An Exploratory Assessment of Service Dog Training for the Treatment of Military Members with Posttraumatic Stress: A Pre-post Interventional Trial Authors: Kalyn C Jannace, Michelle Nordstrom, Shannon Fichter, Daniel B Cooper, Rebecca N Schulz, Drake R Ryff, Tiffany Riggleman, David Carlisle, Paul F Pasquina Published: 2026 ABSTRACT AbstractObjectiveTo assess the impact of a Service Dog Training Program (SDTP) on PTSD, anxiety, and depression symptoms among military beneficiaries experiencing PTSD.DesignNonrandomized, nonblinded, pre-post interventional clinical trial conducted between 2019 and 2022; analysis complete 2023.SettingRecruitment, screening, consent, and enrollment were conducted between March 2019 and August 2022. Active duty military, veterans, and military dependents were recruited from clinics at 2 major military hospitals in the National Capital Region.ParticipantsReceived care in the Military Health System, ages 18 and 65 years, clinical diagnosis of PTSD or posttraumatic stress, or 30 or higher on the PTSD Checklist-5 (PCL-5); diagnosis of an anxiety disorder, unspecified trauma, unspecified anxiety, or adjustment disorder.InterventionsParticipants completed assessments weekly for 4 weeks prior to beginning the intervention, which consisted of twice-weekly SDTP sessions for 6 additional weeks for a total study time of 10 weeks.Main Outcome and MeasuresPrimary: PTSD, anxiety, and depression as measured with the PCL-5 (0-80, higher scores indicate higher symptoms), Generalized Anxiety Disorder-7 (GAD-7, 0-21, higher scores indicate higher symptoms), and Patient Health Questionnaire-9 (PHQ-9, 0-27, higher scores indicate higher symptoms).ResultsThe 118 SDTP participants were, on average, 37.8 (±SD, 11.6) years old, male (57.6%), white (56.8%), employed (82.2%), and had a history of military service (97.5%). Post-SDTP intervention, mean score differences were −5.8 (95% CI, −8.4 to −3.1), −1.5 (95% CI, −0.7 to −2.3), and −1.2 (95% CI, −2.0 to −0.4) for the PCL-5, GAD-7, and PHQ-9, respectively.ConclusionsFollowing completion of the SDTP, participants reported significantly lower psychological impairment. The results of this study support the use of a structured SDTP as an engaging adjunct to existing clinical care for military-connected individuals experiencing PTSD and other psychological symptoms.KEYWORDS: Animal-assisted therapy, Chronic posttraumatic stress disorder, Posttraumatic, Posttraumatic stress disorder, Post traumatic stress disorder, PTSD, Rehabilitation, Service dogs, Stress disorders Over half of military service members (SM) report trauma exposure during service.1 Exposure to a traumatic event significantly increases the risk of mental health conditions, with approximately 30% of combat veterans experiencing PTSD compared with just 6% of civilians.2 PTSD symptoms can vary widely among individuals while being comorbid with other mental health conditions.1,3, 4, 5, 6 Further, PTSD severity has been associated with poor resilience, sleep disturbances, chronic stress, and pain interference.7, 8, 9, 10, 11 Given the complex biopsychosocial impact, no single treatment is effective for all individuals experiencing PTSD, with the dropout treatment rate as high as 50%.12, 13, 14 Exploring novel, engaging, adjunct therapeutic modalities is pivotal in improving biopsychosocial outcomes.15, 16, 17 A growing body of evidence suggests that having a service dog or engaging in an animal-assisted therapy (AAT) program as an adjunct to PTSD treatment improves psychological and psychosocial functioning while reporting high participation rates.18, 19, 20, 21, 22, 23, 24 Service dog training programs (SDTPs) are a form of AAT, where participants actively train a service dog to be placed with a veteran. SDTPs focus on behavioral shaping during each dog training session, with emphasis on patience reinforcement, emotional regulation, and focused attention, which has been purported to improve the social skills and well-being of the participant.25 Given the engaging nature of AAT and the focus on behaviors and symptoms associated with PTSD, the impact of SDTPs on