ASDAS Improvement and Worsening Calculator
ASDAS Improvement and Worsening Calculator
The two ASAS-endorsed change thresholds on one pair of ASDAS values: 1.1 for clinically important improvement, 2.0 for major improvement, and 0.9 in the other direction for clinically important worsening.
ASDAS change
2 ASDAS values → change categoryBaseline ASDAS-CRP 3.80, current ASDAS-CRP 2.10
A difference, read against three published thresholds
a fall of ≥ 1.1 is clinically important improvement · a fall of ≥ 2.0 is major improvement · a rise of ≥ 0.9 is clinically important worsening
- 1.1 and 2.0
- the ASAS improvement thresholds. Derived in the NOR-DMARD registry and cross-validated there and in the ASSERT trial database, then chosen by ASAS members voting after ROC analyses rather than taken off a curve
- 0.9
- the ASAS worsening threshold, added in 2018 from 1,169 patients. Sensitivity 0.38, specificity 0.96 against the patients’ own report of worsening. The alternatives tested were 0.6 (0.55 and 0.91) and 1.1 (0.33 and 0.98)
- there is no major worsening
- ASAS published one worsening threshold and not two. A rise of 3 points has the same published name as a rise of 0.9
- absolute, not relative
- the criteria are stated in ASDAS points. The relative improvement is printed beside the result because it is informative, but no published percentage threshold exists for the ASDAS and the percentage is not what the criteria test
- both values, one version
- an ASDAS-CRP baseline against an ASDAS-ESR follow-up measures the difference between two formulas as well as the change in the patient. On this site’s own example patient the two versions differ by 0.80, which is most of the way to the 1.1 threshold on its own
Worked example
Baseline ASDAS-CRP 3.80, current ASDAS-CRP 2.10
3.80 − 2.10 = 1.70 points of improvement
1.70 is 1.1 or more, so this meets clinically important improvement
1.70 is below 2.0, so it does not meet major improvement
Relative improvement: 1.70 / 3.80 = 44.7% — printed because it is informative, not because any published threshold uses it
The current value of 2.10 is separately in the high-disease-activity state, because 2.1 is the lower bound of that band. The change category and the state are two different statements about the same patient
Room for a 2.0-point fall: the baseline of 3.80 is 1.80 above 2.0, so major improvement was arithmetically reachable here. From a baseline of 1.80 it would not have been
Two ends, one answer: 1.5 → 0.4 and 5.0 → 3.9 are both a fall of exactly 1.1 and both clinically important improvement, although the first ends in inactive disease and the second in very high disease activity
The three thresholds and where each came from
| Change | Published name | Source and cohort | Operating characteristics |
|---|---|---|---|
| fall ≥ 2.0 | Major improvement | Machado 2011; NOR-DMARD, cross-validated in NOR-DMARD and ASSERT | ASAS consensus vote after ROC analyses |
| fall ≥ 1.1 | Clinically important improvement | Machado 2011, same work | ASAS consensus vote after ROC analyses |
| between −0.9 and +1.1 | — no published name | neither paper defines this interval | — |
| rise ≥ 0.9 | Clinically important worsening | Molto 2018; 1,169 axSpA patients in an international longitudinal study | sensitivity 0.38, specificity 0.96 |
The same change from four starting points
| Baseline | Current | Change | Change category | State of the current value |
|---|---|---|---|---|
| 1.5 | 0.4 | 1.10 | Clinically important improvement | Inactive disease |
| 5.0 | 3.9 | 1.10 | Clinically important improvement | Very high |
| 3.0 | 1.0 | 2.00 | Major improvement | Inactive |
| 1.8 | 0.0 | 1.80 | Clinically important improvement only | Inactive |
An absolute change, two improvement thresholds, and a worsening threshold that misses most worsening
The ASDAS arrived in 2009 with coefficients and no cut-offs. Machado and colleagues supplied both sets two years later from the NOR-DMARD registry, cross-validated in NOR-DMARD and the ASSERT trial database: three state boundaries at 1.3, 2.1 and 3.5, and two change thresholds, a fall of 1.1 or more for clinically important improvement and 2.0 or more for major improvement. The method is worth knowing, because it was not a statistic alone — ROC analyses against several external criteria, then debate and a vote among ASAS members. A reader looking for the Youden point that produced 1.1 will not find one.
The worsening side took until 2018 and behaves differently. Molto and colleagues studied 1,169 patients with axial spondyloarthritis in an international longitudinal study of stable patients; 127 of them, 10.9%, reported worsening at the second visit. Three candidate thresholds were tested against that self-report and ASAS endorsed a rise of 0.9, which carries a specificity of 0.96 and a sensitivity of 0.38. Printed plainly: the criterion almost never calls worsening in a stable patient, and it misses nearly two in three patients who say they are worse. That is a deliberate choice for a trial endpoint and it is the wrong instrument for reassurance. There is no published major worsening category; a rise of 3 points has the same name as a rise of 0.9.
