Endogenous theta stimulation during meditation predicts reduced opioid dosing following treatment with Mindfulness-Oriented Recovery Enhancement

Veterans report higher rates of chronic pain than the general population, and a substantial proportion receive long-term opioid therapy (LTOT). High-dose LTOT is theorized to induce maladaptive neuroplastic changes in prefrontal, limbic, and striatal circuitry that diminish self-regulatory capacity and may drive opioid dose escalation. In response to opioid-related harms, the VA has expanded use of complementary and integrative therapies, including mindfulness-based interventions (MBIs). MBIs typically train focused attention (sustaining attention on targets such as the breath, monitoring for mind-wandering, and reorienting attention) and open monitoring (nonjudgmental meta-awareness of moment-to-moment experience). Advanced open monitoring can produce alterations in self-referential processing (e.g., ego dissolution, softened body boundaries, and a sense of unity), which recent research indicates can occur even among novices. EEG investigations of meditation identify frontal midline alpha and theta synchronization as primary biomarkers of mindfulness. Alpha (8–13 Hz) and theta (4–8 Hz) activity in frontal and midline regions have been linked to attentional control and self-referential processing changes. This study tested whether a mindfulness-based intervention for opioid misuse (MORE) alters alpha and theta spectral power and FMT coherence during meditation, and whether these neural changes relate to reduced opioid dosing and shifts in self-referential processing.

Prior work indicates MBIs can ameliorate pain and reduce opioid use, and that mindfulness modulates neurophysiological oscillations in alpha and theta bands, particularly over frontal midline regions. Meta-analyses and reviews have identified frontal midline alpha/theta synchronization as core EEG signatures of meditation, associated with attentional regulation and altered self-referential processing. Neuroimaging studies suggest sustained FMT activity enhances anterior cingulate cortex (ACC) and insula function and connectivity after weeks of practice; preclinical models indicate theta-range stimulation can promote oligodendrocyte output and myelination in ACC and reduce anxiety. Mindfulness-induced self-transcendent states (e.g., ego dissolution, oneness) may reflect integration among default mode, dorsal attention, and salience networks, potentially shifting valuation from drug-related rewards toward natural rewards—relevant for reducing addictive behaviors in chronic opioid users. Within the VA, adoption of MBIs has increased alongside safety initiatives to curb opioid prescribing, providing context for evaluating MBIs like MORE as opioid-sparing strategies.

Design and participants: Veterans with chronic pain receiving long-term opioid therapy were recruited and randomized after baseline assessments to either Mindfulness-Oriented Recovery Enhancement (MORE) or an active supportive group (SG) psychotherapy control. Participants provided informed consent, completed demographic and clinical measures, and underwent urine toxicology screening. Exclusion criteria included prior brief MBI exposure and active validity or psychosis per Mini-International Neuropsychiatric Interview. Sample sizes reported include MORE N = 62 and SG N = 28 in the demographics table. Assessors were blinded to group assignment. Procedures: At pre- and post-treatment (after 8 weeks), participants completed a laboratory mindfulness practice with EEG recording. Daily opioid use was assessed via Timeline Followback. Participants were instructed they would receive a behavioral treatment targeting pain, stress, and opioid-related problems (mindfulness training or supportive psychotherapy). Post-treatment assessments repeated the EEG meditation session. Interventions: MORE comprised eight weekly 2-hour psychologist-led group sessions teaching mindfulness (mindful breathing, body scan), cognitive reappraisal, and savoring in relation to pain, craving, and negative affect. Instructions included inducing nondual states with decreased self-referential processing and softened body boundaries. Home practice entailed daily 15-minute guided mindfulness and pausing before opioid dosing to practice mindfulness at the breakpoint to disrupt habitual dosing. The SG control entailed eight weekly 2-hour Rogerian psychotherapy sessions emphasizing emotional expression and discussion of chronic pain and opioid-related topics; participants journaled 15 minutes daily and were instructed not to practice mindfulness. EEG acquisition: EEG was recorded from 10 midline scalp sites (including F3, F4, FC1, FC2, Fz, Cz, CP1, CP2, Pz) using active Ag/AgCl electrodes and actiCHamp amplifier at 500 Hz sampling rate, 0.489 µV resolution, 140 Hz amplification cutoff, impedances < 20 Ω. Vertical EOGs were recorded. During a 10-minute mindfulness meditation (two 5-minute blocks: eyes open and eyes closed with identical instructions), EEG spectral frequency power and coherence were averaged across blocks at pre- and post-treatment. Meditation instruction: focus on present-moment thoughts, feelings, and body sensations nonjudgmentally without reacting. EEG preprocessing and features: Data were notch-filtered at 60 Hz, low-pass filtered at 40 Hz (Butterworth), and high-pass filtered at 0.1 Hz (2nd order). The PREP pipeline was applied (detrending, harmonics notch filtering, linked-earlobe rereferencing, bad channel interpolation; <10% interpolated). Power spectral density for theta (4–8 Hz) and alpha (9–13 Hz) was computed via Welch periodogram and averaged over regions of interest (ROIs): frontal (F3, F4, Fz), frontal midline (FCz, FC1, FC2), and parietal (Pz, CP1, CP2). Spectral coherence (functional connectivity) was computed as squared coherence magnitude between ROIs, Fisher Z-transformed and averaged within bands. Sensitivity analyses removed visually identified artifacts pre- and post-PREP. Outcomes: Primary neural outcomes were changes in theta and alpha power and theta coherence (with emphasis on FMT coherence) during meditation from pre- to post-treatment. Clinical outcome was change in morphine milligram equivalents (MME) over time (pre-treatment to 4-month follow-up). Self-referential processing changes were assessed (e.g., NADA and perceived body boundaries scales). Statistical analysis: Repeated-measures ANOVAs tested treatment (MORE vs SG) by time (pre vs post) effects on spectral power (theta, alpha) including ROI factors and interactions; similar RM-ANOVAs tested coherence changes. Linear mixed-effects models assessed treatment by time effects on opioid dosing over the treatment phase, with random intercepts and adjustment for pre-treatment MME. Path analysis (AMOS 24) tested mediation of treatment effects on opioid dose reduction by changes in spectral power (Sobel test for indirect effects). Regression models examined associations between spectral power changes and post-treatment self-referential processing controlling for baseline. Pearson correlations related daily mindfulness practice duration to spectral metrics; one high-duration outlier (≥3 SD) was removed.

