# Increasing Resiliency in U.S. Air Force Personnel: A Multi-Site Trial Protocol

## TL;DR

This study tests how well a stress management training program improves resilience in U.S. Air Force personnel through different delivery methods.

## Contribution

The study introduces a two-arm randomization approach to evaluate accessibility and effectiveness of in-person versus computer-based training.

## Key findings

- The study will assess resilience, stress, anxiety, and quality of life at multiple time points post-intervention.
- Mixed-effects models will analyze longitudinal trends to understand intervention effectiveness over time.
- Results will compare real-world settings and randomized control trial outcomes to evaluate training adaptability.

## Abstract

The purpose of this study is to examine the efficacy of the Stress Management and Resilience Training (SMART) in increasing the resilience of U.S. Air Force personnel. We aim to recruit up to 500 active component Air Force personnel and provide a two-arm randomization modality to make SMART more accessible and adaptive to the personnel's schedules. Two-arm randomization will be used to assign three sites for participants to choose in-person or computer-based training (CBT) and two sites where participants are randomized into their training type (in-person or CBT). The use of two-arm randomization will enable the examination of the difference between real-world settings within the framework of causal inference, as well as, differences based upon self-selection and a randomized control trial. We propose to examine the intervention effects at 12, 24 and 36-weeks post-intervention. Initial analysis will include descriptive statistics to characterize demographic status, military grade, duty location, and military occupation. The objectives of our analyses will include testing and estimating the intervention effects by comparing pre-post intervention changes in resilience, stress, anxiety, and QoL at each follow-up. Scores will also be pooled to test for overall intervention effects over time. Intervention effectiveness will be reported by comparing mean or median effects using 95 % confidence intervals and effect size estimates. An analysis of the longitudinal trend over the study period will be conducted by simultaneously examining data from all follow-ups using mixed-effects models in which random effects will be used to characterize between and within-subject variations.