PASI 75 and PASI 90 Response Calculator
PASI 75 and PASI 90 Response Calculator
Relative improvement from baseline: (baseline − current) / baseline × 100. An absolute PASI and a PASI response answer different questions, and treat-to-target guidance now names both.
PASI relative improvement from baseline
Baseline and current PASI to % improvementBaseline PASI 21.6, current PASI 3.2
Formula
PASI 75 = improvement ≥ 75% · PASI 90 = ≥ 90% · PASI 100 = complete clearance
- it is a RELATIVE figure
- the denominator is the patient’s own baseline, so the same percentage means different absolute residues at different baselines. A PASI 90 from a baseline of 12 leaves 1.2 points; from a baseline of 40 it leaves 4.0. Both are PASI 90 responders and only one is nearly clear
- why relative at all
- because the absolute PASI is non-linear and inter-observer variation is substantial, so a stated point change is hard to interpret. Expressing the change as a proportion of the patient’s own starting point cancels part of the rater’s systematic bias, which is the argument for it
- the absolute targets
- EuroGuiDerm’s chapter names final-outcome targets of “PASI ≤ 2, DLQI < 2 or PGA clear or almost clear”. BADBIR’s analysis of 13,422 patients and more than 23,000 paired scores derived the thresholds by receiver operating characteristic curve against the relative endpoints: an absolute PASI of 2 or below was concordant with PASI 90 in 90 per cent of assessments, and an absolute PASI of 4 or below with PASI 75 in 88 per cent. Values of 3 and 5 are in circulation as targets; the two documents read for this page give 2 and 4
- refuses at a zero baseline
- with no psoriasis at baseline the quotient is undefined, and an engine that returned 0 or 100 there would be inventing an answer. This page returns nothing and says why
- negative answers are kept
- a current PASI above baseline gives a negative improvement. It is displayed as such
- what it does not fix
- a relative improvement inherits every blind spot of the index underneath it. A palmoplantar psoriasis with a baseline PASI of 1.5 cannot usefully be assessed by PASI response at all, whatever happens to the patient’s hands
Worked example
Baseline PASI 21.6, current PASI 3.2
(21.6 − 3.2) / 21.6 × 100 = 18.4 / 21.6 × 100 = 85.2% improvement
That clears PASI 75 and falls short of PASI 90
PASI 90 from this baseline would need a current score of 21.6 × 0.1 = 2.16, so 3.2 misses it by about one point
PASI 75 needed 21.6 × 0.25 = 5.4, which 3.2 comfortably clears
The residual 3.2 is above the absolute target of PASI 2 or below that EuroGuiDerm names, so this patient is a PASI 75 responder who has not reached the absolute target — the distinction the two endpoints exist to make
Run the same 85.2% from a baseline of 6.0 instead: the current score would be 0.89, comfortably inside the absolute target. The percentage is identical and the clinical state is not
Set the baseline to 0 and the calculator returns nothing, because (0 − 0) / 0 is undefined
What each relative endpoint leaves behind, by baseline
| Baseline PASI | PASI 75 leaves | PASI 90 leaves | Does PASI 90 reach an absolute PASI of 2 or below? |
|---|---|---|---|
| 6 | 1.5 | 0.6 | Yes |
| 12 | 3.0 | 1.2 | Yes |
| 20 | 5.0 | 2.0 | Just — exactly 2.0 |
| 21.6 | 5.4 | 2.2 | No |
| 30 | 7.5 | 3.0 | No |
| 40 | 10.0 | 4.0 | No |
The published targets, relative and absolute, with their source
| Target | What it means | Source as read |
|---|---|---|
| PASI 50 | 50% improvement from baseline | 2011 European consensus: not achieving it was defined as treatment failure |
| PASI 75 | 75% improvement from baseline | 2011 European consensus: “Treatment success was defined as an improvement of 75% or more in PASI” |
| PASI 90 and PASI 100 | 90% and 100% improvement | EuroGuiDerm chapter: “higher treatment goals (e.g. PASI 90 or PASI 100) are aimed for” with the newer biologics |
| Absolute PASI 2 or below | a current score, not a change | EuroGuiDerm names “PASI <= 2, DLQI < 2 or PGA clear or almost clear” as final-outcome targets; BADBIR derived PASI 2 or below as concordant with PASI 90 in 90% |
| Absolute PASI 4 or below | a current score | BADBIR: concordant with PASI 75 in 88% of assessments |
Two endpoints, two questions, and the baseline decides which matters
A PASI response is a proportion, not a score. It asks how much of the psoriasis a patient started with has gone, and the answer depends entirely on what they started with. That is a feature when the question is whether a drug works — it cancels part of each assessor’s systematic bias and gives a number that compares across trials — and a liability when the question is how the patient is now.
