Interactive demonstration · synthetic data
Hospital discharge summary.
Three medication changes, two requests to the practice and a hospital-arranged follow-up, after an admission with heart failure.
Each scenario has its own address, and the address keeps your place, so you can share it or come back to it. Reset at any time.
Step 1 of 3 Open an item to see the passage it came from, then accept, edit or reject it. 0 of 4 decided.
Source document
Numbers mark the passages each item cites.Admission details
Mr Holt was admitted on 26 July 2026 with worsening breathlessness and ankle swelling. A diagnosis of acute decompensated heart failure was made.4 He responded to intravenous diuretics and was discharged home on 31 July 2026.
Investigations
Echocardiogram showed reduced left ventricular systolic function. Creatinine rose to 168 µmol/L during admission (baseline 102 µmol/L) and was 131 µmol/L on discharge.
Medication changes
Furosemide 40 mg each morning has been started.1 Bisoprolol has been increased from 2.5 mg to 3.75 mg once daily.1 Ramipril 5 mg has been withheld because of the rise in creatinine; please do not restart it until renal function has been reviewed.1
Requests to the GP practice
Please arrange urea and electrolytes within 7 days of discharge2 and review the result before considering restarting ramipril3. Please update the repeat prescription record to reflect the changes above.1
Follow-up
We will review Mr Holt in the heart failure clinic in approximately 6 weeks; the appointment will be sent by the hospital. He has been advised to weigh himself daily and to seek advice if his weight increases by more than 2 kg in 3 days.
Medicines on discharge
| Medicine | Instruction | Status |
|---|---|---|
| Furosemide 40 mg tablets1 | One each morning | New |
| Bisoprolol 3.75 mg | Once daily | Dose increased |
| Atorvastatin 20 mg tablets | One at night | Unchanged |
| Ramipril 5 mg capsules | Withheld | See above |
Proposed work 4
Each item shows the findings and the exact passages it comes from. Open one to decide on it.
Item 1 of 4: Medication reconciliationReconcile the medication record with the discharge summaryAction needed
From the document
- Medication startedMissing from record
Furosemide 40 mg each morning
Record Not on the medication record.
Why Reedley says this New medicine stated in the medication changes and listed as “New” on discharge.
- Dose changedDose differs
Bisoprolol 2.5 mg → 3.75 mg once daily
Record Record shows bisoprolol 2.5 mg once daily.
Why Reedley says this Both the previous and new doses are stated explicitly.
- Medication withheldActive on record
Ramipril 5 mg withheld pending renal function
Record Record shows ramipril 5 mg as an active repeat.
Why Reedley says this Explicit instruction not to restart until renal function has been reviewed.
Proposed work
Three changes in hospital are not reflected in the record: furosemide started, bisoprolol increased, ramipril withheld.
Verified when: Active medication and latest issues match the discharge regimen. Source: Clinical record, by design; simulated here.
Limit It does not show that the patient has collected or is taking the medicine (issued is not taken).
Reedley does not change the medication record. A prescriber makes the changes in the clinical system.
Reedley proposes; a named person decides. Nothing happens until you do.
Item 2 of 4: InvestigationUrea and electrolytes taken by 7 AugustDue within 7 days
From the document
- Request to practice
Urea and electrolytes within 7 days of discharge
Due by 7 Aug 2026 (7 days from 31 Jul).
Why Reedley says this Direct request addressed to the GP practice, with a time window.
Proposed work
The hospital asked for renal function within 7 days of discharge.
Verified when: A U&E result for this patient received within the 7-day window. A booking alone is not enough. Source: Pathology, by design; simulated here.
Limit A received result is not a reviewed result. The GP’s review is the next item, and this item is designed to close only after that review is recorded (the development build closes it on receipt).
Reedley does not book appointments. The practice books through its own system.
Reedley proposes; a named person decides. Nothing happens until you do.
Item 3 of 4: Clinical reviewReview the U&E result before any ramipril restartClinician decision
From the document
- Request to practice
Review the result before any ramipril restart
Why Reedley says this Direct request that depends on the blood test above.
Proposed work
The hospital asked the practice to review renal function before considering a restart.
Verified when: A review and decision recorded by a named prescriber, with the date. Source: Clinical record, by design; simulated here.
Limit It shows that a decision was recorded, not what the decision should be. Reedley does not assess it.
Reedley does not decide whether to restart a medicine. It shows what the record and the letter say.
Reedley proposes; a named person decides. Nothing happens until you do.
Item 4 of 4: Coding proposalAdd heart failure to the problem listProposal
From the document
- DiagnosisNot on problem list
Acute decompensated heart failure
Record Heart failure is not on the problem list.
Why Reedley says this Stated explicitly as a diagnosis in the admission details.
Proposed work
The diagnosis is stated in the letter and is not on the problem list.
Verified when: A coded problem entry exists for this patient. Source: Clinical record, by design; simulated here.
Reedley proposes the entry; a person accepts it and records it in the clinical system.
Reedley proposes; a named person decides. Nothing happens until you do.
For information · no action proposed
- Heart failure clinic in about 6 weeksArranged by the hospital. No practice action proposed.
- Daily weights advisedPatient advice given in hospital. Shown for context.
What the record shows (synthetic)
Current medicines
- Ramipril 5 mg capsules One daily Repeat
- Bisoprolol 2.5 mg tablets One daily Repeat
- Atorvastatin 20 mg tablets One at night Repeat
Problems
Hypertension · Hypercholesterolaemia
Work queue
0 open · 0 closed · Graham HOLT
A person’s report moves an item to actioned; verification comes from the record where Reedley can read it; where it cannot, a named person’s structured statement is accepted today, and that rule is under review. Booked is not attended, sent is not read and issued is not taken.
No work yet. Accept a proposal in the document review to create a work item.
History
Append-only. Who did what, when, and on what basis. Times are simulated.
Document received · Discharge summary · Northfield General Hospital (fictional)
Identity matched · Graham HOLT
Name, date of birth and NHS number on the document match the patient record.
Classified · Discharge summary · General medicine
Document title, discharge date and a medicines-on-discharge table (p.1–2).
Identified · 6 findings, 4 proposals
Awaiting review by a named person.
Identified · 6 findings, 4 proposals
What the demonstration does not show
Real accuracy. These outputs were written in advance. They do not show how often the product extracts information correctly from real documents; that has not yet been measured.
Real integration. There is no connection to EMIS Web, SystmOne or any laboratory, and nothing is written to any record. “Source checks” are simulated to show how verification is meant to work once approved access exists.
Clinical rules. Intervals, routing rules and wording are illustrative. In the product, clinical rules are designed to be owned by named clinicians and versioned; today they are unsigned drafts with no named owner.