A Pilot Replication Study of Implementing an Equine-Assisted Services Program Within a VA Residential Substance Use Disorder Treatment Program

Abstract

Introduction Veterans have high rates of substance use disorders and other mental health conditions including post-traumatic stress disorder. Effective treatments for these conditions exist; however, high attrition rates and residual symptoms after completing treatment are common. Complementary treatment approaches could enhance treatment engagement and/or response among this population. We previously reported a study of one such intervention, an equine-assisted learning, and psychotherapy incorporating horses intervention provided to veterans admitted to a Veterans Health Care Administration residential substance abuse treatment program. The first aim of this study was to replicate the previous study assessing the safety, feasibility, and preliminary outcomes of this intervention. The second aim was to examine the effect of participants attending multiple intervention sessions. Materials and Methods Participants were 94 veterans who participated in one to six sessions of a 3–4-hour program consisting of both equine-assisted learning and psychotherapy incorporating horses. Pre- and post-session administration of the Positive and Negative Affect Scale, State-Trait Anxiety Inventory, and Craving Experience Questionnaire was utilized to assess changes in affect, anxiety, and craving. Wilcoxon signed-rank or paired two-tailed t-tests were utilized for pre- to post-session comparisons of the outcome measures for sessions 1–4. Generalized linear mixed-effects (GLME) models were constructed to determine the impact of dosage. GLME models were constructed to determine the impact of dosage. Results As with our previous study, the intervention was safe and feasible to utilize for this population. There were statistically significant pre- to post-session improvements, with medium-to-large effect sizes, for sessions 1–3 for negative affect and sessions 1 and 2 for positive affect, anxiety, and craving. The GLME models revealed no statistical significance for any of the predictors. Conclusions Taken together, this study and our previous investigation of this equine-assisted services intervention suggest that it is safe and feasible to utilize for veterans admitted to a residential substance abuse treatment program and we have now found short-term benefits in two separate studies. Thus, a randomized controlled trial of this intervention is warranted to demonstrate cause and effect and determine whether longer-term benefits are associated with the intervention. The finding that there was no additional benefit from attendance at more than two intervention sessions suggests that dose–response relationship studies of equine-assisted services interventions for veterans are needed.