Participants for this study were recruited from Combat Stress (CS). CS is a national charity in the UK that provides mental health services to veterans. Recruitment was done by writing to veterans who had accessed services from CS between January 2016 and January 2017 asking them to pass on information about the study to their partners providing they consented and were in an intimate relationship. A minimum sample size of 31 had been identified to detect a 0.5% effect size using the General Health Questionnaire (GHQ-12) between the start and the end of intervention at 80% power and 5% significance level. This is based on the prevalence rates as observed in a previous study profiling mental health needs of this group of participants⁷

An audit was undertaken to establish where the highest number of veteran referrals in the past twelve months (January 2016-January 2017) were populated across the UK. This helped determine where the intervention would be piloted across the UK. Based on this information and demand for support, nine city locations across the UK were chosen.

An options review was carried out of existing programmes on which to base the current one on. When conducting this review, several criteria were used to screen potential interventions. Firstly, the potential intervention needed to consider the military context specifically. Secondly, they were not explicitly about providing therapy and would replicate existing mainstream NHS services. Thirdly, the intervention needed to go beyond peer support and involve a therapist to support a structured programme of strategies to support the partner and also the environment within which the partner and veteran live. Fourthly, the intervention needed to be operationalised practically. As such, the intervention needs to be time limited and deliverable in a limited number of sessions to decrease the likelihood of participants dropping out. In addition, the treatment needed to be based in the community to make it more accessible and provide face to face contact. Finally, the intervention needs to have efficacy beyond the end of the intervention. Following this review, two psychoeducation support programmes SAFE (Support and Family Education Programme, Sherman, 2008) and Homefront Strong²⁰ which have been found to be effective and widely accepted by US military families were used to guide the development of this UK specific programme (The Together Programme). TTP. The development of TTP was done in accordance to The Medical Research Council guidance into the development of complex interventions²¹.

TTP consisted of ten sessions, organised into two categories: firstly, psychoeducation about PTSD and common mental health difficulties in veterans and secondly self-management strategies and skills training for participants themselves. Although each session was structured and dyadic in its design, there was ample opportunity for participants to share in their experiences. These sessions were delivered across five consecutive weeks. A mid programme review was also facilitated by study coordinator by phone to check in with participants. A CBT protocol was used to underpin the programme.

The first session “understanding PTSD and mental illness. How can I help?” And “living alongside mental illness and PTSD”, provides psychoeducation and normalises the reactions of partners. It provides opportunity for partners to share their experiences living alongside mental illness and discusses how these can be qualitatively different. A CBT hot cross bun model is outlined to illustrate how trauma experiences can impact upon veterans’ appraisals of everyday situations and the impact on partners are explored too. Practical strategies like grounding are introduced and are given as a homework activity for partners to do with the veteran, if willing.

Second, “understanding anger and PTSD” and “finding me again” uses the anger episodes model²² to help partners gain an understanding about anger and PTSD. Behavioural and cognitive strategies are presented to partners to help the veteran manage anger at each stage. In addition, caregiver responses are discussed using the Karp model²³ and reflected upon with an emphasis on partners taking care of their needs. The group practices progressive muscle relaxation at the end of the session.

Third, “communicating and reconnecting with partner” reviews communication styles and provides techniques for partners to communicate assertively with the veteran. Reconnecting and developing intimacy in the relationship is also addressed. Part two of session three “How to explain PTSD to children and others” begins by considering the experiences of stigma of partners. Internal and external stigma are differentiated, and the group are encouraged to identify those relevant to themselves in an individual task. Effective ways for talking to children and others about PTSD are explored. Partners are given specific techniques and tips for talking to children. A mindfulness exercise is practiced with the group at the end of the session.

In the fourth session “supporting my partner with low mood and depression” defines depression and the group are instructed about identifying early warning signs that may indicate risk for suicide and how to create a personalised crisis plan with the veteran. Emphasis was placed on removing potential blocks in communication and correcting myths about suicide. Value based behavioural strategies were given to partners to help address veteran’s low mood and depression. Effective means of coping with emotions were explored as a group, including; thought diffusion and compassion-based techniques.

Finally, “moving forward” begins with a review of the themes covered in the programme. The group engages in a reflective discussion about how to make sense of their experiences living alongside PTSD, empowering partners to focus on qualities they have developed. A wellness plan is explored and given as a tool to help them to individually maintain their wellbeing. A list of legitimate sights is also presented²⁴ and the group are invited to add any they fee relevant as a reminder about self-care. The session closes with a goodbye exercise and mindfulness visualisation to empower and give partners hope.

Service user involvement was sought by interviewing 20 partners to learn about what they would value from a support programme. and give consideration to practical needs like; length of programme and sessions, distance of travel to group, preference for telephone support, size of group, potential obstacles and enablers for attendance, expectations and goals of attending a programme of this kind. These partners were attending a one off psychoeducation session of two hours which is tailored for significant others of veterans engaging in a six week residential treatment programme. None of the partners who had taken part in these interviews were included in the target population during the intervention.

Clinicians with relevant expertise were also consulted with during the development of the intervention about the what themes should be covered and the format of the programme.

Participants were considered eligible for this pilot study if the veteran had been diagnosed with a mental health difficulty related to their military service. Participants were required to be in an intimate relationship with the veteran during recruitment, which was determined during screening by the project coordinator.