psychological outcomes and PTSD warrants further investigation in military populations. Preliminary evidence has found service dogs and SDTPs to be associated with positive biopsychosocial outcomes.18,19,22,26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36 A nonrandomized longitudinal clinical trial found that veterans experiencing PTSD who received service dogs reported significantly improved PTSD, depression, and anxiety symptoms 3 months after receiving the dog compared with those remaining on a waitlist.20 Further, psychosocial function improved among the intervention group, with lower social isolation and anger and higher companionship, well-being, life satisfaction, and resilience.20 Other retrospective studies found significantly reduced pain, psychotropic and sleep prescriptions, dosages, and fewer mental health care encounters following participation in SDTP or possession of a service dog.37,38 A pilot feasibility study randomizing 29 participants to either an SDTP intervention or a waitlist found those receiving the intervention had significantly reduced PTSD symptoms compared with the participants on the waitlist, suggesting SDTPs improve psychological outcomes.19 Further longitudinal research with a larger sample is needed to explore the potential psychological health benefits of SDTPs for service members and veterans experiencing PTSD and psychological health symptoms. To address the limitations associated with observational studies and trials with small sample sizes, we conducted a pre-post, nonrandomized interventional study of an SDTP among Military Health System (MHS) beneficiaries experiencing PTSD or posttraumatic stress (PTS). The main objective of this study was to assess the impact of an SDTP on PTSD, anxiety, and depression symptoms among MHS beneficiaries and, secondarily, on psychosocial health. Methods This prospective longitudinal, non-randomized, pre-post interventional study measured psychological and biopsychosocial changes before, during, and after engaging in a structured SDTP. This study received human subject approval from the Walter Reed National Military Medical Center Institutional Review Board. Recruitment and inclusion criteria From March 2019 to August 2022, participants were recruited from Walter Reed National Military Medical and Alexander T. Augusta Military Medical Centers through referrals from Occupational Therapy, Psychiatry, Recreational Therapy, Social Work clinics, and Intensive Outpatient Programs for PTSD. Eligible individuals had to receive care in the MHS, be between the ages of 18 and 65, and have either (1) a clinical diagnosis of PTSD, (2) PTS symptoms defined as a score of 30 or higher on the PTSD Checklist-5 (PCL-5) at screening, or (3) had a clinical diagnosis of an anxiety disorder, unspecified trauma, unspecified anxiety, or adjustment disorder. Individuals were excluded if they were actively suicidal or homicidal, had been in an inpatient psychiatric unit within the last 30 days, demonstrated the inability to independently answer study questionnaires, had a fear of, or allergy to, dogs, participated in a clinical SDTP within the last 6 months, or were engaged in another PTSD study. Procedures Warrior Canine Connection’s (WCC) mission-based trauma recovery SDTP served as the intervention for this study, and sessions were executed by WCC trainers. Dog trainers were certified to ensure the same training techniques were used regardless of whether the trainer and dog were available for each session. Intervention sessions were held at 1 of 2 military hospitals; the research team remained consistent, and the sessions were located separately from any other clinical intervention. Participating in the SDTP did not replace existing clinical care. Dogs were Golden or Labrador Retrievers at least 18 months of age at the beginning of training. Study participants acted as their own controls, first completing a 4-week baseline period with no SDTP activities followed by a 6-week intervention period consisting of 1-hour SDTP sessions twice a week for a total of 12 intervention sessions. The participants did not train the