# Economic Evaluation in Health: faculty workshop blueprint

**Wednesday 23 September 2026, 09:30 to 16:30**
SAMVAD (Board Room, First Floor, Medical College Building), AIIMS Bhopal, Saket Nagar.
Enquiries: hta@aiimsbhopal.edu.in
School of Public Health, with the Regional Resource Centre for HTA
About 25 participants: faculty, senior and junior residents, and research project staff, from mixed
departments.

This is the run sheet for facilitators. Everything needed for the day is in this repository, which is
self-contained and works offline. The five teaching blocks came from the postgraduate course
*Introduction to Economic Evaluation in Health* (a separate repository) and were adapted here for a
faculty audience; that course's own blueprint stays with it.

---

## 1. What the day is for

Participants leave able to pick up a published economic evaluation in their own field and say
whether they believe it. The framing is **reading and appraisal**, not justifying a department's
purchases. Cervical cancer stays the worked example in the taught blocks. Participants bring their own
specialty through the paper each group holds all day.

## 2. The mechanic

- **8 groups of 3** (one group of 4 if 25 turn up), formed **by specialty cluster** from the
  department field on the registration form. Two groups per cluster.
- **Mix levels within each group.** Put at least one faculty member or senior resident with the junior
  residents and project staff in every group, so the reading is shared rather than left to whoever is
  most junior. Sort the registration list by designation before you assign tables.
- Each group gets **one paper at registration** (the paper pack), the **appraisal checklist**, and
  its paper's one-page sheet with a prompt for every session.
- In the **last 3 minutes of each session**, groups apply what was just taught to their paper and
  fill in the checklist rows. The facilitator says so; the last slide of each new deck prompts it.
- By **Session 9** the checklist is full. The report-back is in **four pairs, 6 minutes each**.

| Cluster | Groups | Papers |
|---|---|---|
| Oncology and surgery | 1, 2 | Thiagarajan 2026 (SLNB, oral cancer); Guleria 2025 (laparoscopic IPOM) |
| Cardiology and medicine | 3, 4 | Uy 2021 (rheumatic mitral valve surgery); Kaur 2022 (diabetes and hypertension screening) |
| Critical care, anaesthesia, neonatology | 5, 6 | Sharma 2016 (Kangaroo Ward Care); Patel 2015 (ward NIV for COPD) |
| Paediatrics, infectious disease, nephrology | 7, 8 | Gupta 2022 (peritoneal dialysis first); Srinivasan 2024 (childhood AML) |

If a cluster is under-subscribed, move a group across rather than leave a paper unread; the pairs
matter more than the specialty match.

## 3. Run sheet

"Skip" means advance past the slide without showing it. Nothing is deleted from the decks.
Poll numbers refer to `poll-bank.html`. The deck file names (`block1`, `block2`, `block3a`, `block3b`,
`block3c`) are inherited from the postgraduate course these were adapted from; they no longer run in
that order, so go by the session number, not the file name.

| # | Clock | Min | Session | Deck | Skip | Polls | Lead |
|---|---|---|---|---|---|---|---|
| | 09:30 | 10 | Registration and welcome | none | | | |
| **1** | 09:40 | 20 | Where economic evidence enters health policy | `slides/opening.html` (12 slides) | | Q1, Q2 | |
| **2** | 10:00 | 40 | What economic evaluation is, and what it is not | `slides/block1.html` (28 slides) | 12, 13, 22, 24; 14 optional | Q3 to Q5 | |
| **3** | 10:40 | 40 | Framing the question | `slides/block3a.html` (18 slides) | 13; 10 and 11 brisk | Q6 to Q8 | |
| | 11:20 | 20 | *Tea* | | | | |
| **4** | 11:40 | 55 | Costing your own service | `slides/costing.html` (17 slides) + worksheet | 4 and 9 brisk (named in Session 3) | Q9 to Q11 | |
| **5** | 12:35 | 40 | Outcomes, discounting and the ICER | `slides/block3b.html` (26 slides) | 8, 11, 17 | Q12 to Q14 | |
| | 13:15 | 50 | *Lunch* | | | | |
| **6** | 14:05 | 40 | Thresholds, opportunity cost and budget impact | `slides/block2.html` (23 slides) | 11 brief (seen in Session 1); 12 and 13 optional | Q15 to Q17 | |
| **7** | 14:45 | 25 | Modelling in half an hour | `slides/modelling.html` (15 slides) | | Q18, Q19 | |
| | 15:10 | 15 | *Tea* | | | | |
| **8** | 15:25 | 30 | Handling uncertainty | `slides/block3c.html` (22 slides) | 18; 5 brief | Q20 to Q22 | |
| **9** | 15:55 | 25 | Group appraisal report-back | none | | | |
| **10** | 16:20 | 10 | Close and feedback | none | | Q23 | |

