AI and Referral Letter Completeness: Specialist-Ready Drafts

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-creators-healthcare-AI-and-referral-letter-completeness-r11a3-adults

1. A clinician asks an AI tool to review a draft neurology referral. What is the most appropriate task to assign to the AI?

   1. Write the clinical reasoning explaining why the patient needs a neurology consultation
   2. Decide whether the referral should be marked routine or urgent
   3. Determine if the patient's condition is medically necessary for specialist evaluation
   4. Compare the draft against the neurology intake checklist and identify missing required items

2. A cardiology referral receives a perfect score from an AI checklist tool. However, the patient actually needed GI evaluation instead. What does this scenario illustrate?

   1. A complete referral can still be the wrong referral based on indication
   2. The checklist should have included different criteria
   3. The AI tool made an error in its assessment
   4. The specialist rejected the referral unnecessarily

3. Which of the following is identified in the lesson as something AI explicitly cannot do when processing referral letters?

   1. Write the clinical reasoning for the referral
   2. Score a referral against specialty-specific intake criteria
   3. Flag items that don't match the specialty's requirements
   4. Identify missing lab results or imaging reports

4. What is the primary reason AI cannot determine urgency level for a specialist referral?

   1. Specialty intake checklists already include urgency classifications
   2. AI is not designed to communicate with specialist offices
   3. AI systems lack access to real-time patient vital signs
   4. Urgency determination requires clinical judgment about the individual patient's condition

5. The lesson emphasizes that the referring clinician 'owns the clinical narrative.' What does this mean in practice?

   1. The AI system cannot access the patient's electronic health record
   2. The specialist must accept referrals only from the referring clinician who wrote the narrative
   3. The clinical narrative belongs to the hospital rather than the individual clinician
   4. The clinician is responsible for documenting the complete medical history in the referral

6. A clinician wants to use AI to improve a referral letter for a patient with suspected inflammatory bowel disease. Which AI task would be appropriate based on the lesson?

   1. Ask AI to decide whether this is a routine or urgent GI consultation
   2. Ask AI to write the indication based on the patient's symptoms
   3. Ask AI to review the draft against GI intake requirements and list missing items
   4. Ask AI to determine if the patient should see a gastroenterologist or colorectal surgeon

7. An AI tool flags that a cardiology referral is missing recent echocardiogram results. What type of information is this?

   1. A concrete data element that can be verified against the intake checklist
   2. A determination of whether the referral is urgent
   3. Clinical reasoning that explains why cardiology referral is needed
   4. A subjective assessment of the patient's cardiac risk

8. Why does the lesson caution that AI checklist scores alone are insufficient for referral quality?

   1. Specialists typically ignore AI-generated checklist scores
   2. AI tools frequently make errors in counting checklist items
   3. Most referral intake checklists are not evidence-based
   4. Checklists cannot capture clinical judgment about indication and urgency

9. A clinician receives an AI-generated checklist showing three missing items for a cardiology referral. The clinician adds the items and resends to AI. What should the clinician expect from this second review?

   1. AI will verify the added items and confirm the checklist is clear
   2. AI will write a new clinical narrative for the clinician
   3. AI will determine the urgency level for the referral
   4. AI will now approve the referral as appropriate

10. The lesson provides an example prompt for using AI with referrals. What specific output format does the example suggest requesting?

    1. A table comparing the referral to three different specialty checklists
    2. A score from 0-100 indicating referral quality
    3. A numbered checklist of missing items the clinician can address
    4. A paragraph summarizing the patient's medical history

11. A referring clinician wants to ensure a gastroenterology referral is ready for specialist review. Which action aligns best with the lesson's guidance on AI use?

    1. Use AI to write the clinical indication and reasoning
    2. Use AI to decide if the patient needs GI versus colorectal surgery
    3. Use AI to compare the draft against GI intake requirements and identify gaps
    4. Use AI to set the appointment priority based on the checklist score

12. The lesson notes that 'the indication and urgency are clinician judgment.' What does 'indication' refer to in this context?

    1. The timeframe for scheduling the appointment
    2. The patient's insurance coverage information
    3. The clinical reason why the specialist consultation is needed
    4. The list of required documents for the referral

13. What would be an inappropriate use of AI in the referral workflow, based on the lesson?

    1. Using AI to identify which labs are missing for a cardiology referral
    2. Using AI to check that demographic information is complete
    3. Using AI to verify all required imaging is attached
    4. Using AI to write the clinical reasoning that justifies the referral

14. A patient referral scores 85% on an AI checklist—it has most but not all required items. What does this indicate to the clinician?

    1. The referral is inappropriate and should be cancelled
    2. The referral is complete and can be sent as-is
    3. There are specific gaps the clinician should address before sending
    4. The AI system is malfunctioning and needs recalibration

15. When might a specialist-ready referral still fail to meet the specialist's needs, even after passing an AI completeness check?

    1. When the referral is sent to the wrong fax number
    2. When the AI used the wrong specialty checklist
    3. When the clinical narrative lacks sufficient detail or accuracy for the specialist to act
    4. When the patient demographic information contains typos