same dog each session; however, the purpose for each session remained consistent, with the tasks adjusted based on the dog’s skill level. Participants completed sessions based on the intervention schedule, and if they missed a session, picked up on the next lesson. Participants were not compensated for participation in the intervention but could receive $50 if they participated in blood draws for the biomarkers portion of the study. Characteristics and outcomes Participant demographic covariates assessed at baseline by self-report included age (continuous and in years), sex (male and female), race (White, Black, Asian/Pacific Islander, Multiracial, American Indian, and other),39 employment status (employed and not employed), military status (active duty/reserve/national guard, veteran, and dependent), deployment history, (0, 1, 2, 3-5, 6, or more) education history (high school, associate’s degree, bachelor’s degree, and graduate degree), and relationship status (married/partner, single, and divorced/separated). Primary outcomes were PTSD, anxiety, and depression symptom severity, which were assessed once weekly during the 4-week baseline period and 6-week intervention period. Primary measures included the self-reported PCL-5 (score range, 0-80; higher scores indicate higher symptoms), Generalized Anxiety Disorder-7 (GAD-7) (score range, 0-21; higher scores indicate higher symptoms), and Patient Health Questionnaire-9 (PHQ-9) (score range, 0-27; higher scores indicate higher symptoms).40,41 Secondary outcomes included the psychosocial domains of sleep, stress, resilience, and pain. The Pittsburgh Sleep Quality Index (PSQI) measured sleep quality (score range, 0-21; higher scores indicate worse sleep quality), the Perceived Stress Scale (PSS) measured perceived stress (score range, 0-40; higher scores indicate higher perceived stress), the Connor-Davidson Resilience Scale (CD-RISC) measured resilience (score range, 0-100; with higher scores indicating better resilience), and the Defense and Veterans Pain Rating Scale (DVPRS) assessed current pain (score range, 0-10; higher scores indicate worse pain) and pain interference in the domains of activity, sleep, mood, and stress (all score ranges, 0-10; higher scores indicate worse interference) in the last 24 hours, with all secondary outcomes assessed at week 4 of baseline and weeks 3 and 6 of the intervention period, and again at 3 months postintervention.42,43 To assess the SDTP’s impact on pain and pain interference, the DVPRS was also assessed directly before, and immediately following, every SDTP participant session. Data analysis A total sample size of 132 was necessary to achieve 80% power and determine a 3.3-point statistical difference in PCL-5 scores pre-post intervention, accounting for 20% attrition. The mean outcome scores from assessments collected during the 4-week preintervention baseline period defined the baseline scores for each participant. The last collected score for each participant during the intervention period defined the postintervention outcome score. As such, if a participant’s last intervention session was in week 3 of the intervention instead of week 6, their final week 3 scores were utilized. Paired t tests compared pre-post outcome scores, and Hedges’ g calculated intervention effect sizes. Linear mixed effects models (LMMs) provided the mean difference and 95% CIs for primary and secondary outcome measures from baseline to the end of participation in the intervention or final intervention time point and to assess any lasting effects of the intervention at 3 months postintervention. LMMs efficiently handle missing assessment weeks and account for the dependency of observations within each participant, increasing study power to detect differences in assessment score means.44,45 LMMs also calculated the mean difference in scores for each measure from week 10 (last intervention week) to week 22 (follow-up). For the DVPRS, LMMs compared scores immediately before and after each intervention session. Hosmer and Lemeshow’s purposeful selection procedure selected covariates for inclusion in each