### The shape of the day, and why it is in this order

The morning takes an ICER apart and builds one; only after lunch does the day ask what an ICER is worth.
So framing comes before costing, costing before outcomes, and the threshold session comes sixth rather
than third. Two consequences to hold in mind:

- **Session 5 has to name a threshold before Session 6 explains it.** The "So what do we fund?" slide
  borrows lambda at about Rs 2,12,307 per QALY and says so on the slide. Say out loud that the number is
  on loan and that Session 6 interrogates it. Do not pre-empt Session 6 by explaining where it comes from.
- **The question raised at the end of Session 2** ("is Rs 54,881 per QALY good value?") stays open until
  Session 6. That is deliberate. Tell them it is parked, or the room will think you forgot.

### The clinical bridge slides

Every group paper but one is a clinical or service decision, while the running teaching example is a
public health programme. Four slides exist to close that gap, using one verified Indian clinical paper
(Gupta 2020, adjuvant trastuzumab, PMID 32045547, open access, and **not** in the paper pack):

| Session | Slide | What it does |
|---|---|---|
| 1 | opening 11, "Two examples, side by side" | Names both running examples and says the method does not change with the size of the decision |
| 2 | block1 25, "The same two questions, on a drug" | Classifies the trastuzumab paper with Drummond's two questions |
| 3 | block3a 15, "The same framing, on a drug" | Its five framing choices against the Reference Case principles |
| 5 | block3b 23, "The same arithmetic, on a drug" | Two effectiveness sources, two ICERs a third apart |

**Do not skip these**, even when running late; they are the answer to "that is public health, I
prescribe". Session 8's trastuzumab slide (block3c 17) then completes the same table with the
probabilities, so it reads as a callback rather than a new example.

The decks and the participant handouts name **sessions by number**, never by clock time, so the day can
slip without contradicting anything in a participant's hand. This clock column is for facilitators, and
matches the printed agenda.

Fill in the **Lead** column; the agenda's facilitator column was deliberately left blank.

### Why these skips

| Deck, slide | What it is | Why skipped |
|---|---|---|
| Block 1, 12 and 13 | the six-box grid filled with cervical cancer studies; partial evaluations | Decided cut: the grid in depth. Slide 11 introduces the six boxes; that is enough |
| Block 1, 14 | classifier widget | Keep if the room has phones out and time allows; otherwise skip |
| Block 1, 22 | QALYs are not DALYs | Decided cut: the QALY versus DALY argument. Session 5 covers the distinction in one line |
| Block 1, 24 | where this lands on your desk | Rewritten for faculty (committees, supervised work), but it now overlaps the opening deck's "where you come in". Show it only if the opening was cut short |
| Block 2, 11 | league tables | Its figures were shown in Session 1. Say "you saw these this morning" and move on |
| Block 2, 12 and 13 | the NICE threshold change | Good material, but optional at 40 minutes |
| Block 3A, 10 and 11 | costing in three steps; capital costs | Keep, but brisk. They now set up Session 4 instead of repeating it |
| Block 3A, 12 | cost classifier widget | Keep if phones are out: it is the only place this widget appears |
| Block 3A, 13 | where Indian cost data comes from | Repeated as costing slide 13, in more detail |
| Costing, 4 and 9 | three steps; annuitising capital | Brisk, not skipped. Session 3 named both; here they are practised |
| Block 3B, 8 | QALYs are not DALYs | Decided cut |
| Block 3B, 11 | why discounting health is contested | Decided cut: the discounting argument. Keep slide 9 (the rule) and 10 (the widget) |
| Block 3B, 17 | extended dominance | Decided cut. On slide 15, reveal the cytology row as dominated; say in one sentence that VIA every 10 years is also never chosen, "for a subtler reason we will not go into today", then go to 18 |
| Block 3B, 21 and 22 | the same decision three ways; why NMB | Optional at 40 minutes. Slide 20 alone carries the formula |
| Block 3C, 18 | scenario analysis and value of information | Time; name them aloud if a group asks |

