Skip to content
Wearables & measurementCohort study4 min read

Chronic low back pain: low pain in the moment, and more activity than sensors show

MoveVitals Editorial Collective · Study published · Reading updated

Original paper:In-Field Ecological Momentary Assessment From Wearable Motion Sensors and Self-Report in a Chronic Low Back Pain Cohort. · JOR spine · 2026 · Original paper ↗

Key takeaway

In 989 adults with chronic low back pain who logged pain three times a day for 7 days, the median pain intensity was only 1 on a 0-10 scale, and people rated their activity higher than wrist, waist and back sensors recorded. Participants were a large research-center cohort tracked for just one week, so these profiles may not reflect pain over months.

Study design

Cohort study

Reading basis

Published abstract

Reading focus

Association & confounding

The research question

Pain and activity in chronic low back pain are usually measured once, from memory, which hides day-to-day swings and invites recall bias. The authors asked what pain and activity look like when captured in real time at home, and how people's own activity ratings compare with what wearable motion sensors record.

Who and how

Cohort of 1007 adults with chronic low back pain meeting National Institutes of Health criteria, enrolled through the University of Pittsburgh LB^(3)P Mechanistic Research Center; 989 submitted data. Over 7 days a custom phone app prompted them three times daily to rate current pain intensity (0-10), pain interference (0-10) and activity level (very light to vigorous), and to report sleep and wake times. They wore ActiGraph GT9X sensors on the wrist and waist plus a custom sensor on the skin over the lumbar (L5) segment. Sensor days were kept if they had at least 10 hours of wear, and people needed at least 4 such days. Activity counts were sorted into sedentary, light and moderate-to-very-vigorous using Freedson Adult cutpoints.

What it found

The authors’ reported results. Read each estimate with its population and uncertainty.

  1. 1
    Reported pain was low in the moment: median pain intensity was 1 (IQR 3) and pain interference 3 (IQR 3), and 54.0% of participants had a median intensity of 0 or 1, even though average pain at enrollment was 5.4 (SD 2.1).
  2. 2
    Pain tended to peak in the evening for 30% of participants (intensity) and 40% (interference). Most rated their activity as moderate (36%) or light (33%).
  3. 3
    Sensors told a different story: sedentary time was 47.3% (wrist), 72.8% (waist) and 82.6% (back sensor, 586 participants). Median steps per day were 9575 (wrist, 884 devices), 4114 (waist, 785 devices) and 3695 (back sensor).
  4. 4
    The same people therefore looked quite different depending on where the sensor sat, and self-rated activity was higher than sensor-derived activity. Waist and back activity counts were below the average of US adults.

Limits & uncertainty

  1. 1
    The abstract describes profiles only: it has no comparison group without back pain, so it cannot say whether these pain or activity levels are unusual for this condition.
  2. 2
    Sedentary and activity classes rely on the Freedson cutpoints, which were not built for people with chronic low back pain; the authors themselves say the mismatch could reflect overestimation by participants or poorly fitting cutpoints.
  3. 3
    Only 7 days were observed, the back sensor had data from just 586 people, and sensor placement changed the numbers, so values from different devices are not interchangeable.
  4. 4
    Based on the abstract only; no adjusted or longitudinal analysis is reported.

What to take from it

A single pain score at a clinic visit can miss how much pain varies through the day, and a self-rating of 'moderate activity' may be higher than a sensor would log. Do not read these medians as a target or a norm for your own back pain; talk with your clinician about pain that limits you.

Reading your own data

If you track steps or activity minutes, keep the device on the same body site, because wrist, waist and back placements gave very different counts here. Compare your own weeks to each other, and consider noting pain at the same times of day to see whether evenings stand out for you.

How to read this evidence

Exposure precedes follow-up outcomes, but confounding and reverse causation can remain. A hazard ratio is not an individual absolute risk or a demonstrated intervention effect.

Authors’ structured abstract English

The attributed English abstract is reproduced under the recorded open licence. It is source material, not a MoveVitals interpretation. The prepared analysis above separates the authors’ findings from our interpretation.

Background

Chronic low back pain (cLBP) is a prevalent and debilitating condition. Gaining insight into the daily experiences of those with cLBP is crucial for developing effective management. Pain and activity are typically assessed at a single time point and often rely on retrospective self-reports, which can be prone to recall bias and may not reflect the day-to-day variability of these experiences. As a part of the University of Pittsburgh LB^(3)P Mechanistic Research Center, this study used ecological momentary assessment (EMA) and wearable devices to collect real-time data in a large cohort of adults with cLBP. The primary aims were to collect and characterize pain and activity profiles of individuals with cLBP.

Methods

This study enrolled 1007 adults with cLBP who met the National Institutes of Health defined criteria. Over 7 days, participants were assessed in their own environment. EMA was gathered in real-time via a custom mobile app, prompting participants three times daily to provide their perceptions of current pain intensity (0-10), pain interference (0-10), and activity level (very light to vigorous). Time of falling asleep and waking was also reported. Participants wore ActiGraph GT9X devices on their wrist and waist. A custom back sensor was also adhered to the skin over the lumbar (L5) segment. Activity counts, wear time, and step counts were calculated, utilizing algorithms provided by ActiGraph. Sensor data were filtered to include at least 4 days of 10 or more hours each. Activity counts were categorized into sedentary, light, and moderate-to-very-vigorous based on Freedson Adult cutpoints.

