Biomarkers are measurable indicators of health status—from sleep quality and activity levels to heart rate variability and stress markers. They transform raw sensor data into actionable health insights that power personalized recommendations, retention predictions, and wellness interventions.
Understanding Biomarkers
A biomarker (biological marker) is any measurable substance, structure, or process in the body that indicates normal or abnormal function, or the presence/risk of disease. In health technology, biomarkers are digital measurements collected from:
- Wearable devices: Smartwatches, fitness trackers, rings, patches
- Smartphone sensors: Accelerometer, screen time, ambient light
- Medical devices: Blood pressure monitors, glucose meters, ECG
- Lab tests: Blood work, genetic testing, diagnostic imaging
- Manual tracking: Weight, symptoms, menstrual cycles
Why Biomarkers Matter
Raw health data alone provides limited value. The real power comes from processing biomarkers into actionable insights:
- Personalized recommendations: "Your sleep quality is declining—reduce workout intensity today"
- Retention prediction: Identify members at risk of churning 2 weeks before they cancel
- Clinical interventions: Detect depression symptoms 3-5 days before subjective awareness
- Insurance underwriting: Wellness scores correlated with claims costs
- Research insights: Population health trends and intervention effectiveness
The 5 Main Biomarker Categories
Health biomarkers are organized into five primary categories, each providing unique insights:
1. Activity Biomarkers
What they measure: Physical movement, exercise intensity, energy expenditure, and daily activity patterns
Key metrics (10 total):
- steps: Total daily step count
- active_hours: Hours with significant physical activity
- active_calories: Calories burned during activity
- intense_activity_duration: Time in high-intensity activity (>6 METs)
- floors_climbed: Vertical movement measurement
- activity_low_intensity_duration: Light activities (1.5-2.9 METs)
- activity_medium_intensity_duration: Moderate activities (3-5.9 METs)
- activity_sedentary_duration: Time spent inactive
- active_energy_burned: Energy during active phases
- total_energy_burned: Resting + active energy expenditure
Collection method: Smartphone sensors (no wearable needed)
Use cases: Fitness apps, corporate wellness, insurance incentives, churn prediction
2. Body Biomarkers
What they measure: Body composition, weight, BMI, and physical measurements
Key metrics (8 total):
- height & weight: Basic anthropometric measurements
- body_mass_index (BMI): Weight/height² calculation
- body_fat: Percentage of total weight that is fat Wearable required
- fat_mass & lean_mass: Body composition breakdown Wearable required
- waist_circumference: Visceral adiposity indicator
- resting_energy_burned: Basal metabolic rate
Collection method: Manual entry + smart scales/wearables for body composition
Use cases: Weight management apps, nutrition tracking, clinical research
3. Vitals Biomarkers
What they measure: Cardiovascular health, respiratory function, and physiological markers
Key metrics (13 total):
- heart_rate_resting & heart_rate_sleep: Heart rate during rest and sleep
- heart_rate_variability (HRV): SDNN and RMSSD measurements
- respiratory_rate & respiratory_rate_sleep: Breaths per minute
- oxygen_saturation & oxygen_saturation_sleep: Blood oxygen levels
- vo2_max: Maximum oxygen uptake during exercise
- blood_glucose: Blood sugar levels
- blood_pressure (systolic & diastolic): Arterial pressure
- body_temperature_basal: Resting body temperature
- skin_temperature_sleep: Skin temperature during sleep
Collection method: Wearables required (smartwatches, chest straps, rings)
Use cases: Clinical monitoring, fitness optimization, stress detection, AFib screening
4. Sleep Biomarkers
What they measure: Sleep duration, quality, stages, and patterns
Key metrics (13 total):
- Basic metrics (smartphone-capable):
- sleep_start_time, sleep_mid_time, sleep_end_time
- sleep_duration
- sleep_debt (weekly average)
- sleep_in_bed_duration
- sleep_regularity (consistency over time)
- Advanced metrics (wearable-only):
- sleep_awake_duration, sleep_interruptions
- sleep_light_duration, sleep_rem_duration, sleep_deep_duration
- sleep_latency (time to fall asleep)
- sleep_efficiency (sleep time / bed time ratio)
Collection method: Basic metrics from smartphones, advanced from wearables
Use cases: Sleep apps, mental health monitoring, circadian optimization, recovery tracking
5. Reproductive Biomarkers
What they measure: Menstrual cycle tracking, fertility windows, and hormonal phases
Key metrics (10 total):
- menstrual_cycle_start_date & end_date: Cycle boundaries
- menstrual_cycle_length & day_number: Cycle progression
- menstrual_phase: Menstruation, follicular, ovulation, luteal
- menstrual_phase dates & lengths: Phase tracking
- fertile_window dates: Optimal conception window
- menstruation_period dates: Active bleeding phase
Collection method: Manual tracking + smartphone data
Use cases: Fertility apps, period tracking, hormone optimization, family planning
Collection Methods: Wearable vs Smartphone
📱 Smartphone Sensors (No Wearable Needed)
What they can track:
- All 10 activity biomarkers
- All 10 reproductive biomarkers
- Basic sleep metrics (duration, timing, regularity)
- Some body measurements (manual entry)
Strengths:
- 100% user coverage
- No hardware required
- Passive collection
- Lower user friction
Limitations:
- No vitals (heart rate, HRV, oxygen)
- No advanced sleep stages
- ~40% of total biomarkers
Best for: Insurance, corporate wellness, mental health apps
⌚ Wearable Devices
What they add:
- All 13 vitals biomarkers
- Advanced sleep stages (light/deep/REM)
- Body composition (smart scales)
- Continuous heart rate monitoring
Strengths:
- High precision physiological data
- Continuous monitoring
- Medical-grade accuracy
- 100% biomarker coverage
Limitations:
- Only ~30% of users own wearables
- User must sync device
- Battery management
- Cost barrier
Best for: Fitness apps, clinical trials, performance optimization
The 30% Problem
Only approximately 30% of the general population owns a wearable device. This creates a fundamental challenge for health applications:
- Wearable-only approach: High precision data for 30% of users, 0% coverage for remaining 70%
- Smartphone-only approach: Good-enough data for 100% of users, no vitals
- Hybrid approach: Baseline smartphone data (100%) + enhanced wearable data (30%)
Recommendation: Use smartphone monitoring as primary source for maximum coverage, enhance with wearables when available. Only Sahha currently offers this hybrid capability.
Industry-Specific Use Cases
🏥 Insurance & Health Plans
Key biomarkers: Sleep quality, activity levels, behavioral patterns, wellness scores
Application: Wellness programs, risk assessment, claims prediction, member retention
Why it works: Wellness scores correlate with healthcare costs. Members with declining sleep/activity show 3.2x higher claim likelihood.
🏋️ Fitness & Gym Chains
Key biomarkers: Activity intensity, exercise frequency, recovery metrics, readiness scores
Application: Personalized workout recommendations, churn prediction, progress tracking
Why it works: Declining activity trends predict churn 2+ weeks in advance. Readiness scores optimize workout intensity.
🧠 Mental Health Platforms
Key biomarkers: Sleep patterns, behavioral regularity, screen time, activity levels
Application: Depression screening, symptom monitoring, intervention triggers, treatment effectiveness
Why it works: Behavioral patterns detect depression symptoms 3-5 days before self-awareness. Sleep disruption strongly correlates with mental health.
💊 Supplement & Nutrition Apps
Key biomarkers: Sleep quality, energy levels, activity patterns, body composition
Application: Personalized supplement recommendations, effectiveness tracking, dosage optimization
Why it works: Sleep biomarkers identify magnesium/melatonin needs. Activity levels inform protein requirements.
🔬 Clinical Research & Trials
Key biomarkers: All categories—comprehensive monitoring for efficacy measurement
Application: Remote patient monitoring, adherence tracking, real-world evidence, safety monitoring
Why it works: Objective biomarkers detect non-adherence early. Continuous monitoring captures intervention effects in real-world settings.
🤖 AI & Machine Learning
Key biomarkers: All available data for model training—behavioral patterns especially valuable
Application: Predictive health models, personalization algorithms, research datasets, health coaching bots
Why it works: Large datasets enable supervised learning. Smartphone data provides behavioral signals unavailable from wearables alone.
From Biomarkers to Intelligence
While biomarkers provide raw measurements, intelligence layers transform them into actionable insights:
Raw Biomarker Data
- 7,500 steps
- 7 hours 23 minutes sleep
- 68 bpm resting heart rate
- 4 sleep interruptions
Value: Descriptive metrics
Question: What does this mean? What should I do?
Intelligence Layers Applied
- Wellbeing Score: 0.68 (Medium)
- Archetype: "Irregular Sleeper"
- Trend: Sleep quality decreasing 3 weeks
- Comparison: Below your baseline by 15%
- Insight: "Poor recovery—reduce workout intensity"
- Prediction: "Churn risk elevated—intervention needed"
Value: Actionable intelligence
Getting Started with Biomarkers
Ready to integrate biomarkers into your application? Here's how to proceed:
- Define your use case: What health outcomes are you trying to influence?
- Identify required biomarkers: Do you need vitals, or is activity/sleep sufficient?
- Assess user coverage needs: Can you assume wearable ownership, or do you need 100% coverage?
- Choose collection method: Smartphone-only, wearable-only, or hybrid approach
- Select a platform: Direct integration vs health data API aggregator
- Consider intelligence layers: Do you need processed insights or just raw data?