### Minute plan for the two hands-on sessions

**Session 4, costing (55 min):** slides 20 minutes, stop at slide 15. Worksheet 30 minutes, same groups,
each group costs an episode from one member's department. Last 5 minutes: three groups give their
total and their answer to question (b), the number they are least sure of. Slide 16 ("And your paper")
is a 30-second reminder for lunch.

**Session 9, report-back (25 min):** four pairs, six minutes each, one minute in hand.

| Pair | Groups | Contrast |
|---|---|---|
| 1 | 1 and 2 | a well-built model with a favourable verdict, against one that concludes "not cost-effective" |
| 2 | 4 and 3 | a conclusion turning on a health-system assumption, against one turning on clinical choices |
| 3 | 5 and 6 | a paper that is not really a cost-effectiveness analysis, against a technically competent one |
| 4 | 7 and 8 | an analysis that ends in a price to negotiate, against one that ends in a cost per DALY |

Two minutes per group: what it compared and concluded; the principle it handles best; the one it
handles worst and whether that changes the conclusion; would you act on it. Then two minutes, led by
the facilitator, on what the contrast shows. **Cut groups off at two minutes.** The facilitator key
(`paper-pack-facilitator.html`) has what each pair should surface.

## 4. Roles

Six resource persons. Suggested split, names to be filled in:

| Role | Covers | Person |
|---|---|---|
| Welcome, opening and close | 09:30 welcome, Session 1, Session 10 | |
| What it is, and framing | Sessions 2 and 3, 10:00 and 10:40 | |
| Costing | Session 4, 11:40 slides and worksheet | |
| Outcomes and the ICER | Session 5, 12:35 | |
| Thresholds and budget impact | Session 6, 14:05 | |
| Modelling, uncertainty | Sessions 7 and 8, 14:45 and 15:25 | |
| Floor and polls | runs Socrative all day; timekeeper; chairs the report-back | |

Everyone not presenting **circulates** during the 3-minute paper slots and the worksheet, carrying the
facilitator key. Nudge, do not tell: repeat the prompt, point at the checklist row, and only name the
issue if a group is still stuck after the session that teaches it.

## 5. Before the day

**Print** (A4):

| Item | File | Copies |
|---|---|---|
| Paper pack, cover sheet | `paper-pack.pdf`, page 1 | 1 per participant |
| Paper sheet for the group's own paper | `paper-pack.pdf`, pages 2 to 9 | 3 or 4 of each, one per group member |
| Full text, for the seven open-access papers | from each DOI | 3 or 4 per group |
| Appraisal checklist | `appraisal-checklist.html` | 1 per participant, plus spares |
| Costing worksheet | `costing-worksheet.pdf` | 1 per participant (groups share one to submit) |
| Facilitator key | `paper-pack-facilitator.pdf` | 1 per resource person. **Do not hand out** |
| Poll bank | `poll-bank.html` | 1 for the floor facilitator |

Group 3 (Uy, *Int J Cardiol*) is the only **abstract-only** paper; its sheet says so and carries more
detail. If any faculty member has institutional access, print the full text for that group too.
Group 8's paper also has a published Comment (PMID 39228042); print it for the facilitator.

