Smartwatch for Osteoarthritis Knee Pain

This study is exploring if a smartwatch, specifically the Embrace Plus, can measure pain levels in people with knee osteoarthritis. The goal is to see if the electrodermal activity (EDA) signals detected by the smartwatch can provide a reliable way to understand your pain. This is a small pilot study to test the methods for future, larger studies. You might be able to join if you are 50 to 80 years old and have severe knee osteoarthritis (Kellgren-Lawrence Grade >= 3). We are looking for 15 participants. Success for this study means showing that the smartwatch can capture enough EDA signals to measure pain and that the study procedures are practical.

Study design
This is an interventional study with a planned enrollment of 15 participants. It is a pilot study to test methods and procedures.
What's involved
You would have your pain measured by a visual analogue scale and by the smartwatch over 40 minutes on one day. You will also perform a 30-second chair test while your pain and electrodermal activity are measured.
Compensation
Not stated in the trial record.
Follow-up
Measurements are taken at 0, 5, and 40 minutes during a single test session.

AI-generated from the public study record. Only the study team can confirm whether you're eligible — confirm details with them before making decisions.

NCT06701461

Measurement of Osteoarthritic Patient Pain Through Electrodermal Activity Signals

Not Yet Recruiting
NAAges 50–80InterventionalDiagnostic
Dartmouth-Hitchcock Medical Center
~15 participants
Updated 2026-03-06 on ClinicalTrials.gov
What's tested:Electrodermal Activity signal

At a glance

Recruiting sites
0 of 1 listed site is recruiting right now
RecruitingSuspended, closed, or not yet open
What they're measuring
:Difference in pain measurements as reported by the participant via a visual analogue scale versus smartwatch reported electrodermal activity sensors
Measured over 40 minutes of test time, occurring all in one day.
+9 more outcomes measured
Osteo Arthritis Knee
1 sites across 1 states
New Hampshire1
  • Peter L Schilling, MD, MS · PRINCIPAL_INVESTIGATOR · Dartmouth-Hitchcock Medical Center

Opens a ready-to-send draft in your own email app — review before sending.

Do you actually qualify for this trial?

Add a private profile and we'll compare every criterion below against your situation — and tell you which ones are met, uncertain, or excluding.

Check eligibility for this trial ~2 min · HIPAA-protected · delete anytime
Eligibility criteria

Inclusion

Kellgren-Lawrence Grade \>= 3

Exclusion

Inflammatory arthropathy (e.g., rheumatoid arthritis), BMI \>=35
Complaints of lower extremity joint pain
Known diagnosis of knee osteoarthpathy
Prior history of knee surgery, knee injections, or injury to knee joints (e.g., meniscus tears, ligamentous injuries), BMI \>=35
Subjects with chronic heart problems, including, but not limited to, chronic hypertension, heart palpitations, a weak or irregular heartbeat, or a previous heart attack,
Subjects taking the following drugs within 12 hours of the experiment or during the experiment: caffeine, alcohol, psychoactive drugs, nicotine or marijuana, other recreational drugs, and medicine of any kind that is not normally taken daily. Such medications include;
NSAIDs
Acetaminophen
Steroidal anti-inflammatory agents
Bronchodilators
Appetite suppressants
Lipase inhibitors
Women who are currently pregnant
Subjects with Raynaud's syndrome
Subjects with any of the following conditions: active skin lesions where EDA sensors are, vertigo or dizziness, and anyone with postural orthostatic tachycardia syndrome (POTS), peripheral neuropathy, seizure disorders, methicillin-resistant staphylococcus aureus (MRSA), impaired circulation, medical implants, open skin lesions, chronic eczema on hands where EDA and electrical stimulator's electrodes are attached, diabetes, and epilepsy.
Participants who have a skin sensitivity to metals, have a pacemaker or defibrillator, or have recent head trauma within the past two weeks (even without loss of consciousness)
Participants who cannot feel physical pain or have a history of self-harm
  • :Difference in pain measurements as reported by the participant via a visual analogue scale versus smartwatch reported electrodermal activity sensors40 minutes of test time, occurring all in one day.

    As a pilot study, the goal is not hypothesis testing but rather examining the feasibility of this approach to measuring patient pain in osteoarthritic patients. Thus, the statistical analysis will be primarily descriptive. The baseline experiment is being done to see whether the smartwatch measurements are at all comparable to prior similar experiments using the thermal grill (Posada-Quintero 2016; Posada-Quintero 2021). For this reason, though not typical for a pilot study, participant responses to the baseline test will use similar analysis to prior electrodermal activity (EDA) based thermal grill studies to see if the smartwatch poses a viable EDA signal collection source. This primarily consists of repeated measures analysis of variance (ANOVA) comparing patient-reported visual analog scale (VAS) scores to EDA output.

  • Patient-reported pain during standardized Osteoarthritis Research Society International (OARSI) 30-second chair test5 minutes

    Patients will be asked to participate in the standardized OARSI 30-second chair test. During the OARSI functional tests, the timestamps of patient reports of pain will be recorded by the experimenter.

  • Electrodermal activity (EDA) signals during standardized Osteoarthritis Research Society International (OARSI) 30-second chair test0 minutes. Concurrent with measurement of patient-reported pain.

    A smartwatch will capture the patient's EDA signals during the 30-second chair test

  • Patient-reported pain during standardized Osteoarthritis Research Society International (OARSI) 40m fast-paced walk test10 minutes

    Patients will be asked to participate in the standardized OARSI 40m fast-paced walk test. During the OARSI functional tests, the timestamps of patient reports of pain will be recorded by the experimenter.

  • Electrodermal activity (EDA) signals during standardized Osteoarthritis Research Society International (OARSI) 40m fast-paced walk test0 minutes. Concurrent with measurement of patient-reported pain.

    A smartwatch will capture the patient's EDA signals during the 40m fast-paced walk test

  • Patient-reported pain during standardized Osteoarthritis Research Society International (OARSI) stair climb test10 minutes

    Patients will be asked to participate in the standardized OARSI stair climb test. During the OARSI functional tests, the timestamps of patient reports of pain will be recorded by the experimenter.

  • Electrodermal activity (EDA) signals during standardized Osteoarthritis Research Society International (OARSI) stair climb test0 minutes. Concurrent with measurement of patient-reported pain.

    A smartwatch will capture the patient's EDA signals during the stair climb test

  • Accuracy of predicting patient pain from electrodermal activity signal (EDA) during standardized Osteoarthritis Research Society International (OARSI) function tests0 minutes. Done after all data has been collected.

    We will treat the processed EDA signal as a binary classifier (i.e., prediction of pain or no pain at any given moment). The accuracy of these predictions will be calculated against the actual patient-reported pain.

  • Percentage of correctly identified pain events (Sensitivity) as measured by the smartwatch (True positive events)0 minutes. Done after all data has been collected.

    In these tests, we will treat the processed EDA signal as a binary classifier (i.e., prediction of pain or no pain at any given moment). The sensitivity of these predictions will be calculated against the actual patient-reported pain.

  • Percentage of correctly identified non-pain events (Specificity) as measured by the smartwatch (True negative non-events)0 minutes. Done after all data has been collected.

    In these tests, we will treat the processed EDA signal as a binary classifier (i.e., prediction of pain or no pain at any given moment). The specificity of these predictions will be calculated against the actual patient-reported pain.