Paper reading · Journal club
Prepare a 10-minute journal club: from paper to eight-slide speaker notes
Decide what the audience needs to understand and discuss, then link each slide’s main claim to the paper. A reviewable handoff includes the sequence, speaker notes, timing, figure sources and open questions—not merely a divided abstract.
This guide addresses turning an authorised paper into a 10-minute journal-club talk. The evidence-table guide covers field extraction and the graphical-abstract guide covers a single visual summary; this page covers selection, sequencing and rehearsal for an oral presentation. Eight slides and the timings below are our teaching plan, not a universal standard.
The example uses the published PACE-UP trial by Harris et al. (2017) and its original Fig 1, alongside Bourne’s 2007 presentation advice. Public licences were checked for both articles. This is source-review and outline teaching: no trial rerun, model execution, generated PPT or customer outcome, and no exercise prescription.
Collect traceable materials first
Record the version and material actually read. An abstract alone does not establish review of the full text or figures.
- Authorised full text, supplements and DOI, with retrieval date.
- Audience, discussion goal and whether the 10 minutes include questions; this example reserves the final 2 minutes for discussion.
- Original tables/figures, captions and individual permissions; free access alone does not permit arbitrary figure changes.
- A fact sheet with source locations, design, analysis unit, denominators, comparisons, time windows, effects and intervals.
- Unreported information, source conflicts and unchecked items; exclude patient rows, credentials and confidential material.
An eight-slide plan ready for review
Speaking and discussion together total 10 minutes. Give each slide one main job and adjust the timings through rehearsal.
| Slide | Minutes | Question answered | Handoff item |
|---|---|---|---|
| 1 | 0.5 | What are we discussing? | Paper citation and one discussion goal. |
| 2 | 1 | Why compare the delivery routes? | Background gap, separate from results. |
| 3 | 1.5 | How was the question studied? | Household randomisation, three arms and population. |
| 4 | 1 | Who contributed outcome data? | Original Fig 1, randomised/follow-up denominators and caption. |
| 5 | 2 | What is the main 12-month result? | Both intervention-versus-control effects, units and 95% CIs. |
| 6 | 1 | What is not established? | The direct comparison and limits on equivalence claims. |
| 7 | 1 | Which limitation affects our judgement? | Recruitment and masking: facts separated from appraisal. |
| 8 | 2 | What should the group discuss? | Qualified conclusion, one question and references; includes discussion. |
Map every example number back to the paper
These are published aggregate results, not a new analysis. Keep the original comparison and time window: a 12-week intervention is not 12 months of ongoing treatment.
| Fact to preserve | Example | Source location |
|---|---|---|
| Design and unit | Three arms, household randomisation; 1023 participants from 922 households. | Abstract; Methods: Randomisation and Masking; Fig 1 |
| Follow-up denominator | 956 participants (93%) provided outcome data; they are not the number randomised. | Abstract; Fig 1 |
| Postal versus control | At 12 months: 642 additional steps/day, 95% CI 329–955. | Table 2; Results: Twelve-mo (main) outcomes |
| Nurse support versus control | At 12 months: 677 additional steps/day, 95% CI 365–989. | Table 2; Results: Twelve-mo (main) outcomes |
| Direct intervention comparison | Reported as 36 steps/day, 95% CI −277 to 349; nonsignificance does not establish equivalence. | Results: Twelve-mo (main) outcomes |
| Recruitment limitation | Recruitment was 10% (1023/10467); high follow-up does not remove generalisability concerns. | Discussion: Study Strengths and Limitations |
Ask AI for an outline, notes and a review table
Suggested request: Using only the paper and figures I am authorised to provide, prepare eight slides for the stated audience, totalling 10 minutes including 2 minutes of discussion. For each slide give one claim, speaking points, source section/table/figure, estimated time and open issues. Preserve design, denominators, reference group, units, time windows and intervals.
Also provide a “fact—source location—slide—wording limit” review table and draft speaker notes. Mark absent information as unreported. Do not generate data, invent citations, redraw real results or describe proposed steps as already executed.
This is an editable task request, not a model-run receipt. Confirm file format, editability and actual tool support separately; an outline does not prove that a usable PPT was generated.
Review four things before presenting
Return to the source
Check each main statement against its section, table/figure and context. Both 642 and 677 are versus control. The paper reports a direct difference of 36; do not replace it with the subtraction of displayed values, 677−642.
Check figures and attribution
Keep Fig 1’s essential labels and caption, crediting Harris et al. (2017), PLOS Medicine, DOI and CC BY 4.0. Fig 1 labels the search population 45–74 years; the title/abstract say 45–75. Preserve both and flag the discrepancy rather than silently harmonising or guessing why.
Check limitations and claim strength
Retain the paper’s 10% recruitment rate and lack of participant/nurse masking. Nonsignificance does not mean equivalence, no effect or applicability to every population.
Rehearse with a clock
Include figure explanations, transitions and questions. Cut secondary detail first; do not remove denominators, intervals or key limitations to fit 10 minutes.
Boundaries for figures and methods
Fig 1 may be used under CC BY 4.0 with attribution; check third-party material separately. This page links the original source without redrawing or publishing a new trial figure. Bourne’s article supports presentation methods, not the trial findings.
One talk does not replace full-text appraisal, risk-of-bias assessment or clinical judgement. No software-speed comparison, AI superiority claim or newly executed research task is demonstrated here.
Common questions
Must there be eight slides?
No. This is a teaching sequence for 10 minutes including discussion. Adapt it to the audience and rehearsal; it is not a one-slide-per-minute rule.
Should the whole paper go into the notes?
Keep claims and sources relevant to the discussion goal, with secondary tables in backup. Notes should explain evidence and limits, not turn the paper into a reading script.
Verified sources
The official sources below were checked on 2026-10-26. Access conditions and update dates are rechecked before publication.
- Bourne (2007): Ten Simple Rules for Making Good Oral Presentations
Full text checked for advice on audience, selection, logic, rehearsal and acknowledgement; not empirical evidence.
- Harris et al. (2017): PACE-UP Cluster Randomised Clinical Trial
Public full text, Table 2, original Fig 1 and CC BY 4.0 checked; published aggregate results, with no study rerun.
Research presentations and handoff
Turn authorised material into a reviewable presentation handoff
In the RigorMed workspace, state your paper rights, audience, time limit and discussion question. Request a source-linked outline, draft notes and review table; confirm actual support and output format, then check the paper. This guide is not a PPT-generation run.