AI Playbook for Return-to-Duty Program Compliance
A DOT-regulated truck driver violated the alcohol policy (0.08 BAC on a pre-employment test). The driver completed a Substance Abuse Professional evaluation and was cleared for return-to-duty after completing treatment. The MRO must conduct a return-to-duty test and set up a follow-up testing program.
When to use this playbook
- Use this playbook when the decision looks like the situation above: A DOT-regulated truck driver violated the alcohol policy (0.08 BAC on a pre-employment test).
- It is a fit when you have source files in hand and need a structured, reviewable analysis — not a generic chat answer about "Return-to-Duty Program Compliance".
- Do not use it as a substitute for licensed, legal, clinical, or authorized official judgment in the domain.
What you'll need
- SAP evaluation and treatment completion documentation
- SAP follow-up testing recommendation
- DOT 49 CFR Part 40 return-to-duty requirements
- FMCSA Clearinghouse registration status
- Employer's return-to-duty checklist
Attachments: Documents (Documents)
The Prompt
You are a Medical Review Officer managing the return-to-duty process for a DOT alcohol violation. I am attaching: Work only from the attached source files. If a conclusion is not supported, say so. Produce: 1. Verify the SAP documentation: is the evaluation and treatment documentation sufficient to authorize a return-to-duty test? 2. Confirm the return-to-duty test requirements: direct observation, which substances, and the required result before the driver can return to safety-sensitive duties. 3. Design the follow-up testing plan: minimum frequency, duration, and direct observation requirements under DOT guidelines. 4. Assess the FMCSA Clearinghouse obligations: what must be reported, by whom, and when? 5. Tell me the complete return-to-duty checklist in sequence and what happens if the driver tests positive on the return-to-duty test. Call out where independent models are likely to disagree, and list follow-up documents a reviewer should request.
What to expect
- SAP documentation sufficiency assessment
- Return-to-duty test requirements
- Follow-up testing plan design
- FMCSA Clearinghouse reporting obligations
- Sequential return-to-duty checklist and positive result consequences
Review before you act
- Validate this output against source files before relying on it: Verify the SAP documentation: is the evaluation and treatment documentation sufficient to authorize a return-to-duty test?.
- Validate this output against source files before relying on it: Confirm the return-to-duty test requirements: direct observation, which substances, and the required result before the driver can return to safety-sensitive duties.
- Validate this output against source files before relying on it: Design the follow-up testing plan: minimum frequency, duration, and direct observation requirements under DOT guidelines.
- Validate this output against source files before relying on it: Assess the FMCSA Clearinghouse obligations: what must be reported, by whom, and when?.
- Confirm every cited figure, date, counterparty, or requirement against the attached originals — models compress and can drop a qualifier.
- Treat disagreement between models as a review item, especially on classification, materiality, and recommended next action.
- Do not authorize an operational, clinical, legal, credit, or enforcement action solely because the models agree.
Why compare models on this
For Return-to-Duty Program Compliance, running the same attachments across independent models is useful because the hard part is classification and completeness, not fluency. The workflow is already designed to surface sap documentation sufficiency assessment; return-to-duty test requirements; follow-up testing plan design; fmcsa clearinghouse reporting obligations. Those are comparison artifacts — they only exist if more than one model runs. Models disagree on whether an irregularity is fatal to custody, whether a prescription explains a result, and whether observation is authorized. Those splits are MRO work, not auto-verification.
See governed multi-model AI on your own prompt
Compare GPT-5, Claude, and Gemini side by side, with human review and a decision record built in.

