All releases stage-a.17 Installed09.10.2026

Formulas in the eCRF, DSUR, Kaplan–Meier and RWE tools

The eCRF gets free formulas and if-then checks with a live tester. Safety builds the DSUR from the study's data, suggests coding and searches the literature with one press. Biostat draws Kaplan–Meier curves, rates per 100 patient-years and lab shift tables, checked against references, and keeps the analysis plan with its estimands and the cost estimate by tariffs. For real-world data: reasons for missing data, the visit document for the medical record, My data for the patient, pack scan and site performance. The form library gets the organisation's library and disease packs, Studio makes the FDA package, and an eCRF version has its validation kit and the study archive.

21new items
7sections
2improvements
2026-10-09_0340archive · 73b0f00

YouEDC Capture

  • Validation and archive of a study. On the new page Validation and archive (the last section of the study settings and a button on the eCRF versions page) every released eCRF version has its validation kit as one ZIP: the specification, the DVP of that very version, the signed snapshot, the changes, the UAT scripts and results with the executor's and the approver's signatures, the traceability, the release record (whether exactly what was tried was released) and the audit trail of the QA environment, every file with its SHA-256. The UAT scripts download as a blank sheet in PDF and Excel to run them on paper. The archive of the whole study is built in the background in ZIP volumes of up to 2 GB and is read without YouEDC: every eCRF version, the data in ODM, Excel and CSV, the data cuts, the casebooks by site, the audit trail with its chain check, the signatures, the team, the validation kits, the files and DICOM when asked. Before the database is locked the archive is marked interim; building it and every download go to the audit trail, and the admin's page that deletes a study for good shows whether its archive was built and downloaded Validation kit and study archive · QA environment of a study
  • Why the data is as it is. An unscheduled visit may ask why the patient came: planned follow-up, worsening or new symptoms, an administrative reason, or other with a note; the site gives the reason with the visit date, and the audit trail keeps every change. A “no data” mark may say why the value is missing: not measured in routine practice, measured but not in the medical record, the patient refused, not applicable, unknown. The export has VISIT_REASON on the Events sheet and a sheet of the “no data” marks with their reasons, the export builder puts <VARIABLE>_NDREASON next to <VARIABLE>_ND, and the data quality report counts the missing values of each variable by reason and the unscheduled visits by why the patient came. A de-identified cut keeps the reason codes and drops the free text. In the demo some patients come to unscheduled visits, and the missing values have reasons Entering data into a form · Events, the patient timeline and the visit calendar · Data quality report · Data export · Export builder and formats for statistics
  • The organisation's form library. A form finished in one study can be proposed to the library of the organisation with “Propose to the library” on the form's page: as a new form or as the next version of one already adopted. A steward of the library (an admin of the organisation or a person the admin names) sees the fields and what changed against the last version and approves the version with an electronic signature over its SHA-256, or rejects it with a reason; with two stewards nobody approves their own version. Approved forms are offered in the Form library of the builder of every study of the organisation and stay linked to the library like the YouEDC standard forms: what the team changed and whether a newer version is out. A form can be retired, another organisation never sees the library, every step is in the audit trail. The demo has an invented sponsor's library with a form you approve yourself. Form library, copying and eCRF from Excel
  • Visit document for the medical record. The doctor fills the registry visit once, and under the visit on the patient card there is a Document for the medical record (PDF) button: an A4 examination note with the site, the visit date, the visit's forms as readable sections, the records since the previous visit and the doctor's signature line. The doctor writes in the full name, date of birth and medical record number by hand: the portal does not keep them. The forms for the unblinded team are never printed, and the randomisation arm is not read. The study chooses the forms and the title of each visit, for example “Discharge summary”. The document is signed by hand, and every printout goes to the audit trail with the SHA-256 of its content Document for the medical record
  • The patient sees their own data in ePRO. The study picks up to six variables (numbers from repeated visits and the patient's questionnaire scores) and switches My data on, and the site's investigator approves it for each patient on the patient card. On the personal link the patient sees a chart and the values with dates, checked ones only: the form is complete and no query is open on the value. There are no norms, no red or green, no comparison with others and no interpretation; forms that could reveal the treatment arm are never shown. A fixed note asks to discuss the values with the doctor, and “How your data helps” says how many patients take part. Switching on, approving and withdrawing go to the audit trail; the diabetes and hypertension demos have it on The patient sees their own data (ePRO)
  • Form packs by disease: type 2 diabetes, arterial hypertension, chronic heart failure and chronic kidney disease. The library has 19 new forms: diabetes history, a log of hypoglycaemia by level, complications, the summary of continuous glucose monitoring; hypertension history, office blood pressure of three readings with the mean, 24-hour monitoring, a home blood pressure diary for the patient, organ damage; heart failure history and phenotype with a check of the ejection fraction, NYHA, echocardiography, NT-proBNP, congestion, the six-minute walk, a log of heart failure events; kidney disease history and categories, kidney function and electrolytes, a log of kidney replacement therapy. Each pack is the forms of the disease with visits for a year: choose it as the template in the new study wizard or add it to a study with “Add the pack” in the form library of the builder; the forms the study already has do not come twice. Form library, copying and eCRF from Excel · eCRF builder
  • Pack scan of drugs and devices. A code field in the eCRF reads the DataMatrix of a drug pack (Russian labelling under Decree 1556, the European FMD) or the UDI of a medical device: from a phone photo (read by the browser itself, nothing installed), a 2D scanner or the digits typed by hand. The portal checks the GTIN's check digit and the dates and fills the product, INN, batch and expiry date in the same form. The product of a GTIN comes from the study's list: the data manager uploads it from Excel or CSV, or the site names a new product once at its first scan. Only the GTIN, the batch and the expiry date are kept; the pack's serial number and crypto elements are left out, and the portal does not query МДЛП. In the demo studies the packs of the background drugs are scanned, with invented trade names Scanning drug packs and devices
  • Formulas in the eCRF. A calculated field can now compute a free formula, the way a spreadsheet does: SQRT(HEIGHT * WEIGHT / 3600), IF(SMOKER = 1, CIGS * YEARS / 20, 0), DAYS(AE_START, AE_END). An “if - then” check can be written as a formula too, with the field the query or the error goes to: “SBP - DBP < 20 AND PATIENT.SITE = "01"”. A formula names the fields of this and other forms and the patient's card, which now has the site number: a field, a check or a visit can apply at some sites only. The builder checks the formula as it is typed, shows it in words, refuses what the form may not name or a circle, and values of a test patient typed under the formula show the result at once. A formula is a small language of its own that the portal reads and never runs as a program; the browser and the server compute alike, changes of formulas and calculated values go to the audit trail, and the DVP shows the formulas with the form codes Data entry checks · eCRF builder · AND / OR conditions and checks across forms

