Psychiatry · Medical Literature
Self-Guided Internet-Based Mindfulness-Informed Stress Management for Generalized Anxiety Disorder: Randomized Controlled Trial With Longitudinal Network Analysis.
Abstract
Generalized anxiety disorder (GAD) is characterized by stress-anxiety reinforcement cycles. Evidence for brief, self-guided, internet-based stress management as an adjunct to pharmacotherapy remains limited. This study aimed to evaluate the efficacy of an 8-week self-guided, internet-based stress management program (iSM) incorporating mindfulness techniques as an adjunct to treatment as usual (TAU) in adults with GAD and to examine individual predictors of response and symptom dynamics. A single-blind, parallel-group, superiority randomized controlled trial was conducted at Tongji Hospital, Wuhan, China. A total of 140 adults with GAD were randomly assigned to iSM+TAU (n=73) or TAU (n=67). Outcome assessors were blinded to group allocation. The iSM intervention consisted of 8 weekly self-guided online modules integrating mindfulness-based training and Baduanjin-based stretching exercises. TAU comprised routine pharmacotherapy. Primary outcomes were posttreatment changes in anxiety and depressive symptoms. Secondary outcomes included sleep quality, somatic symptoms, social functioning, mindfulness, rumination, and perceived stress. Exploratory cross-lagged panel network (CLPN) and random-intercept cross-lagged panel modeling (RI-CLPM) analyses were used to examine temporal symptom dynamics. All 140 randomized participants were included in the intention-to-treat analysis. Compared with TAU, iSM+TAU showed greater reductions at posttreatment in anxiety (Cohen d=-0.277, 95% CI -0.521 to -0.033) and depressive symptoms (Cohen d=-0.309, 95% CI -0.592 to -0.026). Significant between-group differences were also observed in somatic symptoms (Cohen d=-0.340, 95% CI -0.604 to -0.077), state anxiety (Cohen d=-0.537, 95% CI -0.849 to -0.224), mindfulness (Cohen d=0.666, 95% CI 0.327-1.006), rumination (Cohen d=-0.344, 95% CI -0.626 to -0.062), and perceived stress (Cohen d=-0.429, 95% CI -0.725 to -0.133), but not in sleep quality or social functioning. No serious adverse events were reported. Median session completion was 7 of 8. In exploratory analyses, higher baseline acting with awareness predicted greater treatment response (standardized β=0.167; 95% CI 0.031-0.335), whereas higher trait anxiety predicted poorer outcomes (standardized β=-0.150; 95% CI -0.234 to -0.002). CLPN and RI-CLPM identified 2 key within-person pathways, a bidirectional association between perceived stress (mean β=0.219; P<.001) and state anxiety (mean β=0.165; P=.02), and a unidirectional effect of mindfulness on subsequent anxiety reduction (mean β=-0.285; P<.001). To our knowledge, this is the first trial to evaluate a brief, self-guided, culturally adapted digital mindfulness intervention as an adjunct to pharmacotherapy in Chinese adults with GAD. The intervention showed clinically meaningful benefits and practical potential as a scalable, resource-efficient approach. Unlike prior studies focused mainly on symptom outcomes alone, this trial combined a randomized design with CLPN and RI-CLPM analyses to provide preliminary insight into symptom change processes over time, adding a more process-oriented analytic perspective that may inform future intervention refinement.
Clinical relevance
In adults with generalized anxiety disorder (GAD) receiving pharmacotherapy, does adding an 8-week self-guided, internet-based mindfulness-informed stress management program (iSM) reduce anxiety and depressive symptoms more than pharmacotherapy alone? __SNAPSHOT__ {"studyDesign":"Single-blind, parallel-group, superiority randomized controlled trial (RCT) with exploratory longitudinal cross-lagged panel network (CLPN) and random-intercept cross-lagged panel model (RI-CLPM) analyses","setting":"Single center — Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China","sampleSize":"n = 140 (iSM+TAU: n = 73; TAU alone: n = 67)","population":"Adults with a diagnosis of generalized anxiety disorder receiving routine pharmacotherapy. Median age 33 years (IQR 26–39); 66.4% female. No significant baseline between-group differences reported.","intervention":"iSM+TAU: 8-week self-guided internet-based stress management program consisting of 8 weekly online modules integrating mindfulness-based training and Baduanjin (traditional Chinese qigong)-based stretching exercises, added to routine pharmacotherapy (TAU).","comparator":"TAU alone: routine pharmacotherapy without the iSM program.","primaryOutcome":"Posttreatment changes in anxiety symptoms (Hamilton Anxiety Rating Scale [HAMA]) and depressive symptoms (Hamilton Depression Rating Scale, 21-item [HAMD-21]) at 8 weeks.","followUp":"8 weeks (posttreatment); 3-month follow-up assessment also conducted (T3). Midintervention assessment at 4 weeks (T1).","fundingSource":"Not reported in available abstract or indexed sources."}
Key findings
The iSM+TAU group demonstrated statistically significant superiority over TAU alone on both primary outcomes — anxiety (HAMA) and depressive symptoms (HAMD-21) — at posttreatment, with an ITT analysis framework. Specific between-group effect sizes (Cohen's d) were reported for several outcomes: somatic symptoms (PHQ-15): d = −0.340 (95% CI −0.604 to −0.077); state anxiety (S-AI): d = −0.537 (95% CI −0.849 to −0.224); mindfulness (FFMQ): d = 0.666 (95% CI 0.327 to 1.006); rumination (RRS): d = −0.344 (95% CI −0.626 to −0.062); perceived stress (PSS-10): d = −0.429 (95% CI −0.725 to −0.133). Cohen's d values for the primary outcomes (HAMA anxiety, HAMD-21 depression) were not separately indexed in available sources and are therefore not reported here. No significant between-group differences were observed for sleep quality (PSQI) or social functioning (SDSS). Median session completion was 7 of 8 modules. A dose-response relationship was observed: participants completing ≥4 sessions showed significantly greater anxiety reduction than those completing <4 sessions (HAMA: F₁,₄₇ = 5.13; P = .03), while incomplete engagers (< 4 sessions) showed no significant separation from TAU controls (HAMA: F₁,₆₂ = 0.31; P = .58). No serious adverse events were reported. In exploratory CLPN/RI-CLPM analyses, two key within-person pathways emerged: a bidirectional association between perceived stress (mean β = 0.219; P < .001) and state anxiety (mean β = 0.165; P = .02), and a unidirectional effect of mindfulness on subsequent anxiety reduction (mean β = −0.285; P < .001). In the iSM+TAU subgroup who completed ≥4 sessions, higher baseline 'acting with awareness' (a mindfulness facet) predicted greater HAMA reduction (standardized β = 0.167; 95% CI 0.031 to 0.335), whereas higher baseline trait anxiety predicted poorer outcomes (standardized β = −0.150; 95% CI −0.234 to −0.002). NNT: not calculable from available data (no binary responder threshold reported). Absolute score changes for HAMA and HAMD-21 were not available in indexed sources.
