LearnGrok
Prompts
PromptIntermediateGetting started

DNA follow-up list for service recovery

Create a safe, policy-led DNA follow-up list from de-identified records for clinic administrators and service leads.

4 min read

Nothing here is clinical advice, and nothing here should touch identifiable patient information.

Use these prompts to turn a de-identified missed-appointment extract into work that can be allocated. They are for clinic administrators and service leads working under an approved local non-attendance policy.

Do not paste names, contact details, dates of birth, NHS numbers, free-text clinical notes, or any other identifiable patient information. Use a local case ID and appointment ID that your team can map back inside its approved system.

Key point

Start with the policy

The records tell you what happened. The local non-attendance policy tells you what administrative action is allowed.

Prepare the two inputs

Before you use the prompts, create two separate documents or exports:

  • A de-identified missed-appointment extract. Keep one row per appointment. Include the fields named in the first prompt, even where the value is blank.
  • The current local non-attendance policy. Include the relevant decision tree, contact-attempt rules, rebooking rules, escalation triggers, owner roles, and stated timescales.

Keep the policy text with the task. A model should not be asked to reconstruct local rules from general practice. If the policy is a long document, paste the sections on non-attendance and the local service appendix. Check that the appendix applies to the service in your extract.

Capabilities and input handling can vary with the model and product configuration. Check the current guidance in the xAI documentation overview before using a workflow with operational data.

Watch out

Remove free text

A notes field often contains identifiers or clinical detail. Exclude it unless your information-governance process has approved a de-identified, administrative extract.

Run the prompts in order

  1. Run Check the missed-appointment extract. Correct duplicate rows, contradictory attendance outcomes and missing decision fields in the source system where possible. Do not treat a clean-looking table as a corrected record.
  2. Run Classify each record under local policy. This is the master list. It gives every appointment one primary route: Contact, Rebook, Escalate or Needs verification.
  3. Split the resulting rows by primary route. Run Create the patient-contact worklist for Contact rows, Prepare the rebooking queue for Rebook rows, and Build the escalation register for Escalate and Needs verification rows.
  4. Run Reconcile the final follow-up lists using the source extract and all four outputs. Do not distribute work until this reports that the lists are ready.

The route labels are deliberately administrative. They do not describe a person's health need, priority of care, or suitability for any treatment. A safeguarding flag or communication flag may trigger a policy route, but the prompt must cite the policy rule rather than make its own judgement.

Decide which list owns the work

Use the master decision list as the record of why each appointment went where it did. The operational lists are for teams to act on.

If the record says Put it in What the team does
Contact Contact worklist Complete only the policy-approved contact action and record its outcome.
Rebook Rebooking queue Apply the stated booking action or obtain the required approval.
Escalate Escalation register Send it to the policy-named owner with the decision required.
Needs verification Escalation register Correct the data or obtain a policy decision before any action.

Do not use Needs verification as a quiet holding area. Give it an owner and a next action. If a rule is unclear, the appropriate local policy owner decides what it means.

Check

Check the counts

The total number of primary routes must equal the number of source appointment IDs. Every difference is an exception to resolve, not a rounding issue.

Check whether the output is wrong

Read five to ten rows from each list against both the source extract and the policy. Look for errors that can appear plausible in a neat table:

  • An appointment marked as cancelled may have been treated as a DNA.
  • A previous-DNA count may use the wrong policy period.
  • A row may cite a policy section that does not actually permit the action shown.
  • A Contact record may have exceeded the policy contact-attempt limit.
  • A Rebook record may already have a booking recorded in the source system.
  • An escalation may name a team that is not named by the policy.
  • The same appointment ID may appear in more than one active worklist.

Pay particular attention to rows labelled Needs verification. If many records land there, the problem may be the extract design or missing policy text, not the records themselves. Do not force a route simply to reduce the exception count.

Note

Keep a decision trail

Save the policy version used, the de-identified input extract, the master decision list, and the reconciliation result in your approved local record location.

When the workflow does not work

Stop if the policy is missing, the source contains identifiable information, or the output assigns actions without a cited policy rule. Remove the data from the conversation, return to your approved source system, and ask the local policy owner or information-governance lead for the missing direction.

If the lists do not reconcile, fix the source extract or the master decision list first. Then rerun only the affected operational list and the final reconciliation. Do not manually patch a distributed list without recording the corrected route and reason.

Copy-ready prompts

6 prompts. Open one to read it, or take the whole pack.