**Set up:**

- Socrative room, with one quiz per session imported from `poll-bank.xlsx` (see its second sheet).
  Test on a phone on the venue Wi-Fi. If Wi-Fi fails, run the questions as a show of hands; the answers
  and explanations still work.
- Decks run offline (reveal.js is vendored), but test on the venue laptop and projector. Press **S**
  for speaker notes; no slide tells the audience to.
- Eight tables, labelled 1 to 8, with the cluster name.
- Assign participants to groups from the registration responses the evening before, and put the
  group number on each name badge.

**PowerPoint copies.** Every deck is also in `slides/` as a `.pptx`, linked from the workshop page, for
anyone who asks for the slides or wants to reuse a few. They are generated from the HTML decks, so if
you edit a deck, rebuild them with `src/pptx-all.sh` rather than editing the PowerPoint by hand.

**Publish the site.** The decks and the widget slides print the address
`sph-aiimsbhopal.github.io/intro-economic-evaluation-health`, and Session 1 shows a QR code for it. Push
this repository to that GitHub repository and turn Pages on (Settings, Pages, Source: GitHub Actions;
the workflow is in `.github/workflows/pages.yml`). Open the address on a phone before the day. If it is
not live, say so on the widget slide and demonstrate the widgets from the laptop instead.

**Check in the facilitator key:** items flagged "Before the day" (Thiagarajan's ICER arithmetic,
Guleria's HTAIn status, Uy's sensitivity analysis, Kaur's open-access status, the Srinivasan Comment). Resolve them if you can; if not, the
key says how to handle the question honestly.

## 6. Open items

1. ~~Venue room and contact e-mail~~. Settled: **SAMVAD (Board Room, First Floor, Medical College Building)** and
   **hta@aiimsbhopal.edu.in**. Both are on the rebuilt brochure and agenda.
2. **Lunch**: confirm; it is on the agenda and the registration panel but not the details bar.
3. **The close**: the brochure promises participants will "take part in producing" HTA evidence. The
   opening deck (slide 9) names four routes. The close needs one concrete next step with a name and a
   date attached, for example an RRC topic-nomination call, a costing-study collaboration, or a
   follow-up appraisal club. Decide which before the day.
4. **Feedback form**: none built yet. A three-question Google Form linked from a QR code on the last
   slide would do.

## 7. What is in this repository

| File | What |
|---|---|
| `index.html` | landing page: the day, with a link to each deck and handout |
| `slides/opening.html` | 10 slides, 09:40. New |
| `slides/costing.html` | 17 slides, 11:40. New |
| `slides/modelling.html` | 15 slides, 14:45. New |
| `slides/block1.html` to `block3c.html` | the five teaching blocks, adapted from the PG course for this audience |
| `handouts/paper-pack.html`, `.pdf` | cover plus eight one-page participant briefs |
| `handouts/paper-pack-facilitator.html`, `.pdf` | the key |
| `handouts/costing-worksheet.html`, `.pdf` | two-page A4 worksheet |
| `handouts/poll-bank.html`, `poll-bank.xlsx` | 23 questions |
| `handouts/appraisal-checklist.html`, `reading-list.html` | from the PG course, footers updated |
| `handouts/papers.py`, `build_paperpack.py`, `build_polls.py` | sources for the generated handouts |
| `widgets/`, `assets/`, `vendor/` | six interactive pages, styles, and reveal.js for offline use |
| `src/` | brochure and agenda sources, and `chk.mjs` |

**What changed in the five adapted decks:** the audience line ("MD Community Medicine") became
"Faculty and senior residents"; cover kickers now carry the session time instead of a block number;
speaker notes and a few slide lines that assumed residents preparing for an exam (dissertation, viva,
PG seat, "students") were rewritten for faculty who supervise work, sit on committees and referee
papers. No teaching content, figure or citation was changed.

All eight decks pass `chk.mjs` with no overflow, no em dashes and no HTTP errors. Every study
cited was checked in September 2026 against its PubMed record (through Europe PMC, or PubMed's own
record page where Europe PMC was rate-limiting), with volume, issue and pages confirmed against
Crossref or the publisher. Worked examples on the
costing and modelling slides are constructed and labelled as such.