About the Crohn's Disease Activity Index (CDAI)
The Crohn's Disease Activity Index is the reference standard for measuring Crohn's disease activity, particularly in clinical trials. It combines eight weighted items — three of them totalled from a seven-day symptom diary (liquid stools, abdominal pain, general wellbeing) with complications, antidiarrhoeal use, an abdominal mass, the haematocrit deficit, and the deficit from standard body weight — into a single score. Below 150 is remission, 150–220 mild, 221–450 moderate, and above 450 severe disease. The commonest scoring error is entering a single day's value where a seven-day total is required, which understates the first three items roughly sevenfold.
Formula
CDAI = 2×(liquid stools, 7-day) + 5×(abdominal pain sum) + 7×(wellbeing sum) + 20×(complications) + 30×(antidiarrhoeal) + 10×(abdominal mass) + 6×(haematocrit deficit) + 1×(% below standard weight)- liquid stools
- Seven-day total count, ×2.
- abdominal pain sum
- Sum of daily 0–3 ratings over 7 days, ×5.
- wellbeing sum
- Sum of daily 0–4 ratings over 7 days, ×7.
- complications
- Count of the six listed categories, ×20.
- antidiarrhoeal
- 1 if used in the past week, ×30.
- abdominal mass
- 0 none, 2 questionable, 5 definite, ×10.
- haematocrit deficit
- Reference (47% men / 42% women) minus measured haematocrit, ×6.
- % below standard weight
- 100 × (1 − actual/standard), floored at 0, ×1.
- The first three items are seven-day diary totals, not single-day values — entering a single day understates them roughly sevenfold and is the classic error.
- The haematocrit term can be negative when the value exceeds the sex-specific reference, exactly as in the original index.
- Being at or above standard body weight contributes nothing; only a deficit adds to the score.
Interpreting the result
A score below 150 is remission and is the threshold trials use to define it. The 150–220 band is mild disease and sits below most trial entry criteria, which typically start at 220. The 221–450 range is moderate disease — the population most induction trials recruit — and above 450 is severe. Beyond these static bands, the CDAI is most often used dynamically: trials define clinical response as a fall of at least 70 points (CDAI-70) or 100 points (CDAI-100) from baseline. Because much of the score is symptom-derived, a raised CDAI should be confirmed to reflect inflammation — with calprotectin, CRP or imaging — before treatment is escalated, since strictures and bile-acid diarrhoea can drive the diary items without active inflammation.
| Score | Band | What it means | Action |
|---|---|---|---|
| < 150 | Remission | Clinical remission — the trial threshold for remission | Continue maintenance therapy; monitor objectively as symptoms and inflammation diverge |
| 150–220 | Mild | Mildly active disease, below most trial entry criteria | Confirm inflammation with calprotectin or CRP; optimise therapy |
| 221–450 | Moderate | Moderately active disease — the usual induction-trial population | Confirm inflammation and exclude abscess before escalating immunosuppression |
| > 450 | Severe | Severely active disease | Urgent objective workup; exclude complications; escalate without delay |
What the CDAI needs (8 inputs)
- Liquid or soft stools (7-day total)
- The week's total number of liquid or soft stools, weighted ×2.
- Abdominal pain (7-day sum)
- Daily 0–3 rating (none to severe) summed over the week, weighted ×5.
- General wellbeing (7-day sum)
- Daily 0–4 rating (generally well to terrible) summed over the week, weighted ×7.
- Complications
- One point each for arthritis/arthralgia, iritis/uveitis, erythema nodosum/pyoderma gangrenosum/aphthous stomatitis, anal fissure/fistula/abscess, other fistula, and fever above 37.8 °C in the past week; weighted ×20.
- Antidiarrhoeal drugs
- Taken in the past week — yes contributes, weighted ×30.
- Abdominal mass
- None (0), questionable (2), or definite (5), then weighted ×10.
- Haematocrit
- The deficit below the sex-specific reference (47% men, 42% women) is weighted ×6; a value above the reference contributes a negative term.
- Body weight
- The percentage below standard body weight is weighted ×1; patients at or above standard weight contribute nothing.
What it returns
- CDAI total
- The weighted sum, typically ranging from about 0 to 600.
- Activity band
- Remission, mild, moderate, or severe.
How it is calculated
The National Cooperative Crohn's Disease Study collected 18 candidate variables prospectively across many patient visits and, at each, recorded the attending physician's overall rating of how the patient was doing. Multiple regression selected the eight variables that together best predicted that global rating, and their regression coefficients became the weights — which is why the multipliers (×30 for antidiarrhoeals, ×7 for wellbeing, ×1 for weight) look arbitrary but are not: each reflects how much that variable moved the physician's assessment. The seven-day diary items exist because single-visit snapshots of stool frequency and symptoms were too noisy; averaging a week of them stabilised the estimate at the cost of the diary burden the index is now known for.
