Apple Health Data Explained: What Each Metric Means and How to Use It

Your iPhone quietly accumulates thousands of data points about your body every day. Most people open the Health app a few times, notice a wall of numbers, and close it again.

That is a missed opportunity. Understanding what each metric actually measures, and where its limits are, turns a passive log into something you can act on. This guide covers the core Apple Health metrics, how they are collected, what the research says about their accuracy, and how to get the data out of your phone so an AI assistant or clinician can help you make sense of it.

---

How Apple Health Aggregates Data

Apple Health is not a single sensor. It is a centralized database called HealthKit that pulls from multiple sources simultaneously. An Apple Watch captures continuous motion, heart rate, and electrical heart signals. Third-party apps like Strava, MyFitnessPal, and sleep trackers can write their own data to the same store. Connected medical devices such as blood pressure cuffs and continuous glucose monitors can contribute too, as long as they support HealthKit integration.

The result is a single timeline that merges inputs from every authorized source, with the most recent or highest-quality reading taking precedence when sources overlap. Apple's developer documentation describes this architecture in detail for developers who need to query the store programmatically.

---

Resting Heart Rate

Resting heart rate (RHR) is the number of times your heart beats per minute while you are fully at rest. Apple Watch estimates it by averaging low-activity readings taken throughout the day, not from a formal sitting measurement. The American Heart Association considers a normal RHR to be between 60 and 100 beats per minute for adults, though trained athletes often run lower.

A sustained upward trend in your RHR, say 5 to 7 beats above your normal baseline over several consecutive days, can signal inadequate recovery, illness brewing, or accumulated stress. That directional signal is more clinically meaningful than any single reading.

---

Heart Rate Variability via SDNN

Heart rate variability (HRV) measures the fluctuation in time between consecutive heartbeats. Unlike many consumer devices that use RMSSD, Apple HealthKit calculates HRV as SDNN, the standard deviation of the inter-beat (RR) intervals between normal heartbeats, typically expressed in milliseconds. Apple's official HealthKit documentation confirms this definition explicitly.

Higher SDNN generally indicates a more adaptable autonomic nervous system, which correlates with better cardiovascular fitness and recovery capacity. Apple Watch records a short HRV sample during each Breathe or Reflect session and also during overnight sleep. Because SDNN is computed over a short epoch rather than a full 24-hour recording (the clinical gold standard), it is best treated as a trend indicator rather than a clinical measure.

One thing worth knowing: if you are comparing your Apple Health HRV number to Oura or Garmin, you may not be comparing the same metric. Apple reports SDNN; Oura reports RMSSD. Both track autonomic function, but the raw numbers are not interchangeable. Your Oura RMSSD and your Apple SDNN could both be "40" and mean different things physiologically.

---

Walking Heart Rate Average

Walking Heart Rate Average is the mean heart rate recorded during periods of low-level activity, specifically when the accelerometer detects a walking pace. It is a quieter signal than exercise heart rate and tends to drop over time as cardiovascular fitness improves. Some researchers consider it a useful proxy for aerobic efficiency because it reflects the heart's workload during everyday movement rather than during a controlled test.

---

VO2 Max Estimation and Its Accuracy

Apple labels this metric "Cardio Fitness" and expresses it in mL/kg/min, the same unit used in clinical VO2 max testing. The estimate is generated by measuring your heart rate response during outdoor walks, runs, or hikes, and requires a heart rate elevation of roughly 30 percent above resting.

How accurate is it? Apple internally validated its algorithm on 755 participants (534 for development, 221 for validation), achieving an intraclass correlation coefficient of 0.86 to 0.89, solid for a consumer device. An independent 2025 validation study published in PLOS One found that Apple Watch Series 9 underestimated VO2 max by a mean of 6.07 mL/kg/min compared to indirect calorimetry (MAPE of 13.31 percent), with the underestimation most pronounced in highly fit individuals. For tracking trends over time within the same person, the device performs adequately. For comparing your number to clinical thresholds or to other devices, keep the margin of error in mind.

---

Sleep Stages: A Relatively New Addition

Apple added automatic sleep stage detection in watchOS 9 (2022), and Apple published an updated algorithm in late 2025 that improves classification accuracy. Sleep is divided into REM, core (light), deep, and awake periods.

A 2024 validation study in Sensors compared Apple Watch, Oura Ring, and Fitbit against polysomnography (PSG), the clinical gold standard, in 30 participants. For binary sleep-versus-wake detection, Apple Watch achieved 97 percent sensitivity. In four-stage classification (wake, light, deep, REM), it overestimated light sleep by about 45 minutes and underestimated deep sleep by about 43 minutes compared to PSG. Oura Ring outperformed both competitors on four-stage accuracy. Apple Watch sleep data is useful for spotting trends in total sleep time and sleep timing consistency; the specific stage durations carry more uncertainty.

---

Activity Rings: Move, Exercise, Stand

The three rings measure active energy burned toward your personalized Move goal, minutes of exercise at or above a brisk walk, and hours of the day that included at least one minute of standing. The exercise ring uses heart rate and motion together to determine activity intensity. These rings are motivational tools designed to encourage daily consistency, not clinical outputs.

