About the Oakland Score for Safe Discharge After Lower GI Bleeding
The Oakland score decides whether a patient with acute lower gastrointestinal bleeding can be safely discharged rather than admitted. It combines seven items available at presentation — age, sex, previous LGIB admission, blood on rectal examination, heart rate, systolic blood pressure and haemoglobin — into a score from 0 to 35, with haemoglobin the heaviest term. A score of 8 or below identifies a low-risk patient with roughly a 95% probability of avoiding an adverse outcome (rebleeding, transfusion, intervention, in-hospital death or 28-day readmission), supporting outpatient management. It is the lower-GI counterpart to the Glasgow-Blatchford score, and it is a triage-to-discharge tool, not a mortality score.
Formula
Oakland = age points + sex points + prior-LGIB points + rectal-exam points + heart-rate points + systolic-BP points + haemoglobin points- age points
- 0 (≤39), 1 (40–69), or 2 (≥70).
- sex points
- 1 if male, else 0.
- prior-LGIB points
- 1 if a previous LGIB admission, else 0.
- rectal-exam points
- 1 if blood found on DRE, else 0.
- heart-rate points
- 0 (≤69), 1 (70–89), 2 (90–109), 3 (≥110).
- systolic-BP points
- 5 (50–89), 4 (90–119), 3 (120–129), 2 (130–159), 0 (≥160).
- haemoglobin points
- 22 (3.6–6.9), 17 (7.0–8.9), 13 (9.0–10.9), 8 (11.0–12.9), 4 (13.0–15.9), 0 (≥16) g/dL.
- Haemoglobin dominates the score — a severe anaemia alone contributes 22 of the 35 possible points, so the discharge decision is largely, though not only, a haemoglobin decision.
- The safe-discharge threshold is 8 or below, inclusive.
- The original derivation reported haemoglobin in g/L; the thresholds here are the equivalent g/dL values used in most implementations.
Interpreting the result
Treat a score of 8 or below as the green light: the patient falls in the low-risk group, with about a 95% probability of no adverse outcome, and can usually be discharged with outpatient follow-up provided the clinical picture agrees. A score above 8 does not meet the safe-discharge threshold and favours admission for observation and investigation. Some centres extend the cut-off to 10 to enlarge the discharge-eligible group at a modest cost in sensitivity; this widens throughput but should be validated locally before adoption. Whatever threshold is used, the score supports the decision — it does not overrule ongoing bleeding, an unstable patient or a clinician's concern.
| Score | Band | What it means | Action |
|---|---|---|---|
| 0–8 | Low risk — safe discharge | ≈ 95% probability of no adverse outcome | Consider discharge with outpatient follow-up, alongside clinical judgement |
| 9–35 | Not low risk | Higher probability of rebleeding, transfusion, intervention, death or readmission | Admit for observation, resuscitation and investigation |
What the Oakland Score needs (7 inputs)
- Age
- 0 points if 39 or under, 1 point at 40–69, 2 points at 70 or over.
- Sex
- 1 point for male, 0 for female — men had a higher risk of adverse outcome in the derivation cohort.
- Previous LGIB admission
- 1 point for a prior hospital admission with lower GI bleeding, 0 if none.
- Digital rectal examination
- 1 point if blood is found on rectal examination, 0 if not — a simple bedside marker of ongoing or recent bleeding.
- Heart rate
- 0 points if 69 or under, 1 at 70–89, 2 at 90–109, 3 at 110 or over.
- Systolic blood pressure
- 5 points at 50–89 mmHg, 4 at 90–119, 3 at 120–129, 2 at 130–159, 0 at 160 or over — the score penalises hypotension, so lower pressures earn more points.
- Haemoglobin
- The heaviest term: 22 points at 3.6–6.9 g/dL, 17 at 7.0–8.9, 13 at 9.0–10.9, 8 at 11.0–12.9, 4 at 13.0–15.9, 0 at 16 or above.
What it returns
- Oakland score (0–35)
- The sum of the seven items. Higher scores indicate a higher risk of an adverse outcome.
- Discharge band
- ≤ 8 supports safe discharge; above 8 does not meet the threshold and favours admission.