1Check the missed-appointment extractUse this first, before assigning any follow-up work. It finds missing fields and duplicate rows that would make the later lists unreliable.
You are checking a de-identified missed-appointment extract for administrative data quality. Do not provide clinical advice, infer health needs, or request names, addresses, dates of birth, contact details, or any other identifiable patient information.

Review the records below. Each record should contain: case ID, appointment ID, service, appointment date, appointment type, attendance outcome, cancellation timestamp, DNA recorded timestamp, previous DNA count within the local policy period, contact attempt count, latest contact outcome, rebooking status, interpreter or communication flag where recorded, safeguarding or vulnerability flag where recorded, and policy reference.

Identify only administrative data issues. Do not fill gaps by assumption. Treat blank, unclear, conflicting, or invalid values as `Needs verification`.

Return two Markdown tables:

1. `Record issues`
Columns: case ID, appointment ID, field with issue, issue found, effect on follow-up decision, required source to check.

2. `Extract summary`
Columns: check, count, affected case IDs, action before classification.
Include checks for duplicate appointment IDs, missing outcome, inconsistent cancellation and DNA status, missing previous-DNA count, missing policy reference, and missing rebooking status.

End with a short section titled `Safe to classify` listing only the appointment IDs with no decision-blocking issue. Do not classify records into contact, rebook, or escalation lists yet.

[Paste de-identified missed-appointment records here]
2Classify each record under local policyUse this once the extract has passed the data checks. This creates the master decision list that separates contact, rebooking and escalation work.
You are creating an administrative follow-up decision list from de-identified missed-appointment records. Apply only the local non-attendance policy supplied below. Do not provide clinical advice, judge urgency, infer risk, or make a decision that is not explicitly supported by the policy or the record. Do not request or include identifiable patient information.

For every appointment record, assign exactly one primary route:
- `Contact`: the policy requires or permits an administrative contact attempt before further action.
- `Rebook`: the policy permits rebooking or placing the appointment in a rebooking workflow.
- `Escalate`: the policy requires review by the named local role, team, or pathway.
- `Needs verification`: a required field, policy condition, or policy reference is missing, unclear, or contradictory.

A record may have a secondary action, but never use a secondary action to replace the primary route. If the policy allows more than one route and gives no order, choose `Needs verification` and explain the conflict. Quote the relevant policy section or rule reference exactly as supplied. Do not invent a policy rule.

Return one Markdown table titled `DNA follow-up decision list`.
Columns: case ID, appointment ID, service, appointment date, primary route, secondary action, policy rule or section, evidence from record, owner role, target timescale stated in policy, decision confidence, verification needed.

After the table, return a `Totals` bullet list with the number of records in each primary route. Then return `Unresolved policy questions` as a numbered list. Include only questions that block a decision.

Local non-attendance policy:
[Paste the relevant current policy text, decision tree, and local owner roles here]

De-identified missed-appointment records:
[Paste checked records here]
3Create the patient-contact worklistUse this for records already routed to Contact. Give the list to the team that manages approved administrative contact attempts.
Create an administrative contact worklist from the de-identified records below. This is not clinical advice. Do not write a clinical assessment, infer why an appointment was missed, recommend a care action, or include identifiable patient information.

Include only records where `primary route` is `Contact`. Follow the supplied local non-attendance policy exactly. If a permitted contact method, contact-attempt limit, wording requirement, or target timescale is absent or unclear, write `Needs verification` rather than guessing.

Return one Markdown table titled `Contact worklist`.
Columns: priority order, case ID, appointment ID, service, policy rule or section, required administrative action, approved contact channel if specified, contact attempt number, target completion date or policy timescale, owner role, standard outcome to record, status, blocker.

Use `Do not contact until verified` in the status column where the record or policy does not clearly authorise the planned action. Sort first by policy timescale, then by appointment date, then by case ID. Do not create message text unless the policy text explicitly requires it.

End with a section titled `Before sending work` containing:
- records with missing contact authorisation or method
- records at or beyond the policy contact-attempt limit
- records whose policy route conflicts with the supplied decision list

Local non-attendance policy:
[Paste relevant policy text here]

De-identified Contact records from the decision list:
[Paste records here]
4Prepare the rebooking queueUse this for records already routed to Rebook. It produces a queue that booking staff can work without having to reinterpret the policy.
Prepare a de-identified administrative rebooking queue from the records below. Apply only the supplied local non-attendance policy. This is not clinical advice: do not select a clinical priority, change a pathway, or infer whether an appointment is needed. Do not include or request identifiable patient information.