LMM, with age, sex, race, employment status, education history, and relationship status included in adjusted models for the DVPRS. Models were not adjusted for the primary outcomes as no variables were selected for inclusion after purposeful selection. Missingness was determined, and the last observation carried forward imputed partially completed outcome scores, while fully missing scores were left missing. All analyses were conducted with R software, and the lme4 R package facilitated LMM fitting (R ver. 4.0.5, lme4 ver. 1.1-29). All hypotheses were evaluated at a 2-sided significance level of 0.05. Results From 2019 through 2022, 177 individuals were screened, with 17 found to be ineligible for participation. Following consent, 21 subjects were lost to follow-up and did not start study participation, and 20 started the study but did not receive the intervention. Of the 119 participants who started the intervention, 94 (79%) completed the final intervention week, and 67 (71%) completed the 3-month postintervention follow-up assessments. One participant withdrew consent and was removed from the analysis. A total of 114 participants had at least 1 baseline PCL-5 assessment and participated in at least 1 SDTP session with at least 1intervention PCL-5 assessment, while 115 had at least 1 baseline GAD-7 and PHQ-9 assessment and participated in at least 1 SDTP session with at least 1 intervention period GAD-7 or PHQ-9 assessment. Forty-five participants (38%) completed all 12 SDTP intervention sessions. Participant characteristics are shown in table 1. On average, SDTP participants were 37.8 (±SD, 11.6) years old, male (57.6%), white (56.8%), employed (82.2%), had a history of military service (97.5%), and had at least 1 military deployment (71.1%). There were no significant differences between those who completed the study and those who did not. Table 1Characteristics of SDTP Participants in the National Capital Region From 2018-2022 (N=118)Table 1 dummy alt textCharacteristicValueAge, y, mean (±SD)37.8 (11.6)Sex, n (%) Male68 (57.6) Female50 (42.4)Race, n (%) American Indian3 (2.5) Asian/Pacific Islander9 (7.6) Black23 (19.5) Multiracial8 (6.8) Other5 (4.2) White67 (56.8)Employment status, n (%) Employed97 (82.2) Not employed17 (14.4)Highest education, n (%) High school33 (28.0) Associate’s degree20 (16.9) Bachelor’s degree32 (27.1) Graduate degree26 (22.0)Relationship status, n (%) Married/partner57 (48.3) Single27 (22.9) Divorced/separated33 (28.0)Military status Active duty/reserve or guard103 (87.3) Veteran12 (10.2) Military dependent2 (1.7)Total military deployments, n (%) 029 (24.6) 124 (20.3) 221 (17.8) 3-524 (20.3) 6 or more15 (12.7)SDTP sessions, mean (±SD)9.6 (2.9) Characteristics of SDTP Participants in the National Capital Region From 2018-2022 (N=118) Primary study outcomes Table 2 reports the overall changes in outcome measures from baseline to final SDTP intervention. Mean PCL-5, GAD-7, and PHQ-9 scores decreased significantly from baseline to the last intervention, with a small to moderate effect on psychological symptoms. Table 3 displays the results of the LMMs assessing changes in the primary outcome scores at each intervention week compared with baseline, and the changes in secondary outcome scores at the midpoint and final intervention weeks compared with baseline. GAD-7 and PCL-5 scores were significantly lower than baseline from study weeks 6 through 10 (intervention weeks 2 through 6), and PHQ-9 scores were significantly lower from study weeks 8 through 10 (intervention weeks 4 through 6). PSQI scores were lower but not statistically significant at study weeks 7 and 10, while PSS scores were significantly lower at both points. CD-RISC scores increased at study week 7, though they did not significantly increase until study week 10. Table 2Participant Outcome Assessment Scores Before and After a 6-week Pre-Post SDTP Intervention in the National Capital Region, 2019-2022 (N=115)Table 2 dummy alt textOutcomeBaselineLast InterventionMean difference (95% CI)P-value*Effect Size (95% CI)†Primary outcomes, mean ± SD (range) PCL-5 (n=114)46.2 (17.3)40.4 (19.7)−5.8 (−8.4 to −3.1)<.001−0.4 (−0.6 to −0.2) GAD-713.3 (4.8)11.8 (5.7)−1.5 (−0.7 to −2.3)<.001−0.3 (−0.5 to −0.2) PHQ-914.9 (5.7)13.7 (5.9)−1.2 (−2.0 to −0.4).006−0.2 (−0.4 to −0.1)Secondary outcomes, mean ± SD (range) PSQI (n=105)12.7 (4.2)12.2 (3.9)−0.5 (−1.2 to 0.3).2−0.1 (−0.4 to 0.1) PSS (n=105)25.8 (6.1)23.6 (6.6)−2.2 (−3.5 to −0.9)<.001−0.4 (−0.6 to −0.2) CD-RISC (n=104)58.3 (19.1)62.0 (19.0)3.7 (1.3-6.1).0030.3 (0.1-0.6) DVPRS pain (n=105)5.0 (2.4)4.5 (2.4)−0.5 (−0.9 to −0.2).003−0.3 (−0.6 to −0.1)DVPRS activity5.3 (2.9)4.7 (2.7)−0.6 (−1.1 to −0.1).02−0.3 (−0.5, −0.04)DVPRS mood5.3 (3.1)4.6 (2.8)−0.7 (−1.2 to −0.2).004−0.3 (−0.5 to −0.1)DVPRS sleep5.3 (3.1)4.5 (2.8)−0.7 (−1.3 to −0.1).02−0.3 (−0.5, −0.05)DVPRS stress5.6 (2.9)4.8 (2.9)−0.8 (−1.3 to −0.23).005−0.3 (−0.53 to −0.1)Bold indicates significant finding at a P-value=.05⁎Paired t test.†Hedges’ g. Participant Outcome Assessment Scores Before and After a 6-week Pre-Post SDTP Intervention in the National Capital Region, 2019-2022 (N=115) Bold indicates significant finding at a P-value=.05 Paired t test. Hedges’ g. Table 3Linear Mixed Models Comparing Weekly Intervention Assessment Scores to Baseline Assessment Scores in SDTP Participants at Large Military Treatment Facilities in the National Capital Region, 2019-2022 (N=115)Table 3 dummy alt textStudy Week, Estimate (95% CI)GAD-7PCL-5 (n=114)PHQ-9Baseline (Wk 1−4)ReferenceReferenceReferenceWk 5−0.57 (−1.15 to 0.00)−1.16 (−2.77 to 0.46)−0.19 (−0.78 to 0.41)Wk 6−0.92 (−1.51 to −0.33)−4.05 (−5.72 to −2.28)−0.34 (−0.95 to 0.28)Wk 7−1.26 (−1.84 to −0.67)−3.79 (−5.42 to −2.15)−0.60 (−1.20 to 0.00)Wk 8−1.69 (−2.32 to −1.06)−4.60 (−6.36 to −2.85)−1.06 (−1.71 to −0.42)Wk 9−1.53 (−2.16 to −0.89)−4.68 (−6.45 to −2.91)−0.91 (−1.57 to −0.26)Wk 10−1.56 (−2.18 to −0.94)−6.38 (−8.12 to −4.63)−1.33 (−1.97 to −0.69)Secondary measuresPSQI (n=105)PSS (n=105)CD−RISC (n=104)Baseline (Wk 1−4)ReferenceReferenceReferenceWeek 7−0.17 (−0.84 to 0.49)−1.31 (−2.34 to −0.29)1.14 (−0.85 to 3.12)Week 10−0.61 (−1.31 to 0.09)−2.12 (−3.22 to −1.02)3.59 (1.47 to 5.70)Bold indicated significant at 2−sided alpha of 0.05. Linear Mixed Models Comparing Weekly Intervention Assessment Scores to Baseline Assessment Scores in SDTP Participants at Large Military Treatment Facilities in the National Capital Region, 2019-2022 (N=115) Bold indicated significant at 2−sided alpha of 0.05. DVPRS score changes at study weeks 7 and 10 are displayed in table 4, with significant decreases in all domains at week 10, but not significant for pain, sleep, and activity at week 7. DVPRS score changes immediately following an intervention session with a dog are shown in table 5. Compared with preintervention sessions, all DVPRS scores decreased significantly postsession with a dog. Table 4Linear Mixed Models Comparing DVPRS Scores During an SDTP Intervention at Large Military Treatment Facilities in the National Capital Region, 2019-2022 (N=105)Table 4 dummy alt textStudy Week, Estimate (95% CI)PainStressSleepActivityMoodBaseline (W 1−4)ReferenceReferenceReferenceReferenceReferenceWk 7−0.02 (−0.33 to 0.29)−0.77 (−1.21 to −0.32)−0.49 (−0.97 to 0.00)−0.39 (−0.80 to 0.01)−0.61 (−1.06 to −0.17)Wk 10−0.60 (−0.93 to −0.27)−0.80 (−1.27 to −0.33)−0.82 (−1.33 to −0.30)−0.67 (−1.11 to −0.24)−0.78 (−1.25 to −0.31)Bold denotes significant results at an alpha of 0.05. Linear Mixed Models Comparing DVPRS Scores During an SDTP Intervention at Large Military Treatment Facilities in the National Capital Region, 2019-2022 (N=105) Bold denotes significant results at an alpha of 0.05. Table 5Linear Mixed Models Assessing DVPRS Score Changes After Each SDTP Intervention Session to 2019-2022 (N=105)Table 5 dummy alt textIntervention Session (Study Wk) to Model Estimate (95% CI)PainStressSleepActivityMoodSession 1 (Wk 5)−0.50 (−0.68 to −0.32)−1.05 (−1.26 to −0.84)−0.36 (−0.57 to −0.16)−0.72 (−0.91 to −0.53)−0.85 (−1.06 to −0.65)Session 2 (Wk 5)−0.51 (−0.67 to −0.36)−1.06 (−1.24 to −0.87)−0.38 (−0.56 to −0.20)−0.74 (−0.91 to −0.58)−0.88 (−1.06 to −0.70)Session 3 (Wk 6)−0.53 (−0.66 to −0.39)−1.06 (−1.22 to −0.91)−0.40 (−0.55 to −0.24)−0.77 (−0.91 