Psychiatry · Medical Literature
Randomized controlled trial of a modular, transdiagnostic, cognitive-behavioral anxiety intervention versus primary care behavioral health treatment as usual.
Abstract
Offering behavioral treatment for anxiety in primary care will help to address the anxiety treatment gap. Modular Anxiety Skills Training (MAST) is a transdiagnostic cognitive-behavioral therapy (CBT) intervention for Primary Care Behavioral Health (PCBH) settings. This multi-site randomized controlled trial compared the effects of MAST versus PCBH treatment as usual (TAU), both delivered by frontline clinicians, on anxiety severity and impairment. Participants (N = 167, 70% male, 89% White, 9% Hispanic, M age 51 [16] years) were recruited from primary care at two Veterans Health Administration healthcare systems. Contrary to our hypothesis, intent-to-treat analyses found no differences by condition at 16-weeks post-treatment. Participants in both conditions improved significantly on overall anxiety severity and impairment (OASIS) as well as anxiety and depression symptom severity, functional disability, and quality of life, with small to medium effect sizes (0.21-0.61). All gains were maintained at 28-week follow-up, and participants receiving MAST demonstrated further significant reduction in OASIS. Post hoc analyses suggested baseline depression severity moderated the condition effect; participants with moderate and severe comorbid depression responded better to MAST than PCBH TAU. While MAST was not superior to PCBH TAU at post-treatment as hypothesized, it may potentially yield a more enduring impact and work better with comorbid depression; however, these findings require prospective replication. This study advances the literature by comparing brief CBT for anxiety in primary care to a strong active comparison comprising behavioral treatment. MAST may offer implementation advantages due to being manualized and modular. Further research on modular CBT for anxiety is warranted.
Clinical relevance
Among Veterans with anxiety presenting to primary care, does Modular Anxiety Skills Training (MAST), a structured transdiagnostic cognitive-behavioral intervention delivered by frontline behavioral health providers, produce superior reductions in anxiety severity and impairment compared with Primary Care Behavioral Health treatment as usual (PCBH TAU)? __SNAPSHOT__ {"studyDesign":"Multi-site randomized controlled trial (RCT), intent-to-treat analysis","setting":"Two Veterans Health Administration (VHA) primary care healthcare systems; country and specific sites not reported in abstract","sampleSize":"n = 167 (arm-level allocation not reported in abstract)","population":"Primary care patients at VHA facilities with anxiety; mean age 51 years (SD 16), 70% male, 89% White, 9% Hispanic; comorbid depression severity noted as a post-hoc moderator variable","intervention":"Modular Anxiety Skills Training (MAST): a transdiagnostic, modular cognitive-behavioral therapy (CBT) intervention designed for Primary Care Behavioral Health (PCBH) settings, delivered by frontline clinicians","comparator":"PCBH treatment as usual (TAU), also delivered by frontline clinicians in the primary care behavioral health setting","primaryOutcome":"Anxiety severity and impairment as measured by the Overall Anxiety Severity and Impairment Scale (OASIS) at 16 weeks post-treatment","followUp":"16 weeks post-treatment (primary); 28-week follow-up assessment also conducted","fundingSource":"Not reported in abstract; prior work by this group was supported by the Department of Veterans Affairs Health Services Research & Development Service (VA HSR&D)"}
Key findings
The primary intent-to-treat analysis found no statistically significant between-group difference on the OASIS at 16 weeks post-treatment (specific between-group effect size, confidence intervals, and p-values not reported in the abstract). This was contrary to the investigators' a priori hypothesis that MAST would be superior. Both MAST and PCBH TAU produced statistically significant within-group improvements from baseline across multiple domains: overall anxiety severity and impairment (OASIS), anxiety and depression symptom severity, functional disability, and quality of life, with small to medium within-group effect sizes (Cohen's d or equivalent range: 0.21–0.61; specific per-outcome values and confidence intervals not reported). All gains were maintained at 28-week follow-up. Additionally, participants in the MAST arm demonstrated a further statistically significant reduction in OASIS scores between the 16-week and 28-week assessment points (magnitude and p-value not reported in abstract). Post-hoc moderation analyses suggested that baseline depression severity moderated treatment response, though the direction, magnitude, and statistical parameters of this moderation effect are not reported in the abstract. NNT/NNH cannot be calculated from the available data.
Practice implications
This RCT did not establish superiority of MAST over PCBH TAU for anxiety in primary care Veterans at the primary 16-week endpoint. Both interventions produced clinically meaningful within-group improvements across anxiety, depression, functional disability, and quality-of-life domains, with effect sizes reaching the small-to-medium range (0.21–0.61). The finding of a late MAST benefit at 28 weeks is hypothesis-generating but must be interpreted cautiously, as it represents a secondary, post-randomization time point. The post-hoc moderation finding regarding baseline depression severity may help identify which patients benefit more from MAST versus TAU, but requires prospective replication before clinical application. Generalizability is substantially limited by the VA-specific, predominantly male (70%), predominantly White (89%) sample; findings may not apply to women, younger patients, non-Veterans, or more racially and ethnically diverse primary care populations. The active, structured nature of PCBH TAU likely narrowed the opportunity to detect a between-group difference. Clinicians should note that, based on this abstract alone, it is not possible to determine whether the within-group effect sizes meet accepted thresholds for minimally important clinical differences on the OASIS. Importantly, only the abstract was available for this review; the full paper may report additional subgroup, safety, or adherence data that could materially alter interpretation. __LIMITATIONS__ Null primary outcome: MAST was not superior to an active, structured comparator (PCBH TAU), which itself may represent a high-quality active control rather than a true 'usual care' condition, limiting ability to detect differential effects.,Small, demographically homogeneous sample (n = 167; 70% male, 89% White, VA-only), severely restricting generalizability to women, younger patients, non-White populations, and non-VA primary care settings.,Post-hoc moderation analysis for baseline depression severity is exploratory and hypothesis-generating only; not pre-specified and subject to inflated Type I error risk.,The late MAST benefit at 28 weeks is a secondary time point finding and may reflect regression to the mean, differential dropout, or natural symptom trajectory rather than true treatment effect.,Blinding status of outcome assessors is not reported in the abstract; open-label or unblinded delivery in PCBH settings is common and introduces performance and detection bias.,Funding source is not confirmed in the abstract; prior work by this group was VA-funded, which may influence study design and population selection.,Within-group effect sizes (0.21–0.61) are reported without confidence intervals or p-value adjustments for multiple outcomes, precluding assessment of precision or family-wise error rate.,Specific arm-level sample sizes, dropout rates, and adherence data are not available from the abstract alone, making attrition bias assessment impossible.,Applicability to non-CBT-trained frontline clinicians or settings without structured PCBH infrastructure is uncertain. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In this VA-based RCT (n = 167), MAST did not demonstrate superiority over PCBH treatment as usual on anxiety severity and impairment at 16 weeks, though both conditions produced significant within-group improvements with small-to-medium effect sizes maintained at 28 weeks. Clinicians practicing in PCBH settings can be reassured that structured behavioral health usual care may be sufficient for many Veterans with anxiety, but should await replication in broader, more diverse populations before adopting MAST as a preferential intervention. The post-hoc finding that baseline depression severity moderated outcomes is intriguing and warrants prospective investigation to guide patient stratification.