About the Harvey-Bradshaw Index for Crohn's Disease Activity
Clinical remission in Crohn's disease is a Harvey-Bradshaw Index below 5, with 5–7 counting as mild disease, 8–16 moderate, and above 16 severe. Five items make up the total, all assessed for the previous day only and none requiring a blood test: general well-being, abdominal pain, the number of liquid stools, an abdominal mass, and any of eight listed complications. It exists as a same-day, bedside simplification of the Crohn's Disease Activity Index, and while the two track each other closely (reported r = 0.93), one HBI point corresponds to roughly 27 CDAI points with wide enough scatter that converting between them for an individual patient is unreliable.
Formula
HBI = general well-being + abdominal pain + liquid stools (count) + abdominal mass + complications (count)- general well-being
- 0–4, patient-rated.
- abdominal pain
- 0–3.
- liquid stools
- Raw count for the previous day, one point each, uncapped.
- abdominal mass
- 0–3, clinician-assessed.
- complications
- Raw count of the eight listed complications, one point each, uncapped.
- Unlike the CDAI, none of the five HBI items carry a weighting coefficient — every point counts equally, which is the main simplification the index makes.
- All items are assessed for yesterday only, not averaged over a week, which removes the CDAI's diary requirement.
- There is no published maximum score, because two of the five items are open-ended counts rather than capped categories.
Interpreting the result
A score below 5 is clinical remission and supports continuing current maintenance therapy. Rising scores through the mild, moderate and severe bands should prompt objective confirmation — calprotectin or CRP — before treatment is escalated, because symptoms alone (which is all HBI measures) can be driven by strictures or bile-acid diarrhoea without active inflammation, or can under-report inflammation that has not yet produced symptoms. A single HBI value is far less informative than a trend, and because the score is uncapped at the high end, a very high number should prompt exclusion of an abscess or other complication rather than being read purely as a severity gradient.
| Score | Band | What it means | Action |
|---|---|---|---|
| < 5 | Remission | Clinical remission | Continue maintenance therapy; confirm with objective markers if escalation is being considered |
| 5–7 | Mild | Mild disease activity | Check calprotectin or CRP; consider optimising current therapy before escalating |
| 8–16 | Moderate | Moderate disease activity | Objective reassessment, exclude abscess/stricture, consider escalation of therapy |
| > 16 | Severe | Severe disease activity | Urgent objective workup, exclude complications, escalate therapy without delay |
Scroll the table sideways for every column.
What the Harvey-Bradshaw needs (5 inputs)
- General well-being yesterday
- Five-point scale from very well (0) to terrible (4), rated by the patient's own sense of how they were.
- Abdominal pain yesterday
- None (0), mild (1), moderate (2), or severe (3).
- Liquid stools yesterday
- One point per liquid stool, with no upper cap — this is the only unweighted, uncapped item, and it is why the index itself has no fixed maximum.
- Abdominal mass
- None (0), dubious (1), definite (2), or definite and tender (3).
- Complications
- One point each for: arthralgia, uveitis, erythema nodosum, aphthous ulcers, pyoderma gangrenosum, anal fissure, new fistula, or abscess. Also uncapped.
What it returns
- Harvey-Bradshaw Index
- The sum of all five items. No fixed ceiling, because liquid stools and complications are both uncapped counts.
- Activity band
- Remission, mild, moderate, or severe, from the thresholds below.
How it is calculated
Harvey and Bradshaw built the index by stripping the CDAI down to same-day clinical assessment: the CDAI's eight items, several diary-based and individually weighted by regression coefficients derived in the 1970s, become five items scored for the previous day only, added together with no weighting at all. The trade for that simplicity is precision — later work quantified it directly, finding that CDAI and HBI correlate well (r = 0.93) but that each one-point change in HBI corresponds to roughly a 27-point change in CDAI with a wide prediction interval, so the same HBI value can map to a meaningfully different range of CDAI values in different patients.