to −0.62)−0.91 (−1.07 to −0.76)Session 4 (Wk 6)−0.54 (−0.66 to −0.42)−1.07 (−1.21 to −0.93)−0.41 (−0.55 to −0.28)−0.79 (−0.92 to −0.66)−0.94 (−1.08 to −0.81)Session 5 (Wk 7)−0.55 (−0.66 to −0.45)−1.08 (−1.20 to −0.96)−0.43 (−0.55 to −0.31)−0.81 (−0.93 to −0.70)−0.97 (−1.09 to −0.85)Session 6 (Wk 7)−0.57 (−0.67 to −0.47)−1.09 (−1.21 to −0.97)−0.45 (−0.56 to −0.33)−0.84 (−0.94 to −0.73)−1.00 (−1.11 to −0.89)Session 7 (Wk 8)−0.58 (−0.68 to −0.48)−1.10 (−1.22 to −0.98)−0.46 (−0.58 to −0.35)−0.86 (−0.97 to −0.75)−1.03 (−1.15 to −0.91)Session 8 (Wk 8)−0.59 (−0.70 to −0.48)−1.10 (−1.24 to −0.97)−0.48 (−0.61 to −0.35)−0.88 (−1.00 to −0.77)−1.06 (−1.19 to −0.93)Session 9 (Wk 9)−0.61 (−0.73 to −0.48)−1.11 (−1.26 to −0.96)−0.50 (−0.64 to −0.35)−0.91 (−1.04 to −0.77)−1.09 (−1.24 to −0.94)Session 10 (Wk 9)−0.62 (−0.77 to −0.48)−1.12 (−1.29 to −0.95)−0.52 (−0.68 to −0.35)−0.93 (−1.09 to −0.78)−1.12 (−1.29 to −0.95)Session 11 (Wk 10)−0.63 (−0.80 to −0.47)−1.13 (−1.33 to −0.93)−0.53 (−0.73 to −0.34)−0.96 (−1.14 to −0.78)−1.15 (−1.34 to −0.96)Session 12 (Wk 10)−0.65 (−0.84 to −0.46)−1.13 (−1.36 to −0.91)−0.55 (−0.77 to −0.33)−0.98 (−1.19 to −0.77)−1.18 (−1.40 to −0.96)Study weeks 1-4 were baseline weeks with no interventions; CI at an alpha of 0.05; significant findings are in bold; model adjusted for age, sex, race, employment status, education history, and relationship status. Linear Mixed Models Assessing DVPRS Score Changes After Each SDTP Intervention Session to 2019-2022 (N=105) Study weeks 1-4 were baseline weeks with no interventions; CI at an alpha of 0.05; significant findings are in bold; model adjusted for age, sex, race, employment status, education history, and relationship status. Three months postintervention, a total of 57 participants completed the primary outcome measures; for secondary outcomes, 47 completed the PSQI, 50 the PSS, 49 the CD-RISC, and 48 the DVPRS and subscales. PCL-5 and GAD-7 scores increased 1.44 (95% CI, −1.25 to 4.12) and 0.29 (95% CI, −0.67 to 1.24), respectively, while PDQ-9 scores decreased −0.22 (95% CI, −1.21 to 0.77), although these results were not significant. PSQI scores increased (0.56; 95% CI, −0.32 to 1.43), while PSS (−0.74; 95% CI, −2.08 to 0.60) and CD-RISC scores decreased (−0.56; 95% CI, −3.22 to 2.09). DVPRS pain (0.19; 95% CI, −0.24 to 0.62), stress (0.02; 95% CI, −0.59 to 0.62), sleep (0.62; 95% CI, −0.03 to 1.27), activity (0.34; 95% CI, −0.22 to 0.90), and mood (0.28; 95% CI, −0.33 to 0.89) increased. Discussion This nonrandomized, pre-post interventional trial found that an SDTP can be a successful, non-invasive adjunct to treatment of psychological symptoms and psychosocial functioning for military beneficiaries experiencing PTSD. Following a 4-week baseline period with no service dog training, then 6 weeks of twice-weekly intervention sessions with a dog, participants reported significantly less PTSD, depression, and anxiety symptoms. Participants also reported significantly reduced stress, pain, and pain interference in the domains of activity, mood, sleep, and stress, and significantly increased resilience. The results of this study build upon the “Little Dog” study, which used the same SDTP intervention with a much smaller sample.19 Both studies found a moderate effect of the intervention on PTSD symptoms; however, the present study’s results reached statistical significance, denoting an observable impact of the intervention on PTSD symptoms. Further, the present study found a small to moderate, but also significant, impact of the intervention on anxiety and depression symptoms. The present study also examined the mean difference in PCL-5, PHQ-9, and GAD-7 scores pre-post SDTP intervention, finding the intervention significantly decreased average scores. For secondary outcomes, this study found a significant increase in resilience and a significant decrease in perceived stress, like the “Little Dog” study, though their results did not reach statistical significance. Similar findings suggest the present study’s larger sample size allowed for the power to detect significant differences in the group and supports that SDTPs