Skip to content

Four workflows, before and after

Each example shows who does which step today and who does it with OpenAdapt. Where a published study measured the "today" time, we cite its source and year. The volumes are assumptions, so replace them with your own.

We don't show an "after" time for these workflows. We measure it with your own cases in your first project.

To estimate what a task costs your team today:

staff hours per month = items per month × minutes per item ÷ 60

1. Referral entered into an EMR over a desktop client or Citrix

A referral arrives by fax or as an electronic referral. A staff member keys it into the EMR, often a desktop EMR that staff reach through Citrix or remote desktop.

Step Today With OpenAdapt
Read the referral and decide what goes in Staff, or an AI tool that reads faxes No change
Find or create the patient Staff search the EMR OpenAdapt searches and confirms it's the right patient before it types. If two patients could match, it stops.
Enter demographics, coverage, and the referral Staff type each field OpenAdapt types the approved fields into the same EMR screens
Confirm it saved Staff trust the screen OpenAdapt reads the referral back through a second path and compares it
Missing or unclear information Staff call the referring office or the patient No change. OpenAdapt stops and hands the referral to staff.

Today, in numbers:

  • A referral takes about 10 to 12 minutes of staff time to enter. A time-and-motion study at UCSF's pediatric access center measured 719 seconds on average for a faxed referral, including calls to the referrer or the patient when information was missing (JAMIA Open, 2020). Calderdale and Huddersfield NHS Foundation Trust reported 10 minutes per referral to move the data into its patient record system (HFMA case study, 2024).
  • Assumption: a 10-physician clinic that receives 20 referrals per physician a week handles about 867 referrals a month. At 11 minutes each, that's about 159 staff hours a month.

What a person still does: unclear faxes, patients who don't match, missing fields, and the phone calls. For a rough guide, when the NHS trust above automated this task with a conventional RPA bot, about 5% of referrals still went to staff. That's a different product, so don't read it as a forecast for OpenAdapt.

Over Citrix, OpenAdapt sees the session window as pixels, the same way your staff do. Our public Citrix test used a stand-in session, so we run the readiness test in your own Citrix environment during the first project. The record check needs a second way to read the referral, such as a report or a read-only login. Without one, OpenAdapt can't mark the entry done and checked.

When we show this workflow, we use made-up patients in a test EMR, such as the open-source OpenEMR. We don't show any commercial EMR vendor's product or branding, and naming an EMR in these docs doesn't imply a partnership.

What to measure: minutes per referral today, referrals per week, how many runs stop for review, and how long a stop takes to answer.

2. Billing audit evidence

Some practices audit their billing every month. They compare the procedures documented in charts with the procedure codes on the month's billing report, and follow up on anything missing.

Step Today With OpenAdapt
Load the month's billing report Physician or billing staff OpenAdapt loads the report spreadsheet
Collect the clinical notes Open each chart in the EMR, one patient and service date at a time OpenAdapt opens the charts and collects the notes
Compare documented procedures with billed codes By hand, in a spreadsheet OpenAdapt lines them up row by row in a workbook
Write up missing codes By hand OpenAdapt drafts a list for billing follow-up
Decide what to submit Physician or billing staff No change. A person reviews each flagged item with its saved evidence.

We didn't find a published time study for this audit, so this example has no "today" time.

This workflow reads records and produces a report. It doesn't write into the EMR.

What to measure: charts reviewed per month, minutes per chart, how many items the audit flags, and how many runs stop for review.

3. Eligibility check in a payer portal (US)

Step Today With OpenAdapt
Open the portal and sign in Staff OpenAdapt, with the password supplied from your computer at run time
Find the member Staff search OpenAdapt searches and confirms the member matches the patient
Read the benefits Staff OpenAdapt reads the agreed fields
Record the result in your practice system Staff type it in OpenAdapt enters it and reads the entry back
The portal shows something the recorded task didn't cover Staff work it out OpenAdapt stops and sends it to staff

Today, in numbers:

  • A portal eligibility check takes medical practices about 8 minutes of staff time, or $4.46 in labor. By phone, it's about 16 minutes and $8.57. A dental portal check takes about 7 minutes (2024 CAQH Index, data year 2023).
  • Assumption: 600 portal checks a month at 8 minutes is about 80 staff hours.

Health plans report that about 96% of medical eligibility checks are already fully electronic (same report). If your practice system gets eligibility that way, keep it. This example is for the checks that still happen in a portal.

Eligibility results are patient data, so this workflow runs on a computer you control. Our Hosted plan is for browser work with non-regulated data only.

What to measure: portal checks per day, minutes per check, and how many runs stop for review.

4. Claim status

Step Today With OpenAdapt
Pick claims past their expected payment date Billing staff or a work-queue report No change
Look up each claim Staff search the payer portal, or call OpenAdapt searches the portal
Record the status Staff type it into the billing system or work queue OpenAdapt enters it and reads it back
Denials, appeals, and phone calls Billing staff No change. Claims that OpenAdapt can't read clearly go to staff.

Today, in numbers:

  • A claim status check in a payer portal takes about 10 minutes of staff time and $5.24 in labor. By phone, it's about 25 minutes and $13.80 (2024 CAQH Index).
  • Assumption: 150 portal look-ups a month at 10 minutes is 25 staff hours.

Health plans report that about 80% of medical claim status inquiries are fully electronic (same report). If your clearinghouse returns claim status that way, keep it. This example is for the portal checks that remain.

What to measure: look-ups per week, minutes per look-up, and how many runs stop for review.