Long-Term Evaluation of an Australian Peer Outdoor Support Therapy for Contemporary War Veterans

Questionnaire data was collected from the nine participants at four time points (pre-Trek, immediately post-Trek, 2-3 month follow-up, and 9-month follow-up). Analysis of individual participant scores on questionnaires over time was assessed using Reliable Change (RC) calculations (37). RC compares pre- and post-intervention scores to determine whether an individual’s positive or negative results reflects meaningful change (i.e., whether any observed change is outside of the expected bounds of typical variation for that measure).

Program Description of Trojan’s Trek. Trojan’s Trek is an annual 6-day South Australian POST held in the Northern Flinders Ranges. Trekkers spend three nights camping and travelling in the Flinders Ranges. The first and last nights are spent at Moolooloo Station. The Trek incorporates a range of different psychological approaches within its program structure including CBT-based group sessions, skill building activities, informal gatherings and one-on-one debriefing activities. Trojan’s Trek is drug and alcohol free and is funded by private and corporate donations and the South Australian Department of Health.

Data Collection. Trojan’s Trek requires participants to individually complete questionnaires pre- and immediately post-Trek in a group setting. Paper copies of follow-up questionnaires (at 2-3 and 9-months) are mailed to participants with a postage-paid return envelope. Participants are asked to complete the four questionnaires within a 2-week timeframe in their choice of setting. The University of South Australia Human Research Ethics Committee exempted this study from ethics approval because it used archival data.

The Reliable Change Index (RCI) was used to examine the data for each of the nine participants. The RCI tells us whether change reflects real change or fluctuations expected as a result of the standard error of measurement for the particular test being used (37). Questionnaire results from immediately post-Trek, 2-3 month follow-up, and 9-month follow-up were compared to baseline measures using RCI calculations where the necessary statistical information was available.

The test-retest reliability of 0.47 (for men) was used to calculate RCIs for the GSE. Test-retest reliability was unavailable for the DASS21, however, it was available for the DASS (42-item version) (r=0.71, r=0.79, and r=0.81 for depression, anxiety, and stress respectively) (43). By multiplying results on the DASS21 by two, scores became comparable to the DASS-42 (41, 44). Test-retest reliability information was not available for LSQ and PNI and so RCI calculations were not conducted. Instead, general comments are made about these findings.

Table 1 shows means (and standard deviations) for DASS21 scores at each measurement point. Mean scores for Depression and Stress Subscales were in the severe range pre-trek, and were in the normal to mild range throughout follow-up. Mean scores on the Anxiety subscale were in the extremely severe range pre-trek, and were reduced to the moderate range throughout follow-up.

Figure 1 shows a stacked bar graph of the distribution of DASS-21 scores across severity bands at each measurement point. Pre-trek, nearly 80% of participants reported severe or very severe levels of Depression, anxiety and Stress, which was reduced post-trek, remaining at less than one third of participants throughout follow-up (as indicated by the dashed line, Figure 1).

Figure 1.

Figure 2 (first three panels) displays results of the Reliable Change analyses for DASS-21 measures. A reliable decrease in Depression occurred for 67% of the sample immediately post-trek and at the 2-3m follow-up. This decrease was maintained by 44% of participants at 9 months. Anxiety was reliably reduced for 78% of participants throughout follow-up. Stress was reliably reduced in 78% of participants at 0 and 2-3 months, and maintained in 67% at 9 months. No participants experienced a reliable increase in DASS-21 measures.

Figure 2.

Table 2 shows means (and standard deviations) for wellbeing measures at each measurement point. All measures demonstrated improved wellbeing following the trek.

The only wellbeing measure with sufficient published statistics to calculate reliable change was Self-Efficacy.

This is displayed in the final panel of Figure 2. Self-Efficacy scores increased for approximately half of the sample during follow-up. No participants experienced a reliable decrease in self-efficacy.

Figure 3 displays the simple change in the remaining wellbeing measures at each of the post-trek time points. The top panel of Figure 3 shows the positive wellbeing parameters. Immediately post-trek, one third of participants reported an increase in positive friend relationships, which increased to nearly 90% at 2-3 months and 100% at 9 months. One third of the sample reported an increase, and 44% reported a decrease in positive family relationships post-trek. This was consistent throughout follow-up. An increase in positive partner relationships was seen for 44% of participants immediately post-trek. This increased to 56% at 2-3 months, dropping to 22% at 9 months. Life satisfaction was increased for 67% of participants. This was consistent throughout follow-up.