Practice implications
This trial provides moderate-quality evidence that a self-guided, 8-week internet-based mindfulness and movement program added to pharmacotherapy produces statistically significant improvements in anxiety symptoms, depression, somatic burden, state anxiety, rumination, perceived stress, and mindfulness compared with pharmacotherapy alone in Chinese adults with GAD. Effect sizes for the most robustly reported secondary outcomes (state anxiety d = −0.54; mindfulness d = 0.67) fall in the small-to-moderate range and may be clinically meaningful as adjunctive benefits, though absolute score changes were not available for independent clinical significance thresholds. Notably, no benefit was established for sleep quality or social functioning, two functionally important outcomes in GAD. The exploratory network analyses offer a mechanistically plausible hypothesis — that mindfulness reduces anxiety via downstream perceived stress modulation — but these are hypothesis-generating only and require confirmatory study. The dose-response finding (≥4 sessions needed for benefit) has clinical relevance for implementation: partial engagement may confer little benefit. Baseline acting-with-awareness may be a useful candidate predictor of treatment response, warranting prospective evaluation. Important population restrictions: the sample was drawn from a single Chinese psychiatric hospital, was predominantly female and relatively young (median age 33 years), and included only patients already on pharmacotherapy. Findings should not be generalized to primary care populations, older adults, those with severe GAD, patients in Western healthcare systems, or those who prefer or are ineligible for pharmacotherapy. Because this is a single-center RCT with only an abstract and partial results available in indexed sources, replication is needed before any practice change is warranted. This summary is based on the published abstract, a preprint, and indexed partial results; the full dataset was not accessible for review. __LIMITATIONS__ Single-center design at a tertiary psychiatric hospital in China, limiting generalizability to primary care, Western, or non-pharmacotherapy-treated populations.,Single-blind design only (outcome assessors blinded; participants and treating clinicians not blinded), introducing potential performance and detection bias.,TAU comparator consisted of pharmacotherapy without structured psychosocial contact, creating a potential attention-control imbalance that may inflate estimated effect sizes.,Absence of an active psychosocial control (e.g., psychoeducation or supportive contact) means observed benefits cannot be attributed specifically to mindfulness or Baduanjin components versus general digital engagement.,Relatively small sample (n = 140) may limit precision of effect size estimates and statistical power for subgroup analyses.,No significant benefit was observed for sleep quality or social functioning, two clinically important GAD outcome domains.,The exploratory CLPN and RI-CLPM network analyses are hypothesis-generating; causal inference from observational longitudinal pathways is not established.,Funding source is not reported, precluding assessment of potential conflicts of interest.,Baduanjin is a culturally specific practice; its acceptability, feasibility, and effectiveness may differ substantially in non-Chinese populations.,Surrogate/intermediate outcomes (mindfulness scores, perceived stress) were used alongside clinician-rated scales; no clinical response or remission rates were reported, limiting direct clinical actionability.,Primary outcome effect sizes (Cohen's d for HAMA and HAMD-21) were not available in indexed sources at time of this summary; clinical significance of scale score changes cannot be fully evaluated.,3-month follow-up data were presented but durability of effects beyond this period is unknown.,Only abstract and partial results were available for this summary; full trial data (including complete statistical tables) were not accessible. __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ A self-guided, 8-week internet-based mindfulness and Baduanjin program added to pharmacotherapy produced statistically significant small-to-moderate reductions in anxiety, depression, somatic symptoms, state anxiety, rumination, and perceived stress in Chinese adults with GAD, compared with pharmacotherapy alone; no benefit was detected for sleep quality or social functioning. Findings are promising but come from a single-center, single-blind trial in a specific cultural and clinical context, with no active control and no reported remission or response rates, limiting generalizability and clinical actionability at this time. This intervention warrants further evaluation in larger, multicenter, actively controlled trials before it can be recommended as a routine adjunct to pharmacotherapy for GAD.