Health Tech Buying Guide 2026

The Health Tech Buying Framework

Consumer health technology ranges from FDA-regulated medical devices with rigorous clinical validation to unregulated wellness gadgets with no evidence base whatsoever — and they are often marketed with nearly identical language. The buyer's responsibility is to distinguish between them before spending money.

We apply a five-question framework to every product we test:

1. What specific health claim is being made? "Tracks your health" is not a claim. "Detects atrial fibrillation with 98.5% sensitivity" is. The more specific the claim, the more evaluable it is.
2. Is there a reference standard? Every accuracy claim requires a gold standard comparison. Heart rate accuracy means nothing without ECG comparison. Sleep staging accuracy means nothing without PSG comparison. Calorie tracking accuracy means nothing without USDA/laboratory reference data.
3. Who produced the evidence? Company-funded studies have a documented publication bias. Independent peer-reviewed studies and FDA submissions carry more weight.
4. Does FDA clearance apply? FDA clearance is required for products making diagnostic claims about specific medical conditions. General wellness claims don't require clearance — which means most consumer wearables are unregulated by FDA regardless of their health-adjacent language.
5. What is the subscription/ongoing cost? Health tech has moved to subscription models. The true cost of a health tracker is the device cost plus 3 years of subscriptions — often 2-3x the initial purchase price.

Wearables: What Actually Matters

Heart Rate Monitoring

Look for: published MAE (mean absolute error) against ECG reference during exercise — particularly at high intensities. Manufacturer-published accuracy claims typically test at rest or moderate intensities, where all devices perform well. The differentiating performance is during high-intensity intervals. For precision heart rate training, consider pairing any wrist wearable with a chest strap for workouts.

Ignore: claims about "research-grade" accuracy without published data.

GPS and Route Tracking

Look for: multi-band GNSS (Multi-Band, Multi-GNSS, or Dual-Frequency GPS in product listings). Single-band GPS is meaningfully less accurate in urban environments. Distance error on a certified course is the standard we use — look for <0.5% error.

Battery Life

Always compare battery life in the same GPS mode. "2 weeks battery" means nothing if that mode doesn't have GPS active. GPS battery life (continuous outdoor GPS) is the relevant metric for runners and cyclists.

Platform Ecosystem

The most important practical consideration is often the ecosystem: Garmin integrates best with Garmin Connect and TrainingPeaks; Apple Watch integrates best with Apple Health and iPhone-first apps; Whoop is platform-agnostic; Oura integrates widely. Choose the ecosystem that works with apps you actually use.

Health Apps: Evaluation Criteria

Calorie and Nutrition Tracking

Key metric: database size and accuracy for common foods. For AI photo recognition: MAE against USDA FoodData Central reference values for a standardized meal set. Ask: does the app log micronutrients, not just macros? Does it integrate with Apple Health/Google Fit? Is the database crowdsourced (verification issues) or professionally curated (more reliable)?

Mental Health and Meditation Apps

Clinical validation standard: randomized controlled trials (RCTs) showing efficacy against validated outcome measures (PHQ-9 for depression, GAD-7 for anxiety, PSS for stress). Several major apps have published RCTs; many have none. Check the app's research page and look for studies in peer-reviewed journals, not internal publications.

Subscription Value Assessment

Calculate 3-year total cost: device + (monthly subscription × 36). Many health apps offer free tiers that cover basic functionality — assess whether premium features are genuinely used before committing to subscriptions.

Medical Devices: Buyer Safety

Consumer medical devices — blood pressure monitors, CGMs, pulse oximeters — should be evaluated against clinical validation standards, not consumer electronics criteria.

For blood pressure monitors: look for AAMI SP10 or ESH/IP validation. This means the device has been tested against a mercury sphygmomanometer reference across a diverse population, and 95% of readings fall within ±5 mmHg of the reference. Many inexpensive blood pressure monitors have not undergone this validation.

For CGMs: look for FDA clearance (required) and published MARD (mean absolute relative difference) against laboratory blood glucose reference. The lower the MARD, the better. Dexcom G8 achieves 7.4-7.6% in our testing and published studies — this is currently the gold standard for consumer CGMs.

For pulse oximeters: look for FDA 510(k) clearance. Consumer wearable SpO2 features are not FDA-cleared as medical devices and should not be used for clinical decisions. Only FDA-cleared pulse oximeters meet the ±2% SpO2 accuracy standard required for clinical use.

Red Flags to Avoid

• "Clinically validated" without specifying what clinical standard, by whom, and where the data is published
• Accuracy percentages without reference to what the comparison was made against
• Health score claims without disclosed calculation methodology
• FDA "registered" (which means little) vs FDA "cleared" or "approved" (which means clinical validation)
• Before/after testimonials without blinded controlled study data
• Privacy policies that allow sharing "de-identified" data with third parties without opt-out