Patient-Centered Pain Care Using Artificial Intelligence and Mobile Health Tools (REACT (AI CBT))

This study will evaluate a new approach for back pain care management using artificial intelligence and evidence-based cognitive behavioral therapy (AI-CBT) so that services automatically adapt to each Veteran's unique needs, achieving outcomes as good as standard care but with less clinician time.

Cognitive behavioral therapy (CBT) is one of the most effective treatments for chronic back pain. However, only half of Veterans have access to trained CBT therapists, and program expansion is costly. Moreover, VA CBT programs consist of 10 weekly hour-long sessions delivered using an approach that is out-of-sync with stepped-care models designed to ensure that scarce resources are used as effectively and efficiently as possible. Data from prior CBT trials have documented substantial variation in patients' needs for extended treatment, and the characteristics of effective programs vary significantly. Some patients improve after the first few sessions while others need more extensive contact. After initially establishing a behavioral plan, still other Veterans may be able to reach behavioral and symptom goals using a personalized combination of manuals, shorter follow-up contacts with a therapist, and automated telephone monitoring and self-care support calls. In partnership with the National Pain Management Program, the investigators propose to apply state-of-the-art principles from "reinforcement learning" (a field of artificial intelligence or AI used successfully in robotics and on-line consumer targeting) to develop an evidence-based, personalized CBT pain management service that automatically adapts to each Veteran's unique and changing needs (AI-CBT). AI-CBT will use feedback from patients about their progress in pain-related functioning measured daily via pedometer step-counts to automatically personalize the intensity and type of patient support; thereby ensuring that scarce therapist resources are used as efficiently as possible and potentially allowing programs with fixed budgets to serve many more Veterans. The specific aims of the study are to: (1) demonstrate that AI-CBT has non-inferior pain-related outcomes compared to standard telephone CBT; (2) document that AI-CBT achieves these outcomes with more efficient use of scarce clinician resources as evidenced by less overall therapist time and no increase in the use of other VA health services; and (3) demonstrate the intervention's impact on proximal outcomes associated with treatment response, including program engagement, pain management skill acquisition, satisfaction with care, and patients' likelihood of dropout. The investigators will use qualitative interviews with patients, clinicians, and VA operational partners to ensure that the service has features that maximize scalability, broad scale adoption, and impact. 278 patients with chronic back pain will be recruited from the VA Connecticut Healthcare System and the VA Ann Arbor Healthcare System, and randomized to standard 10-sessions of telephone CBT versus AI-CBT. All patients will begin with weekly hour-long telephone counseling, but for patients in the AI-CBT group, those who demonstrate a significant treatment response will be stepped down through less resource-intensive alternatives to hour-long contacts, including: (a) 15 minute contacts with a therapist, and (b) CBT clinician feedback provided via interactive voice response calls (IVR). The AI engine will learn what works best in terms of patients' personally-tailored treatment plan based on daily feedback via IVR about patients' pedometer-measured step counts as well as their CBT skill practice and physical functioning. The AI algorithm the investigators will use is designed to be as efficient as possible, so that the system can learn what works best for a given patient based on the collective experience of other similar patients as well as the individual's own history. The investigator's hypothesis is that AI-CBT will result in pain-related functional outcomes that are no worse (and possibly better) than the standard approach, but by scaling back the intensity of contact that is not resulting in marginal gains in pain control, the AI-CBT approach will be significantly less costly in terms of therapy time. Secondary hypotheses are that AI-CBT will result in greater patient engagement and patient satisfaction. Outcomes will be measured at three and six months post recruitment and will include pain-related interference, treatment satisfaction, and treatment dropout.

AI CBT engine will make recommendations to step-down or step-up intensity of CBT FU based on what patient reports and what other similar patients report. Stepped care model.

Controls receive 10 hour-long standard telephone CBT sessions, a pedometer/log after baseline, and a Patient Handbook.

AI CBT engine will make recommendations to step-down or step-up intensity of CBT follow-up based on what patient reports and what other similar patients report. Stepped care model.

Controls receive 10 hour-long standard telephone CBT sessions, a pedometer/log after baseline, and a Patient Handbook.

The Roland Morris Disability Questionnaire (RMDQ) is a 24-item checklist designed for patients to identify the level of disability and functional status associated with chronic low back pain. Patients are instructed to endorse items that describe their functional status that day. Scores range from 0-24, with higher scores indicating more disability.

An 11-point Numeric Rating Scale (NRS) for pain severity, with 0 representing "No pain" and 10 representing the "Worst pain imaginable." Patients were asked to rate their level of pain on average in the last week.

Pain-related interference was measured using the Brief Pain Inventory - Short Form (BPI). Scores range from 0-10, with higher scores indicating more interference.

Depression symptom severity was assessed using the 9-item Patient Health Questionnaire (PHQ-9). Scores range from 0-27, with higher scores indicating more depression symptom severity.

Inclusion Criteria: Back pain-related dx including back and spine conditions and nerve compression and a score of >=4 (indicating moderate pain) on the 0-10 Numerical Rating Scale on at least two separate outpatient encounters in the past year At least 1 outpatient visit in last 12 months At least moderate pain-related disability as determined by a score of 5+on the Roland Morris Disability Questionnaire At least moderate musculoskeletal pain as indicated by a pain score of >=4 on the Numeric Rating Scale Pain on at least half the days of the prior 6 months as reported on the Chronic Pain item Touch-tone cell or land line phone. Exclusion Criteria: COPD requiring oxygen Cancer requiring chemotherapy Currently receiving CBT Suicidality Receiving surgical tx related to back pain Active psychotic symptoms Severe depressive symptoms Can't speak English Sensory deficits that would impair participation in telephone calls Patient not planning to get care at study site PCP not affiliated with study site Limited life expectancy (COPD requiring oxygen or Cancer requiring chemotherapy Active psychotic symptoms, suicidality, severe depressive symptoms (Beck Depression Inventory (BDI) score or 30+) Substance use disorder or dependence, active manic episode, or poorly controlled bipolar disorder as identified by MMini International Neuropsychiatric Interview Severe depression identified by chart review of diagnoses and mental health treatment notes Cognitive impairment defined by a score of <=5 on the Six-Item screener Current CBT or surgical treatment related to back pain.

Piette JD, Newman S, Krein SL, Marinec N, Chen J, Williams DA, Edmond SN, Driscoll M, LaChappelle KM, Kerns RD, Maly M, Kim HM, Farris KB, Higgins DM, Buta E, Heapy AA. Patient-Centered Pain Care Using Artificial Intelligence and Mobile Health Tools: A Randomized Comparative Effectiveness Trial. JAMA Intern Med. 2022 Sep 1;182(9):975-983. doi: 10.1001/jamainternmed.2022.3178.

MacLean RR, Buta E, Higgins DM, Driscoll MA, Edmond SN, LaChappelle KM, Ankawi B, Krein SL, Piette JD, Heapy AA. Using Daily Ratings to Examine Treatment Dose and Response in Cognitive Behavioral Therapy for Chronic Pain: A Secondary Analysis of the Co-Operative Pain Education and Self-Management Clinical Trial. Pain Med. 2023 Jul 5;24(7):846-854. doi: 10.1093/pm/pnac192.

Mattocks KM, LaChappelle KM, Krein SL, DeBar LL, Martino S, Edmond S, Ankawi B, MacLean RR, Higgins DM, Murphy JL, Cooper E, Heapy AA. Pre-implementation formative evaluation of cooperative pain education and self-management expanding treatment for real-world access: A pragmatic pain trial. Pain Pract. 2023 Apr;23(4):338-348. doi: 10.1111/papr.13195. Epub 2022 Dec 29.

Piette JD, Krein SL, Striplin D, Marinec N, Kerns RD, Farris KB, Singh S, An L, Heapy AA. Patient-Centered Pain Care Using Artificial Intelligence and Mobile Health Tools: Protocol for a Randomized Study Funded by the US Department of Veterans Affairs Health Services Research and Development Program. JMIR Res Protoc. 2016 Apr 7;5(2):e53. doi: 10.2196/resprot.4995.