The relative endpoints came first. The 2011 European consensus defined treatment success as a PASI 75 and treatment failure as not reaching a PASI 50, with the zone between decided on quality of life. When the interleukin-17 and interleukin-23 inhibitors made near-complete clearance routine, PASI 90 and PASI 100 became the goals the EuroGuiDerm chapter now names. But a relative endpoint has a structural problem at both ends of the baseline range. At a high baseline a PASI 90 can leave extensive disease: from a baseline of 40 it leaves 4.0 points. At a low baseline it becomes arithmetically unstable, because one point of inter-observer variation — against a published inter-rater intraclass correlation whose lower confidence bound is 0.44 — swings the percentage by tens of points.
So absolute targets have partly displaced relative response in treat-to-target guidance. BADBIR, the British Association of Dermatologists Biologics and Immunomodulators Register, analysed 13,422 patients and more than 23,000 paired assessments and found an absolute PASI of 2 or below concordant with PASI 90 in 90 per cent of cases, and an absolute PASI of 4 or below concordant with PASI 75 in 88 per cent. The EuroGuiDerm chapter names an absolute PASI of 2 or below among its final-outcome targets. Both sets of numbers are in current use and they can disagree on the same patient, which is why this page prints the implied residual score for each relative endpoint alongside the percentage.
Two practical points. The baseline must be a genuine pre-treatment baseline: a score recorded after a few weeks of potent topical steroid is already a treated score, and every response measured against it understates the systemic drug. And a relative improvement inherits every blind spot of the PASI underneath it — there is still no term for site, symptoms, nails or joints, so a palmoplantar psoriasis cannot usefully be followed this way at all. No patient-reported outcome measure is computed anywhere on this site. The Dermatology Life Quality Index is owned and licensed by Cardiff University, whose own page states that its measures may not be reproduced in any publication and charges per patient for non-clinical use, so it is named here and never scored. This page computes the published number and names the guideline or trial that defines its thresholds. It renders no clinical decision.
Frequently asked questions
How is PASI 75 calculated?
PASI 75 is not a score but a response: an improvement of 75 per cent or more from the patient’s own baseline PASI. The arithmetic is (baseline − current) / baseline × 100, and 75 or above meets it. From a baseline of 21.6, PASI 75 needs a current score of 5.4 or below.
What is the difference between PASI 90 and an absolute PASI of 2?
PASI 90 is a 90 per cent fall from baseline; absolute PASI 2 is a current score. They agree about 90 per cent of the time — that figure is from BADBIR’s 13,422-patient analysis — and they part company at high baselines, where a PASI 90 responder starting at 40 still carries 4.0 points of psoriasis.
Which target does current guidance use?
Both. The EuroGuiDerm psoriasis chapter records the shift towards PASI 90 and PASI 100 as treatment goals and also names final-outcome targets of “PASI ≤ 2, DLQI < 2 or PGA clear or almost clear”. Absolute targets of PASI 3 and PASI 5 are in circulation elsewhere; the documents read for this page give 2 and 4.
Why does this calculator refuse a baseline of zero?
Because a relative improvement from nothing is undefined — the division has a zero denominator. Returning 0 or 100 per cent there would be inventing an answer, so the page returns nothing and says why.
Can the improvement be negative?
Yes, and it is shown as such. A current PASI above the baseline means the psoriasis has worsened, and the percentage is negative. The result is not floored at zero, because a worsening score is a finding rather than an error.
Related calculators
References
- EuroGuiDerm. Disease severity and treatment goals, psoriasis vulgaris living guideline chapter, September 2023 — the 2011 European consensus definition and the PASI 50/75 treatment goal algorithm. guidelines.edf.one
- Mahil SK, Wilson N, Dand N, et al. Psoriasis treat to target: defining outcomes in psoriasis using data from a real-world, population-based cohort study (the British Association of Dermatologists Biologics and Immunomodulators Register, BADBIR). Br J Dermatol. 2020;182(5):1158–66.
- Fredriksson T, Pettersson U. Severe psoriasis — oral therapy with a new retinoid. Dermatologica. 1978;157(4):238–44. doi:10.1159/000250839. The derivation publication; the index was that trial’s severity measure.
- DermNet. PASI score — the four regions, the 0–6 area categories and the three signs on 0–4. dermnetnz.org/topics/pasi-score
- Inter-rater concordance study of the PASI (Psoriasis Area and Severity Index). An Bras Dermatol. 2010 — 20 patients scored independently by three first-year dermatology postgraduates; intraclass correlation 0.729 (95% CI 0.440–0.882).
- Cardiff University. Quality of life questionnaires — the DLQI’s ownership, the prohibition on reproducing the measures, and the per-patient fee schedule for non-clinical use. cardiff.ac.uk/medicine/resources/quality-of-life-questionnaires
Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/