How iOS 15 Walking Steadiness Calculates Fall Risk

Apple’s Walking Steadiness feature in iOS 15 utilizes iPhone motion sensors to passively evaluate mobility, classifying a user's fall risk probability over the coming year. By analyzing metrics such as gait speed, step length, asymmetry, and double support time during everyday movement, the operating system converts continuous balance measurements into actionable risk tiers: OK, Low, and Very Low.

Sensor Data Collection

Walking Steadiness relies on the custom sensor fusion of the iPhone’s built-in accelerometer and gyroscope. The device captures movement data automatically when carried in a pocket or attached securely near the waist during flat-ground walking. Over multiple days, the algorithm filters out irregular movements, such as carrying groceries or sudden stops, to isolate a baseline representation of continuous, natural gait.

Key Walking Balance Measurements

The algorithm calculates stability by processing four primary gait parameters:

  • Step Length: The distance between the heel strike of one foot and the heel strike of the opposite foot. Shorter step lengths often indicate decreased forward propulsion or compensatory balance strategies.
  • Walking Speed: The rate of forward movement over time. A decline in steady walking velocity correlates clinically with reduced physical fitness and functional mobility.
  • Walking Asymmetry: The percentage of time in which the duration or length of steps taken by one foot differs from the other. Higher asymmetry frequently points to unilateral pain, injury, or neurological imbalance.
  • Double Support Time: The percentage of a single gait cycle during which both feet are touching the ground simultaneously. A higher double support time is a universal clinical indicator of guarded walking and fear of falling.

Predictive Fall Risk Calculation

To translate these sensor measurements into a mathematical probability of falling, Apple mapped gait algorithms against real-world clinical mobility benchmarks, including datasets derived from the Apple Heart and Movement Study. Traditional physical therapy metrics—such as the Timed Up and Go (TUG) test and the 6-Minute Walk Test—served as validation targets.

The algorithm applies a scoring model that integrates the four balance metrics over a rolling period (typically measured across consecutive days and weeks). Rather than reacting to a single erratic day, the system evaluates long-term trends to prevent false positives.

Classification Tiers

The calculated score assigns the user to one of three fall risk probabilities:

  • OK: Indicates stable balance. The probability of falling within the next 12 months remains low.
  • Low: Mobility data reveals subtle balance degradation. Users in this category are estimated to have an increased risk of falling over the next 12 months compared to the average population.
  • Very Low: Severe balance degradation is detected across multiple gait parameters. The user has a statistically significant, high probability of experiencing a fall within the next year.

When a user transitions into the "Low" or "Very Low" ranges, iOS 15 generates an alert and provides evidence-based mobility exercises within the Health app designed to rebuild stability and reduce the likelihood of future falls.