Longevity & AgingResearch PaperOpen Access

Wearable Sensors Reveal How Chronic Low Back Pain Patients Really Move During Daily Tasks

Body-worn sensors tracked spine and hip motion in 954 adults with chronic low back pain, showing hip-dominant movement and clear differences by sex and age.

Sunday, November 15, 2026 0 views
Published in JOR Spine
Physical therapist attaching small wireless sensors along the spine of a middle-aged patient bending forward in a bright clinic

Summary

Researchers at the University of Pittsburgh used four wearable motion sensors to measure how 954 adults with chronic low back pain move during bending, twisting, sit-to-stand, and lifting tasks. During twisting and side bending, the trunk moved the most. During functional tasks such as lifting and standing up, the hips did most of the work. When bending forward, people relied mainly on the hips, with the lower back contributing slightly more in the first part of the motion. Women generally showed greater range of motion and higher speeds than men, and people under 60 moved more dynamically than older adults, except at the hip during forward and backward bending. The data create a reference set for future work to define pain subtypes. No healthy comparison group was included yet.

Detailed Summary

Chronic low back pain is among the costliest and hardest-to-treat chronic conditions, and patients with similar pain can move very differently. Objective movement measures could help sort patients into subgroups and guide treatment. Optical motion capture, the gold standard, is impractical in most clinics, which makes wearable inertial sensors an attractive alternative.

This study is part of the NIH HEAL Initiative's Back Pain Consortium and the University of Pittsburgh LB3P cohort. Adults with chronic low back pain (pain for at least 3 months, on at least half the days of the past 6 months) wore four sensors at T1/T2, T12/L1, L5/S1, and the right thigh. They performed self-paced maximal movements (axial rotation, lateral bending, flexion/extension) and fast-paced tasks (rotation, bending, combined rotation-flexion, five times sit-to-stand, and lifting a 3.6 or 5.85 kg box). The team calculated range of motion, velocity, acceleration, and lumbopelvic rhythm, the ratio of lumbar to hip motion across quarters of the movement. Usable data came from 954 participants (mean age 58.6, 60% female, mean BMI 31.4).

Movement patterns depended on the task. During isolated twisting and side bending, the trunk showed the greatest motion, while the hip contributed most during forward and backward bending. During functional tasks such as combined rotation-flexion, sit-to-stand, and lifting, overall excursion was lower and the hip had the largest share. Lumbopelvic rhythm showed a hip-dominant pattern, with slightly greater lumbar contribution early in flexion. Women generally had larger ranges of motion and higher velocities than men. Participants under 60 moved more dynamically, except at the hip during flexion/extension, where those 60 and older showed more excursion. Most variables were not normally distributed, so medians and interquartile ranges are reported.

The results provide reference values for chronic low back pain and support person-specific approaches to understanding biomechanics. Future work will add asymptomatic controls and combine the movement data with behavioral and biological measures to identify distinct phenotypes that might predict treatment response.

Caveats include the descriptive design with no healthy comparison group, so the findings cannot show which patterns are specific to pain. Sample sizes varied by test because of partial completion and sensor problems, and no data were excluded, so outliers remain in the summaries. Sex and age comparisons were descriptive, and the cohort came from a single health system and was mostly White. Sensor-derived angles also carry measurement error compared with optical motion capture.

Key Findings

  • Analyzed IMU data from 954 adults with chronic low back pain across self-paced and fast-paced functional tasks.
  • Trunk dominated motion in twisting and side bending, while hips contributed most in lifting, sit-to-stand, and combined rotation-flexion.
  • Lumbopelvic rhythm was hip-dominant, with slightly greater lumbar contribution in the first quarter of flexion.
  • Females generally showed greater range of motion and higher velocities than males.
  • Participants under 60 moved more dynamically, except at the hip during flexion/extension, where those 60 and older moved more.

Methodology

Prospective observational cohort (LB3P, NIH BACPAC/HEAL) analyzing baseline enrollment-visit data from 954 adults with chronic low back pain. Four IMUs (T1/T2, T12/L1, L5/S1, right femur) captured ROM, velocity, acceleration, and lumbopelvic rhythm during nine standardized tasks. Data are reported as medians and IQRs, stratified by sex and age (<60 vs ≥60).

Study Limitations

There is no asymptomatic control group, so the findings cannot separate pain-related movement from normal variation. The analysis is descriptive, with varying sample sizes per test, outliers retained, and a mostly White, single-region cohort with a mean BMI of 31. IMU-derived measures, though validated against optical capture, may be less accurate than the gold standard.

Enjoyed this summary?

Get the latest longevity research delivered to your inbox every week.

Enter your email to subscribe: