Smartphone-Based Ecological Momentary Assessment of Pain in Older Adults Undergoing Auricular Point Acupressure for Chronic Low Back Pain: Secondary Analysis of a Randomized Controlled Trial

BACKGROUND: Chronic low back pain (cLBP) is a prevalent and disabling condition in older adults. Auricular point acupressure (APA), a nonpharmacologic intervention, has shown promise in managing cLBP. However, its impact on daily symptom fluctuations remains unclear. Ecological momentary assessment (EMA), which collects real-time data, offers a method to capture these fluctuations. OBJECTIVE: This secondary analysis evaluated the effects of APA on EMA-reported pain intensity and pain interference among older adults with cLBP and examined associations among APA practice patterns, recall-based assessments, and EMA engagement. METHODS: Data were drawn from a published 3-arm randomized controlled trial of 272 adults aged ≥60 years with cLBP, randomized to targeted auricular point acupressure (T-APA), nontargeted auricular point acupressure (NT-APA), or education control. For this analysis, only participants who completed at least one EMA entry during the 4-week intervention were included. A total of 61 participants were excluded due to missing EMA data, resulting in a final analytic sample of 211 (T-APA: 72; NT-APA: 74; and control: 65). EMA-reported pain intensity and interference were collected using a smartphone app 3 times daily over 29 days. Linear mixed-effects models assessed the effects of group assignment and APA practice behaviors on EMA outcomes, adjusting for demographics, smoking, opioid use, and baseline recall-based pain. Spearman correlations assessed associations between EMA and 7-day recall measures. RESULTS: The overall compliance rate was 44.8%, with an attrition rate of 54%. There were no significant differences in compliance or attrition across treatment groups. Older participants showed significantly lower compliance, though attrition was not associated with age. Pain intensity reported via EMA was significantly lower than recall-based pain, while EMA-reported pain interference was higher. EMA and 7-day recall pain outcomes were strongly correlated (Spearman =0.53-0.95; P