patient and current guidance; HumanPulse does not replace professional medical judgment.