Free scoping read. A written assessment of your manuscript in two business days, at no cost.

Start one

Home›Services›Data, analytics & AI›Interpretation & visualisation

Statistical Analysis and Reporting

Interpretation, Reporting and Visualisation

Pubrica provides statistical interpretation, guideline-compliant reporting and publication-ready data visualisation. The service supports academic and clinical researchers, pharmaceutical and biotech companies, public bodies and publishers who have completed an analysis and need it written up. Our statisticians, subject experts and scientific illustrators work together to interpret the results, prepare reports to the guideline your study design requires, and draw figures to the journal’s specification. Deliverables include the interpretation, the report with its completed guideline checklist and the final figures, prepared so that each figure and the statistic underneath it say the same thing.

  • Interpretation, not description — what the numbers mean in context, what they support and what they do not support
  • Reports built to the guideline your design requires: CONSORT for trials, PRISMA for reviews, STROBE for observational work, TRIPOD for prediction models
  • Figures drawn to publication specification — resolution, colour, type size at final print size and file format, so they are not sent back at production
  • The right figure for the result, which for clinical work usually means CONSORT diagrams, Kaplan–Meier curves, forest plots and heatmaps rather than default chart-library output
  • Uncertainty reported, because a point estimate without its interval presents the result with more confidence than the data allow
What this service delivers Included and excluded

Statistical and contextual interpretationIncluded

Guideline-compliant reportingIncluded

Publication-specification figuresIncluded

Dashboards and executive summariesIncluded

Re-running the analysisBiostatistics

Making a null result look positiveNever

Results → interpreted → reported → drawn → submitted

Every interpretation and reporting order
  • Guideline named on the order
  • Uncertainty reported with every estimate
  • Figures to publication specification
  • Interpretation separated from conclusion
  • Peer-reviewed before delivery
CONSORT, PRISMA, STROBE, TRIPODPlus ICH, GCP, GPP 2022 and EQUATOR
5–20 business daysBy package · expedited quoted separately
Statisticians and subject expertsWorking together, not in sequence
ISO 9001:2015Quality management certified · cert. H3CR9TCU20251212INDQM10N

Service Scope

Six Components of Every Reporting Project

Each project is scoped on the order with the reporting guideline named. The interpretation of what the findings mean for your field remains your conclusion; we report what the results statistically support.

Statistical and contextual interpretation Interpretation of what the findings show, which patterns and relationships they reveal and what they can and cannot support, written as a narrative that sets out the reasoning.

From the interpretationThere was a statistically significant difference between groups (p = 0.03).The intervention group had a 21% lower event rate (absolute difference 4.2 percentage points, 95% CI 0.4 to 8.0). The interval reaches close to no difference, so the direction is clearer than the magnitude.

Reporting to the guideline your design requires Structured documents from full research reports to executive summaries, built to CONSORT, PRISMA, STROBE or TRIPOD as the design demands — which is what makes them compliant, transparent and reproducible.

From the reporting checkCONSORT 2025: 31 of 37 items present. Missing: 6b changes to outcomes, 11a who was blinded, 14b why the trial ended, 17b absolute and relative effect sizes, 19 harms, 24 protocol access. Each located and drafted.

Figures built to specification Resolution, file format, embedded fonts, type size at final print size, column width and colour, prepared to the journal’s published specification. These are the most common reasons a figure is returned at production.

From the figure check72 dpi PNG · 6 pt labels · RGB · 1,400 px wide600 dpi TIFF at 90 mm single-column width, 8 pt Arial at final size, CMYK, fonts embedded. The content is unchanged; the file is now acceptable.

The right figure for the result Kaplan–Meier curves with numbers at risk, forest plots with weights visible and CONSORT flow diagrams whose numbers reconcile with the text, each chosen to match the structure of the result.

From the figure noteSurvival difference shown as a bar chart of mean follow-upKaplan–Meier with numbers at risk beneath the axis and censoring marked. A bar chart of means discards the timing, which is the entire finding in a survival analysis.