YouEDC Studio

  • From the beta testers' suggestions in YouEDC Studio: an organisation adds its own items to the start-up and lock checklists, and they appear in all its DM projects; the Overdue visits tool shows the missed visits separately with their reason (they no longer count as overdue) and exports them on their own Excel sheet DM projects · Overdue visits and missing pages
  • Package for FDA in the CDISC section. After a build the block Package for FDA appears: the m5/datasets folders as FDA expects them (SDTM and ADaM in .xpt, define.xml with the CDISC style sheet, the annotated CRF acrf.pdf and the reviewer's guides). The annotated CRF is built from the form and the mapping specification: every field shows its SDTM variable in a coloured box, every form its domains, fields not submitted are marked [NOT SUBMITTED], with contents by visit and by form. define.xml now points to the aCRF pages of every collected variable and to the cSDRG and ADRG, and visits, arms and elements have Origin Protocol; define.html is the CDISC style sheet's view and opens in any browser. The cSDRG and ADRG drafts follow the sections of the PHUSE templates, filled with what the build knows, the sponsor's gaps in yellow, in Word and PDF. Next to them a readiness list: ASCII, SAS V5 limits, SSTDTC in TS, check errors. The agency's validator is run by the user CDISC: SDTM, ADaM, Define-XML
  • Studio tools light up from the project's data. A DM project's page has a new Project data strip: the data manager uploads the study export from their EDC once (ZIP, Excel, CSV or SAS XPT, MainEDC as it is too), and the portal sees which data it holds: patients, visits with dates, AEs and their seriousness, medications, medical history, labs, randomisation, queries, the audit trail. The tools that data allows are lit, the others are grey, open as before and say what is missing. Only the structure of the export is kept, with no values, patient numbers or the file itself; the upload goes to the project's audit trail. The Studio home lights up the tools from all your DM projects, and demo projects get their data map by themselves DM projects · YouEDC Studio

YouEDC Safety

  • Ready-made literature searches. The portal takes the study products from the reference safety information and the suspect products of the cases, drops the dose and the placebo from the name and prepares a search in every connected database with the usual safety words. The specialist presses one button: the searches are saved, run weekly, and the first run starts at once. With no products yet, typing the name is enough. Literature is now in the YouEDC Safety menu, and the study's Safety page points to the articles to screen and to the searches not started yet. Demo projects still never go online, also on “Run now”. YouEDC Safety
  • “Search now” and clearing the excess in literature. The button at the top of the Literature tab starts every active search of the study at once, and a bar with a running mouse shows how many are done and which one is running; the articles appear when all are done. If a query was too wide, the articles it found and nobody screened yet go with one button and a required reason in the audit trail; decisions, cases, hand-entered articles and the run log stay. On youedc.com PubMed search works now: the hosting DNS did not find the NCBI address, the portal asks another DNS. YouEDC Safety
  • A coding helper on the case. On the Events and assessment tab the portal proposes the MedDRA term and the seriousness criteria under each event. The terms come from the dictionaries the specialist already reads: the study's adverse event dictionary, their own MedDRA subscription and the built-in CTCAE, with the coders' earlier decisions. The seriousness criteria come from the words of the event and the narrative (“hospitalised”, “life-threatening”, “died”) with the sentence quoted; a negation such as “no hospitalisation” does not count. These are rules and a dictionary, without AI: nothing leaves the server. The specialist presses Take or Not this, the event changes only by their decision, and every decision goes to the audit trail with its source YouEDC Safety
  • DSUR builder. On the study's DSUR tab the development safety update report (ICH E2F) is built from the study's data: the cases are taken as they were known on the data lock point, and the portal builds the line listing of serious adverse reactions (7.2), the tabulation of every SAE since the start of development by the MedDRA system organ class of your coding (7.3), the exposure from YouEDC Capture without unblinding, the RSI changes, deaths, treatment withdrawals, SUSAR timeliness, the literature, signals and the cases after the data lock point, and drafts every section of ICH E2F. People only correct the text: the places in [square brackets] are highlighted and counted. Issued for signature, signed by the author and approved by another person; the approved report is kept as signed in PDF, Word and Excel and goes to the TMF inbox. YouEDC Safety