- Theta power: Significant treatment x time interaction, F(1,60) = 7.33, p = 0.009. MORE showed greater increases in theta power than SG with moderate to large effect sizes across ROIs: frontal d = 0.65, frontal midline d = 0.68, parietal d = 0.70. - Alpha power: Significant treatment x time interaction, F(1,60) = 6.29, p = 0.02. MORE showed greater increases than SG with effect sizes: frontal d = 0.57, frontal midline d = 0.60, parietal d = 0.67. No treatment x time x ROI interaction, indicating increases across all ROIs. - Theta coherence: Significant treatment x time x ROI interaction, F(5,60) = 3.69, p = 0.003, driven by increased frontal midline theta (FMT) coherence in MORE vs SG, F(1,60) = 4.53, p = 0.023. - Opioid dosing: Linear mixed model showed a significant treatment x time interaction, F(1,110.27) = 5.50, p = 0.02. Estimated marginal mean MME decreased from 94.6 ± 36.1 mg at pre-treatment to 79.7 ± 36.16 mg at 4-month follow-up in MORE, versus 100.35 ± 40.54 mg to 92.18 ± 40.58 mg in SG. - Mediation: Increases in frontal theta power correlated with decreases in opioid dose by 4-month follow-up (r = -0.29, p = 0.049). Path analysis with Sobel test indicated a significant indirect effect of MORE on opioid dose reduction via increased frontal theta power (z = 1.97, p = 0.048). - Practice dose-response: Greater minutes of daily mindfulness practice were associated with increases in theta coherence (r = 0.35, p = 0.004), central alpha power (r = 0.35, p = 0.04), and showed a trend with frontal theta power (r = 0.33, p = 0.06). - Self-referential processing: Neural changes during meditation were associated with alterations in self-referential processing (e.g., self-transcendence and perceived body boundary changes), consistent with Figure 4 associations.

The findings indicate that an 8-week course of Mindfulness-Oriented Recovery Enhancement induces robust increases in theta and alpha power and enhances frontal midline theta coherence during meditation in veterans on long-term opioid therapy. These oscillatory changes align with prior evidence linking frontal midline rhythms to attentional control and altered self-referential processing. Clinically, MORE produced greater reductions in opioid dosing over time than an active supportive psychotherapy, and this effect was partially mediated by increases in frontal theta power during meditation, supporting a neurocognitive mechanism whereby mindfulness practice augments prefrontal inhibitory control over opioid dose escalation. The observed associations between practice time and neural metrics, and between frontal theta changes and self-transcendent experiences, suggest that deepening meditative states may generate endogenous theta stimulation in prefrontal circuits, potentially reshaping valuation processes toward natural rewards and away from drug-related reinforcement. These results underscore the relevance of combining endogenous (meditation) with exogenous neuromodulatory approaches (e.g., neurostimulation or neurofeedback) to further enhance therapeutic effects for opioid misuse and related disorders.

After 8 weeks of MORE, participants exhibited increased theta and alpha power and enhanced frontal midline theta coherence during meditation, which predicted decreases in opioid dosing for most patients post-treatment. The reduction in opioid dose was partially mediated by increases in frontal theta power, suggesting endogenous theta stimulation during mindfulness practice as a plausible mechanism. Adequately powered, full-scale clinical trials are warranted to test MORE and other MBIs as opioid-sparing interventions, and future work may explore augmenting endogenous meditation-driven effects with exogenous neurostimulation or neurofeedback.

- EEG montage was low-density, precluding precise source localization of neural generators. - Participants took prescribed opioids on the day of EEG recording; acute pharmacological effects may have influenced neurophysiology despite sensitivity analyses controlling for dose and recent benzodiazepine/THC use. - Modest sample size limits generalizability and power to detect smaller effects; larger samples are needed. - Temporal precedence cannot be fully established; decreases in opioid dosing might have contributed to EEG changes rather than vice versa. Multiple follow-up EEG assessments would clarify directionality. - Increases in EEG power and FMT coherence may mark meditation skill rather than constitute causal mechanisms; causal tests with exogenous theta-burst stimulation could help disentangle mechanisms. - Some inconsistencies in reported sample sizes across sections may reflect reporting limitations in the presented text.