Uncertainty shown Intervals, spread and precision carried into both the text and the figures, so a reader can judge how much the result will bear rather than taking the point estimate at face value.

From the figure standardEvery estimate plotted with its interval. Where intervals are too wide to plot legibly, they go in the table and the figure caption says so — rather than a clean-looking chart that implies precision that is not there.

Independent review before delivery A second reviewer checks the interpretation against the output and the reporting against the checklist, because an interpretation is best checked by someone who did not write it.

From the reviewSecond read: three statements stronger than the data supports, one figure whose axis started above zero, and a subgroup result reported without noting it was not pre-specified. All corrected before delivery.

Samples are illustrative and anonymised. Hover any one of them to play it again.

Types of Reports and Figures

Report Types and Data Visualisation Formats

The report type is chosen for the audience, and the figure type for the structure of the result.

Quantitative interpretationNumerical results analysed and explained — effect sizes, intervals, precision and what the analysis can and cannot support.
Qualitative interpretationNon-numerical data — interview, free-text and thematic material — analysed for themes and context, with the analytic approach stated rather than implied.
Research reports and manuscriptsFull structured reports and journal-ready manuscripts with methodology, findings, discussion and conclusions, built to the applicable guideline.
Clinical study reports and regulatory documentsSubmission-format documents where the structure is prescribed and the reporting burden is heavier than a manuscript’s.
Dashboards and executive summariesHigh-level views for decision-makers, with denominators visible and the uncertainty carried into the summary rather than dropped at the last step.
Publication figuresCONSORT flow diagrams, Kaplan–Meier curves, forest plots, heatmaps, scatter and regression plots, and data tables — drawn to the journal’s specification.

Also available. Storyboarding, custom templates and chart libraries, reformatting existing visualisations, automated and macro-driven reporting for recurring outputs, and delivery in PowerPoint, Word or PDF. What we will not do: re-run your analysis inside this service — that is biostatistics and is quoted separately — or report a result more strongly than the output supports.

Who we serve

Six Client Groups We Support with Research Reporting

Each client group needs the same evidence presented differently, depending on whether a reviewer, a regulator, a board or a funder has to act on the findings.

Academic researchers and scholarsComplex datasets reported for journal submission, with clarity, reproducibility and precision.
Clinical researchers and healthcare professionalsTrial and outcome studies reported to regulatory and guideline standards, with the full transparency those standards require.
Pharmaceutical and biotech companiesR&D outcomes and trial data presented for internal review, regulatory submission and scientific communication — three audiences, one evidence base.
Corporations and analystsReporting and visualisation that turns analytics into decisions, with the uncertainty surviving into the summary rather than being lost at the last step.
Government and non-governmental organisationsInterpretation and reporting for policy, programme evaluation and funding assessment, where the finding has to be defensible in public.
Publishers and communication professionalsPublication-ready figures, data tabulation and reports to international standards, prepared at volume and to house specification.

Compliance and Standards

Reporting Guidelines by Study Design

Each report is checked against the guideline that matches your study design. The guideline is named on the order so that compliance can be verified.

CONSORT (2025 statement)

Randomised trial reporting. The items most often missing are outcome changes, blinding, harms and absolute effect sizes.

PRISMA2020

PRISMA 2020 (27 items and the flow diagram)

Systematic reviews and meta-analyses, with the flow diagram numbers reconciled against the text rather than drawn separately.

STROBEobservational

STROBE (And RECORD for routine data)

Observational studies, where the reporting burden falls on confounding, missing data and how the sample was actually assembled.

TRIPOD+AI, 2024

TRIPOD+AI (BMJ, 2024)

Prediction model development and validation, including machine learning models.

ICMJE Recommendations (January 2026)

Authorship, declarations, data transparency and the statements a journal will check before it reads anything else.

GPP2022

Good Publication Practice (GPP 2022)

For industry-sponsored publication — authorship, writing support and the disclosures that accompany them.

EQUAnetwork

The EQUATOR Network (Where the guideline is found)

Where a study design has a reporting guideline not listed here, we identify it through the EQUATOR Network.