Due to careful consideration of participants privacy and confidentiality, no veterans were permitted to attend the group sessions. Other family members were also not eligible to attend because the content of the programme had been specifically written for partners in intimate relationship with the veteran. Intimacy is defined as the quality of being close, the ability to self-disclose and the desire to be affectionate with another person²⁵. This definition was used to determine if a partner was in an intimate relationship with the veteran during screening.

Data for this study was collected at three time points, before attending the programme, upon completion and three-months after completing the programme. At the beginning of TTP participants were asked to provide demographic information related to age, sex, no of dependents, if they had served in the military themselves, employment status, quality of life, living arrangements and length of relationship with veteran. In addition, seven standardised measures were administered. At the end of the programme participants were asked to complete the same questionnaires and provide qualitative feedback in a programme evaluation. Measures were again sent to participants three months after to assess for change in outcomes. This questionnaire was sent three times to elicit data.

Common mental health difficulties (CMD), such as anxiety and depression were explored using The General Health Questionnaire (GHQ-12)²⁶, this is a 12-item self-administered instrument. A cut off score of 12 or above was used to identify rates. Participants completed the Secondary Traumatic Scale, (STSS)²⁷, a 17 item self-report measure of secondary trauma symptoms using a five-point Likert scale. A cut score of 28 and above was used to indicate secondary trauma. In addition, a three item Alcohol Use Disorders Identification Test (AUDIT -C)²⁸ with a cut of score of five and above was used to indicate problematic drinking.

The seven item Dyadic Adjustment Scale (DAS)²⁹ and Relationship Assessment Scale (RAS)³⁰ were administered upon invitation to attend the programme, after screening. The functioning measures included perceived social support (MSPSS)³¹ and general self-efficacy (GSE).

Qualitative data was collected asking participants to suggest three most helpful things about TTP and three areas that could be improved. Content analysis was then used to elucidate the most frequently mentioned themes. Content analysis was performed by looking at the most frequently endorsed positive and negative statements made by participants. These were then grouped into common themes and the most prevalent themes reported

All participants in this study were female. Unfortunately, because no male partners expressed interest in attending the programme or were screened for this pilot study it was not possible to elucidate what barriers to care there may be for this population. The average age of participants attending was 47 years. Over half of the sample had been in the relationship with their partner for over nine years with half of participants having dependents. Most of the sample (63%) were in employment. A small proportion of participants had served in the military themselves (11%), all of which had not seen conflict in their military careers (Table 1). 45% of participants identified their quality of life to be low before taking part in the intervention.

Baseline mental health outcomes in Table 1 show 79% of participants met diagnostic criteria for anxiety and depression and 94% met criteria for secondary PTSD. A smaller proportion of participants (20%) met caseness for alcohol misuse at the start of this study.

Changes in primary and secondary outcome measures between the start of treatment and at three months follow up are presented in Table 3. These results demonstrate significant score reductions for CMD. At three months follow up, scores had fallen from above the cut off score of 12 to subthreshold levels of anxiety and depression (17.1 to 14.0, p<0.05). A medium effect size (0.48) was observed.

Significant reductions were observed in secondary traumatic stress symptoms. Changes in mean secondary traumatic scale scores reduced from being above the cut off score of 38 to below the threshold score (45.1 to 40.7, p<0.05). A medium effect size was observed (0.47). Overall these reductions appeared to be similar to those measured for anxiety and depression highlighting a global improvement in mental health and wellbeing after engaging in the five-week programme.

Improvements in secondary outcomes such as relationship satisfaction were observed for participants after completing the programme (17.9 to 23.0, p>0.05) however these were not found to be significant statistically. Modest improvements in participants ratings of alcohol usage, social support, self-efficacy after completing the programme were observed however, these changes did not reach statistical significance.

This pilot profited from strong ecological validity. This was because all of the participants were in relationships with veterans who had a mental health diagnosis, thus the sample used in this study is likely to be representative of the clinical population of help seeking veterans. Furthermore, we had high rates of programme completion and were able to successfully follow up 86% of those who completed the programme and no differences were present between participants were followed up or not. Finally, using a manualised approach increased our confidence in fidelity of the intervention being received. However, some groups or individuals would have benefited from a more flexible approach like having access to more than one telephone support session.

A number of limitations need to be considered when interpreting the described findings; We did not employ a randomised control trial design. Therefore, because we were not able to use both randomised and control conditions it is uncertain how much influence confounding variables may have had on the gains observed. For example, we did not control for whether participants had received support prior attending TTP. However, this was a pilot study and given the promising findings reported it would be impossible to conduct further controlled study to formally test the efficacy of TTP.

Secondly, the sample of participants included in this study may only represent partners of the ‘most ill’ veterans, for example data suggests 82% of veterans seeking this source of help have PTSD and comorbidities of three or more mental health needs and physical needs³⁴.

It is noted only partners of veterans actively seeking help or where the veteran had consented for their partner to have access to information about TTP were eligible for attending this study. As a result, we may be excluding partners the opportunity to engage in the programme who might need support. Unfortunately, it is not known if invitation responses may have influenced participants interest and engagement in the programme. Greater understanding of veteran’s altitudes towards their partners engagement in support is needed and educating the veteran about what the programme entails may allay concerns.

The aim of this study was to explore the feasibility of offering support to the partners of veterans living alongside veterans with PTSD. We used a noncontrolled design, and while limitations exist because of the lack of randomisation or control group, the data presented suggests cautious optimism for the efficacy of TTP to support the needs of individuals in romantic relationships living alongside veterans with mental health difficulties.