DIPSS Calculator for Myelofibrosis

Five risk factors, and anaemia is the only one worth two points — the change that turned the fixed-at-diagnosis IPSS for myelofibrosis into a score that can be applied at any point in the illness.

Last updated 8 Oct 2026

DIPSS (Passamonti 2010)

5 factors → 0–6, anaemia counts double

Age over 65 yearsThe published factor is age older than 65 years. Age does not change during the illness in a way that matters over the time-scales involved, but note that unlike every other item it cannot improve — a patient rescored after their 66th birthday carries this point permanently.

Haemoglobin under 10 g/dLTHE ONLY TWO-POINT ITEM IN THE INSTRUMENT, and the single change that distinguishes DIPSS from the fixed IPSS it was built from. Passamonti and colleagues state the reason in their own discussion: it ‘allows us to assign a greater weight to anemia in the score’, because anaemia acquired during follow-up carried more hazard than the other factors did. 10 g/dL is 100 g/L. A post-transfusion value is not the patient’s haemoglobin.

White cell count above 25 ×10⁹/LLeucocytosis above 25 ×10⁹/L. Note this is the TOTAL white cell count and not a neutrophil count, and that the direction is upwards — leucopenia, which also occurs in advanced myelofibrosis, scores nothing here.

Circulating blasts 1% or morePERIPHERAL BLOOD blasts, from the differential, with the threshold at 1% — low enough that it is easy to dismiss as noise. One blast in a hundred counted cells scores the point. Note that the later MIPSS70 instruments use a 2% threshold for the same variable, so the two are not interchangeable.

Constitutional symptomsWeight loss of more than 10% in the preceding six months, unexplained fever, or drenching night sweats — the same B-symptom construct used in lymphoma staging. This is the item most likely to change between two scorings of the same patient, which is exactly what a dynamic score is for.

4pointsExample

Age 71 (1); haemoglobin 9.2 g/dL (2); white cells 14 ×10⁹/L (0); circulating blasts 0% (0); drenching night sweats (1)

The five factors, the double weight on anaemia, and the four categories

DIPSS = age >65 (1) + haemoglobin <10 g/dL (2) + white cells >25 ×10⁹/L (1) + circulating blasts ≥ 1% (1) + constitutional symptoms (1)
Range 0 to 6

Low 0 · intermediate-1 1–2 · intermediate-2 3–4 · high 5–6

the 2 on anaemia: the one asymmetric weight in the instrument, and the reason DIPSS is a different score from the fixed IPSS for myelofibrosis rather than the same score applied later. An implementation that gives every item one point produces a 0-to-5 total and misplaces every anaemic patient by one band
dynamic means rescorable: the 2009 IPSS for myelofibrosis was derived and validated for use AT DIAGNOSIS only. DIPSS was fitted with the factors as time-dependent covariates, so it can be applied at any point in the illness and its hazard ratios describe the consequence of MOVING into a worse category during follow-up: 4.13 from low to intermediate-1, 4.61 from intermediate-1 to intermediate-2, 2.54 from intermediate-2 to high, and 1.94 per point treated as a continuous time-dependent covariate
blasts at 1%, not 2%: peripheral blood blasts of 1% or more. The threshold is deliberately low. The later MIPSS70 family uses 2% for the same variable, so the two cannot be scored from one reading of the differential
the low-risk group is a single score: zero. Any one factor other than nothing puts a patient in intermediate-1, and because anaemia is worth two points, anaemia alone does it. No median survival can be quoted for the low-risk group because the median was not reached in the derivation cohort
DIPSS-plus, and why it is not computed here: Gangat and colleagues added three factors to the DIPSS CATEGORY — platelets below 100 ×10⁹/L, red cell transfusion need, and an unfavourable karyotype — and reported median survivals of 15.4, 6.5, 2.9 and 1.3 years for its four groups. It is not computed on this site: the mapping from the DIPSS category to the points those three factors are added to could be read only in a secondary summary, and the same summary rendered the IPSS survival figures inverted. The DIPSS category is surfaced above as the number DIPSS-plus consumes, so a reader with the primary paper can carry on from here
the cohort and its era: 525 patients with primary myelofibrosis diagnosed between 1980 and 2008, collected by the International Working Group for Myelofibrosis Research and Treatment across centres in Pavia, Barcelona, Florence, Milan and Rochester, Minnesota. Ruxolitinib was not licensed until 2011, so every survival figure here predates JAK inhibition entirely, as well as momelotinib and current transplant practice
what a median is: a median is the middle of the derivation cohort and not a forecast: half of that cohort reached the figure and half did not, and a figure measured in a cohort treated decades ago is a historical measurement rather than an expectation

Worked example

Age 71 (1); haemoglobin 9.2 g/dL (2); white cells 14 ×10⁹/L (0); circulating blasts 0% (0); drenching night sweats (1)
1 + 2 + 0 + 0 + 1 = 4 points
4 sits in the 3–4 intermediate-2 band, median survival 4 years in the 1980 to 2008 derivation cohort
Note that half this total is the anaemia item. Score haemoglobin as one point — which an implementation that treats every factor equally will do — and the total is 3 and the band is still intermediate-2; but take the same patient with no night sweats and the equal-weight version gives 2 (intermediate-1) against the correct 3 (intermediate-2)
Correct the anaemia with transfusion and the haemoglobin item does not go away: the factor is the patient's haemoglobin, not the post-transfusion figure
Rescore the same patient six months later with resolved sweats and a normal haemoglobin and the total is 1, intermediate-1. That is what DYNAMIC means, and the published hazard ratios are for movement between categories rather than for a category fixed at diagnosis
The DIPSS category for this patient is 2 on the 0-to-3 scale, which is the number DIPSS-plus adds its karyotype, platelet and transfusion points to

The five factors and the one that counts double

Maximum 6. Read straight out of Passamonti 2010, whose discussion states the reason for the asymmetry in its own words. The four one-point items together reach only 4, so the high-risk band cannot be reached without the anaemia point.
Risk factor	Points
Age over 65 years	1
Haemoglobin under 10 g/dL	2
White cell count above 25 ×10⁹/L	1
Circulating blasts 1% or more	1
Constitutional symptoms	1

The four categories and what they predicted, 1980 to 2008

From 525 patients with primary myelofibrosis diagnosed between 1980 and 2008. The hazard ratios are the dynamic part: they describe what happened to patients who MOVED into the category during follow-up, which is a different statement from a category assigned once at diagnosis. Every figure predates ruxolitinib, licensed in 2011. Every outcome figure here describes the cohort the index was derived in and not the patient in front of you; the spread within one stratum is wider than the gap between strata.
Risk category	DIPSS score	Median survival	Hazard ratio for moving up from the category below
Low	0	Not reached	—
Intermediate-1	1 – 2	14.2 years	4.13
Intermediate-2	3 – 4	4 years	4.61
High	5 – 6	1.5 years	2.54

Why anaemia counts twice, and what dynamic actually means

Myelofibrosis has had a prognostic score since 2009, when the International Working Group for Myelofibrosis Research and Treatment published an IPSS built on five features measured at diagnosis: age over 65, haemoglobin under 10 g/dL, white cells above 25 ×10⁹/L, circulating blasts of 1% or more, and constitutional symptoms. That score was derived and validated for use at diagnosis and nowhere else, which is a real limitation in a disease that is often followed for a decade before anything changes.

DIPSS is the same five features refitted with the factors as time-dependent covariates, so it can be applied at any point. Refitting them that way changed one weight: haemoglobin under 10 g/dL became worth TWO points, with everything else still worth one, because anaemia acquired during follow-up carried more hazard than the rest. The paper says so plainly — the double weight exists to give anaemia greater weight in the score — and that single asymmetry is the whole difference between DIPSS and its predecessor. An implementation that gives every item a point produces a maximum of 5 instead of 6 and puts every anaemic patient a band too low.