---

Mobility Metrics

Mobility metrics, including walking speed, step length, double support time, and stair speed ascent and descent, appear under the Mobility category. These are derived from accelerometer and barometer data. They are particularly relevant for older adults: a 2019 study in JAMA Network Open found that digital gait metrics captured by consumer devices correlated with fall risk. Walking asymmetry and cardio fitness low-level alerts may prompt users to discuss changes with a clinician.

---

Blood Oxygen (SpO2)

Blood oxygen measurement is available on Apple Watch Series 6 and later. The sensor uses red and infrared light reflected from the wrist to estimate peripheral oxygen saturation. Apple explicitly notes that the measurement is designed for general wellness purposes and is not intended to replace clinical pulse oximetry. In 2024, Apple removed SpO2 from new Watch Ultra and Series 9 models in the US temporarily during a patent dispute; the feature was later restored via software update. Normal SpO2 readings are generally considered 95 percent or above.

---

ECG and AFib History

The ECG app, available on Apple Watch Series 4 and later, generates a single-lead electrocardiogram by having the wearer place a finger on the Digital Crown to complete a circuit. The Apple Heart Study, published in the New England Journal of Medicine in 2019 and conducted by Stanford University with over 400,000 participants, demonstrated that the irregular-pulse detection algorithm had an 84 percent positive predictive value for atrial fibrillation when confirmed against simultaneous ECG patch monitoring.

Separately, Apple Watch can passively flag AFib History, the estimated percentage of time the heart was in an irregular rhythm, available in watchOS 9 and later on Series 4 and newer hardware. These features do not diagnose AFib or any other condition; they are intended to prompt a follow-up conversation with a physician when irregular patterns are detected.

---

Apple's Privacy Model

Apple Health data is encrypted on-device when your iPhone is locked. Per Apple's legal privacy documentation, health and activity data synced to iCloud uses end-to-end encryption when you have iOS 12 or later and two-factor authentication enabled, meaning Apple cannot access it. Health data is excluded from unencrypted iCloud backups entirely; it only appears in local backups if you have set an iTunes encryption password.

---

Exporting Your Data

To export everything Apple Health has collected: open the Health app, tap your profile picture in the upper right corner, scroll to the bottom, and tap Export All Health Data. After a brief processing period, you receive a ZIP archive containing an export.xml file. This file is comprehensive but not human-readable without a parser. Third-party apps and AI health assistants can ingest this data to produce trend summaries, flag anomalies, or prepare summaries you can bring to a clinician.

---

Limitations Compared to Garmin and Oura

Apple Health does not currently offer a composite recovery score comparable to Garmin's Body Battery index or Oura's Readiness Score, which synthesize HRV, resting HR, sleep, and activity load into a single daily number. Apple Watch sleep stage accuracy lags Oura Ring on four-stage classification, particularly for deep sleep detection (50.5 percent sensitivity for Apple Watch versus 79.5 percent for Oura in the 2024 Sensors study cited above). Garmin devices calculate HRV using a full-night average, which aligns more closely with clinical SDNN methodology than Apple's short-epoch measurements. For users whose primary goal is sleep optimization or recovery tracking, dedicated devices may offer more actionable outputs.

---

Using AI to Interpret Your Apple Health Data

Exporting your Apple Health archive and sharing it with an AI health assistant unlocks a layer of analysis that the native app's charts cannot provide. A well-designed assistant can surface multi-metric correlations, for example identifying that your resting HR tends to spike two days after nights when your sleep efficiency drops below 80 percent, and help you frame questions to bring to your doctor. Vidaya's Vaya Chat is built for exactly this kind of longitudinal data review, available at $10 per month or $89 per year.

---

Frequently Asked Questions

What is the difference between HRV and resting heart rate in Apple Health? Resting heart rate is the average number of beats per minute during calm activity. HRV (reported as SDNN) measures the variability between individual beats. Both trend downward during poor recovery, but HRV is often considered a more sensitive early warning because it reflects autonomic nervous system balance rather than average workload.

How accurate is Apple Watch sleep tracking? For detecting whether you are asleep versus awake, Apple Watch achieves about 97 percent sensitivity. For distinguishing between sleep stages, accuracy is lower: a 2024 study found it overestimates light sleep by roughly 45 minutes and underestimates deep sleep by about 43 minutes compared to polysomnography.

Can I share my Apple Health export with my doctor? Yes. The XML export can be shared directly, but most clinicians will not have a tool to read raw XML. Using an AI health assistant or a clinical health data viewer to generate a summary document is a more practical approach.

Does Apple see my health data? With two-factor authentication enabled on iOS 12 or later, health data synced to iCloud is end-to-end encrypted. Apple states it cannot decrypt, view, or access this data. Without two-factor authentication, standard iCloud backup encryption applies, where Apple holds the keys.

Why does my Cardio Fitness score seem lower than my actual fitness? The 2025 PLOS One validation study found Apple Watch tends to underestimate VO2 max in highly fit individuals. If your true VO2 max is above average for your age, the displayed "Cardio Fitness" number may read lower than a laboratory test would show.

What Apple Watch models support ECG and blood oxygen? ECG is available on Series 4 and later. Blood oxygen (SpO2) is available on Series 6 and later, Apple Watch Ultra, and SE (2nd generation and later), subject to regional availability and applicable hardware and software versions.

---