Responsible Use of AI

Expert Interpretation and Accountable Data Visualisation

Software tools can produce a plausible chart from almost any dataset, including a misleading one. Interpretation determines what the reader concludes, so it is carried out and checked by our statisticians and subject experts.

What we use it forWe use AI tools to draft plotting code, reformat figures to a journal or template specification, and check that figures, tables and text agree with each other and with the underlying data.
What we do not use it forWe do not use them to decide what a result means, to choose a scale, baseline or exclusion that changes the impression, or to write an interpretation that has not been checked against the analysis it describes.
What goes in your declarationThe tool, the stage it was used at, and the human oversight applied — because journals, funders and the ICMJE January 2026 update all now expect it, and an undeclared tool found later is treated as a research integrity matter rather than a style preference.

Why we publish no accuracy figure of our own

Accuracy figures for AI tools are usually reported by the companies that sell them.

Pubrica therefore does not quote a self-generated accuracy figure, because it would carry the same conflict of interest. Instead, we tell you which tool was used on your project, at which stage, and what a human reviewer changed afterwards.

Methodology

Reporting Methodology: Five Steps with Independent Review

The interpretation is drafted by somebody who understands the analysis and checked by somebody who did not write it.

Step one

Consultation and requirements

We review your research goals, study design, data sources and target, whether a journal, a regulator or an internal decision, which determines the guideline, the structure and the figures.

Step two

Analysis and interpretation

Statisticians and subject experts work together on what the output means, what it supports and where its limits lie, instead of a statistician handing numbers to a writer.

Step three

Drafting the report

We prepare a structured report with methodology, findings, discussion and conclusions, built to the guideline your design requires and to the journal or regulatory format.

Step four

Designing the figures

We build figures, dashboards, tables and infographics suited to the result and to the publication specification for resolution, colour, type size and format.

Step five

Review and finalisation

A second reviewer checks the interpretation against the statistical output and the reporting against the checklist, with anything overstated brought back before delivery.

Request a Free Sample Interpretation and Figure Redraw

The result written up as it would appear in your target journal, with the uncertainty reported properly, plus the figure redrawn to that journal’s specification and a note on what would have to change across the rest. Two business days, no charge.

What you provide
The outputStatistical results, in any form
The designWhat kind of study this is
The targetJournal, regulator or internal
The guidelineCONSORT, PRISMA, STROBE, TRIPOD or none
The audienceWho has to act on it
The deadlineAnd what depends on it

Please send the statistical output rather than a summary of it. An interpretation based on a summary inherits anything the summary left out, and the omitted details are often the most important.

Our experts

Your Reporting and Visualisation Team

A named team of subject specialists stays with your project from scoping to delivery. Your project coordinator introduces each member before work begins, and the handover records which of them completed each stage.

Subject matter expertClinical or scientific expert matched to your topic

Reviews the scientific content for accuracy and confirms that each conclusion is supported by the evidence. Matched to your field rather than allocated from a rota.

OwnsScientific accuracy and interpretation
Reads againstThe study protocol and current literature
FieldMatched to your topic
BiostatisticianStatistical input where the work reports analysed data

Checks that statistical methods, tables and conclusions are consistent with the analysis plan and that the analysis can be reproduced.

OwnsStatistical accuracy
Reads againstICH E9 and the statistical analysis plan
ToolsR, SAS, Stata, Python
Data specialistData handling, cleaning and documentation

Prepares, cleans and documents the data so that every step from raw data to final output can be traced and reproduced.

OwnsData handling and documentation
Reads againstThe data management plan and applicable data standards
OutputDocumented, reproducible datasets
Second reviewerIndependent check of code and outputs

Independently checks code, tables and outputs against the analysis plan. Discrepancies are logged and resolved before delivery.

OwnsIndependent quality control
Reads againstThe analysis plan and outputs
Signed off byProject lead

Sample work

Sample Work: Interpretation and Figure Before and After Revision

Step through the changes one at a time. The analysis itself was not changed; the revisions affect what the write-up claims and what the figure shows.

