CASE STUDY · HEALTHCARE INTEROPERABILITY & DATA EXCHANGE
Connecting three disconnected EHR systems into one patient view.
- 12 minutesChart-review time down from 12 minutes to 2 minutes in the pilot
01 · PROJECT OVERVIEW
The client and the brief.
An anonymised regional hospital network. Name withheld under confidentiality.
An integration engine with FHIR-compatible interfaces, patient matching and a read-only unified view. Two of the three EHRs were connected in the pilot; the third is planned for phase two.
- 1Clinicians checked 3 systems to see one patient's history, taking about 12 minutes per chart review.
- 2Each system held a part of the story: encounters in one, results in another, notes in a third.
- 3Clinicians knew the gaps, and some relied on memory or on asking colleagues.
- 4The network wanted a single view without replacing any of the three EHRs.
02 · WHAT WENT WRONG
What went wrong, and how we fixed it.
Every engagement hits problems. These were the ones that mattered here, and what we changed so they did not come back.
- PROBLEM 01Early on, the patient matching produced a handful of suspected duplicates and a few missed links
Root cause: one EHR stored names and dates in a different format, and another held old addresses. We tightened normalisation of names and dates, added a manual review step for borderline matches, and ran a reconciliation report each week with a clinical records lead. We also made the view display which source each item came from, so clinicians could verify quickly.
- PROBLEM 02Trust grew once the matching errors were visible and fixable
03 · THE PROCESS
How we ran it.
- 01Discovery (first weeks)
We interviewed clinicians about what they look for during a chart review, so the unified view would show that first. We also mapped each EHR's interface options and its version.
- 02Build
Two-week iterations. Each ended with a demo to a small group of clinicians and the IT team.
- 03Pilot
A single department used the unified view next to the existing systems. We held a short weekly review with clinicians and fixed issues in order of patient-safety impact.
- 04Cadence
Weekly check-ins with the network's integration lead, plus a shared board of open issues.
04 · RESULTS
What changed.
Measured against the way the work ran before the engagement began.
- NOW 01
Clinicians open one timeline for the two connected systems rather than logging into each.
- NOW 02
A records lead owns the match-review queue and reviews it on a schedule.
- NOW 03
New systems are assessed for interface support before purchase.
- NOW 04
The IT team monitors feed health and acts on alerts, instead of hearing about gaps from clinicians.
- 12 minutes
Chart-review time down from 12 minutes to 2 minutes in the pilot
One patient timeline across two systems
Third system scheduled for phase two
05 · WHAT WE BUILT
How it works.
- An integration engine receives data from the connected EHRs through their supported interfaces and normalises it to FHIR resources.
- A patient matching service links records using demographics and identifiers, with a review queue for uncertain matches.We favoured caution: an unmatched record is an inconvenience, a wrongly merged record is a safety risk.
- Terminology mapping aligns local codes to standard code sets, with unmapped codes surfaced rather than dropped.
- The unified view is read-only.It shows a timeline of encounters, results, medications and allergies, and links back to the source system.
- Access is role-based, every view is logged, and data in transit and at rest is encrypted.
- Monitoring tracks feed delay and failed messages per source.
- Controls and handover
- Clinical safety drove several design choices. The view never writes back to a source system. Every patient timeline shows the source of each item and the time it was last refreshed, so a clinician can tell when data may be stale. We trained a small group of clinical champions first, and they trained their colleagues. IT received runbooks for restarting feeds, reading the match-review queue and onboarding the third EHR.
06 · QUICK ANSWERS
The short version.
- What problem did the client have?
- Clinicians checked 3 systems to see one patient's history, taking about 12 minutes per chart review. Each system held a part of the story: encounters in one, results in another, notes in a third. Clinicians knew the gaps, and some relied on memory or on asking colleagues. The network wanted a single view without replacing any of the three EHRs.
- What did Brandsmashers build?
- An integration engine with FHIR-compatible interfaces, patient matching and a read-only unified view. Two of the three EHRs were connected in the pilot; the third is planned for phase two.
- What were the results?
- Chart-review time down from 12 minutes to 2 minutes in the pilot; one patient timeline across two systems; third system scheduled for phase two.
YOUR TURN
Have a similar requirement?
Tell us what is stuck. We will come back with a plan and matched, pre-vetted engineers.