How it is calculated
Oakland is a points score derived by logistic regression on a national audit of acute lower GI bleeding, with the coefficients converted to integer weights. The outcome it was fitted against is a composite of clinically meaningful events — rebleeding, red-cell transfusion, therapeutic intervention (endoscopic, radiological or surgical), in-hospital death and readmission within 28 days — so the score is calibrated to the exact bundle of things that would make discharge unsafe. Because anaemia and haemodynamic compromise are the strongest signals of a bleed that will not behave, haemoglobin and the vital-sign terms carry most of the weight, while age, sex, prior bleeding and rectal findings fine-tune the estimate.
Facts & figures
| Variable | Categories → points |
|---|---|
| Age | ≤39: 0 · 40–69: 1 · ≥70: 2 |
| Sex | Female: 0 · Male: 1 |
| Previous LGIB admission | No: 0 · Yes: 1 |
| Rectal exam | No blood: 0 · Blood: 1 |
| Heart rate | ≤69: 0 · 70–89: 1 · 90–109: 2 · ≥110: 3 |
| Systolic BP (mmHg) | 50–89: 5 · 90–119: 4 · 120–129: 3 · 130–159: 2 · ≥160: 0 |
| Haemoglobin (g/dL) | 3.6–6.9: 22 · 7.0–8.9: 17 · 9.0–10.9: 13 · 11.0–12.9: 8 · 13.0–15.9: 4 · ≥16: 0 |
Total range 0–35; the safe-discharge threshold is ≤ 8.
Evidence
Derivation and validation — UK national audit
2017 · n = 2,336Derived from a UK-wide audit of acute lower gastrointestinal bleeding admissions and validated internally, using a composite adverse-outcome endpoint of rebleeding, transfusion, therapeutic intervention, in-hospital death and 28-day readmission.
The model discriminated adverse outcomes with an area under the ROC curve around 0.84. A threshold of 8 or below identified low-risk patients suitable for discharge with high sensitivity for excluding adverse outcomes.
US external validation
2020External validation in a US cohort of adults with acute lower GI bleeding, testing the ≤ 8 threshold for safe discharge.
At a threshold of 8 or below the score was highly sensitive for excluding adverse outcomes (around 98%) with correspondingly modest specificity, and discriminated adverse outcomes with an AUROC near 0.87 — supporting its use to identify low-risk patients across health systems.
How it compares
Oakland Score vs Glasgow-Blatchford Score
They are territory-matched siblings: Glasgow-Blatchford identifies low-risk upper GI bleeds for discharge, Oakland does the same job for lower GI bleeds — use the one that matches the suspected source.
Both were built to find patients who can avoid admission, both lean heavily on haemoglobin and haemodynamics, and both are deliberately tuned for high sensitivity. The difference is anatomical: Glasgow-Blatchford was derived in upper GI bleeding and includes urea, while Oakland was derived in lower GI bleeding and adds a rectal-examination finding and prior-LGIB history. Applying the wrong one to the wrong territory is the main pitfall.
Oakland Score vs ABC Score
Different endpoints: Oakland answers 'can this lower GI bleed go home?', while ABC answers 'what is this patient's 30-day mortality?' across both upper and lower bleeding — pick the score whose question matches the decision.
ABC spans upper and lower GI bleeding and predicts death, driven by age, comorbidity and organ dysfunction. Oakland is confined to lower GI bleeding and predicts a composite adverse outcome centred on the discharge decision. A young patient with a self-limiting bleed can be Oakland-low and ABC-low together, but the two are measuring different things and are best used side by side rather than interchangeably.
Pearls & pitfalls
- Haemoglobin dominates the score — a markedly low haemoglobin can push a patient over the threshold on its own, which is clinically appropriate for a bleed.
- It is a discharge score for lower GI bleeding, not a mortality score and not for upper GI bleeding; for suspected upper GI bleeding use Glasgow-Blatchford.
- The rectal-examination item requires an actual DRE — omitting it and assuming 'no blood' will underscore the patient.
- Score the presenting, post-initial-resuscitation values; the systolic BP and heart-rate terms reflect the patient in front of you.
- The extended ≤ 10 threshold discharges more patients but at some cost in sensitivity, and should be validated in the local population before it is adopted.
- A low score does not override active bleeding, haemodynamic instability, significant comorbidity or a poor social situation — clinical judgement retains a veto.
Critical actions
- Complete a digital rectal examination and a full blood count before scoring — both feed directly into the result.
- Discharge Oakland ≤ 8 patients with clear safety-netting and outpatient follow-up rather than routine admission, where the clinical picture agrees.