Facts & figures
| Harvey-Bradshaw Index | CDAI | |
|---|---|---|
| Items | 5, equally weighted | 8, individually weighted |
| Data source | Same-day clinical assessment | 7-day symptom diary plus exam findings |
| Remission threshold | < 5 | < 150 |
| Typical use | Clinic, ward, routine follow-up | Clinical trial eligibility and endpoints |
| Correlation with the other | r = 0.93 (Best, 2006) | Same figure, same study |
Scroll the table sideways for every column.
High correlation does not mean interchangeable values — regression found roughly a 27-point CDAI change per HBI point, with wide enough prediction limits that the same HBI can correspond to a meaningfully different range of CDAI scores.
Evidence
Derivation — Harvey and Bradshaw
1980Published as a short letter proposing a simplified, same-day alternative to the Crohn's Disease Activity Index, replacing its diary-based, individually weighted items with five equally weighted items assessed for the previous day only. The index was proposed on the basis of a high correlation with the CDAI in the authors' own data.
Later independent analysis confirmed the correlation the original index was proposed on: a Spearman/Pearson correlation of 0.93 between HBI and CDAI, described by that analysis as showing the two indices are 'purported... to give essentially the same information' — a claim that same analysis went on to test more rigorously.
Quantifying the CDAI-HBI relationship — Best
2006 · n = 224224 single-visit assessments of patients with Crohn's disease, drawn from the datasets originally used to develop both indices, plotted against each other to model CDAI from HBI directly rather than relying on the correlation coefficient alone.
Regression of CDAI on HBI predicted an increase of about 27 CDAI points per HBI point, with 95% prediction limits wide enough to diverge noticeably at higher scores. The paper's own conclusion: 'a good but far from perfect relationship' — CDAI remains preferred for clinical trials, HBI for routine, easier use.
How it compares
Harvey-Bradshaw vs Crohn's Disease Activity Index (CDAI)
Use HBI for routine clinical care and CDAI when a trial-grade, diary-based instrument is specifically required — the two correlate well but are not interchangeable at the individual-patient level.
CDAI needs a 7-day symptom diary and eight individually weighted items; HBI needs a same-day assessment across five equally weighted items. Direct comparison found a correlation of 0.93 between them, but modelling CDAI directly from HBI showed a roughly 27-point CDAI change per HBI point with wide prediction limits — 'good but far from perfect' in the analysis's own words. Remission thresholds reflect the same relationship: CDAI below 150, HBI below 5.
Harvey-Bradshaw vs Faecal calprotectin and CRP
The Harvey-Bradshaw Index measures symptoms and the biomarkers measure inflammation; when they disagree, the biomarker should usually drive the decision.
A Harvey-Bradshaw Index in the remission range does not establish that the bowel is not inflamed, and a raised index does not establish that it is. Symptoms in Crohn's disease are frequently generated by things immunosuppression will not fix — bile-acid diarrhoea after ileal resection, a fibrotic stricture, small-intestinal bacterial overgrowth, or coexisting functional symptoms. Faecal calprotectin and CRP track mucosal inflammation more directly and are what treat-to-target strategies escalate on. The index remains valuable for tracking how the patient actually feels over time, which no biomarker reports.
Pearls & pitfalls
- Liquid stools and complications are uncapped raw counts, not capped categories like the other three items — this is why the index has no published maximum.
- HBI and CDAI are correlated but not interchangeable: about 27 CDAI points per HBI point, with a wide enough spread that converting one to the other for an individual patient is unreliable even though it works well on average.
- The index measures symptoms only. It does not distinguish active inflammation from a stricture or bile-acid diarrhoea producing the same symptoms — check calprotectin or CRP before treating the score as inflammation.
- All items are yesterday-only. Do not average over the week, which is a CDAI convention, not an HBI one.
- A high score, particularly driven by the uncapped items, should prompt a specific look for abscess or fistula rather than being read only as a severity gradient.
Critical actions
- Confirm a rising HBI with an objective marker (calprotectin or CRP) before escalating therapy — symptoms and inflammation diverge in Crohn's disease.