The thresholds are absolute, in ASDAS points, and that gives the index a character worth understanding. A fall from 1.5 to 0.4 and a fall from 5.0 to 3.9 are both exactly 1.1 and both meet clinically important improvement, although one patient has arrived in inactive disease and the other is still in the very-high band. Conversely a patient whose baseline is below 2.0 cannot reach major improvement however well they do, because the index has a floor at zero — which is why trials restrict entry by baseline score, and why both the current value and the headroom above 2.0 are printed beside the change here. The relative improvement is printed too, because it is informative, but no published ASDAS threshold is a percentage.
One trap is specific to this calculation. Both values must come from the SAME version of the index. ASDAS-CRP and ASDAS-ESR use different coefficients and a different inflammatory transform, and on the ASDAS-CRP and ASDAS-ESR page’s own example patient they differ by 0.80 — most of the way to the 1.1 threshold before the patient has changed at all. Pick one version per patient and keep to it. The two formulas are not interchangeable. The 2022 ASAS-EULAR recommendations name an improvement in ASDAS of 1.1 or more as their continuation criterion and ASDAS of 2.1 or more as their high-activity criterion, and also record that in TICOSPA a treat-to-target strategy aiming at an ASDAS below 2.1 was not significantly superior to usual care. The rheumatoid equivalents of this page are the EULAR DAS28 response criteria, which needs both a change and a current value, and ACR20, ACR50 and ACR70, which is a conjunction over seven measures rather than a difference. This states what the instrument measured in a named cohort; the clinician in front of the patient decides what follows.
Frequently asked questions
What counts as worsening?
A rise of 0.9 points or more is the ASAS-endorsed definition of clinically important worsening, from 1,169 patients studied by Molto and colleagues in 2018. Against the patients’ own report of worsening it has a sensitivity of 0.38 and a specificity of 0.96, so it rarely misclassifies a stable patient and misses most patients who describe deterioration. No major worsening threshold has been published.
Can I compare an ASDAS-CRP baseline with an ASDAS-ESR follow-up?
No. The two versions use different coefficients and a different inflammatory transform, and the published position is that the formulae are not interchangeable. On a single worked patient they differ by 0.80 points, which is most of the 1.1 improvement threshold on its own. Use the same version at both visits.
Why can a patient with a low baseline not achieve major improvement?
Because the threshold is an absolute fall of 2.0 points and the ASDAS does not go below zero. A patient starting at 1.8 can at most fall by 1.8. This is why trials restrict entry by baseline score, and why the headroom above 2.0 is printed beside the result here.
Is there anything published between a fall of 1.1 and a rise of 0.9?
No, and the gap is deliberate rather than an oversight: the improvement thresholds were set in 2011 and the worsening threshold in 2018, and neither names the interval between them. A change of 0.8 is a change that falls below both validated cut-offs, which is not the same statement as stable disease.
Related calculators
References
- Machado P, Landewé R, Lie E, et al. Ankylosing Spondylitis Disease Activity Score (ASDAS): defining cut-off values for disease activity states and improvement scores. Ann Rheum Dis. 2011;70(1):47–53.
- Molto A, Gossec L, Meghnathi B, et al. An Assessment in SpondyloArthritis International Society (ASAS)-endorsed definition of clinically important worsening in axial spondyloarthritis based on ASDAS. Ann Rheum Dis. 2018;77(1):124–7.
- Machado PM, Landewé R, van der Heijde D; Assessment of SpondyloArthritis international Society (ASAS). Ankylosing Spondylitis Disease Activity Score (ASDAS): 2018 update of the nomenclature for disease activity states. Ann Rheum Dis. 2018;77(10):1539–40.
- Lukas C, Landewé R, Sieper J, et al. Development of an ASAS-endorsed disease activity score (ASDAS) in patients with ankylosing spondylitis. Ann Rheum Dis. 2009;68(1):18–24.
- Ramiro S, Nikiphorou E, Sepriano A, et al. ASAS-EULAR recommendations for the management of axial spondyloarthritis: 2022 update. Ann Rheum Dis. 2023;82(1):19–34.
- Assessment of SpondyloArthritis international Society. ASDAS calculator. asas-group.org. (ASAS’s own instrument page: both formulas in rounded form as text, the state and improvement cut-offs, the instruction that a CRP below the limit of detection or below 2 mg/L is replaced by “the fixed value of 2 mg/l (0.2 mg/dl)”, and the note that since 2024 the CRP-based index is called simply ASDAS.)
- Machado P, Navarro-Compán V, Landewé R, van Gaalen FA, Roux C, van der Heijde D. Calculating the ankylosing spondylitis disease activity score if the conventional C-reactive protein level is below the limit of detection or if high-sensitivity C-reactive protein is used: an analysis in the DESIR cohort. Arthritis Rheumatol. 2015. doi:10.1002/art.38921.
- AScalate: Treat-to-target in Axial Spondyloarthritis — statistical analysis plan, NCT03906136. ClinicalTrials.gov document archive. (The only document read that states both the ASDAS-CRP and the ASDAS-ESR coefficients to three decimal places.)
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/