YouEDC Sites

  • Site performance from real work. The portal counts how each site of the directory worked in the organisation's studies: patients per month, days from activation to the first patient, entry delay, queries per 10 pages and the days to answer them, the share of patients kept, the monitor's queries per 10 pages verified at SDV, deviations per 10 patients. The profile is on the site's card, and the Site performance page puts every site side by side, sorted by any column, with an Excel file. These are numbers of the site, none of a doctor or a member of the team. A sponsor gets a profile only with the site's consent: an organisation admin records it with the day and the document, and the file For a sponsor holds only the sites that agreed. A person's demo studies now belong to a demo organisation with its own directory Site performance profile

YouEDC Trials

  • The monitoring plan is signed. A monitor or the project manager signs a version of the plan with the password as its author, another person who approves visit reports approves it with their own signature or returns it with a note; the approved plan goes to the TMF as a Word file, and when the plan's settings change later, the Monitoring plan tab asks to sign it again. Risk-based SDV: the plan sets the share of verified forms for a site's high, medium and low risk, and a site that has verified less shows in the sites table and goes to the monitor's desk YouEDC Trials

YouEDC Biostat

  • Biostat: time to event, exposure-adjusted rates, lab shift tables and a check against references. The study report gains Kaplan–Meier curves by arm to the first adverse event, the first serious one and early discontinuation, with the 95 % CI, the patients at risk and the median; adverse events per 100 patient-years (EAIR) with the exact Poisson CI, also for the most frequent terms; lab shift tables, first result against the worst later one, by the study's own normal ranges. The new page “Checked against references” runs the Biostat methods on published data (NIST StRD, the example of R's survival package, SciPy) and shows how many digits agree; its PDF is a record for the validation file. The CI of a mean now takes the exact Student's t quantile instead of a table. Everything stays descriptive, without p-values; a biostatistician reviews the numbers before decisions. YouEDC Biostat: the study's statistics by itself
  • Cost estimate by tariffs. The sites record the healthcare resource use in the new library form Healthcare resource use (CDASH HO): hospital stays with bed-days, day hospital, visits, procedures, emergency calls, with the DRG (КСГ) or service code. The data manager, the project manager or the statistician uploads the ОМС tariff table of their region and year in Excel or CSV (the portal ships no real tariffs). The Cost estimate page in YouEDC Biostat counts the cost per patient and per patient-year by arm, lists the records it could not price and why, and the Excel file carries the same numbers with the key of each. It is an estimate by tariffs, not the actual costs, and the page says so. The diabetes and hypertension demos have the form and an invented tariff table Cost estimate by tariffs
  • Analysis plan and estimands. The question of the study is written before the data as an estimand of ICH E9(R1): population, treatment and comparator, variable, intercurrent events with a strategy for each and the summary measure, and around it the elements of a target trial: time zero, confounders, a negative control. Every element that is data names its eCRF field. The design check reads the plan against the eCRF's visit schedule: eligibility, treatment and confounders must be collected no later than time zero (else immortal time, or a confounder the treatment has already affected), the outcome after it, an observational study keeps the start date of the treatment and names a negative control, the summary measure suits the variable's type. Each save is a new version that never changes, with its SHA-256, the eCRF version and the number of patients at that moment; a change after the first patient needs a reason and goes to the audit trail. In the diabetes and hypertension demos the plan was written before the first patient Analysis plan and estimands

Portal

Improvements and fixes

  • From the beta testers' suggestions: when a study is not created, the top of the form shows what stops it, with links to the fields; the invitation letter for a doctor has a QR code for the phone; a site's principal investigator can be picked from the same owner's other studies; the data manager's approval panel is highlighted only when it can be filled in now; the patient card has the hint “How to add another visit”
  • The bottom of every page and every PDF shows the version and build of the system, e.g. “build stage-a.17”: a screenshot or a printout tells at once which build it came from