Systematic reviews and meta-analysis: forest, funnel and L’Abbé plots
Read every part of a forest plot, recognise heterogeneity and publication bias, and interpret the L’Abbé plot — a named item on the RCGP data-interpretation list.
Reviewed and kept current
Last editorial review: 28 September 2026 · Next scheduled refresh: March 2027.
Clinical vignette
Begin with the patient. Read the case, weigh the options, then work through the unit.
Mrs K, aged 82, attends with her daughter. The daughter has read that a falls-prevention intervention “reduces falls by 8%”. She asks whether her mother should be referred.
The figure comes from a meta-analysis of eight randomised trials. The forest plot gives a pooled odds ratio of 0.92. The diamond extends from 0.81 to 1.04.
Mrs K lives alone and is keen to stay independent. The daughter wants to know whether the evidence shows that the intervention works, and whether it is worth her mother’s time and effort.
Options to weigh
- Tell them that the intervention reduces falls by 8%.
- Explain that the diamond crosses the line of no effect, so the pooled result is not statistically significant.
- Look at the heterogeneity (I²) and the funnel plot before drawing any conclusion.
- Tell them that the meta-analysis proves the intervention does not work.
Reflect
How would you explain this result so that Mrs K and her daughter can decide, and what would you be careful not to say?
What this unit covers
Pocket-sized and self-contained: three AKT-style questions, the answers explained, an expanded discussion with hyperlinked guidance, a decision algorithm and a certificate.
By the end of this unit you will be able to
Five steps, in order
Answer first, then learn. Each section unlocks as you progress, and your progress is saved in this browser.
1Three single best answer questions
Commit to an answer for all three single best answer questions, then submit. Sections 2–4 open on submission; your certificate opens at 2/3 or better.
Question 1 of 3
A meta-analysis of eight RCTs of an intervention to prevent falls presents a forest plot. The pooled odds ratio is 0.92, and the diamond extends from 0.81 to 1.04.
What is the correct interpretation?
Question 2 of 3
A funnel plot from a systematic review of a new antidepressant shows many small studies with large benefits but a noticeable absence of small studies showing no benefit or harm.
What is the most likely explanation?
Question 3 of 3
A L’Abbé plot is drawn for 12 trials of an antiemetic. For each trial, the event rate (vomiting) in the treatment group is plotted on the y-axis against the event rate in the control group on the x-axis. Most points lie below the diagonal line of equality.
What does this indicate?
2Answers explained
Question 1
Correct answer E. The intervention has no statistically significant effect on falls
The diamond is the pooled estimate: its centre is the point estimate and its width is the 95% CI. It crosses the line of no effect (OR = 1), so the pooled result is not statistically significant.
A forest plot alone cannot show publication bias (that is the funnel plot’s job), and heterogeneity is judged from I² and the scatter of the individual studies.
Question 2
Correct answer B. Publication bias
A funnel plot shows each study’s effect size (x-axis) against its precision or size (y-axis). Without bias, the points form a symmetrical inverted funnel: large studies cluster near the true effect at the top and small studies scatter widely at the bottom. A missing corner of small “negative” studies — asymmetry — suggests publication bias (or other small-study effects).
Question 3
Correct answer C. The antiemetic reduces vomiting in most trials
Points on the diagonal mean the event rate was the same in both arms. Below the line means the treatment arm had a lower event rate than the control arm. When the event is harmful (vomiting), that favours the treatment. Points above the line would favour control.
Wide scatter of points around the line suggests heterogeneity; the spread along the x-axis shows the range of baseline risk.
3Expanded discussion, worked examples & guidance
1. Systematic review vs meta-analysis
A systematic review answers a focused question using a pre-registered protocol (e.g. on PROSPERO), a comprehensive search (several databases plus grey literature and trial registries), explicit inclusion criteria, duplicate screening and data extraction, and risk-of-bias assessment. Reporting follows PRISMA, including a flow diagram. A meta-analysis is the optional statistical pooling of the results.
2. Anatomy of a forest plot
- Each row is one study. The square is its point estimate; the size of the square reflects its weight (usually larger, more precise studies); the horizontal line is its 95% CI.