6 changes, each explained

As received. Statistical output and a survival figure from a completed trial.

  • “A significant difference was observed (p = 0.03).”
  • Survival shown as a bar chart of mean follow-up
  • No confidence intervals anywhere
  • Subgroup result reported in the abstract
  • Figure: 72 dpi PNG, 6 pt labels, RGB
  • CONSORT declared; 6 items unreported

As delivered. Interpreted, reported to CONSORT and redrawn to specification.

  • 21% lower event rate; absolute difference 4.2 points (95% CI 0.4 to 8.0) — direction clearer than magnitude
  • Kaplan–Meier with numbers at risk and censoring marked
  • Intervals on every estimate, in text, table and figure
  • Subgroup moved out of the abstract — not pre-specified, and labelled so
  • 600 dpi TIFF, 90 mm column, 8 pt at final size, CMYK, fonts embedded
  • All 37 CONSORT items addressed; the 6 missing ones drafted

The analysis was not re-run. Every number here comes from the output as supplied. Extract redacted; field and values changed.

Redacted extract · field and values changedPremium package · CONSORT 2025

Finished work

A delivered figure, shown with the client’s permission.

Grouped bar chart comparing the percentage of bleeding-related adverse events, including haemorrhage, contusion, haematoma, rectal and gastrointestinal haemorrhage, purpura, cerebral haemorrhage, haematochezia, epistaxis and subarachnoid haemorrhage, across paroxetine, citalopram, escitalopram, fluoxetine, fluvoxamine and sertraline, with group sizes in the legend.

Data visualisation · comparative safety analysis

Bleeding-related adverse events across six SSRIs, grouped by event type with each drug’s sample size stated in the legend, so that rates can be compared directly.

Benefits

Benefits of Accurate Interpretation and Reporting

Each benefit can be checked in your own report and figures. None is a promise that the paper will be accepted.

01 Findings stated at the correct strengthOverstatement draws closer scrutiny from reviewers, and understatement can obscure a real result. We report each finding at the strength its interval supports.
02 Compliance with the declared reporting guidelineReviewers check declared compliance, and journals increasingly require the checklist. Unreported items are a common finding and are straightforward to correct before submission.
03 Figures that present the result accuratelyThe figure type matches the structure of the result and shows uncertainty, rather than discarding information to look clean.
04 Figure files to journal specificationResolution, format, fonts and colour are prepared to the journal specification. A figure returned at proof stage can cost a fortnight for a problem that takes an hour to fix at submission.
05 Independent internal reviewAn interpretation is best checked by someone who did not write it, so overstatement is corrected internally before the paper reaches a journal reviewer.
06 One evidence base for regulatory and journal reportingReporting is built to the guideline and to the journal format together, so the regulatory document and the manuscript draw on the same evidence and do not need to be reconciled later.

Packages

Interpretation and Reporting Service Packages

The price depends on how much interpretation is needed, which guideline applies and how many figures must be built rather than adjusted. We therefore provide the free sample first.

Basic

Basic

Researchers wanting concise insight and a clear summary

Quoted per projectPriced on output volume

Delivery 5–7 business days

  • ✓Interpretation with the essential statistical reporting
  • ✓Concise report, two to three pages
  • ✓Basic charts, graphs and tables
  • ✓Uncertainty reported with every estimate

Suited to projects where the analysis is complete and only the write-up is needed.

View a sample Request a quote
Standard

Standard

Academic projects, journal submissions and structured organisational reporting

Quoted per projectPriced on guideline and figures

Delivery 10–12 business days

  • ✓In-depth interpretation with the methodology explained
  • ✓Reporting structured to journal or industry standards
  • ✓Advanced visualisation — survival curves, forest plots, heatmaps
  • ✓Figures built to the target journal’s specification
  • ✓Guideline checklist completed and gaps drafted

Suited to journal submissions that require a reporting guideline.

