Research planning and methods-readiness watcher
Turn a rough clinical study idea into a protocol-ready checklist.
StudyVahti maps early study ideas to PICOTS, protocol/reporting standards, bias domains, estimand logic, feasibility assumptions, and places where statistical or methodological review is needed before protocol drafting.
What this is, and when to use it
What. StudyVahti turns a rough clinical study idea into a protocol-ready checklist. It maps your idea to PICOTS, the reporting standard your study will be judged against (STROBE, CONSORT, STARD, PRISMA…), the bias domains to plan for, and an analysis-ready variable codebook, in your browser, nothing uploaded.
When to use it. At the very start, once you have a research question but before you write the protocol or collect data. It is a design-stage tool; when your protocol is registered and an observational study is moving into analysis, continue in StudyVahti Vault to keep the variables, estimand, plan, decisions, outputs, claims, and audit trail together.
Who it’s for. Master’s and PhD students planning a clinical or epidemiological study, and their supervisors, who can use the readiness sheet as a shared checklist in a supervision meeting instead of catching the gaps at the thesis defence.
How. Fill in the study identity, PICOTS, variables, bias checks, and feasibility. StudyVahti scores readiness out of 100, names the next gate to close, and exports a codebook and readiness sheet. The codebook is tidy and analysis-ready by construction (snake_case names, one concept per column, a documented type and role per variable) so the data imports cleanly into R (tidyverse) or Python (pandas) with no renaming or reshaping later. Clean names up front are the cheapest analysis you will ever do.
When should I use StudyVahti?
Early, after you have a research question but before you write the protocol or collect data. It is a planning-stage tool, not an analysis or write-up tool.
Who is StudyVahti for?
Master’s and PhD students designing a clinical or epidemiological study, and their supervisors — the readiness sheet works as a shared checklist between student and supervisor.
Does StudyVahti write or approve my protocol?
No. It maps your idea to the reporting standards and flags where methods review is needed. It does not approve studies, appraise design quality, or replace a qualified methodologist. Any causal or inferential study should still involve one.
How does it make my R or Python analysis easier?
It builds a tidy variable codebook (snake_case names, one concept per column, and a documented type and role for each variable) so your data reads straight into the tidyverse or pandas without renaming or reshaping. Fixing variable names before data collection is far cheaper than fixing them in a half-finished analysis.
Which reporting standard does my study need?
StudyVahti maps it for you: STROBE for observational studies, CONSORT / SPIRIT for randomized trials, STARD / TRIPOD-AI for diagnostic-accuracy and prediction studies, PRISMA for systematic reviews, and COREQ / SRQR for qualitative work.
Is my data uploaded anywhere?
No. StudyVahti runs entirely in your browser and saves nothing to a server — the study idea, the codebook, and the readiness sheet stay on your device.
- PICOTS
- Population, Intervention, Comparator, Outcome, Timing, Setting — the structured skeleton of a well-specified research question. Vague PICOTS is where most study designs go wrong.
- Estimand
- The exact quantity you are trying to estimate: which effect, in whom, at what time, and how you handle events like death or treatment switching. Naming it up front prevents a vague analysis.
- Time-zero (index date)
- The moment follow-up starts for every participant. Getting it wrong is the usual cause of immortal-time bias.
- Immortal-time bias
- A bias where a stretch of time during which the outcome could not occur is credited to a treatment group, inflating its apparent benefit. A well-defined time-zero avoids it.
- Target-trial emulation
- Designing an observational study as if you were running the randomized trial you ideally would — it forces explicit eligibility, treatment, and time-zero definitions.
- ROBINS-I
- A structured tool for assessing risk of bias in non-randomized studies of interventions.
1. Study Identity
Define the clinical question, study pathway, and intended estimand before asking for artifacts.
2. PICOTS
StudyVahti is strict here because vague populations, comparators, timing, or settings become vague analyses.
3. Variables And Codebook
Use stable variable names, roles, and coding notes so data provenance and analysis assumptions stay inspectable.
Recommended packs update from the current study setup.
Variable audit updates as rows are added.
Variable packs append modular rows and skip duplicate names. Format: variable_name | label | type | role | allowed values or coding note
4. Bias And Causal Validity
These checks are mapped to target-trial thinking, nonrandomized-study bias domains, and transparent reporting expectations.
5. Feasibility And Power Readiness
StudyVahti records sample, event, precision, and governance assumptions. It does not replace a formal sample-size calculation.
Readiness Sheet
Generated locally from the fields above.
Planning Cockpit
Gate status and open planning decisions generated from the same fields.
Local Artifacts
Scaffolds only. Review before using in a protocol, analysis plan, or repository.
Generate the readiness sheet to create local artifacts.