- The vertical line of no effect is at 1 for ratios (log scale) or 0 for differences.
- The diamond at the bottom is the pooled result: its centre is the estimate and its width is the CI.
- Labels such as “favours treatment | favours control” tell you which side is good.
Worked example 1 — reading a forest plot
Five trials of SGLT2 inhibitors vs placebo for HF hospitalisation: individual RRs 0.70 (0.55–0.89), 0.65 (0.50–0.85), 0.81 (0.60–1.09), 0.72 (0.58–0.90), 1.02 (0.60–1.73). Pooled RR 0.72 (0.65–0.80); I² = 12%.
- Three studies are individually significant; two are not (their CIs cross 1).
- The pooled diamond lies wholly below 1 → significant 28% relative reduction.
- I² of 12% → low heterogeneity; pooling is reasonable.
Answer. Pooling non-significant small trials can yield a precise, significant overall result — that is the purpose of meta-analysis. (Illustrative numbers.)
3. Heterogeneity
Heterogeneity is variation in results between studies beyond what chance would explain. Look at the forest plot (do the CIs overlap?), the χ² (Cochran’s Q) test, and I² — the percentage of variation due to heterogeneity rather than chance: about 25% low, 50% moderate, 75% high.
- Fixed-effect model: assumes one true effect; suitable when heterogeneity is low.
- Random-effects model: assumes the true effect varies between studies; gives a wider CI and more weight to small studies.
With high heterogeneity, look for explanations (different populations, doses, durations) through subgroup or sensitivity analysis — pooling may be inappropriate.
4. Funnel plots and publication bias
A symmetrical inverted funnel is reassuring. Asymmetry suggests publication bias, poor-quality small studies, or true small-study effects. Egger’s test formalises this. The same “funnel” shape is used for practice-level control charts, where practices outside the control limits are true outliers (Unit 11).
5. L’Abbé plots
Worked example 2 — interpreting a L’Abbé plot
x-axis = control-group event rate; y-axis = treatment-group event rate; one point per trial; diagonal = no effect.
- Harmful outcome (death, vomiting): points below the diagonal favour treatment.
- Beneficial outcome (pain relief ≥50%): points above the diagonal favour treatment.
- Points spread widely along the x-axis show that baseline risk varies between trials; tight clustering around a line parallel to the diagonal suggests a consistent effect.
Answer. Always check the axes and whether the event is good or bad before deciding which side “wins”.
- The diamond’s width, not its position alone, determines significance.
- A large square is a heavily weighted study, not necessarily a large effect.
- Forest plots of ratios use a logarithmic x-axis, so 0.5 and 2 are equidistant from 1.
References and further reading
Guidance checked at the last editorial review (28 September 2026). Always use the current version.
4Decision algorithm
Appraising a systematic review and meta-analysis
Start
Is the question focused (PICO), and was the protocol pre-registered?
Decision
Was the search comprehensive (multiple databases, grey literature, no language restriction)?
Decision
Was the risk of bias of the included studies assessed?
Decision
Is heterogeneity high (I² > 50–75%, non-overlapping CIs)?
Decision
Is the funnel plot asymmetrical?
Outcome
Read the diamond: does its CI cross the line of no effect? Convert to absolute benefit for your patients.
Revision summary; apply clinical judgement and local pathways.
5Certificate of completion
You have met the pass mark. Enter your details to issue your personalised certificate of completion, then save it as a PDF for your appraisal or revalidation portfolio.
Your certificate is issued immediately as a PDF. Your details stay in your browser and are not sent to MD Acumen.
Your completion and score have also been recorded by your organisation’s learning platform.
MRCGP AKT · CAS Programmes · For institutions · Legal & regulatory · enquiry@mdacumen.com
Educational content for healthcare professionals; apply clinical judgement and current local guidance. MRCGP is a registered trademark of the Royal College of General Practitioners. These units are independent, curriculum-aligned revision resources and are not endorsed by the RCGP.
© 2026 MD Acumen Ltd · Registered in England and Wales · Company no. 16538952 · ICO registration ZC105897