- Admit patients scoring above 8 for observation, resuscitation and investigation of the bleeding source.
- Resuscitate and reassess an unstable patient before relying on the score; it assumes initial resuscitation has occurred.
- Document that the score supported the decision, and that ongoing bleeding or instability was considered independently of the number.
Why this score exists
Oakland and colleagues designed the score around a decision rather than a diagnosis: the question was not how severe the bleed is in the abstract but whether this particular patient can go home safely. That framing explains the choice of a composite outcome built from the events that would make discharge a mistake, and the deliberate emphasis on haemoglobin and haemodynamics, which are the values that separate a self-limiting bleed from one that will declare itself. The threshold of 8 was chosen to keep sensitivity high, accepting lower specificity, because the cost of wrongly discharging a serious bleed outweighs the cost of a cautious admission.
About the creator
First author, 2017 derivation study
Derived the score from a UK national audit of lower gastrointestinal bleeding to identify patients safe for outpatient management.
Senior author
Provided the prediction-model methodology; he is a leading author of the TRIPOD reporting standard for such models.
Limitations
- It applies only to acute lower gastrointestinal bleeding, not upper GI bleeding.
- Specificity at the ≤ 8 threshold is low, so it errs towards caution and admits some patients who would have done well.
- It predicts a composite adverse outcome rather than mortality specifically, so it is not a prognostic death score.
- Haemoglobin's dominance means a chronic anaemia unrelated to the acute bleed can inflate the score.
- It was derived largely in UK audit populations, and although externally validated, local calibration is worth checking before adopting an extended threshold.
If you are the patient
The Oakland score helps doctors decide whether someone who has had bleeding from the lower bowel (bright or dark blood from the back passage) needs to stay in hospital or can safely go home with follow-up. It adds up seven things measured when you arrive — your age, sex, whether you have been admitted for this before, whether there is blood when the doctor examines the back passage, your pulse, your blood pressure and, most importantly, your blood count (haemoglobin) — to give a score out of 35. A score of 8 or less means you are in the low-risk group, with about a 95% chance of no serious problem, so going home is usually reasonable. A higher score means it is safer to be admitted for a closer look. It is about whether it is safe to go home, not about longer-term risk, and the doctor will always weigh it against how you actually look and feel.
Frequently asked questions
What is the Oakland score?#
The Oakland score is a seven-item tool for acute lower gastrointestinal bleeding that predicts the risk of an adverse outcome and identifies patients who can be safely discharged. It uses age, sex, previous LGIB admission, rectal-examination findings, heart rate, systolic blood pressure and haemoglobin, scoring 0 to 35.
What Oakland score is safe for discharge?#
A score of 8 or below marks a low-risk patient with roughly a 95% probability of avoiding an adverse outcome, supporting discharge with outpatient follow-up. Some services extend the threshold to 10 to discharge more patients, at a small cost in sensitivity.
Is the Oakland score for upper or lower GI bleeding?#
Lower GI bleeding. It was derived and validated specifically in acute lower gastrointestinal bleeding. For upper GI bleeding, the equivalent safe-discharge tool is the Glasgow-Blatchford score.
Which variable matters most in the Oakland score?#
Haemoglobin. It carries up to 22 of the 35 possible points, so a severe anaemia contributes most of the score on its own — appropriately, since a low haemoglobin is the strongest signal of a significant bleed.
How is the Oakland score different from the Glasgow-Blatchford score?#
Both find low-risk patients who can avoid admission, but Glasgow-Blatchford is for upper GI bleeding and includes urea, while Oakland is for lower GI bleeding and adds a rectal-examination finding and previous-LGIB history. Choose the score that matches the suspected bleeding source.
Does a low Oakland score guarantee it is safe to go home?#
No. It identifies a low-risk group and supports the discharge decision, but it does not override active bleeding, an unstable patient, significant comorbidity or clinical concern. It is a decision aid, not a rule that replaces judgement.
References
Original / primary reference
Clinical practice guidelines
- Oakland K, Chadwick G, East JE, et al. Diagnosis and management of acute lower gastrointestinal bleeding: guidelines from the British Society of Gastroenterology. Gut. 2019;68(5):776-789.
- Sengupta N, Feuerstein JD, Jairath V, et al. Management of Patients With Acute Lower Gastrointestinal Bleeding: An Updated ACG Guideline. Am J Gastroenterol. 2023;118(2):208-231.