have a positive impact on psychosocial health. With each intervention week, symptom scores for all primary outcomes were lower, suggesting that as participation in the SDTP increased, the overall impact on primary symptoms decreased. There was also a significant improvement in secondary outcome scores, except for the domain of sleep. The SDTP intervention also demonstrated a small improvement in PSQI, although these results were not statistically significant; therefore, the impact of an SDTP on sleep continues to warrant further investigation. All other secondary outcomes were significantly reduced from preintervention to postintervention, with a small to moderate impact of the intervention found for the domains of resilience, perceived stress, and pain, to include pain interference with activity, mood, stress, and sleep. This SDTP intervention also improved pain and pain interference in the domains of activity, mood, stress, and sleep directly before and after each SDTP session. These findings were significant in reported pain and pain interference across all domains. The most substantial drops were observed in the domains of mood and activity, which is consistent with previous studies that demonstrated service dogs and AAT to be associated with reduced psychological symptoms and improved psychosocial functioning.19,20,38,46, 47, 48 Further, with each additional intervention session, the reduction in scores presession versus postsession increased, suggesting a dose-response relationship is likely present with SDTP, though this study did not have the power to detect this because of overlapping confidence intervals. Finally, primary outcome scores 3 months postintervention increased for PCL-5 and GAD-7 and showed a small decrease for the PHQ-9; however, these findings were not significant. Secondary outcome measures trended similarly. This could suggest a somewhat lasting impact of the intervention on psychological and psychosocial health; however, because of the substantially decreased response rate at the 3-month follow-up, it is also possible the effects of the intervention were waning, and study analysis was not powered enough to detect this difference. Study limitations This study was not without limitations. First, the coronavirus 2019 (COVID-19) pandemic started in the middle of recruitment and during existing intervention sessions, and despite great efforts to ensure continuity of study execution, some variation in SDTP sessions did occur. To maintain social distancing and infection control procedures, participants were staggered, and there was variation in SDTP session intervals and weekly assessment completion times. Some participants also may have missed a week of intervention before coming back and completing the following week, which impacted the overall sample available for analysis. This may bias measures toward the null, eliminating or attenuating the impact of the intervention. Further, evidence is mixed regarding the impact of the pandemic itself on the mental health of military-connected individuals.49, 50, 51 As such, it is not possible to know if the pandemic may have impacted the mental health of the participants. This was a pre-post study where all participants served as their own controls. No control group means the counterfactual comparison used to estimate intervention effects was within participants who all received the intervention. The effects observed could be because of a placebo effect or observation bias, though observation bias can also be present in control groups. The intervention was that participants had universal health coverage and may have better controlled psychological symptoms because of a consistent care regimen. Even with COVID-19 pandemic disruptions, most (85%) participants received at least 1 intervention session, with most completing at least 11 of the 12 sessions. This suggests that the intervention was well received and provides quantitative support to the qualitative findings from the “Little Dog” study.19 Finally, this study specifically considered how an SDTP may improve psychological health symptoms in the immediate