Clarrio.ai / Evidence & methodology
A practical methodology for evaluating longitudinal health data, patient-reported experience and program outputs, with separate perspectives for payers, teams and users.
More data does not automatically mean stronger evidence. Define the intended decision, assess whether the sources represent the relevant population and preserve the limitations through every output. This is an evaluation approach, not a claim of demonstrated Clarrio.ai outcomes.
01 / Evidence & methodology
Write down who the program includes, what question it addresses and which decision the output supports. A clinic review, a natural-history study and a payer appraisal require different designs. Define baseline, observation windows, variables and review responsibilities before selecting the sources.
Stakeholder perspectives02 / Evidence & methodology
Keep source, units, instrument version and collection time with each observation. Distinguish missed reports from missing connections and distinguish no observation from a negative finding. Review whether changes in devices, follow-up or population membership affect comparisons. Harmonization should make differences explicit, not erase them.
Source-aware records03 / Evidence & methodology
Longitudinal views can describe observed changes. They do not by themselves establish why those changes happened. Selection, confounding and regression to the mean may matter; a suitable comparator and analysis plan depend on the question. Report uncertainty and limits to generalisability rather than turning a pattern into an intervention claim.
Benefits and evaluation04 / Evidence & methodology
Payers need evidence relevant to their population and decision. Teams need inspectable outputs and accountable review workflows. Patients and members need meaningful questions, manageable burden and understandable feedback. Operational completion, patient experience and health outcomes are different dimensions; none should be substituted for the others.
Payer and user comparisonExternal methodology references
These external references inform evidence evaluation. They do not constitute Clarrio.ai validation or endorsements.
Provenance, quality and relevance to the research question and population.
Read external guidanceStudy planning, population definition, data collection, quality and analysis.
Read external guidanceChoosing measures appropriate to their intended context of use; this guidance concerns medical-product labeling claims.
Read external guidanceImplementation alongside functioning services, with attention to feasibility, acceptability and equity.
Read external guidanceConnect this topic to a program, product workflow or evidence requirement.