Key takeaways
- Decide early what moves as structured data, what moves as documents and what stays in a read-only archive.
- Build a field-by-field map, including appointment types, providers, insurance plans and balances.
- Run at least two test loads and validate with counts, sample charts and front desk spot checks.
- Keep the old system's records accessible, because patients' access rights do not end when you switch.
What a migration really moves
A data migration is the part of an EHR or practice management switch that staff remember for years, for good or bad. The new system can be excellent and still fail at the front desk if future appointments arrive with the wrong visit type, insurance plans land under the wrong payer or allergies arrive as a scanned page instead of a list. Most problems trace back to decisions made, or not made, in the first weeks of the project.
| Data type | Usually migrated as | Watch for |
|---|---|---|
| Demographics and contacts | Structured fields | Duplicate patients, old phone numbers, guarantor links |
| Insurance coverage | Structured, mapped to new payer list | Plans mapped to the wrong payer ID; inactive plans marked active |
| Future appointments | Structured, mapped to new visit types and providers | Visit lengths, locations and provider templates |
| Problems, medications, allergies | Structured where codes match | Free-text entries that cannot map |
| Visit notes and letters | Documents or PDF | Indexing by date and type so they can be found |
| Balances and ledgers | Summary balances, detail in archive | Totals that do not reconcile |
| Recall and reminder lists | Structured | Lists lost between systems, causing missed follow-ups |
| Older history | Read-only archive | Access for staff and for patients' records requests |
Patients will notice gaps
Patients see far more of their records than they used to. In 2024, 65% of individuals were offered and accessed their online medical records or patient portal, up from 25% in 2014[1]. When a migration drops a medication list or a past visit, patients notice in the portal and call the office. Plan for those calls and for a clear answer about where older records are.
Source: Office of the National Coordinator for Health IT (ASTP/ONC), HHS, 2024[1]
Show the data
| Item | Value |
|---|---|
| 2014 | 25% |
| 2024 | 65% |
The legal duty continues too. A covered entity must act on a patient's request for access no later than 30 days after receipt, with one extension of no more than 30 days[2]. Records that stay in the old system are still part of what patients can request, so your archive must remain searchable and printable after cutover.
Decide the scope first
Migrating everything as structured data is expensive and rarely worth it. A common approach is to bring active patients' demographics, insurance, future appointments, active problems, medications, allergies and recent results across as structured data, bring a set period of notes and letters as documents, and keep the rest in a read-only archive. Agree that scope in writing with the new vendor and any conversion company before mapping starts.
- Define active patient, for example seen in the last three years or with a future appointment.
- Choose the document look-back period by specialty and by how often clinicians open older notes.
- Decide how long the old system stays available read-only, and who pays for it.
- Confirm how records will be produced for requests from the archive.
Build the field map
The map is a spreadsheet that lists every source field, its destination field, any transformation, and who signed off. Front office data needs the most attention because it drives the schedule and billing on day one. Map these lists explicitly: providers and their identifiers, locations, appointment types and lengths, payers and plans, referral sources, patient statuses and recall types.
Example: a fictional family practice found its old system had 140 appointment types, many unused for years. Instead of migrating them all, the team mapped them to 22 types in the new system, and future appointments using retired types were assigned to the nearest match and flagged for a scheduler to review.
Test loads, not one big bang
Extract a full copy
Take a complete extract from the old system into a secure location agreed with both vendors.
Transform with the map
Apply the field map and code translations. Log every record that cannot be mapped.
Load into a test environment
Never into the live system first. Use the new vendor's test practice.
Validate
Run counts, open sample charts and have staff check what they use every day.
Fix and repeat
Correct the map and the source data, then load again. Plan for at least two full test loads.
Rehearse the timing
Time the final test load so you know how long cutover will take.
Validation checklist
- Counts match: active patients, future appointments, active insurance records and documents, old versus new.
- Twenty sample charts compared field by field by a clinician and a front desk lead.
- Every provider's schedule for the next two weeks matches the old system, slot by slot.
- Ten patients with complex coverage show the right primary and secondary plans.
- Account balances reconcile to the old system's aged trial balance in total.
- Recall lists produce the same patients due in the next three months.
