A randomized controlled trial comparing mindfulness to escitalopram for anxiety: In-person and remote, synchronous delivery pre and post COVID-19 pandemic.

For adults with anxiety disorders, Mindfulness-Based Stress Reduction (MBSR) delivered via videoconference was comparably effective to in-person MBSR, but MBSR-VC did not meet the threshold for non-inferiority compared to the antidepressant escitalopram (ESC) delivered by videoconference. In a randomized controlled trial with 202 participants, MBSR-VC and ESC-VC showed similar average improvement on the Clinical Global Impression of Severity scale (1.39 vs 1.51 points), but non-inferiority was not supported. In-person treatments had a greater impact on social anxiety than their video versions. ESC-VC received higher satisfaction ratings and had a greater effect on panic symptoms than MBSR-VC. Remotely delivered MBSR is a viable option for anxiety disorders, though social anxiety may benefit more from in-person care.

Study at a glance

Abstract

During the pandemic, a randomized controlled trial (RCT) of Mindfulness-Based Stress Reduction (MBSR) versus the antidepressant, escitalopram (ESC) added a second phase using a synchronous, videoconference (VC) format for both interventions, enabling randomized and naturalistic comparisons. The original in-person RCT demonstrated non-inferiority of MBSR to ESC. In the second phase, we hypothesized that MBSR-VC would be non-inferior to ESC-VC, and that MBSR-VC would be non-inferior to in-person MBSR (MBSR-IN). Adults with anxiety disorders were recruited for a 3-center RCT. Primary (Clinical Global Impression of Severity: CGIS) and secondary outcomes (satisfaction, other measures for anxiety subtypes) were compared between MBSR-IN vs MBSR-VC, ESC-IN vs ESC-VC and MBSR-VC vs ESC-VC. MBSR-VC (n = 100) and ESC-VC (n = 102) did not significantly differ in mean CGI-S change at endpoint (1.39 vs 1.51, p = 0.17) but non-inferiority of MBSR-VC was not supported. MBSR did not vary by delivery format (VC: 1.3 vs. IN: 1.3, p = 0.77) and MBSR-VC's non-inferiority to MBSR-IN was demonstrated within the pre-specified margin of -0.40 (95 % CI: -0.34 to 0.25). ESC-IN and ESC-VC were not significantly different in CGI-S change in unadjusted comparisons (1.4 vs 1.5, p = 0.53), but the average CGI-S at endpoint was significantly lower in the ESC-VC group in a multivariable longitudinal model. Other findings included in-person treatments having a greater impact on social anxiety compared to their VC version, ESC-VC having higher satisfaction ratings and a greater impact on panic symptoms than MBSR-VC. Remotely delivered MBSR demonstrated comparable effectiveness to in-person MBSR for anxiety disorders, providing support for this delivery approach. Social anxiety symptoms improved more with in-person care. Clinicaltrials.gov: NCT03522844.