Cluster randomised trial, 20 clinics

Clinical trial poster

A cluster randomised trial of text message reminders for childhood vaccination, laid out as a full A0 portrait board. Every block a trial poster is expected to carry is here and filled in: the abstract, the methods as a labelled protocol rather than a paragraph, a CONSORT participant flow whose counts reconcile, the primary outcome with its confidence interval, and three charts with numbered captions and a CONSORT flow.

Create a poster with OneCraftA0 Portrait, printed at 841 by 1189 millimetres

The whole board

The poster at full size, exactly as it prints. Every number, citation and caption on it was written for this example, so the layout is being judged on real content.

NU
Text message reminders raise childhood vaccination completion: a cluster randomised trial
S. Gupta ¹, P. Singh ¹,², R. Okoro ³, D. Chen ²
1 Centre for Global Health, Northgate University · 2 National Primary Care Development Agency · 3 Kaduna Regional Health Service
Abstract
Vaccination series in rural districts are often started and not finished. We tested whether an automated text message sent three days before each due dose raises the proportion of children completing the full schedule by 14 months. Twenty clinics were randomised, ten to reminders and ten to usual care, and 1,980 mother and infant pairs were enrolled. Completion rose from 61.2% under usual care to 82.4% with reminders, a difference of 21.2 percentage points (95% CI 15.9 to 26.5, p<0.001).
Introduction
Coverage of the first dose in this region is above 90%, but completion of the full schedule falls below 65%. The gap is not refusal: caregivers who drop out say they intended to return and lost track of the date once the child was well. Mobile phone ownership in the enrolled districts is 87%, which makes a reminder cheap to deliver and easy to scale on infrastructure that already exists. The evidence for it in low resource rural settings is thin and mostly uncontrolled, and the few trials that exist measure coverage at a single dose rather than completion of the whole schedule.
METHODS
Design. Parallel cluster randomised controlled trial, 20 clinics, allocation 1:1 by computer generated sequence held off site.
Participants. 1,980 mother and infant pairs presenting for the first dose, infant aged under 8 weeks, household holding at least one mobile phone.
Intervention. One SMS in the caregiver stated language, sent three days before each due dose, naming the child, the dose and the clinic.
Comparator. Usual care: the printed card and the verbal return date given at the visit.
Outcome. Completion of the full schedule by 14 months, verified against the clinic register, assessor blind to allocation.
Acknowledgements
Funded by the Regional Research Council under grant RRC-2026-118. We thank the clinical and administrative teams at the twenty participating clinics, who carried the register work alongside a full caseload, and the caregivers who agreed to be contacted for fourteen months. The funder had no role in the design, the analysis or the decision to publish.
Results
82.4%
completed with reminders
n=921, 759 of 921
61.2%
completed with usual care
n=921, 564 of 921
+21.2
percentage point difference
95% CI 15.9 to 26.5, p<0.001
Reminders
Usual care
Dose 1
Dose 2
Dose 3
Dose 4
0255075100
Figure 1. Completion by arm at each dose in the schedule. Error bars are 95% confidence intervals.
PARTICIPANT FLOW
Clinics assessed
26
No register
6
Clinics randomised
20
Pairs enrolled
1,980
Lost to follow-up
138
Analysed at 14 months
1,842
Results
Completion was 82.4% with reminders and 61.2% under usual care, a difference of 21.2 percentage points (95% CI 15.9 to 26.5, p<0.001). The arms matched at the first dose, given at enrolment, and separated from the second dose onward.
Difference
07142128
Under 5 km
5 to 10 km
Over 10 km
Figure 2. Difference in completion by distance from the clinic, in percentage points.
Reminders
Usual care
Week 8
Week 16
Week 24
Week 40
Week 56
0255075100
Figure 3. Cumulative completion by week since the first dose, both arms.
Discussion
One message per dose closed roughly a third of the gap between the first dose and the last. The effect grew with distance, which fits a barrier of remembering a date rather than unwillingness to travel. Phone ownership here was 87%.
Contribution
The first cluster randomised test of dose level reminders in rural primary care in this region, with completion verified against the clinic register rather than caregiver report. The message templates and the analysis script are released under an open licence.
Future work
A twelve month follow-up across six districts is planned, powered to detect a difference in the second year rather than only in the first, with a cost per completed schedule reported alongside the effect. A nested qualitative study will ask the caregivers who received every message and still did not return what the message would have needed to say. We will also test whether one message before the final dose performs as well as one before every dose, since the cost of the intervention scales with the number sent.
Conclusion
One text message before each due dose raised full schedule completion by 21 percentage points at a cost of roughly $0.34 per child, and the benefit was largest for the households furthest from a clinic, which are the households usually hardest to reach. The intervention needs no new staff, no new equipment and no change to the schedule itself.
Completion 82.4% with reminders, 61.2% with usual care
Difference 21.2 points, 95% CI 15.9 to 26.5, p<0.001
Largest gain beyond 10 km from the clinic
Runs on existing staff and existing phone numbers
REFERENCES
1.Wakadha H, et al. Vaccine 2013;31(6):987-993.
2.Eze P, Lawani LO. BMJ Glob Health 2021;6(7):e005035.
3.Mekonnen ZA, et al. Syst Rev 2019;8:154.

Block by block

What each block on the board is for, in the order a reader walks it.