- Duplicate patient check run, with merges decided before go-live.
Sign off each item with a name and date. If an item fails, decide whether it blocks go-live or can be fixed after, and write that down too.
Cutover weekend
Choose a cutover date away from your busiest weeks, and avoid the first weeks of January, when coverage changes for many patients at once. Freeze changes in the old system at a set time, take the final extract, load, validate the critical counts and check the next clinic day's schedule before staff arrive. Keep the old system available read-only from the moment of the freeze.
- Book lighter schedules for the first two weeks after go-live.
- Print or export the first three days' schedules from the old system as a fallback.
- Have the conversion team on call for the first clinic day.
- Keep a shared issue log and review it twice a day for the first week.
Contracts, security and records
Everyone who handles the extract is handling protected health information. A covered entity may let a vendor handle it only with satisfactory assurance, in a business associate agreement, that the vendor will safeguard it[3]. That includes a separate conversion company, not just the EHR vendors. Agree how the extract is transferred, where it is stored, and when every copy is destroyed.
Small practices face an extra challenge: many run less common systems. Only 32% of physicians in solo practice used an EHR from one of the five market-leading developers in 2024[4], so conversion tools may not exist off the shelf for your source system. Ask for references from practices that moved from your exact system. Keep the project's decisions, sign-offs and destruction certificates with your compliance records; HIPAA requires covered entities to retain required policy and compliance documentation for six years[5].
Common mistakes that surface later
- Recall lists left behind, so patients due for annual visits or screenings are never contacted.
- Guarantor links broken, so statements go to the patient instead of the parent or responsible party.
- Inactive insurance migrated as active, causing eligibility rejections in the first weeks.
- Referral sources dropped, so the practice loses track of which offices send it patients.
- Patient communication preferences and opt-outs not migrated, so texts reach people who asked to stop.
- Old system access canceled too early, before records requests and audits are covered.
Each of these is cheap to prevent with a line in the field map and expensive to fix after go-live. Ask the conversion team to show you each one in the test load.
The first 30 days after go-live
Keep the issue log open for a month and review it weekly with both vendors. Watch three signals: eligibility rejections, patients calling about missing information in the portal, and schedulers reporting appointments with the wrong type or length. A spike in any of them usually points to a mapping rule rather than a training problem. At day 30, run the validation counts again, confirm recall lists are producing the expected patients, and agree in writing when the old system will move to archive-only access.
What the front desk tells patients
Example script: 'We moved to a new records system this month. Your recent visits, medications and allergies are in the new portal. Older notes are kept safely in our previous system, and we can send you copies if you need them. Could you confirm your mobile number and pharmacy so we know they are up to date?'
Use the first visit after go-live to confirm demographics, insurance and pharmacy with every patient. It turns the migration into a data clean-up, and patients generally appreciate being asked.
Frequently asked questions
Should we migrate all historical data into the new EHR?
Usually not. Migrate active patients' key structured data and a set period of documents, and keep older history in a read-only archive that staff can search and print.
How many test loads do we need?
Plan for at least two full test loads into a test environment, with validation and fixes between them, before the final cutover.
What happens to records requests after we switch?
Patients can still request records held in the old system. Keep the archive accessible and make sure staff know how to produce copies within the required time.
Who needs a business associate agreement during a migration?
Any vendor that handles the patient data extract, including a separate conversion company, in addition to the old and new system vendors.
References
Office of the National Coordinator for Health IT (ASTP/ONC), HHS (2024). Individuals' Access and Use of Patient Portals and Smartphone Health Apps, 2024 (opens in a new tab) ↑
Legal Information Institute, Cornell Law School (current). 45 CFR § 164.524 - Access of individuals to protected health information (opens in a new tab) ↑
Legal Information Institute, Cornell Law School (current). 45 CFR § 164.502 - Uses and disclosures of protected health information: General rules (opens in a new tab) ↑
Office of the National Coordinator for Health IT (ASTP/ONC), HHS (2024). Office-based Physician Electronic Health Record Adoption, 2008-2024 (ONC Data Brief) (opens in a new tab) ↑
Legal Information Institute, Cornell Law School (current). 45 CFR § 164.530 - Administrative requirements (opens in a new tab) ↑