Figure 3.

The lower panel of Figure 3 shows the negative wellbeing parameters. Nearly 80% of participants reported a decrease in negative friend relationships at 0 months and 2-3 months post-trek. This reduced to 67% at 9 months. Nearly 80% of participants reported decrease in negative family relationships at 0 months. This reduced to just over half at 2-3 months and one third at 9 months. Nearly 80% of participants reported decreases in negative partner relationships at 0 months, 100% at 2-3 months, returning to 80% at the end of follow-up.

The small sample size means that the generalizability of results to the overall veteran population is limited. Sample size limitations could be overcome in future studies by integrating Trojan’s Trek participant groups across different years into a single sample for analysis. Missing data-points further decreased available data for analysis. Methods of mean substitution and interpolation were implemented to maximise the available data-set. These methods may have influenced the results.

The absence of a control group further limits the study. The archival nature of the data meant that it was not possible to include a control group. Future studies could include a matched wait-list control to help determined whether the passing of time alone had a healing effect on participants rather than therapeutic components related to Trojan’s Trek.

External factors or life circumstances may have influenced questionnaire scores, especially at 2-3 and 9 month follow-ups. Trekkers may have been undergoing other therapeutic interventions between the conclusion of the trek and follow-up and it was not possible to assess whether the observed positive changes resulted from the Trek alone or from additional treatments. It is also possible that Trojan's Trek is self-selective in that those who agreed to participate may be more likely to engage in therapy as a whole. As Trojan’s Trek is alcohol and drug free, positive changes experienced by veterans during the trek may be attributable to the break from these substances. Future studies should further assess these potential influencing factors.

The assessment tools chosen by Trojan’s presented problems during analysis. Neither the PNI nor the LSQ have sound psychometric properties, meaning that the validity and reliability of these tools is unknown. The varied PNI results may have come about as a result of the questionnaire itself, rather than a lack of positive outcome from the Trek. The current researchers have encouraged Trojan’s Trek to use more psychometrically sound and well-regarded assessment tools to improve the meaning of results, as the PNI and LSQ lacked comprehensive psychometric properties and scoring instructions.

The current study is the first long-term evaluation of an Australian POST program to date, and more specifically the first long-term evaluation of Trojan’s Trek. The results provide limited support for the effectiveness and benefits of Trojan’s Trek for contemporary post-deployed war veterans. Given the fact that only one group has run annually to date, the feasibility of running additional and more regular groups should be considered. Having rapid access to a Trojan's Trek group would improve the accessibility and hence the likely effectiveness of this approach to assisting traumatised veterans.

The fact that the researcher did not become actively involved in the data collection from participants and so did not contaminate the underlying focus on peer relationships in Trojan’s Trek increases confidence in the conclusion that it was the peer relationships rather than the presence of the researcher that led to the observed benefits. Rapport and trust relationships were established between participants and staff and the researcher did not disrupt this. It is likely that participants would not have been as open in their personal sharing with an independent researcher as they were with Trojan's Trek staff.

A further strength was that the study did not only consider group statistical differences, but also the clinical significance of Trojan’s Trek for individuals. RCIs were used to determine clinical significance and determine a greater meaning of results where possible despite the small sample size.

Despite limitations associated with the questionnaires that Trojan’s Trek administered to participants, the questionnaires were quick to complete with questions that were easy to understand, consequently minimising participant burden. It is estimated that the four questionnaires took approximately ten minutes to complete.

Findings have great applicability in assisting war veterans who are experiencing difficulty reintegrating back into society. This research has the potential to help veterans who are struggling to find appropriate mental health care for psychological issues.

Results of the current study lend support to POST as an alternative or complementary treatment for post-deployed veterans. Previous research (28) showed positive results over a three-month period. The current researchers showed that these benefits could be maintained over a longer period of time. This suggests that Trojan’s Trek could be considered as a ‘first-step’ treatment approach for veterans seeking psychological assistance.