Title band, authors and QR
The full study title runs across the band with the institution mark on the left and a QR code on the right. Author superscripts are generated from the affiliation list, so moving an author never leaves a stale number behind.
Abstract and introduction
The abstract states aim, method, result and the size of the effect in one block. The introduction gives the gap the study fills: first dose coverage above 90% but full schedule completion below 65%, and the reason caregivers give for it.
Methods as a labelled protocol
Five labelled entries covering design, participants, intervention, comparator and outcome. This is the block that most often gets written as a paragraph and most often goes unread when it is.
Key numbers with intervals
Three stat blocks carry 82.4%, 61.2% and the 21.2 point difference. The two arm blocks carry their denominators (n = 921 analysed in each arm) and the difference block carries the 95% CI, 15.9 to 26.5, underneath rather than beside it.
Participant flow
A CONSORT style spine with exclusions branching to the side: clinics assessed, excluded, randomised, pairs enrolled, lost to follow-up and analysed. Counts reconcile at every step.
Figures and results
Three numbered charts: completion by dose, effect by distance from the clinic, and cumulative completion by week. Captions are generated with their figure numbers so renumbering cannot break the pairing.
Discussion, contribution and future work
What the result means, what is new, and what the limits are, including the point that high phone ownership in this district limits how far the effect size travels.
Conclusion and references
A short conclusion, four checked takeaway points, and an auto-numbered reference list in journal style at the foot of the column. The three references are invented for this fictional study, because the reference block is required on this layout; on your board, replace them with your sources.
How to adapt this board
Swap the two arms for yours and keep the difference as the third key number, because a reader looks for the effect before anything else. Keep the CONSORT counts in the order your analysis reports them, so the flow reads down the page the way the trial ran. Then retype the three chart figures with your own values: the first is the outcome by dose or visit, the second is the effect split by whatever your subgroup is, and the third is the cumulative curve. Nothing on the board is tied to vaccination, and every block resizes, so a trial with two figures rather than three loses one block and gives the space to methods.

What makes this board work

The numbers reconcile

The participant flow starts at 26 clinics assessed, drops 6 without a register, randomises 20, enrols 1,980 pairs, loses 138 and analyses 1,842. A reader who checks the arithmetic finds it holds, which is the first thing a reviewer at a poster session does.

Methods is a protocol, not prose

Design, Participants, Intervention, Comparator and Outcome each get a bold label and one sentence. Nobody standing at a board reads a method paragraph, and the labelled form lets a reader jump straight to the one thing they came to check.

Every estimate carries its interval

The three key numbers show the count they came from and the confidence interval underneath. A headline percentage with no denominator and no interval is the most common fault on a real poster board.

Questions people ask

What has to be on a clinical trial poster?

Title, authors and affiliations, abstract, methods, a participant flow diagram, the primary outcome with its confidence interval, figures with captions, discussion, conclusion and references. Registration number and funding usually sit in the acknowledgements.

Do I need a CONSORT diagram on a poster?

For a randomised trial, yes. Reviewers look for it, and it answers who was excluded and why in less space than a paragraph. The counts must add up, because that is what gets checked.

How much text should a trial poster carry?

Enough that each block is full and nothing is padded. This board runs about 1,100 words across all sections. The failure to avoid is a half empty box, which reads as an unfinished study.

Can I change the layout after generating?

Yes for the blocks, no for the figure media. Every section, figure, key number and reference list drags, resizes and deletes. A figure's media is fixed by the layout: on this board every figure is a chart, which you can switch among ten chart kinds, but it cannot be swapped for an uploaded image.

Which numbers must agree with each other?

The two arm percentages, the difference block, the analysed count in the CONSORT flow and the bars in Figure 1 all describe the same 1,842 people. If one changes, the other three change. Check them in that order before printing, because a reader at the board will do exactly that.

Where do the trial registration and funder go?

In the acknowledgement section at the foot of the left column, one line each: the registry name with the id, then the funder and grant number. They are not results, so they do not belong in the results zone, but a reviewer expects to find them without hunting.

Can the CONSORT flow show more than six boxes?

No. The flow block on every poster layout has six nodes, one of which is an aside for exclusions. Fold the reasons for exclusion into that one aside with counts separated by commas, and keep the main path as assessed, randomised, the two arms and analysed.

Build your own in about a minute

The button below opens the generator with this use case already described. Change the wording to match your own, generate, then edit anything you like.

Make my clinical trial poster

Other poster examples

Nursing research poster

A ward based trial of four hourly against two hourly repositioning on high specification foam, laid out as an A0 landscape board in four columns. It reports both outcomes a nursing trial has to report: whether patients were harmed, and how much nursing time the change gives back.

Systematic review poster

A review of four day week trials in office based work, laid out as an A0 portrait board in journal style. The screening flow reconciles from 1,842 records to 22 included studies, and the central figure is not the effect but how the outcome was measured, because that is what the review actually found.

Materials science poster

A materials science poster is read composition first, so this A0 portrait example puts the weight percent line at the top of its methods protocol where a metallurgist looks for it. Two charts split the two properties: tensile results in one, hardness and the measured silicon length in the other, with the 14 percent strength gain as the key number.

Want the steps in the builder? Read Add charts and diagrams, then choose the template, theme and size. For everything this generator can do, see the poster maker.

Sources

Written and checked by the OneCraft team. Last checked .