Facts & figures
| Item | Weight | Source |
|---|---|---|
| Liquid/soft stools (7-day total) | ×2 | Diary |
| Abdominal pain (7-day sum, 0–3/day) | ×5 | Diary |
| General wellbeing (7-day sum, 0–4/day) | ×7 | Diary |
| Complications (0–6) | ×20 | Clinical |
| Antidiarrhoeal use | ×30 | Clinical |
| Abdominal mass (0/2/5) | ×10 | Clinical |
| Haematocrit deficit | ×6 | Laboratory |
| % below standard weight | ×1 | Anthropometry |
Weights are the regression coefficients that best predicted a physician's global assessment in the derivation cohort — not clinical rankings of importance.
| Score / change | Meaning |
|---|---|
| < 150 | Remission |
| 220–450 | Typical trial entry (moderate disease) |
| > 450 | Severe disease |
| Fall of ≥ 70 points | Clinical response (CDAI-70) |
| Fall of ≥ 100 points | Clinical response (CDAI-100) |
Evidence
Derivation — Best (NCCDS)
1976 · n = 112The National Cooperative Crohn's Disease Study collected 18 predictor variables prospectively across 187 visits of 112 patients with Crohn's disease of the small bowel, colon or both. At each visit the attending physician rated overall how well the patient was doing, and multiple regression derived an equation predicting that rating from eight selected variables.
The resulting index set values of 150 and below as quiescent disease, values above 150 as active, and values above 450 as extremely severe — the thresholds still in use. The eight-variable equation reproduced the physician's global assessment closely enough to serve as a standardised, comparable measure of activity.
Rederived coefficients — Best
1979A re-analysis of the National Cooperative Crohn's Disease Study data to recompute the eight regression coefficients underlying the index.
Confirmed and refined the original weights, consolidating the CDAI as the standard activity measure for Crohn's disease trials rather than replacing it.
Endpoints review — Sandborn
2002A structured review of the activity indices and efficacy endpoints used in clinical trials of medical therapy for adult Crohn's disease, examining how the CDAI is applied to define remission and response.
Codified the endpoints now standard across drug trials: remission as a CDAI below 150 and clinical response as a fall of at least 70 or 100 points from baseline (CDAI-70 and CDAI-100), while noting the index's known weaknesses in perianal and fistulising disease.
How it compares
CDAI vs Harvey-Bradshaw Index (HBI)
Use the CDAI when a trial-grade, diary-based measure is required, and the Harvey-Bradshaw Index for routine clinical care — the two correlate well but are not interchangeable for an individual patient.
The CDAI needs a seven-day symptom diary, a haematocrit and a weight, and combines eight individually weighted items; the Harvey-Bradshaw Index is five equally weighted items assessed for the previous day, with no diary or bloods. They correlate at about 0.93, but modelling one from the other found roughly a 27-point CDAI change per HBI point with wide prediction limits — 'good but far from perfect' — so HBI cannot substitute for an actual CDAI value in an individual. Remission thresholds mirror the relationship: CDAI below 150, HBI below 5.
CDAI vs Endoscopic activity (SES-CD) and faecal calprotectin
The CDAI measures how the patient feels; endoscopic and biomarker measures show whether the bowel is inflamed, and the two agree less often than is comfortable.
Clinical remission by CDAI and mucosal healing are different endpoints, and treating the first as evidence of the second is the commonest misuse of this score. A patient can be below 150 with ulceration on ileocolonoscopy, and can be above 220 from a stricture or bile-acid diarrhoea with no active inflammation at all. Current practice is treat-to-target: symptomatic response measured by an index such as the CDAI, confirmed against an objective marker — faecal calprotectin, CRP, or endoscopic scoring with SES-CD. Where the two diverge, the objective measure should drive the treatment decision.
Pearls & pitfalls
- The first three items are seven-day diary totals. Entering a single day's figure understates them roughly sevenfold and is by far the commonest scoring error.
- The weights are regression coefficients, not importance rankings — antidiarrhoeal use carries ×30 because that is what best fit the physician's rating in 1976, not because it is clinically the heaviest feature.
- Much of the score is symptomatic, so it rises with strictures, bile-acid diarrhoea or functional overlay that do not respond to immunosuppression — confirm inflammation objectively before escalating.
- It performs poorly in isolated perianal or fistulising disease; use a fistula-specific measure there rather than forcing the CDAI.
- The haematocrit term can legitimately be negative, and being above standard weight contributes nothing — both are features of the original equation, not errors to correct.
Critical actions
- Confirm a raised CDAI reflects active inflammation — check calprotectin or CRP and consider imaging — before escalating therapy.
- Exclude an abscess before starting or increasing immunosuppression, particularly with a high score driven by the complications item or an abdominal mass.
- Remember that strictures and bile-acid diarrhoea inflate the index without active inflammation and do not respond to immunosuppression.
- Use the CDAI itself, not a converted Harvey-Bradshaw value, wherever a protocol specifies CDAI eligibility or endpoints.
- Track change from baseline (CDAI-70 / CDAI-100) rather than a single absolute value when assessing response to a treatment change.
