Therapeutic area · Diabetes, obesity, pre-diabetes
HbA1c and lipid panels summarize months in one number. Clarrio.ai adds the glucose, weight, sleep, activity and nutrition context that sits between those numbers. Align clinical, treatment and behavioral information on one shared longitudinal timeline, where available and appropriate to the program.
What Clarrio.ai captures
Longitudinal health intelligence
Align clinical observations, medication or intervention history, and daily-life context without losing source meaning. Available data and review boundaries depend on the program.
One person / one shared timeline
Illustrative research data · Synthetic demonstration
Health and disease trajectories
Fasting glucose
Treatments and interventions
Exposure · dose · reported experience
Behavior and daily life
Average daily activity
Week 4 / Disease125 mg/dL · Illustrative laboratory record
Week 4 / TreatmentReview · No dose change recorded
Week 4 / Behavior29 min · Routine resumed
Separate measures and independently scaled channels; values appear in their own units. Temporal alignment does not establish causality.
The evidence gap
Diabetes and obesity care generates rich continuous data, including CGM traces, smart scales and activity trackers, but it is fragmented across vendor apps and rarely reaches research datasets in a harmonized form.
With new therapies changing weight and glycemic trajectories, sponsors and clinicians need to understand real-world response over time: persistence, side effects, lifestyle change and what happens after dose adjustments or discontinuation.
Type 2 diabetes
Time-in-range and glucose variability change weekly; HbA1c reflects a long average.
Obesity therapies
Real-world persistence and weight trajectories after starting or stopping therapy.
Pre-diabetes
Early lifestyle change is visible in daily data long before labs move.
Type 1 diabetes
Sleep, exercise and illness context around hypoglycemia events.
Weekly mean glucose (mg/dL) over 120 days
Observations2
Events in periodNot visible
Events with no nearby visit3 of 3
Illustrative · synthetic data
How Clarrio.ai addresses the gap
Device integration
CGM, scales and wearables harmonized to one timeline.
Lab harmonization
Results mapped to LOINC with units, comparable across labs.
Lifestyle check-ins
Nutrition, appetite and side-effect questionnaires.
Cohort analytics
Trajectories grouped by therapy, start date or program.
Data captured
Select a layer. Exact sources depend on deployment configuration and data permissions.
Captured from patients or caregivers in the Patient App.
Research & operational outputs
Metabolic dashboard
Biomarkers, sleep, activity and weight trends together.
Therapy trajectory view
Weight and glucose change since therapy start, across a cohort.
Persistence summary
Who continues, pauses or stops, with reported reasons.
Research dataset
Person-week variables with a data dictionary.
Pharmaceutical & biotech
Real-world persistence, tolerability and trajectory evidence for metabolic therapies.
Hospitals & endocrinology
One view of device and lab data before each review.
CROs
Remote device data with fewer site visits.
Payers
Program engagement and follow-up completeness.
Longevity clinics
Metabolic domain profiles within preventive programs.
Program evaluation
These are the KPIs we agree with customers before deployment, measured against their own baseline. Data coverage, patient experience and clinical outcomes answer different questions. Higher engagement alone does not demonstrate better health or lower utilization.
Explore benefits and evidence| Lever | How it is measured |
|---|---|
| Follow-up completion | Scheduled follow-ups with data relative to due |
| Reporting burden | Manual hours exporting device data |
| Cohort retention | Participants active at month 6 and 12 |
| Data usability | Share of person-weeks meeting analysis criteria |
Illustrative use case
Medical affairs, illustrative
Month 0
Participants starting therapy enroll; scale and wearable paired.
Months 1–6
Weekly weight, monthly side-effect and appetite check-ins.
Month 4
Persistence summary highlights pauses; reasons captured in-app.
Month 12
Trajectory dataset supports an observational publication plan.
Illustrative · synthetic data
Clarrio.ai has joined Mayo Clinic Platform's Solutions Studio, supporting its mission to advance interoperable, data-driven healthcare innovation. This does not imply endorsement or clinical validation.
Explore Metabolic Health Intelligence
Explore Metabolic Health IntelligenceEvidence interpretation
Device wear, meal context and intermittent laboratory results represent different measurement layers. Comparisons need consistent windows and source methods. Neither a sensor pattern nor a program check-in establishes the effect of a routine or service.
Benefits in practice
Pharma & biotech
Routine records can describe encounters and prescriptions while leaving daily function, symptom burden and experience under-observed. Adding remote data can fill selected gaps, but a large dataset is not automatically decision-ready. Population selection, questionnaire suitability, exposure timing and missing observations can all change the interpretation of a longitudinal analysis.