and shortly after participation. However, what was not measured were longer-term outcomes and whether the effect of the intervention was long-lasting. Additionally, the participants did not take the service dog home and may not have had their own service dog at home, which could lead to a slow return to baseline psychological functioning. Given disruptions to recruitment and retention resulting from the COVID-19 pandemic, this study was not powered to capture differences 3 months postintervention. Future studies should examine outcomes beyond the 3 months post period and include questions regarding how health behaviors may have changed or been influenced because of the intervention. This would include determining if a participant had a service dog at home or sought to obtain a service dog following the intervention. Conclusion Following a SDTP intervention, participants reported significantly lower psychological and psychosocial impairment, including improved pain and pain interference. The results of this study support the use of a structured SDTP as an engaging adjunct to existing clinical care for military-connected individuals experiencing PTSD and other psychological symptoms. Disclaimers The identification of specific products or scientific instrumentation is considered an integral part of the scientific endeavor and does not constitute endorsement or implied endorsement on the part of the authors, DoW, or any component agency. The views expressed in this article are those of the authors and do not necessarily reflect the official policy of the Department of War or the U.S. Government. The opinions and assertions expressed herein are those of the author(s) and do not reflect the official policy or position of the Uniformed Services University of the Health Sciences or the Department of War. The views expressed in this article are those of the authors and do not necessarily reflect the official policy of The Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc. Disclosure The authors have no conflicts of interest to report. Acknowledgments Acknowledgments The authors would like to acknowledge the contributions of the Big Dog Team, especially those of Samantha Klock and Morgan Hunt, of the Henry M. Jackson Foundation, and Rick Yount and Meg D. Olmert from the Warrior Canine Connection, which provided the intervention and service dogs in training for this research. Data statements Data Access, Responsibility, and Analysis: Michelle Nordstroma,c and Kalyn C. Jannacea,b had full access to all the data in the study and take responsibility for the integrity of the data and accuracy of the data analysis. Kalyn C. Jannacea,b was responsible for overseeing all statistical analysis for the study and for final approval of results. Drake R. Ryffa,b, Daniel B. Coopera,b, and Rebecca N. Schulza,b all conducted the statistical analysis for this study. aCenter for Advanced Research for Military Optimization, Readiness, and Rehabilitation, Department of Neurology and Rehabilitation, Uniformed Services University for the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814, USA bHenry M. Jackson Foundation for the Advancement of Military Medicine, 6720A Rockledge Drive, Bethesda, MD 20817, USA cWalter Reed National Military Medical Center, 4494 Palmer Road North, Bethesda, MD 20814, USA Data Sharing Data from this study will be shared with other researchers only through the execution of an appropriate data sharing agreement and only from those participants who consented to their data being shared. All data will be de-identified and include no data that would be linked directly to a participant. A limited data dictionary will be shared with the elements outlined in the data sharing agreement. All statistical/analytical code used to arrive at the results presented in this study can be shared. Footnotes Supported by Congressionally Reprogrammed RDT&E funds, award HU00011520016. The funding organization only provided the funding for the project and did not have a role in study design, execution, or analysis. Clinical Trial Registration No.: Registration Number: NCT03907254. References