Follow-up after digital consultations
Personal proposal by Per-Arne Andersen
H7 · Spring 2027
https://thesis.uya.no/proposals/close-the-follow-up-gap/

Turn consultation notes into confirmed follow-up tasks with an owner and due date.

TECHNICAL
- Extract follow-up actions with Qwen2.5-7B-Instruct and one fixed prompt.
- Build a confirmation screen with source sentence, owner, deadline and completion status.

DATA (planned)
30 fictional consultation notes
Write 150–250 words per note with 0–4 actions. Label action, owner and deadline separately; include 10 notes with missing owners or dates.
Qwen2.5-7B-Instruct: https://huggingface.co/Qwen/Qwen2.5-7B-Instruct
Transformers generation settings: https://huggingface.co/docs/transformers/main_classes/text_generation

METHOD
Use 10 notes for prompt development and 20 held out. Report action precision/recall and owner/date accuracy; compare with manual extraction by 6 students.

OUTPUT
Task-list prototype, labelled notes and extraction results.

BACHELOR
Task: Build a structured follow-up form that links actions to their source note.
Data: 30 fictional consultation notes (planned)
Write 150–250 words per note with 0–4 actions. Label action, owner and deadline separately; include 10 notes with missing owners or dates.
Requires: Clinical review of scripts; recruit students. No patient notes.
Method: Enter actions manually from the fictional notes; check missing-owner/date handling and observe 4 task sessions.
Output: A working follow-up checklist with tested validation rules.
Use relevant literature to justify the established approach; a new research contribution is not the aim of this proposal.

MASTER
Study whether confirmation reduces model extraction errors without creating false responsibility. Compare raw, human-confirmed and manual outputs; analyse unknown owners/dates separately and use source evidence to explain corrections.
Intended contribution: Evidence about how confirmation changes accuracy and accountability in follow-up work.

STARTING PAPERS
Amershi et al. (2019) — Guidelines for Human-AI Interaction: https://www.microsoft.com/en-us/research/wp-content/uploads/2019/01/Guidelines-for-Human-AI-Interaction-camera-ready.pdf
Select and justify the correction, explanation or uncertainty design being tested.
Buçinca et al. (2021) — To Trust or to Think: https://arxiv.org/abs/2102.09692
Derive a competing explanation based on verification effort and overreliance.

Search Scopus or Web of Science and ACM Digital Library using the topic query, then follow citations to the thesis start date. Record searches and compare methods, data, findings and limitations in literature-matrix.csv. Use that review to confirm or revise the gap and choose a current comparator. The linked papers are starting points.

START HERE
Tools: Qwen2.5-7B-Instruct, Python, Streamlit
1. Write 3 fictional notes and manually label every action, including explicit missing values. Confirm access to inference compute and record model, software and hardware versions.
2. Create the development/held-out split before trying prompts; do not inspect held-out results during tuning.
3. Pilot action confirmation on 3 development notes and fix the scoring rules.
Literature search: clinical consultation follow up task extraction accountability
Study controls:
- For generation, use do_sample=False and num_beams=1; record model/tokenizer revisions, runtime, quantisation, prompt and output limit. Keep failed outputs.
- Before collecting participant data, agree consent, storage and withdrawal handling with the supervisor. Use participant codes, not names, in study files.
- Include note dates, negated/historical actions and genuine no-action cases. Stratify the 10-development/20-held-out split, keeping scenario families separate. Clinical reviewers define action boundaries and acceptable unknowns.
- Report raw model, human-confirmed and manual output separately, including invented owners/dates. Student participants do not define the clinical answer key; never automatically dispatch follow-ups.
- For the master’s study, use the research task above to define the factors and comparisons in this pilot plan. Preregister one primary outcome and feasible scope after the literature review; do not add every possible model or interface variant.

REQUIRES
Clinical review of scripts; recruit students. No patient notes.

--------------------

NON-TECHNICAL
- Prepare paper consultation notes and follow-up lists.
- Ask staff to assign responsibility and identify actions at risk of being missed.

DATA (planned)
12 scripted handovers + 8 staff interviews
Create 4 clear cases, 4 with missing owners and 4 with missing deadlines. Record intended actions separately.


METHOD
Think-aloud review; count missed actions and code responsibility disputes and handover failures.

OUTPUT
Handover checklist and documented causes of missed follow-up.

BACHELOR
Task: Find where follow-up actions lose their owner or deadline.
Data: 12 scripted handovers (planned)
Create 4 clear cases, 4 with missing owners and 4 with missing deadlines. Record intended actions separately.
Requires: Recruit health professionals and review the fictional cases.
Method: Use paper handovers in 4 staff walkthroughs; describe missing information and responsibility hand-offs.
Output: A revised handover checklist and workflow map.
Use relevant literature to justify the established approach; a new research contribution is not the aim of this proposal.

MASTER
Explain why action tables resolve some handover failures but leave others untouched. Contrast information-format problems with disputed responsibility using matched narrative/table cases.
Intended contribution: An explanatory model separating representation failures from ownership failures.

STARTING PAPERS
Vessey & Galletta (1991) — Cognitive Fit: https://pubsonline.informs.org/doi/10.1287/isre.2.1.63
Test whether a representation helps one kind of task more than another.
Orlikowski & Gash (1994) — Technological Frames: https://dl.acm.org/doi/10.1145/196734.196745
Compare how roles interpret the purpose, operation and use of the same system.

Search Scopus or Web of Science and ACM Digital Library using the topic query, then follow citations to the thesis start date. Record searches and compare methods, data, findings and limitations in literature-matrix.csv. Use that review to confirm or revise the gap and choose a current comparator. The linked papers are starting points.

START HERE
Tools: LibreOffice Writer/Calc, audio recorder with consent; no programming required.
1. Prepare a pilot with 2 examples from: 12 scripted handovers + 8 staff interviews. Write the task questions and a reference answer sheet.
2. Write a recruitment message, information sheet and consent form for the participants named above. Agree privacy handling with the supervisor before contact.
3. Pilot one session after approval; revise unclear questions, freeze the task sets and coding categories, then recruit the planned sample.
Literature search: clinical consultation follow up task extraction accountability qualitative scenario study
Study controls:
- Before collecting participant data, agree consent, storage and withdrawal handling with the supervisor. Use participant codes, not names, in study files.
- Pilot separately, then freeze the questions and coding plan. Check objective answer keys independently; keep an audit trail of coding, including disagreements.
- For comparisons, counterbalance order and case assignment; do not show a person both versions of one case. Report participant-level findings, not repeated tasks as independent people.
- Narrative and action-table versions must contain identical facts, including the same missing responsibilities. Do not repair missing information in only one version.

REQUIRES
Recruit health professionals and review the fictional cases.

Bachelor: apply established methods and evaluate a practical solution or study. Master: position a research question in current scientific literature, investigate a mechanism or unresolved problem, and explain the contribution. Final scope is agreed with me.
Study sizes are proposed; participant recruitment and planned materials are not already arranged.