Algorithm-informed Decision-making to Advance Pain Treatment

Official title

Algorithm-informed Decision-making to Advance Pain Treatment (ADAPT Study)

Overview

Researchers are exploring how certain personal factors, such as symptoms, health history, and lifestyle, may help predict which treatments will work best for someone. The purpose of this study is to evaluate how well a study algorithm chooses the best treatments for a patient when considering their unique pain characteristics.

Detailed description

During the treatment stages participants may be placed into the following interventions:

1. Physical Therapy \& Exercise -10, standardized sessions over 8 weeks delivered individually and in-person by a physical therapist 2. Duloxetine - FDA approved for musculoskeletal pain, self-administered SNRI medication taken daily over 8 weeks. 3. Acupressure- a mobile health program- self-administered mHealth program with daily practice over 8 weeks. 4. PRISM- cognitive-behavioral therapy plus resilience training- 8, weekly, digitally delivered pain self-management modules with corresponding weekly 15-minute standardized telehealth coaching sessions.

After participants are placed in an initial assignment based on a combination of the use of the phenotype based algorithm and randomization, they have the option of continuing on that treatment, switching to another (to be assigned by another randomization combined with use of the algorithm) or adding a second treatment to the first (to be assigned by another randomization combined with use of the algorithm).

Analysis of results will not be a function of the arms listed here, but on whether those allocations were based on the algorithm assigning them initially to their presumed "most effective" treatment; or to a potentially less effective treatment (2nd, 3rd or 4th, choice). These "ranked phenotypic" assignments are masked from participants and researchers until after the trial is completed.

Interventions

• Device Algorithmically informed randomization (taking into account phenotypes)
  After Phenotypic data is collected from participants, the algorithm will use that data to "rank" the likelihood of success of different treatments for an individual; it will then,randomize participants based on two patterns of assignment. And at visit 3, it will re-randomize those participants who wish to add an additional treatment modality or transfer from one modality to another.
• Behavioral Physical Therapy (PT) and Exercise
  A high-contact intervention (10 in-person sessions) involving individualized programs tailored to participant needs and guided by physical therapy best-practice guidelines. Treatment emphasizes home exercise, progressive low-intensity conditioning, and functional endurance.
• Behavioral Cognitive-Behavioral Therapy for Pain (PRISM-CBT)
  A moderate-contact telehealth intervention (8 virtual sessions) integrating standard CBT techniques with positive activity modules to enhance engagement, mood, and self-efficacy. Weekly modules focus on pacing, cognitive reframing, relaxation, gratitude, and acts of kindness.
• Behavioral mHealth Acupressure Self-Management:
  A low-contact, self-guided intervention delivered via the MeTime mobile application, supported by brief staff instruction. Participants practice daily self-acupressure at ten standardized points for approximately 30 minutes per day, stimulating key acupoints associated with pain modulation.
• Drug Duloxetine
  This is a serotonin-norepinephrine reuptake inhibitor (SNRI) FDA-approved for chronic pain. A low-contact pharmacologic intervention consisting of an 8-week dose escalation of 60 mg duloxetine, followed by taper as indicated. Participants initiate medication at home after a standardized orientation, with remote safety and adherence check-ins.

Primary outcome measures

• Change in Pain, Enjoyment and General Activity (PEG) scores [Time frame: T2 (approximately week 6) - T3 (approximately week 15)]

Secondary outcome measures (4)

• Change in Patient-Reported Outcomes Measurement Information System (PROMIS) Pain Interference scores [Time frame: T2 (approximately week 6) - T3 (approximately week 15)]
• Patient Global Impression of Pain (PGIC) [Time frame: T3 (approximately week 15)]
• Change in ecological momentary assessment (EMA)-derived mean Pain Interference [Time frame: T2 (approximately week 6) - T3 (approximately week 15)]
• Change in Oswestry Disability Index (ODI) scores [Time frame: T2 (approximately week 6) - T3 (approximately week 15)]

Eligibility criteria

Inclusion criteria

• Participants must meet the definition of Chronic Low Back Pain (cLBP) described in the NIH Task Force Report on Research Standards for cLBP, i.e., low back pain present at least six months, and present more than half of those days.2
• Individuals must have a pain interference score of ≥60 on PROMIS Pain Interference. The normal population mean for pain interference is 50. Participants must be 1 standard deviation (SD) above the population mean (>=60) for inclusion.
• Individuals must be willing and eligible to receive at least three of the four proposed interventions.
• Able to stand for at least 10 minutes

Exclusion criteria

• Red flags suggesting the presence of an identifiable cause of low back pain, including fever, weight loss, initiation of pain following major trauma, immunosuppression, IV drug use, recent bacterial infection, severe or persistent sensory or motor involvement of bowel, bladder, or lower extremity.
• Unable to read, write of speak fluent English
• Visual or hearing difficulties that would preclude participation
• Active uncontrolled drug/alcohol addiction
• Individuals receiving new disability or compensation or involved in litigation in the past 12 months.
• Pregnancy or breastfeeding
• Individuals on high doses of opioids (over 100 Oral Morphine Equivalents \[OME\] per day).
• Upcoming scheduled back surgery, back surgery within the last year, or more than one back surgery in the past.
• History of discitis osteomyelitis (spine infection) or spine tumor
• History of an autoimmune disorder such as ankylosing spondylitis, rheumatoid arthritis, polymyalgia rheumatica, psoriatic arthritis, or lupus
• History of cauda equina syndrome or spinal radiculopathy
• Diagnosis of schizophrenia or schizoaffective diagnosis
• Diagnosis of any vertebral fracture in the last 6 months
• Osteoporosis requiring treatment (vitamin D and calcium supplements are acceptable)
• Cancer

• History of any bone-related cancer or cancer that metastasized to the bone
• Currently in treatment for cancer or plan to start cancer treatment in the next 12 months
• History of any cancer treatment in the last 24 months
• Life expectancy is less than 2 years
• Current/planned (in the next 2 years) enrollment in another study of a device or investigational drug that would interfere with this study, this may include participation in a blinded trial.
• Any other diseases or conditions that would make a patient unsuitable for study participation as determined by the site principal investigators. This would include but not be limited to severe psychiatric disorders, active suicidal ideations or recent suicide attempts

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: No

Study design

Allocation

Randomized

Model

Sequential

Masking

Quadruple blind

Primary purpose

Health services research

Study locations

United States · 1 center

• University of Michigan — Ann Arbor

Identifiers

NCT: NCT07731750 · HUM00273626 · UC2AR086182

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