- Exclude abscess before starting or increasing immunosuppression, especially with a high score driven by the complications item.
- Do not use HBI as a proxy CDAI value for trial eligibility or endpoints; use the CDAI itself if that is what is required.
- Track the trend over visits rather than acting on one score in isolation.
- If the HBI conflicts with objective inflammation markers, trust the objective marker and investigate the discrepancy — do not average the two.
Why this score exists
Harvey and Bradshaw's stated purpose was practical rather than psychometric: the CDAI was already established as the trial standard by 1980, but its diary requirement and weighted items made it unwieldy for a normal clinic visit, so they built an index a clinician could complete from the previous day's symptoms and a brief exam. That design goal — speed and simplicity over precision — is exactly what later analysis confirmed: HBI tracks CDAI well but not exactly, which is the expected result of deliberately discarding a week of diary data and several regression weights in exchange for a five-item bedside score.
About the creator
First author, 1980 index
Proposed a five-item, same-day simplification of the CDAI that needed no symptom diary and no blood tests.
Co-author
Co-proposed the simplified index in the 1980 Lancet letter from which it takes its name.
Limitations
- Purely symptom- and exam-based; it does not include any objective inflammatory marker, so it can be discordant with actual mucosal disease activity in either direction.
- The correlation with CDAI, while strong, leaves enough scatter that HBI cannot reliably substitute for a CDAI value in an individual patient, only on average across a population.
- Two of five items are open-ended counts, which means a small number of very frequent liquid stools or several complications can dominate the total in a way the three capped items cannot.
- Not designed to distinguish Crohn's disease activity from symptom-alike conditions such as bile-acid diarrhoea, small intestinal bacterial overgrowth, or a fixed stricture.
- Derived and popularised well before biologic therapy was standard practice; it was not built with treat-to-target, calprotectin-driven strategies in mind, though it remains widely used alongside them.
If you are the patient
The Harvey-Bradshaw Index is a quick way for your Crohn's disease team to check how active your symptoms have been, using how you felt yesterday, any tummy pain, how many loose or watery stools you had, whether a mass can be felt in your abdomen, and whether you have any of a specific list of complications like joint pain, mouth ulcers or a skin problem. A score under 5 means remission. It is a symptom score, not a direct measure of inflammation, so your team will usually also check a stool or blood test (calprotectin or CRP) alongside it, because it is possible to feel well with ongoing inflammation, or to have symptoms from something other than active disease, such as a stricture. It correlates well with a more detailed research score called the CDAI but the two do not always give exactly matching results for the same person.
Frequently asked questions
What is the Harvey-Bradshaw Index?#
A five-item clinical score for Crohn's disease activity — general well-being, abdominal pain, liquid stools, abdominal mass, and complications — all assessed for the previous day, with no laboratory tests or diary required.
What HBI score indicates remission?#
Below 5. Scores of 5–7 indicate mild activity, 8–16 moderate activity, and above 16 severe activity.
How does HBI compare with the CDAI?#
They correlate well (reported r = 0.93), but not perfectly. Regression analysis found roughly a 27-point change in CDAI for each 1-point change in HBI, with wide prediction limits — described in that analysis as 'a good but far from perfect relationship'. CDAI remains the standard for clinical trials; HBI is the practical clinic tool.
Does the Harvey-Bradshaw Index have a maximum score?#
No formal maximum. Two of its five items — liquid stools and complications — are uncapped counts rather than fixed categories, so the total has no published ceiling.
Can HBI replace the CDAI for a clinical trial?#
Not reliably. While the two correlate well on average, the scatter around that relationship is wide enough that HBI cannot substitute for an actual CDAI measurement in an individual patient, which is why trials requiring CDAI eligibility or endpoints still calculate it directly.
Does a low HBI mean there is no inflammation?#
Not necessarily. HBI measures symptoms only. Mucosal inflammation and symptoms diverge often enough in Crohn's disease that objective markers like faecal calprotectin or CRP are checked alongside the index rather than instead of it.