Harmonizing data after the fact means months of cleaning. Capture pushes consistency to the point of entry: one locked, approved form for every site, edit checks that raise queries as values are typed, and calculated fields that never depend on local arithmetic.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Sites
5
Open auto-queries
11
Out-of-range values
3
Forms on approved version
100%
queries
queries
queries
queries
What to know first
The problem
The same protocol, run at five sites, produces five slightly different datasets. One site records weight in kilograms and another in pounds. One writes dates as day-month, another as month-day. One treats "not done" as a blank and another as a zero. A site reads a severity scale one way, another reads it differently. Each is a small thing; together they are the reason cross-site analysis takes weeks of reconciliation.
The causes fall into four groups: format (units, date formats, decimals), definition (what counts as a baseline, an event onset or a missed visit), version (different CRF versions in use) and practice (how a measurement is actually taken). Software can attack the first and third directly, and help with the second and fourth by making the intended definition visible at the field.
This page is about the data-consistency problem. For the operational side of running many sites, see multi-site clinical trial management, and for early-phase networks EDC for multi-site early-phase trials.
Cause and control
| Source of inconsistency | Example | Control in Capture |
|---|---|---|
| Version drift | Site B still uses last month's form | Forms move draft to approved; approved forms are locked and only approved forms appear in the casebook |
| Out-of-range values | Systolic 182 typed instead of 128 | Range high, range low and custom value edit checks raise an auto-query at entry |
| Local arithmetic | BMI worked out by hand with rounding differences | Calculated fields (BMI, BSA, eGFR, QTc, age from DOB) are computed and read-only |
| Unit ambiguity | Creatinine in mg/dL at one site, umol/L at another | Units set on the question; lab tables carry unit, result and reference-range columns |
| Participant identification | Duplicate or inconsistent IDs | Site-level numbering such as 001-0042 and QR-code enrollment |
| Hidden site drift | One site systematically differs | By-site filtering and exports make a site comparison a filter, not a project |
These controls reduce variation at entry. Unit conversion between systems of measurement is a design decision; see the lab unit converter page to plan it.
At the point of entry
A form is drafted, reviewed, approved and then locked. Sites cannot add a variant. Edit checks fire as the value is entered, so a coordinator fixes a typo while the participant is still in clinic, not six weeks later.
Systolic blood pressure
SYSBPAbove the range set for this field. Auto-query raised
Weight
WEIGHTHeight
HEIGHTBMI
BMICalculated, read-only
26.8 kg/m2 CalculatedCreate a study, add sites, enter a deliberately wrong value and see the query. Free, every feature, no credit card.
Running the sites
Participants can enroll by QR code, specific to a site or to the whole study, with no app to install. Participant numbers are site-level, so 001-0042 tells you at a glance which site a record belongs to and prevents number collisions. Site coordinators work in their own portal, see names for their own participants, and the rest of the team works with coded IDs. See QR-code participant enrollment.
For harmonization, the reviewing side matters as much as the entry side. Data can be filtered by site and exported by site, so a data manager can compare distributions of the same variable across sites, spot the site that enters temperatures in Fahrenheit, and send a targeted note. The query management workflow then closes the loop in the record.
International studies add language, regulatory and device differences. Capture does not claim multilingual participant forms on this page; if your sites need them, confirm the current status before you design around it. The international multi-site page covers what can be planned for.
Playbook
Do it once at design, and the data stay comparable for the whole study.
Write down what counts as baseline, event onset, a missed visit and a not-done assessment. Put the definition into the field hint.
Choose one unit per variable and set it on the question. Decide how a missing or not-done value is recorded.
Add range and custom rules for the variables that matter most, and set priorities so critical queries are visible.
Approve the forms and test with sample data in the sandbox. Approved forms are locked for live use.
One CRF completion guide, one training session recording, one list of common queries.
Filter by site, compare distributions and queries, and retrain where one site drifts.
Every site enters data on the same approved, locked form. Edit checks raise auto-queries at entry, calculated fields are computed and read-only, and data can be filtered and exported by site.
No. Forms move from draft to approved and approved forms are locked for live use. Changes are deliberate amendments.
Site-level numbering, for example 001-0042, with QR-code enrollment that is site-specific or study-specific and needs no app install.
This page does not claim automatic unit conversion. Units are set on the question, and lab tables carry unit and reference-range columns. Decide the unit policy at design time.
Yes. Data can be filtered by site and exported by site, and site coordinators have their own portal.
This page makes no multilingual claim. Confirm current language support with us before you design a multi-country study around it.
Yes. Capture is suitable for Phase 1, 2 and 3 multi-site studies.
Open the free sandbox with every feature, no credit card and no time limit, and add sites to a test study.
Keep exploring
Multi-site clinical trial management
Running many sites.
Edit checks software
Rules that raise queries at entry.
Query management software
Closing the loop on discrepancies.
QR-code participant enrollment
Site and study QR codes.
International multi-site studies
Planning across countries.
Lab unit converter
SI and conventional units.
Free sandbox, every feature, no credit card. Build the form once and test it across sites.