Handle intake form changes during open referrals with a fictional version decision table, partial-completion cases, effective dates, and buyer tests.
Changing an intake form during an open referral should preserve which version was presented and completed, while making the decision about any new information explicit rather than silently replacing earlier answers.
A new form can improve future intake while creating uncertainty for referrals already in progress. Some people have submitted the old version; others have a saved draft or a link they have not opened. The buyer needs to see how those different states are handled before approving a midstream template change.
Behavioral health intake form version open referrals
Separate the form definition from each person's response instance. Identify the version actually used, the date it was offered, and the completion state. A template's current label should not be the only evidence of what a person saw previously. Clinical leadership should determine the clinical meaning of changed questions and the appropriate response to missing or newly required clinical information.
Create a four-state version table
The fictional update replaces V1 with V2 on November 12. Person A submitted V1 on the 10th; B saved a V1 draft; C received a V1 link but has not started; D first receives a form on the 13th. These are four different decisions. A universal replace-everything action could erase A's original context or leave B's partial work uninterpretable. The table should state the approved treatment for each state and who determines any follow-up.
Describe the actual change before choosing a response
Distinguish a spelling correction, a changed response choice, a new administrative question, and a clinically important question. They need not have the same transition rule. Map old and new questions carefully if results will be compared. A field with the same name can change meaning, while a renamed field can retain the same meaning. Preserve the source response instead of converting answers through an undocumented assumption.
Handle partial completion as its own case
If B resumes a saved draft after the update, ask which questions appear and whether earlier answers still have their original meaning. If the organization chooses to request a new form, explain what happens to the earlier draft and what staff tell the person. Avoid asking the person to repeat sensitive information simply because the team cannot locate it. A new requirement should lead to the approved review, not a fabricated answer copied from a superficially similar field.
Test historical readability and later reports
Open A's completed V1 after V2 becomes current. The record should still make the earlier question and answer understandable. For reporting, identify whether a measure uses V1 responses, V2 responses, or a reviewed mapping. Unknown or incompatible responses should remain visible rather than being treated as no. A current template display is insufficient evidence of historical readability if it retroactively changes the wording around old answers.
Work through the fictional acceptance exercise
Use a four-state V1-to-V2 transition table for submitted, partially completed, unstarted, and newly issued intake forms. All examples below are fictional.
| Exercise | Starting evidence | Required action | Failure to expose |
|---|---|---|---|
| Submitted V1 | A completed the old version before the change | Preserve the original question-and-answer context | V2 wording silently reinterprets old answers |
| Saved draft | B resumes a partially completed V1 | Apply the approved draft transition rule | Saved answers appear under changed questions |
| Unopened link | C opens an older invitation after V2 launches | Explain which version is actually presented | The invitation date is assumed to prove the version |
| Cross-version report | A report combines V1 and V2 responses | Use a reviewed mapping or separate the results | Changed choices are treated as equivalent |
Boundary 1: Submitted V1
Historical responses need to remain intelligible on their original terms, even if staff request additional information later.
Boundary 2: Saved draft
The decision needs to account for both the person's prior work and the meaning of the updated questions.
Boundary 3: Unopened link
A sent link and a rendered form may be different facts. The demonstration should establish the actual behavior.
Boundary 4: Cross-version report
An apparently consistent field name does not establish comparable information across the version boundary.
A changed response choice cannot inherit a saved answer
V1 asks whether a person can attend daytime appointments, with Yes and No choices. B saves a V1 draft answering No. V2, activated November 12, replaces that question with a three-choice availability field: mornings, afternoons, or evenings. A's submitted V1 and D's newly issued V2 remain separate instances under the transition table.
When B resumes, the organization applies its reviewed rule to preserve the old draft and request the new availability answer explicitly. B selects evenings on V2. The earlier No remains attached to the earlier daytime question; it is not mapped automatically to evenings. No could also have meant irregular availability or another constraint that V2 does not capture.
A report using the old question can show B's historical No. A report using the new field can show B's newly supplied evenings answer from November 12 onward. A combined availability report requires the accepted cross-version definition and should distinguish missing V2 responses from actual answers.
Copying No into a newly numbered choice would be technically tidy but could assign an arbitrary time preference. Displaying the V2 wording around the saved V1 answer would change what the record appears to say B was asked. The resolved transition preserves both response contexts and identifies exactly which new fact was obtained, rather than treating a template update as patient-supplied information.
Observe the work before setting a target
AHRQ’s flowchart tool examines process handoffs.
AHRQ describes observation in simulated settings.
Rehearse the four states in synthetic records before changing a live form. Capture the wording shown at initial issue and after the update, then inspect the saved results outside the editing screen. Have the relevant administrative and clinical owners approve the transition choices. Document any manual follow-up needed for open referrals so staff know which people need additional information and which completed responses remain usable.
Questions to settle with the buying committee
Should a new form replace every open referral's old form? Decide separately for submitted, partial, unstarted, and newly issued instances. Preserve earlier evidence and apply an explicit transition rule.
Can answers be copied when question wording changes? Only use an appropriate reviewed mapping where meaning supports it. A similar field name alone does not establish equivalent answers.
Who decides whether a changed clinical question needs follow-up? Authorized clinical leadership should define the clinical response. Administrative template updates should not silently make that decision.
What should be tested after the new version is activated? Reopen historical responses, resume a saved draft, open an older invitation, and review cross-version reporting with synthetic data.
Bring the specimen to a full-platform review
Use the transition table during the platform configuration review. The acceptance decision should address historical meaning and people already in process, as well as the appearance of the new form.
Ease offers a unified CRM, EHR, and RCM platform for behavioral health. For the version-transition table, what historical question-and-answer context remains available to authorized CRM, EHR, and RCM staff after V2 becomes current?
Related reading
- Referral Documents Arrive Late: Route Them Correctly
- Intake Reassignment: Make Queue Ownership Explicit
Frequently Asked Questions
Should a new form replace every open referral's old form?
Decide separately for submitted, partial, unstarted, and newly issued instances. Preserve earlier evidence and apply an explicit transition rule.
Can answers be copied when question wording changes?
Only use an appropriate reviewed mapping where meaning supports it. A similar field name alone does not establish equivalent answers.
Who decides whether a changed clinical question needs follow-up?
Authorized clinical leadership should define the clinical response. Administrative template updates should not silently make that decision.
What should be tested after the new version is activated?
Reopen historical responses, resume a saved draft, open an older invitation, and review cross-version reporting with synthetic data.
