Post-combat adaptation: improving social support and reaching constructive growth

ABSTRACT

Background and Objectives: Posttraumatic stress disorder, a commonly researched mental health outcome associated with trauma, does not develop in the majority of survivors. More common trajectories of adaptation include resilience, and posttraumatic growth (PTG). The objectives of the current study were to: (1) describe posttrauma adaptation profiles in a sample of Israeli male military veterans (N = 448); and (2) to explore the protective factors that promote constructive PTG within two profiles of posttrauma adaptation.

Methods: The study used secondary data to estimate latent profile mixture models and a series of logistic regression analyses.

Results: Demographic controls, combat related variables, endorsement of coping strategies, and reports of improvement in social support were not significant predictors of constructive growth in the resilient class. However, those in the struggling growth subset of the sample who reported improvement in perceived social support increased the odds of reaching constructive growth.

Conclusion: These findings highlight the importance of tailored clinical interventions that account for more complex profiles of posttrauma adaptation; and further, provide evidence that adaptation takes place over time. Finally, these findings call for future research to continue to explore the quality of PTG and the contexts in which protective factors promote positive adaptation.

KEYWORDS:

• Combat
• social support
• posttraumatic growth
• latent profile analysis

Acknowledgements

The authors wish to thank the Peace of Mind Team at METIV: Israel Center for the Treatment of Psychotrauma and the soldiers who participate in the Peace of Mind program. Without them, this project would not have been possible.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 A waitlist control design was utilized during the original data collection. Participants were allocated to the waitlist or intervention groups using a matching procedure in which unit members were compared on demographic and combat variables. Due to limitation in the control data, and the focus of the current study which precludes an evaluation of the efficacy of the intervention, the control group data was not included in the presented analysis. Rather, only data collection during the intervention phase was used, this included three combat units that had moved off of the waitlist control and completed the entire intervention.

2 Given the subsamples selected for the second phase of analysis were a non-random subset of the population, and given these subsamples were selected because of initial reports of growth, the data used in the regression models was not independent of the outcome variable of interested (i.e. constructive growth). As such, there was concern of a non-corrected selection bias in a typical logit model. To address this concern, a Heckman probit model was selected as a more appropriate method of estimation, as the Heckman regression is specifically designed to test, and account for, selection bias (Bushway, Johnson, & Slocum, 2007). However, there was no selection bias found in the Heckman estimation. These models are available upon request.

3 The latent profile models presented here do not test the assumption of local independence. An alternative model that allows continuous indicators in the profiles to covary within latent class are available upon request.

4 The equality of coefficients between models for resilience and struggling subgroups were tested. There was no significant difference between the coefficients for improvement in social support between models (b = 0.656, p = 1.18).