01
Define the population, observation window, outcome concept and intended decision before selecting sources.
02
Combine appropriately scoped Patient App reports with available clinical and device data in the Platform; keep reported experience separate from recorded events.
03
Use source-aware review to inspect context and quality. Agree variable definitions, analysis methods and limitations with the study team.
Clinical development, medical affairs and HEOR teams can work from a common record while asking different questions. Clinical operations needs coverage by site and timepoint; an evidence analyst needs definitions and provenance; medical affairs needs an account of the population and study limitations. Those outputs should not be treated as interchangeable.
Participants have a structured way to contribute the experience that encounters may miss. Questions should be meaningful, proportionate and appropriate to the population. Explain the study purpose, who reviews submissions and whether reporting receives an individual response; a research diary is not an emergency service.
A payer-facing evidence package should connect patient experience to a defined decision question and relevant population. Include baseline characteristics, follow-up, missingness and the analysis method. Patient-reported experience can complement other evidence, but it does not alone establish comparative effectiveness, economic value or a coverage recommendation.
Workflow illustration · not a customer result or live record
Assess suitability against the research question: provenance, completeness, accuracy and relevance. Predefine the analysis rather than choosing a favorable pattern after viewing the results. Publication and payer use require appropriate scientific review.
Explore evidence and evaluationPayers & health economics
Claims and encounter datasets can describe recorded utilization without fully capturing function, experience or the practical burden of a program. Patient reports add a different perspective, but they can be incomplete or drawn from a population unlike the members covered by a payer. The task is to establish whether those observations are suitable for a specific appraisal question, not to turn engagement into a savings claim.
01
Specify the relevant covered population, decision, comparator where needed and evaluation horizon.
02
Review provenance, participation, measurement windows and missingness before examining results.
03
Combine appropriately governed patient experience with relevant clinical or utilization evidence; document assumptions and limits for expert appraisal.
HEOR and population-health teams can examine how reported experience varies across the observation period and which groups are represented. A useful output explains the denominator behind a measure and the observations excluded from analysis. Governance must also define the permitted purpose of any linkage to payer-held data.
The member perspective should remain visible in a program evaluation: burden, everyday function, access and what participants consider meaningful. Reporting should not become an unexplained condition of receiving services. Explain the collection purpose and permitted uses in a way members can understand.
Decision support means an inspectable evidence package, not an automatic reimbursement verdict. Include the target population, relevant baseline differences, data sources, follow-up and limitations. Cost and utilization conclusions need appropriate data and analysis; neither can be inferred from app activity or observation coverage.
Workflow illustration · not a customer result or live record
Separate program participation, patient-reported experience, clinical outcomes and economic endpoints. Define each before collection, use an appropriate comparison where the question requires one and examine whether findings transfer to the covered population.
Explore evidence and evaluationPatients, members & caregivers
A person can complete many questions without understanding why they are being asked or who will read the answers. Device observations may look precise while missing personal context. A useful longitudinal journey starts with the person's priorities and makes the reporting purpose, burden and review arrangements understandable.
01
Agree what the care or research program needs to learn and what participation involves.
02
Use appropriate reports alongside available observations; distinguish the patient's own answers from caregiver observations.
03
Discuss a dated interval with the responsible professional, including what is missing and what matters to the person.
Teams gain context for a more specific conversation, but should not assume that a missed report means a person disengaged. Accessibility, fatigue, language and connectivity may affect reporting. Define support and review arrangements before introducing new questions.
Patients and members can keep a dated account of experiences they might otherwise forget. Caregivers can contribute observations when appropriately scoped, without replacing the person's own perspective. Personal patterns invite discussion; they are not instructions or proof that a particular routine produced an outcome.
A program's apparent reach can hide who could not participate. Payers and commissioners should examine access, reporting burden and representativeness alongside outcome evidence. Patient experience is an important perspective in appraisal, not a shortcut to an economic conclusion.
Workflow illustration · not a customer result or live record
Ask whether reporting is understandable, manageable and meaningful. Examine completion alongside accessibility and support needs. A high completion rate alone does not show that people feel heard or that a program has improved their health.
Explore evidence and evaluationRelated solutions
Investigate disease trajectories, treatment exposure and patient experiences in real-world settings. Clarrio.ai aligns clinical, treatment and behavioral observations to support defined research questions.
See solutionConnect clinical information with longitudinal treatment and patient-generated data for research and appropriate healthcare workflows. A shared timeline preserves source context for professional review.
See solutionRequest a demo
Explore how clinical, treatment and behavioral information align around your program and the research or review questions your team needs to answer.
Opens the Calendly booking page in a new tab.
Prefer to write first? Send an inquiry and the team will reply.