Electronic health records and electronic medical records differ mainly in scope. An electronic medical record (EMR) is one practice's digital chart. An electronic health record (EHR) is meant to gather and share information across every clinician involved in a patient's care. For data extraction, both hold typed fields and scanned pages that need reading.
How do electronic health records and electronic medical records differ?
Electronic health records and electronic medical records differ in how far their information is meant to travel. The federal health IT office describes an EMR as a digital version of one clinician's paper chart, and an EHR as a record built to reach beyond the organization that first collected it.
The clearest official description is a 2011 post on the federal health IT site. The ONC explainer "EMR vs EHR: What is the Difference?" says electronic medical records "are a digital version of the paper charts in the clinician's office," and that "the information in EMRs doesn't travel easily out of the practice." It says EHRs "focus on the total health of the patient" and "are designed to reach out beyond the health organization that originally collects and compiles the information."
For someone pulling data out of these systems, the practical differences look like this:
| Electronic medical record (EMR) | Electronic health record (EHR) | |
|---|---|---|
| Scope | One practice or facility's chart | A patient's care across clinicians and settings |
| Sharing | Mostly stays inside the practice | Designed to be shared with other providers |
| Typical contents | Visit notes, orders, results, medications from that office | The same, plus records received from outside: referral letters, hospital summaries, outside lab reports |
| Scanned pages | Paper intake forms and outside records scanned in | Often more, because outside documents arrive by fax and file upload |
| What it means for extraction | Fewer sources, more consistent layouts | More document types and layouts in one chart, more dating and duplicate work |
None of these lines is sharp. A small practice's system may share data widely, and a large system may still hold most outside records as images.
What EMR abstraction looks like in practice
EMR abstraction is the work of reading a patient's electronic chart and copying defined facts into a form, registry or dataset. In practice it mixes two tasks: copying values that already sit in typed fields, and reading values from documents stored inside the chart as scanned pages or attached files. The second task usually takes most of the time.
A typical EMR abstraction session starts with the typed parts of the chart. Problem lists, medication lists, vital signs and lab results entered in the system can often be read or exported directly. Then the abstractor opens the document tab and works through scanned items: an outside discharge summary, a consult letter, a faxed lab report. Each of those has to be read, dated and matched to the abstraction form's definitions.
The definitions matter more than the system. A form that asks for "date of diagnosis" needs a rule for which document counts, whether the chart is called an EMR or an EHR. The general method, with forms and data dictionaries, is in the guide to medical record abstraction. For hospital stays, the facts most often abstracted come from one document, which the post on discharge summary extraction breaks into fields.
EMR extraction by software follows the same split. Typed fields come out through the system's own reports or interfaces. Scanned pages need reading, and every value read from them should carry a link back to its page.
Scanned pages inside an EHR chart
Scanned pages inside an EHR chart are images of paper documents filed into the electronic record. They exist because much of what reaches a practice still arrives as paper or fax: referral packets, outside hospital records, signed consent forms, insurance cards and handwritten intake forms. The system stores the picture, but the facts on it are not in typed fields.
Health data standards plan for these images. The HL7 FHIR R4 DocumentReference resource states that its scope includes clinical notes and "scanned paper", and lists "Scanned Paper, and digital records of faxes" among the documents it describes. In other words, an electronic record can store a scanned page with a label, a date and a patient, while the content of the page stays an image.
For anyone extracting data, that means an electronic chart is often a mix:
- Typed data: values entered directly, which can be queried.
- Text documents: notes typed or dictated, which can be searched but are not split into fields.
- Images: scans and faxes, which need OCR and extraction before anything on them can be searched or counted.
Scanned pages also carry the usual paper problems: skewed pages, faint fax print, handwriting and several documents filed under one label. Dating and splitting them follows the same rules as a paper record.
Handling these pages means handling health information. Schema Care will not take real patient data until a business associate agreement is signed with the customer, and its compliance status table says which safeguards are in the design, in progress or planned.
When does the EHR and EMR distinction stop mattering?
The EHR and EMR distinction stops mattering once you are working with a specific chart. What decides the work is what the chart holds: which fields are typed, which documents are scanned, and how they are dated. The ONC explainer itself notes that some people use the two terms interchangeably, and says its office uses EHR "almost exclusively."
The distinction still helps in two places. When you plan an abstraction project, "EHR" hints that the chart will hold more outside documents, which means more scanned pages, more layouts and more duplicates to sort. When you request records from another facility, asking for the full record rather than the practice's own notes can be the difference between getting outside reports or not.
Everywhere else, the useful questions are concrete. Can the typed data be exported, and in what form? How are scanned documents labelled and dated? Who will read the images, and how will each extracted value point back to its page? Questions about connecting systems, such as FHIR interfaces and integration with an existing EHR, are covered in the developer integration category.