Healthcare document extraction for administrative teams

Administrative, not clinical

Healthcare paperwork falls into two kinds. Clinical documents inform care; administrative documents decide who pays and how much. docRefinery is for the second kind. It reads what is printed on a claim form or an explanation of benefits and returns it as data. Procedure and diagnosis codes come back exactly as printed. docRefinery does not choose, correct or validate them against coding rules, and it plays no part in clinical decisions.

What billers key from each form

The 20-type catalogue covers medical bills. 1500 claim forms, explanations of benefits, referrals, prior authorisation and intake forms are set up as a workflow or a custom type, trained at no charge.

Standard claim forms and a fixed payer list

Some of these forms are standard by design. The 1500 claim form is a single paper claim form for use by all third-party payers, and the National Uniform Claim Committee maintains its data set and physical layout. Payers issue explanations of benefits from a small set of templates, and a practice deals with the same payers month after month. Intake forms are the practice's own. So a billing team sees a limited set of layouts, each learned once and remembered after that.

Coded fields and free text need different trust

Handwriting, checkboxes and dense remittance tables can all sit in the same document. Confidence thresholds are set per field, and here that matters more than anywhere else. A member ID or an allowed amount should clear a high bar before anyone relies on it; a free-text reason for referral can pass at a lower one. Values under their threshold go to a person, and a field the form leaves blank comes back null rather than a guess. Workflow rules can add checks within a document, such as paid plus adjustments plus patient responsibility equalling the billed amount on a remittance line.

Billers decide what gets posted

Billers stay in charge of what is posted. In the review queue, a reviewer sees the source page next to each held value with the cited region highlighted, corrects it, and adds or removes table rows. Every change is kept on the run, and exports carry the corrected values.

What billing teams get beyond the fields

Patient data and deployment

Patient documents need care about where they are processed. Enterprise plans support private VPC and fully air-gapped on-premises deployment, so documents need not leave your network. That is a deployment capability, not a compliance certificate. Data retention is configurable per plan, from 7 days on Starter to custom terms on Enterprise, and data is encrypted in transit and at rest. docRefinery makes no HIPAA or other regulatory compliance claim. A SOC 2 Type II audit is in progress and no report has been issued yet; SSO is planned for Enterprise and not available yet; there are no data-residency commitments yet. Evaluate on hosted plans with synthetic documents, and talk to us about your obligations before sending real patient data.

Posting to practice management

Your code takes the fields from the API (extract a document) and posts them, or a biller exports from the dashboard; the Excel export keeps each remittance table on its own sheet, line by line. One format and your own field names mean the mapping is written once for every payer. No practice management or EHR connector exists today and there are no webhooks; integrations are in design on the roadmap.

Where healthcare teams need something else

• You need clinical decision support, results interpretation or any clinical use.
• You want medical codes assigned or audited. docRefinery extracts the codes printed and stops there.
• Your procurement requires a compliance attestation or a SOC 2 report today.
• Your remittances already arrive as electronic files. That data needs no document extraction.
• Most of your documents are handwritten clinical notes.