Medical Writing
From search questions to interpretation of results, we retain the links between evidence screening, analytical methods and result sources.
This service covers literature searches, Field evidence reviews, target evidence and competitive landscape reviews, preclinical experimental statistics, and meta-analysis, survival analysis and prediction models. Statistical deliverables include analysis code, results tables and figures, and methods descriptions that allow verification. The service supports all development stages and the three service areas, and can also be commissioned separately.
01Medical Writing
Each service below specifies its scope, deliverables, specifications and timeline, and pricing basis.
02Medical Writing
Searches, evidence synthesis and statistical modelling have different input requirements. Define the research question before selecting the workflow.
Reporting checklists describe the methods and results completed. They do not ensure that a study is free from bias or replace a study protocol. Interpretation of results should reflect the strength of the evidence.
Evidence and statistics can be commissioned separately or included in medical education, research publication or clinical documentation projects. Clients may commission only evidence searches, statistical analysis or statistical review, with the remaining work performed by the client or a third party.
Input materials, analysis scope, deliverables and acceptance criteria are specified for each project; no fixed packages or additional starting prices are assumed.
03Medical Writing
Search records should show where, when and how searches were performed. Statistical reports should describe the analysis population, methods and uncertainty.
Field evidence reviews are named according to their methods; materials without systematic searching and screening are not called systematic reviews merely because they include a reference list. Statistical figures should state data sources and validation methods; internal validation does not replace external validation.
| Standard, version and location | Items to check |
|---|---|
| PRISMA 2020, items 5–8, 13, 16 and 20. | Eligibility criteria, information sources, complete search strings, selection processes, synthesis methods, and selection and synthesis results. |
| PRISMA extension for reporting literature searches in systematic reviews (PRISMA-S, 2021), items 1, 8 and 12–16. | Databases and platforms, complete search strategies, update methods, search dates, search peer review, record counts and deduplication methods; state truthfully if search peer review was not performed. |
| Strengthening the Reporting of Observational Studies in Epidemiology (STROBE, 2007), items 7, 12, 16 and 19. | Variable definitions, statistical methods, main results and limitations. |
| TRIPOD+AI reporting guideline for multivariable prediction models (2024), items 12a, 12c, 12e and 23a in the 27-item checklist. | Data splitting, model development and internal validation, performance evaluation methods, and performance results with confidence intervals; applicable to regression and machine learning methods. |
| ARRIVE 2.0；Percie du Sert N, et al. The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. PLoS Biol 2020;18(7):e3000410. PMID 32663219 | Experimental design, sample size and blinding. |
| SYRCLE risk of bias tool; Hooijmans CR, et al. SYRCLE's risk of bias tool for animal studies. BMC Med Res Methodol 2014;14:43. PMID 24667063 | Risk of bias assessment in systematic reviews of animal studies. |
04Medical Writing
The following reproductions of public data demonstrate statistical presentation formats. All are labelled: Not a Yanto Health project.
Reproduced using the public survival::lung dataset. There are 228 participants: 138 men and 90 women, with 165 deaths. Shading indicates 95% confidence intervals and tick marks indicate censoring. Numbers at risk count participants still under follow-up and event-free immediately before each time point.
Reproduced from the public aggregate trial dataset dat.bcg, including 13 trials. The restricted maximum likelihood (REML) random-effects pooled risk ratio (RR) is 0.49, with a 95% confidence interval of 0.34–0.70 and heterogeneity I² of 92.2%; the 95% prediction interval is 0.15–1.55. Intervals use a normal approximation. The prediction interval retains heterogeneity in the true effect of a new study; the pooled estimate does not replace interpretation of differences between studies.
Reproduced using the public MASS::biopsy dataset, with 699 out-of-fold predictions divided into 5 groups by predicted probability rank; ties are split in original record order. Observed rates within groups have 95% Wilson confidence intervals. The Brier score is 0.035; connecting lines aid reading only, and out-of-fold predictions have not been recalibrated. This example shows internal validation, not external validation or clinical effectiveness.
05Medical Writing
Literature searches deliver a search strategy, database-specific search logs, eligible reference list and bibliographic table. Systematic reviews additionally involve predefined eligibility criteria, screening, data extraction, evidence appraisal and synthesis. The requirement confirmation form should specify whether these subsequent steps are included.
The current literature search service specifies no more than 5 working days in total. This must not be extended to a general timeline for a complete systematic review or meta-analysis; those timelines are confirmed for each project.
Statistical analysis delivers code, results tables and figures, and methods descriptions that allow verification. The client or a third party can check results with the same data, code and execution environment. Model selection and interpretation within a project should have an explicit rationale.
With existing statistical results, work can begin with result review or manuscript preparation. Where the client writes the manuscript, statistical review alone can be commissioned. If existing materials are insufficient for verification, the additional materials needed and the scope must first be defined.
The statistical figures on this page reproduce public data; they are not client projects or evidence of treatment effects. Different figures may use different datasets. Sample sizes, models and validation methods in the examples cannot be transferred directly to commissioned projects.
Get started · Submit a request
Please provide the research question, status of data or literature, proposed analyses, target deliverables and deadlines. We first issue a requirement confirmation form, followed by an itemised quotation and schedule.
This section describes our services. It does not concern any specific medicinal product and does not constitute drug advertising. The stated timelines, quality control and resolution procedures reflect standard working practices. Final deliverables, timelines, prices and each party's obligations are governed by the contract signed by both parties.
Writing and statistical support for research publications follows the recommendations of the International Committee of Medical Journal Editors (ICMJE) and Good Publication Practice 2022. Our contribution is disclosed in the acknowledgements in line with ICMJE and GPP 2022; we are not listed as authors. Funding sources are also disclosed truthfully. We do not promise publication outcomes.
