Skip to content

Clinical & Data Science

The clinical and data science function is led by the Founding Machine Learning Scientist, who owns the ML strategy, clinical data analysis, research methodology, and data quality standards. This role bridges the gap between clinical science and the technical platform.

Team Lead

  • Founding Machine Learning Scientist: Owns ML model development, clinical data analysis, data collection design, and research methodology. Reports directly to the Founder/CEO.

Responsibilities

The ML Scientist owns:

  • Data science & ML: Model development, clinical data analysis, feature engineering
  • Clinical data model: What data to collect, form design, validation rules
  • Data quality: Ensuring accuracy, identifying outliers, missing data patterns
  • De-identification: HIPAA Safe Harbor compliance in data pipelines
  • Research methodology: Study design, statistical analysis, evidence base
  • Research partnerships: IRB submissions, academic collaborations, publications

This function does NOT own: - Platform engineering and infrastructure (owned by Founding Technical Lead) - Business development and clinic relationships (owned by Commercial Lead) - Company strategy (owned by Founder/CEO)

Medical Advisory Board

We maintain a clinical advisory board of external menopause experts:

  • Gynecologists (2-3): Diverse practice settings
  • Internal medicine physicians (1): Primary care menopause management
  • Nurse practitioner (1): Advanced practice perspective
  • Researcher (1): Academic menopause research

The board meets quarterly to: - Review data findings - Validate clinical approaches - Advise on research partnerships - Provide clinical guidance on new features

Data Collection & Form Design

Process for New Data Elements

When engineering or a customer wants to collect new data:

  1. Clinical assessment: Does this data help answer a clinical question?
  2. Validation: Is there evidence supporting its collection?
  3. Design: How do we ask the question clearly?
  4. Testing: Does the form work in practice with providers?
  5. Implementation: Engineering builds it
  6. Monitoring: Track data quality

Current Data Collection Areas

Patient Demographics - Age, gender identity, race/ethnicity - Insurance, occupation, education (optional)

Menopause Status & Symptoms - Years since last menstrual period - Vasomotor symptoms (hot flashes, night sweats) - Genitourinary atrophy (vaginal dryness, dyspareunia) - Mood (depression, anxiety, mood swings) - Cognitive (memory, concentration) - Sleep (insomnia, quality)

Medical History - Contraindications to hormone therapy - Risk factors (VTE, stroke, cancer risk) - Prior treatments and outcomes - Comorbidities

Treatment Data - Current treatments (HT, non-hormonal options, lifestyle) - Dosage, formulation, duration - Adherence and tolerability - Side effects

Outcome Measures - Symptom resolution/improvement - Quality of life - Healthcare utilization - Treatment changes

Data Quality Standards

We maintain high standards for data accuracy:

  • Completeness: Required fields filled out
  • Consistency: Data doesn't contradict itself
  • Accuracy: Age and dates make sense
  • Clinically plausible: Values within expected ranges
  • Uniqueness: No duplicates

When data quality issues are identified: - Provider is notified - Data is flagged (not deleted) - Provider can correct and resubmit - We track quality metrics to help providers improve

Provider Relationships

Clinic Onboarding

The clinical team leads clinic onboarding:

  1. Initial contact: VP Clinical or Clinical PM meets with clinic
  2. Needs assessment: Understand their workflow, patient population, use cases
  3. Form customization: Adjust data collection forms to clinic needs (within clinical standards)
  4. Training: Train clinic staff on using MenoTime
  5. Early support: Extra support during first weeks
  6. Feedback: Regular check-ins to gather feedback

Provider Support

Ongoing support includes:

  • Form assistance: Help with data entry questions
  • Data interpretation: Explain reports and findings
  • Feedback loop: Listen to provider experience, feed back to product
  • User adoption: Tips for integrating MenoTime into workflow

Feedback Mechanism

Providers regularly provide feedback through: - Feature requests - Bug reports - User experience feedback - Clinical questions

Clinical team collects and synthesizes this feedback for product prioritization.

Provider Success Metrics

We track:

  • Clinic activation: How quickly clinics start submitting data
  • Data quality: Completeness and accuracy of submissions
  • Retention: How many clinics remain active over time
  • Satisfaction: NPS score, qualitative feedback
  • Utilization: Are clinics using reports and insights?