Results

Out of 1007 participants, 989 submitted EMA data (58.8 ± 16.5 years old; 40% male and 60% female; mean pain intensity at enrollment of 5.4 (SD 2.1) and a median of 5 (interquartile range [IQR] 3) on a 0-10 scale; mean PROMIS Pain Interference T -score at enrollment of 60.5 (SD 7.5) and a median of 61.2 (IQR 9.6)). The median reported pain intensity level from the EMA was 1 (IQR = 3), while pain interference was 3 (IQR = 3). More than half of the participants reported a median pain intensity of either 0 or 1 (54.0%) and a median pain interference between 0 and 3 (57.4%). Most participants self-reported their activity levels as moderate (36%) or light (33%). Based on pain ratings during each day, most participants had their pain intensity (30%) and pain interference (40%) peaking in the evening. ActiGraph data from 884 wrist-worn and 785 waist-worn devices were analyzed. Wrist data showed a median of 1 765 325 (IQR 796 995) activity counts/day and 9575 (IQR 4228) steps/day. Waist data showed 358 390 (IQR 223 758) activity counts/day and 4114 (IQR 3146) steps/day. The percentage of daily sedentary activity was 47.3% for wrist and 72.8% for waist. The back sensor data from 586 participants showed a median of 340 345 (IQR = 223 399) activity counts/day and a median of 3695 (IQR = 2743) steps/day. The percentage of time spent in daily sedentary activity was 82.6%. Both ActiGraph devices and the back sensor indicated that the majority of the time was spent in sedentary activity level, which is lower than the activity level reported in the EMA.

Conclusions

Despite having cLBP with self-reported moderate pain levels, participants generally reported periods of relatively low levels of pain intensity and interference in their EMA. In addition, their EMA-reported activity levels differed from the sensor data. Participants self-reported higher levels of activity compared to the activity levels calculated by the wearable sensors. This suggests that participants overestimated their activity levels on EMA, or that the activity level cut-points may need to be re-evaluated for the cLBP population. Additionally, sensors placed on different body locations showed varying activity and step counts. The activity counts calculated from the waist ActiGraph and the back sensor from this cohort were lower than the average activity counts in the US adult population. Further research is needed to better quantify these differences for people with cLBP to develop a more comprehensive understanding of the pain experience.

Sources & attribution

In-Field Ecological Momentary Assessment From Wearable Motion Sensors and Self-Report in a Chronic Low Back Pain Cohort.

Bell KM, Alfikri Z, Anderst W, Clark WW, Cook HA, Darwin J, Dicianno BE, Greco CM, Hoydick J, Jakicic JM, McKernan GP, Parmanto B, Patterson CG, Piva SR, Roos RE, Schneider MJ, Sundaram M, Vo NV, Zhou L, Sowa GA.

JOR spine · 2026-02-01

DOI: 10.1002/jsp2.70156 ↗PubMed · 41635465 ↗Europe PMC · Recorded licence: cc by ↗

Abstract © the original authors, reproduced under the recorded open licence. Translations support reading; use the English record to check terminology and numerical results. Our methodological interpretation is separate from the authors’ conclusions.

MoveVitals editorial interpretation · educational content · no individual clinical review recorded

Sources checked 2026-11-01

Practical guides

Read in context

·Cohort study

Activity–stroke and heart attack curves differ by accelerometer processing method

Among 90,237 UK Biobank adults free of cardiovascular disease at baseline, followed until December 2022 for 1,298 strokes and 2,031 myocardial infarctions, the shape of the link between moderate-vigorous physical activity (MVPA) and each outcome depended on how the accelerometer data were processed. For stroke, hazard fell in a straight line only with the machine-learning metric; for myocardial infarction, two metrics gave a curved decline and activity counts a straight one. The abstract gives no hazard ratios, so it shows that methods disagree, not which one is right, and the findings are observational associations.

Preventive Medicine

In-depth explainersRead study →
·Validation study/review

Smartwatch data as forensic evidence: heart rate is best validated, the rest uneven

A systematic review and meta-analysis of studies on smartwatches from four major brands, framed around digital forensics, found that heart rate was both the most studied and the most accurate measure. Step count, distance, energy expenditure, activity classification and sleep were less consistently validated, and results varied by brand, protocol and reporting. The abstract gives no pooled numbers, and most studies were run in healthy people in laboratories, so it cannot say how a particular watch performs for a particular person in daily life.

Forensic Science International

In-depth explainersRead study →
·Overview of reviews

Activity trackers: about 1,800 more steps a day, but small effects on other outcomes

An umbrella review of 39 systematic reviews and meta-analyses, covering 163,992 participants, found that activity trackers were associated with more physical activity (SMD 0.3–0.6), equating to about 1,800 extra steps and 40 minutes more walking per day, and with a bodyweight reduction of about 1 kg. Effects on blood pressure, cholesterol, glycosylated haemoglobin, quality of life and pain were typically small and often non-significant. The abstract does not say what else came with the trackers in these programs or how much the reviews overlap, so the gains cannot be credited to the device alone.

The Lancet Digital Health

In-depth explainersRead study →
·Validation study/review

Seven sleep trackers: good at spotting sleep, weak on wake and sleep stages

In a sleep laboratory, 34 healthy young adults wore a subset of seven consumer sleep trackers for three nights. Against polysomnography every device detected sleep very well (sensitivity 0.93 or higher) but caught wake poorly (specificity 0.18–0.54), and five of the seven matched or beat research actigraphy (specificity 0.39). Sleep-stage estimates were inconsistent, and the participants were healthy adults aged 18 to 35 in a fixed 8-hour bed window, with each device tested in only 11 to 20 of them, so the results say little about people with sleep problems.

Sleep

In-depth explainersRead study →
Wearables & measurement →