The other consequence of the double weight is that anaemia alone is enough to leave the low-risk group, which is a single score of zero. From there the gradient is steep: median survival was not reached in the low-risk group, was 14.2 years at 1 to 2 points, 4 years at 3 to 4, and 1.5 years at 5 to 6. The hazard ratios published alongside those medians — 4.13, 4.61 and 2.54 for each step up — are specifically for patients who MOVED into a worse category during follow-up, which is the statement a dynamic score is built to make and the reason rescoring is the point rather than an afterthought.

Two thresholds invite error. The blast threshold is 1%, not 2%: the later MIPSS70 family uses 2% for the same variable, so one reading of a differential cannot score both. And the haemoglobin is the patient’s, not the post-transfusion figure. DIPSS-plus, which adds platelets below 100 ×10⁹/L, transfusion need and an unfavourable karyotype to the DIPSS category, is named here and deliberately not computed, because the mapping from category to points could only be read in a secondary summary that was demonstrably unreliable elsewhere on the same page. Every outcome figure here describes the cohort the index was derived in and not the patient in front of you; the spread within one stratum is wider than the gap between strata. The derivation cohort was diagnosed between 1980 and 2008, so every figure here predates ruxolitinib. Every laboratory threshold here is method- and laboratory-dependent, so the reader’s own laboratory’s reference interval takes precedence. A risk stratum is not a diagnosis and not a plan. This page computes the index, names the stratum and reports what it predicted in the derivation cohort; what follows is a decision for the treating team with the patient.

Frequently asked questions

Why is anaemia worth two points in DIPSS?

Because when the five factors were refitted as time-dependent covariates, anaemia acquired during follow-up carried more hazard than the others. The authors state in their discussion that the extra point exists to give anaemia greater weight. It is the only asymmetric weight in the instrument and the only difference from the 2009 fixed-at-diagnosis IPSS for myelofibrosis.

What does dynamic mean here?

That the score may be applied at any point in the illness and rescored when things change, rather than only at diagnosis. The published hazard ratios — 4.13, 4.61 and 2.54 for each step up — describe what happened to patients who moved into a worse category during follow-up, which is a different quantity from the risk of a category assigned once at the start.

What is the maximum DIPSS score?

Six: 2 for anaemia and 1 each for age over 65, white cells above 25 ×10⁹/L, circulating blasts of 1% or more and constitutional symptoms. Low risk is 0, intermediate-1 is 1 to 2, intermediate-2 is 3 to 4 and high is 5 to 6. Because the four one-point items only reach 4, the high-risk band cannot be reached without the anaemia point.

Does this page compute DIPSS-plus?

No, and that is deliberate. DIPSS-plus adds platelets below 100 ×10⁹/L, red cell transfusion need and an unfavourable karyotype to the DIPSS category, with median survivals of 15.4, 6.5, 2.9 and 1.3 years for its four groups. The mapping from the DIPSS category to the points those factors are added to could only be found in a secondary summary that got other figures on the same page demonstrably wrong, so it is not reproduced. The DIPSS category is shown above as a number for anyone working from the primary paper.

Are the DIPSS survival figures still accurate?

They are accurate as a measurement of the cohort they came from and they are not a forecast. The 525 patients were diagnosed between 1980 and 2008; ruxolitinib was licensed in 2011 and momelotinib much later, and transplant practice has changed. Every outcome figure here describes the cohort the index was derived in and not the patient in front of you; the spread within one stratum is wider than the gap between strata.

Related calculators

References

Passamonti F, et al. A dynamic prognostic model to predict survival in primary myelofibrosis (DIPSS): a study by the International Working Group for Myelofibrosis Research and Treatment. Blood. 2010. 525 patients diagnosed 1980–2008.
Cervantes F, et al. New prognostic scoring system for primary myelofibrosis, from the International Working Group for Myelofibrosis Research and Treatment. Blood. 2009. The fixed-at-diagnosis parent of DIPSS.
Gangat N, et al. DIPSS plus: a refined Dynamic International Prognostic Scoring System for primary myelofibrosis incorporating karyotype, platelet count and transfusion status. J Clin Oncol. 2011. Named here and NOT computed.

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/