Include only records where `primary route` is `Rebook`. Preserve the service, appointment type, and any policy-approved rebooking condition exactly as recorded. If the policy does not state whether rebooking is permitted, what appointment type to use, who must approve it, or the applicable timescale, classify the row as `Needs verification`.

Return one Markdown table titled `Rebooking queue`.
Columns: queue order, case ID, appointment ID, service, original appointment type, policy rule or section, rebooking action, approval required from, policy timescale, booking constraint stated in record or policy, owner role, queue status, verification needed.

Allowed queue statuses are: `Ready for booking`, `Awaiting approval`, `Needs verification`, and `Do not rebook pending escalation`. Do not invent dates, appointment slots, specialties, or patient preferences. Sort `Ready for booking` records by policy timescale, then original appointment date.

End with `Queue controls`, a bullet list covering duplicate bookings to check, records with an existing rebooking status, and records blocked by missing approval.

Local non-attendance policy:
[Paste relevant policy text here]

De-identified Rebook records from the decision list:
[Paste records here]
5Build the escalation registerUse this for records already routed to Escalate, plus records that cannot be safely classified because the policy or data is unclear.
Build an administrative escalation register from the de-identified records below. Apply only the local non-attendance policy provided. This is not clinical advice. Do not assess clinical risk, state that a person is safe or unsafe, or recommend a clinical intervention. Do not include identifiable patient information.

Include records where `primary route` is `Escalate` or `Needs verification`. Escalate only to the owner, team, or pathway named in the policy or supplied record. If neither names an owner, state `Escalation owner not specified, verify locally`. Do not substitute a role based on assumption.

Return one Markdown table titled `Escalation register`.
Columns: escalation order, case ID, appointment ID, service, escalation reason, triggering record field or policy rule, required decision, policy-named owner, policy timescale, information missing, interim administrative status, escalation status.

Allowed interim administrative statuses are: `Awaiting policy decision`, `Awaiting data correction`, `Awaiting named owner review`, and `Action prohibited pending review`. Sort by stated policy timescale. If no timescale is stated, place the record after timed cases and mark it `Needs local timescale`.

Then return `Escalation handover notes` as a numbered list. Each item must contain the case ID, the exact decision needed, and the specific evidence available. Keep each item to two sentences or fewer.

Local non-attendance policy:
[Paste relevant policy text here]

De-identified Escalate and Needs verification records:
[Paste records here]
6Reconcile the final follow-up listsUse this before distributing work. It checks that every missed appointment has one accountable route and that no record appears in the wrong queue.
Reconcile the de-identified missed-appointment source list against the Contact worklist, Rebooking queue, Escalation register, and master decision list below. This is an administrative completeness check only. Do not provide clinical advice, infer missing information, or include identifiable patient information.

Match records using appointment ID first and case ID second. Each source appointment must have exactly one primary route in the master decision list: `Contact`, `Rebook`, `Escalate`, or `Needs verification`. A record may appear in an operational list only if that list matches its primary route, except that `Needs verification` records may appear in the Escalation register.

Return three Markdown tables:

1. `Reconciliation exceptions`
Columns: appointment ID, case ID, exception type, source route, list where found, likely cause, action required, owner role.

2. `Route totals`
Columns: route, source count, operational-list count, difference, pass or fail.

3. `Distribution-ready lists`
Columns: list name, included appointment IDs, excluded appointment IDs, reason for exclusion, release status.

Use only these release statuses: `Ready to distribute`, `Hold for correction`, and `Hold for policy review`. End with a single line in this format: `Overall release status: [status]`. Set it to `Ready to distribute` only if there are no unmatched records, duplicate primary routes, route mismatches, or unresolved decision-blocking fields.

De-identified source missed-appointment list:
[Paste source records here]

Master decision list:
[Paste decision list here]

Contact worklist:
[Paste contact worklist here]

Rebooking queue:
[Paste rebooking queue here]

Escalation register:
[Paste escalation register here]

Last checked against xAI’s own pages on 2026-08-21. Grok changes quickly; anything version-specific should be confirmed upstream before you rely on it.

More in Getting started

Found something out of date?

Grok changes quickly and this page is a snapshot. If something here is wrong, or you know a better resource, send it over.

Suggest a link →

Advertise on LearnGrok

$420.69one-time, for a 30-day run

Square works best. PNG, JPEG or WebP, up to 2 MB.

Stripe on the next step. Live once approved.