Data Analysis & Research Support

Population Health Analytics

Our analytics team (clinical + data) produces:

  • Benchmark reports: How does your clinic compare to standards?
  • Trend analysis: How are outcomes changing over time?
  • Treatment patterns: What treatment approaches are providers using?
  • Outcomes analysis: Which treatments work best for different populations?

Research Partnerships

We partner with academic medical centers for research:

  • Data access: Providing de-identified data to researchers
  • IRB support: Helping with institutional review
  • Publication support: Helping researchers publish findings
  • Research agenda: Informing what questions to ask with our data

Our Internal Research

Clinical team works on:

  • Algorithm validation: Does our de-identification work as intended?
  • Outcome prediction: Can we predict which treatments work best?
  • Literature reviews: Validating clinical approaches
  • Quality improvement studies: What practices lead to better outcomes?

Clinical Decision Support

(Future roadmap item)

If we build clinical decision support features:

  • Recommendations based on symptom profile
  • Treatment guideline decision trees
  • Risk assessment algorithms

These would be validated by the clinical team and advisory board before launch.

Compliance & Ethics

HIPAA in Clinical Context

The clinical team ensures:

  • De-identification is clinically valid (not just technically compliant)
  • Patient privacy is protected at every stage
  • Data use follows HIPAA minimum necessary principle
  • Audit logs are adequate for healthcare standards

Research Ethics

For research partnerships:

  • Ensure IRB approval is obtained
  • Verify proper informed consent
  • Ensure research serves patient benefit
  • Maintain ethical standards in data sharing

Clinical Ethics

We follow clinical ethics principles:

  • Beneficence: Research benefits patient care
  • Justice: Data use serves diverse populations fairly
  • Non-maleficence: We don't harm through inappropriate data use
  • Autonomy: Patients/providers know how their data is used

Collaboration with Engineering

The clinical and engineering teams work closely:

Clinical → Engineering

Clinical team provides: - Data requirements (what to collect) - Validation rules (what's acceptable) - Form designs (how to ask questions) - Analysis specifications (what insights matter)

Engineering → Clinical

Engineering provides: - Technical feasibility assessment - Performance and scalability guidance - Security/compliance implementation details - Infrastructure and tool recommendations

Joint Decisions

Together, teams decide: - Data model design - De-identification approach - Security controls - Validation algorithms - Reporting capabilities

Team Meetings

Weekly Clinical Sync

  • Time: Tuesday, 11 AM PT
  • Duration: 30-45 min
  • Attendees: Clinical team + invited clinicians
  • Agenda: Data quality issues, research updates, provider feedback, upcoming work

Bi-weekly Clinical Product Sync

  • Time: Thursday, 1 PM PT
  • Duration: 30 min
  • Attendees: Clinical PM, Engineering PM, VP Eng
  • Agenda: Feature prioritization, requirements refinement

Monthly Advisory Board Meeting

  • Time: First Friday, 1-2 PM PT
  • Attendees: Advisory board members, VP Clinical, CEO
  • Agenda: Findings review, clinical guidance, research opportunities

Ad-hoc Engineering-Clinical Sync

  • When: For major data model changes, new research partnerships, complex algorithms
  • Duration: 1-2 hours
  • Format: Working session to align on approach

Provider Communication

Regular Updates

We keep providers informed through:

  • Monthly newsletter: Highlights from data, new features, tips
  • Quarterly business review: Clinic-specific performance, goals
  • Annual summit: In-person gathering of providers, presentations, networking

Feedback Channels

Providers can reach out through: - In-app support: Questions about forms or using MenoTime - Email: support@timelessbiotech.com - Dedicated clinic contact: VP Clinical or Clinical PM for major concerns - Feedback portal: Asana or form for feature requests

Goals & Metrics

The clinical team drives:

  • Clinical accuracy: 100% (zero clinical errors in data model)
  • Data quality: Target 90%+ completeness on key fields
  • Provider satisfaction: NPS >50
  • Research partnerships: 2-3 active partnerships per year
  • Publications: 2-4 peer-reviewed publications per year

Last updated: February 2025