Your research. Our infrastructure.
Health research runs on a layer of work that is nobody's research question: the capture forms, the pipelines, the consent trail, the model that has to survive review. We build and run that layer, so the question stays yours.
A database that fits the protocol.
Capture built around the study, not the tool.
We design and host study databases — REDCap where it fits, something purpose-built where it does not. Instruments, branching logic, validation, roles and audit trail, set up so the data arrives analysis-ready instead of needing a month of cleaning first.
- REDCap instrument design, branching logic and validation rules built with you.
- Roles, consent states and an audit trail configured before the first participant.
- Exports shaped for your analysis, so cleaning is not a second project.
The analysis that would have taken months.
Models you can defend to a reviewer.
Predictive models, classification and pattern discovery across large clinical datasets — built so the method is reportable. Every result traces back to the records that produced it, because a number you cannot explain is a number you cannot publish.
- Predictive and prognostic models trained on your cohort, not a generic one.
- Automated extraction and coding of variables buried in free text.
- Methods documented to the standard a reviewer will ask for.
Participants who actually finish.
Collection people do not abandon.
Web platforms and mobile apps for data collection, follow-up and participant engagement. Built for the person filling it in on a phone in a waiting room, because a questionnaire nobody completes is a dataset nobody can use.
- Participant-facing apps and portals for enrolment and follow-up.
- Reminders, scheduling and patient-reported outcomes that cut loss to follow-up.
- Straight into your study database, with no manual re-entry.
A specialist when you need one.
Complexity, deadlines, and the gap between.
Sometimes the constraint is not the data but the time, or a method nobody on the team has used before. We work alongside your group for one phase or a whole project: analysis plan, implementation, and the figures that go into the paper.
- Analysis plans written with you, before the data locks.
- Survival, longitudinal, hierarchical and missing-data methods.
- Publication-ready figures and tables, and the code behind them.
Sensitive data, handled properly.
Built for data you cannot afford to leak.
Participant data carries obligations that outlast the study. Storage, access control, pseudonymisation and deletion are designed in from the start, and the whole system can run inside your own institution's environment when that is what your approval requires.
- Runs on-premise or in your own tenancy where approvals require it.
- Pseudonymisation, access control and retention set by the protocol.
- Every access logged, so a data-handling question has an answer.