All Studies · Anxiety & Stress · Feldman (2025)

Feldman (2025)

Moderate · RCT Secondary Analysis · Secondary Analysis (RCT-derived) · n = 147
Quality Index (Adjusted QI) 0.762
0 Low < 0.56 Moderate 0.56–0.77 High ≥ 0.78 1
Study Details
ConditionGeneralized anxiety disorder (GAD) — therapeutic relationship
Clinical domainAnxiety & Stress
PopulationAdults with GAD (subsample n=147; 71% women, mean age 33.2 ± 13.3)
Sample sizen = 147
Country / SettingUSA (Boston, MA and New York, NY)
Protocol clusterKY-GAD Protocol
Duration12 weeks; twelve 120-min weekly group sessions in groups of 3-6 (secondary analysis of Simon 2021 RCT)
Control typeActive Control
ComparatorCBT (n=65) + Stress Education attention control (n=26); 3-arm design
Outcomes & Effect Sizes
Primary outcomeTherapeutic relationship (alliance, credibility, expectancy)
Scales usedWorking Alliance Inventory (WAI: goal, task, bond), Credibility and Expectancy Questionnaire (CEQ), PSWQ (worry), HAM-A (anxiety), QIDS-SR (depression)
Key resultWAI and CEQ scores significantly higher for CBT than SE at midpoint; no significant differences between KY and CBT on alliance or credibility. WAI at midpoint did not predict change in symptoms. Task subscale showed largest differences vs SE for both KY and CBT.
Effect sizeCBT > SE on WAI/CEQ (significant); KY ≈ CBT on WAI/CEQ (not significantly different); WAI did not predict outcome change
Retention147/226 (65%) completed WAI and CEQ measures (KY=56, CBT=65, SE=26)
SafetyNot separately reported (see Simon 2021 parent trial)
View PDF
Methodological Summary

Secondary analysis of 3-arm RCT (Simon 2021); ANOVA + longitudinal linear mixed effects models; N=147 subsample; therapeutic relationship as process variable

KY Protocol Components

Therapeutic Relationship Analysis in KY vs CBT vs SE for GAD (Simon 2021 — Secondary Analysis) PARENT TRIAL: Simon et al. (2021) — 226 adults with GAD randomized to 12 weeks of KY, CBT, or Stress Education. SUBSAMPLE: N=147 who completed WAI and CEQ measures (KY=56, CBT=65, SE=26). KY PROTOCOL: Same as Simon 2021 — yoga theory, philosophy, physical postures/exercises, breathing/relaxation, meditation, mindfulness practices. 20-min daily home practice (audio-guided movements + coordinated breathing + mind-body awareness). KEY MEASURES: Working Alliance Inventory (WAI: goal, task, bond subscales), Credibility and Expectancy Questionnaire (CEQ). KEY FINDINGS: • WAI and CEQ scores were significantly higher for CBT than SE by midpoint (week 6). • No significant differences between CBT and yoga on WAI or CEQ. • Largest WAI subscale differences vs SE were in 'task' for both CBT and yoga. • WAI at midpoint did NOT predict change in worry, anxiety, or depression from midpoint to post-treatment. IMPLICATION: Patient ratings of therapeutic relationship were similar between yoga and CBT, suggesting yoga as a credible treatment with comparable alliance quality.

Quality Item Scores — 1 fail · 2 partial · 3 pass · ★ critical
★A1
3
A2
3
★B1
2
★B2
2
B3
3
B4
2
★B5
3
B6
2
★B7
2
★B8
2
B9
2
JP1
3
Critical fails0
Raw QI0.806
SAF0.946
Adjusted QI0.762
Final ratingModerate