How this instrument works
The Pneumonia Severity Index, also called the PORT score after the Pneumonia Patient Outcomes Research Team that derived it, estimates 30-day mortality risk in adults with community-acquired pneumonia and sorts them into five risk classes. It was built to help decide who can be treated safely as an outpatient and who needs hospital admission, since over-admitting low-risk patients and under-admitting high-risk ones are both real costs of getting that call wrong.
The score works in two steps. Step 1 is a gate: a patient 50 or younger, with none of five listed comorbidities (neoplastic disease, liver disease, congestive heart failure, cerebrovascular disease, renal disease) and none of five vital-sign abnormalities (altered mental status, respiratory rate ≥30/min, systolic blood pressure <90 mmHg, temperature <35°C or ≥40°C, pulse ≥125/min), is automatically Class I — the lowest-risk group — without ever reaching the lab values or the point tally. Step 2 only runs for everyone else: it sums age in years (with 10 points subtracted for women), plus weighted points for nursing-home residence, each comorbidity, each vital-sign abnormality, and several lab and imaging findings — arterial pH, BUN, sodium, glucose, hematocrit, PaO2, and pleural effusion on chest x-ray. The resulting point total sorts the patient into Class II (≤70 points), III (71-90), IV (91-130), or V (>130).
Each class corresponds to a mortality-risk band from the original derivation and validation cohorts, rising from well under 1% in Class I to roughly a quarter of patients in Class V. Because it was built and tested on adults specifically being evaluated for community-acquired pneumonia, it isn't intended as a general-purpose severity check for other lung infections, and it doesn't replace clinical judgment about things it can't measure directly, like social support for safe outpatient care or the trajectory of a patient's condition over the preceding hours.
- Enter age in years, and set sex, nursing home residence, and history of neoplastic disease, liver disease, congestive heart failure, cerebrovascular disease, and renal disease.
- Set the five vital-sign fields — altered mental status, respiratory rate ≥30/min, systolic blood pressure <90 mmHg, temperature <35°C or ≥40°C, and pulse ≥125/min — based on the patient's presentation.
- Set the lab and imaging fields — arterial pH <7.35, BUN ≥30 mg/dL, sodium <130 mEq/L, glucose ≥250 mg/dL, hematocrit <30%, PaO2 <60 mmHg (or O2 sat <90%), and pleural effusion on chest x-ray.
- Read the comorbidity and vital-sign flag counts, whether the Class I gate is met, the Step-2 point total, and the resulting PORT risk class (1 through 5).
Worked example — 80-year-old nursing-home resident with multiple abnormalities
An 80-year-old man, a nursing-home resident with active neoplastic disease, presents with altered mental status, a respiratory rate ≥30/min, systolic blood pressure under 90 mmHg, arterial pH under 7.35, BUN ≥30 mg/dL, and PaO2 under 60 mmHg. He fails the Class I gate immediately on age alone (80 > 50), so Step 2 applies: 80 (age) + 10 (nursing home) + 30 (neoplastic disease) + 20 (altered mental status) + 20 (respiratory rate ≥30) + 20 (SBP <90) + 30 (pH <7.35) + 20 (BUN ≥30) + 10 (PaO2 <60) = 240 points.
That total is well above the Class IV ceiling of 130, so this patient lands in Class V — the highest-risk PORT class. By contrast, a 45-year-old with no listed comorbidities and normal vital signs meets the Class I gate outright — age ≤50, zero comorbidity flags, zero vital-sign flags — and is classified Class I without Step 2 ever running, regardless of what any labs might show.
Questions
What is the Pneumonia Severity Index used for?
It estimates 30-day mortality risk for an adult with community-acquired pneumonia and helps guide the decision between outpatient treatment, general hospital admission, and closer monitoring. It was developed and validated by the Pneumonia Patient Outcomes Research Team and published by Fine and colleagues in the New England Journal of Medicine in 1997. It is meant to be used by clinicians alongside their own judgment and the patient's full clinical picture, not as a standalone or self-diagnosis tool.
Why does the calculator check a Class I gate before tallying any points?
Because the original derivation found that patients who are 50 or younger, with none of five specific comorbidities and none of five vital-sign abnormalities, have a low enough mortality risk to be classified Class I without needing lab values at all. This gate can classify someone Class I even if their labs are never drawn — the point-based Step 2 only applies to patients who don't meet all three gate conditions.
Why does the score subtract 10 points for women?
The original PSI derivation cohort found sex-based differences in mortality risk at a given age, and built a 10-point subtraction for women directly into the age component of the score. It's part of the validated point structure rather than an adjustment added later.
What is the approximate mortality risk for each PORT class?
In the original derivation and validation studies, reported 30-day mortality rose sharply across the five classes — roughly a few tenths of a percent in Class I, under 1% in Class II, up to around 3% in Class III, roughly 4-10% in Class IV, and around 27% in Class V. These figures come from the derivation cohort and later validation studies; exact percentages vary somewhat between published cohorts, so treat them as an approximate risk gradient rather than a precise per-patient probability.
Does a low PORT class mean it's safe to treat someone at home?
A low class (I or II) is one input supporting outpatient management, not a decision by itself. Clinicians also weigh factors the score doesn't capture directly — ability to take oral medication and fluids, home support, substance use, and clinical judgment about trajectory — before deciding on site of care. The original authors and subsequent guidelines both frame PSI as decision support, not a replacement for that judgment.
Is the Pneumonia Severity Index the same as CURB-65?
No — they're two different, commonly used pneumonia severity tools. CURB-65 is a shorter, five-variable score often used for a quick bedside estimate, while PSI/PORT uses more variables (including labs and comorbidities) and was validated with a specific focus on identifying very-low-risk patients suitable for outpatient care. Different studies favor one or the other for different purposes; neither is simply a shorthand version of the other.
Who derived the PSI and where was it published?
Michael J. Fine and colleagues on the Pneumonia Patient Outcomes Research Team derived and validated it using a large cohort of patients with community-acquired pneumonia, publishing the results in the New England Journal of Medicine in 1997 (336(4):243-250). It has since been incorporated into pneumonia management guidelines and clinical calculators including MDCalc.
References
- Fine 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(4):243-250 (PMID 8995086)
- MDCalc — Pneumonia Severity Index (PSI/PORT Score)
Read this first: This instrument computes a screening figure from population formulas — it is not a diagnosis, and it cannot see the whole picture a clinician can. Use it to inform a conversation, not to replace one.