View a sample Request a quote
Premium

Premium

Clinical trials, regulatory submissions, grants and high-impact publication

Quoted per projectPriced on scope and review rounds

Delivery 15–20 business days

  • ✓End-to-end interpretation including advanced statistical reporting
  • ✓Full guideline-compliant reporting — CONSORT, STROBE, PRISMA or TRIPOD
  • ✓Custom figures, interactive dashboards and infographics
  • ✓Executive summary written for the decision-maker
  • ✓Multiple review rounds with a second statistical reader

Expedited delivery is available and is quoted separately, because it affects the depth of review; we explain how when we quote.

View a sample Request a quote

Send one result and one figure, and we will interpret and redraw them free of charge. Recurring reporting, and automated or templated outputs, are quoted against a schedule.

Quality Assurance

Our Service Commitments

Each commitment below states what we check, its limits and the remedy if the work does not meet the level on your order.

The Pubrica guarantee

Conclusions Proportionate to the Statistical Evidence

The interpretation is checked against the statistical output by a second reader before delivery, and the reporting is checked against the guideline named on your order. Both are testable against documents you already have.

01

The guideline is named on the order

CONSORT, PRISMA, STROBE or TRIPOD, at its current version, so that the report is checked against a defined standard.

02

Uncertainty accompanies every estimate

In the text, the tables and the figures. A point estimate on its own claims more precision than the study produced.

03

A second reader before you see it

The interpretation is reviewed by someone other than its author, because the author is least able to see where it overstates.

04

Free rework within the agreed scope, or a refund

Where the work does not meet the level on your order we do it again at no charge within the agreed scope, and where you would rather not, we refund it.

Client Testimonials

“They turned raw trial data into a guideline-compliant report with clear figures. The reporting checklist found six items we had declared and not done.”
Dr KennethClinical researcher

USA

Interpretation and reporting FAQs

Interpretation and Reporting FAQs

If your question is not answered here, ask an expert — no quote required.

Researchers preparing manuscripts, clinicians and sponsors compiling regulatory submissions, institutions and students presenting theses, and organisations that need decisions taken from analysis.

Most clients come to us with a completed analysis and a write-up that has not started, or one that does not yet present the result clearly.

Journal-ready manuscripts, clinical study reports, white papers, regulatory documents, business reports and academic submissions, each to the guideline and format that applies.

Where a study has a reporting guideline, it is named on the order and the checklist is completed and delivered with the report.

Yes — CONSORT for trials, PRISMA for reviews, STROBE for observational studies, TRIPOD for prediction models, plus ICH, GCP and journal-specific requirements.

We check every item of the declared guideline. Identifying unreported items before submission is routine and avoids having them raised in peer review.

For clinical and epidemiological work: CONSORT flow diagrams, Kaplan–Meier curves, forest plots, heatmaps, scatter and regression plots, and properly formatted data tables.

The figure type follows the shape of the result. A survival difference shown as a bar chart of means has discarded the timing, which was the finding.

It is drafted against the statistical output rather than a summary of it, and checked by a second reader who did not write it — against the output and against the checklist.

The most common correction is a statement worded more strongly than the interval supports. It is an easy error to make and difficult to see in your own writing.

Not inside this service. If the interpretation stage turns up something that looks like an analysis problem, we raise it rather than quietly correct it.

Re-analysis is biostatistics and is quoted separately, deliberately — the person interpreting a result should not also be the person who produced it.

Related services

Related Data Analytics Services

This service turns a completed analysis into a report. The services below produce the analysis or take the report through to publication.

Request a Free Sample Interpretation and Figure Redraw

The result written up as it would appear in your target journal with the uncertainty reported properly, the figure redrawn to that journal’s specification, and a note on what the rest would need. Two business days, no charge.

Get a free quote

Regulations

Check your project against this rule

  1. 1Your work
  2. 2Your details
  3. 3Send
What do you need?

Material you send is held under signed confidentiality, with access limited to the assigned team and encrypted transfer. It is not used to train anything and is not passed to third-party tools. Ask for the agreement before you send anything you would rather cover first.