Why this score exists
The index was a deliberate act of standardisation: before it, trials could not be compared because each defined 'active disease' differently. Best and the NCCDS group chose to anchor the score to the one benchmark they had — an experienced physician's overall judgement — and let regression decide which variables and weights reproduced it, rather than assembling items by clinical intuition. That is why the weights are statistical artefacts of the derivation cohort rather than a ranking of clinical importance, and why the diary items, awkward as they are, were kept: they were what stabilised the estimate. The design succeeded at its actual goal, comparability across trials, which it still delivers; its impracticality in clinic is the price of that origin, and the reason the Harvey-Bradshaw Index was later built to strip it down.
About the creator
First author, 1976 derivation study
Derived the CDAI for the National Cooperative Crohn's Disease Study, and later published the regression relating CDAI to the Harvey-Bradshaw Index.
Senior author
Co-investigator on the National Cooperative Crohn's Disease Study that produced the index.
Limitations
- The seven-day diary requirement and the need for a haematocrit and body weight make it impractical for routine clinic use, which is why simpler indices replaced it there.
- It is heavily symptom-weighted and correlates only modestly with endoscopic or mucosal inflammation, so a low CDAI does not confirm mucosal healing.
- It performs poorly in isolated perianal and fistulising Crohn's disease, for which it was never well suited.
- The weights are fixed artefacts of a 1970s derivation cohort assessed against physician judgement, predating biologic therapy and treat-to-target strategies.
- Placebo response rates on the CDAI in trials are high, partly because so much of the score is subjective — a recognised limitation of using it as a trial endpoint.
If you are the patient
The CDAI is a detailed score doctors use — mostly in research studies — to measure how active someone's Crohn's disease is. It adds together several things, including a diary you keep for a week of how many loose stools you have, how much tummy pain you get, and how well you feel each day, along with whether you have certain complications, whether you take anti-diarrhoea medicines, whether a mass can be felt in your abdomen, a blood count, and your weight. A score under 150 means the disease is settled (remission), and higher scores mean more active disease. Because it needs a week-long diary and blood tests, in ordinary clinic visits doctors often use a shorter score (the Harvey-Bradshaw Index) instead. Whichever is used, your team will usually also check a stool or blood test, because how you feel does not always match how much inflammation is present.
Frequently asked questions
What is a normal or remission CDAI score?#
Below 150 is remission. Scores of 150–220 are mild, 221–450 moderate, and above 450 severe. Clinical trials typically enrol patients in the 220–450 range.
What does CDAI-70 or CDAI-100 mean?#
They are response endpoints used in trials: a fall of at least 70 points (CDAI-70) or at least 100 points (CDAI-100) from the baseline score, indicating clinical response to treatment.
Why are the first three items seven-day totals?#
Because single-visit snapshots of stool frequency, pain and wellbeing were too variable. Summing a week of diary entries stabilised the estimate — at the cost of the diary burden. Entering one day's value instead of the weekly total is the classic scoring error.
How does the CDAI compare with the Harvey-Bradshaw Index?#
They correlate well (about r = 0.93) but are not interchangeable — roughly 27 CDAI points per HBI point, with wide scatter. The CDAI is the trial standard; the HBI is the practical clinic tool because it needs no diary or bloods.
Is a low CDAI proof of mucosal healing?#
No. The CDAI is largely symptom-based and correlates only modestly with endoscopic activity, so objective measures such as faecal calprotectin, CRP or endoscopy are needed to confirm true healing.
How long does it take to complete a CDAI?#
It cannot be completed at a single visit. Three of its items — liquid stool count, abdominal pain and general wellbeing — are seven-day totals recorded prospectively in a diary, so the score always describes the week before the appointment rather than the day of it. That requirement is the main reason the Harvey-Bradshaw Index exists: it substitutes single-day equivalents and can be scored in the room.
Why is the CDAI criticised as a trial endpoint?#
Because most of its weight sits on subjective symptoms. The diary items dominate the total, which makes the score sensitive to placebo response and to symptoms that are not driven by inflammation at all — bile-acid diarrhoea after ileal resection, a fibrotic stricture, or coexisting functional symptoms all raise it. Regulators and trialists have moved toward co-primary endpoints pairing patient-reported outcomes with endoscopic healing for exactly this reason.
Does the CDAI include any laboratory or objective measures?#
Only two, and both are weak. Haematocrit contributes via its deviation from a sex-specific reference, and body weight via its deviation from standard weight. There is no CRP, no faecal calprotectin and no endoscopic component, so a patient can sit in the remission range with active mucosal inflammation. Pair the score with an objective marker rather than treating it as a measure of disease activity on its own.
References
Original / primary reference
Validation and endpoints
- Best WR, Becktel JM, Singleton JW. Rederived values of the eight coefficients of the Crohn's Disease Activity Index (CDAI). Gastroenterology. 1979;77(4 Pt 2):843-846.
- Sandborn WJ, Feagan BG, Hanauer SB, et al. A review of activity indices and efficacy endpoints for clinical trials of medical therapy in adults with Crohn's disease. Gastroenterology. 2002;122(2):512-530.
- Best WR. Predicting the Crohn's disease activity index from the Harvey-Bradshaw Index. Inflamm Bowel Dis. 2006;12(4):304-310.