Quantitative and Qualitative Outcomes of Transactional Analysis Psychotherapy with Male Armed Forces Veterans in the UK presenting with Post-Traumatic Stress Disorder

This paper presents findings from a two-year research project conducted within a live-in residential charity setting in the UK, examining clinical outcomes of TA psychotherapy among 15 male armed forces veterans presenting with severe PTSD (post-traumatic stress disorder) and other comorbid disorders. Outcomes were measured for short-term (24 sessions) and long-term (52 sessions) transactional analysis (TA) treatment using the quantitative CORE-OM (Evans, Mellor-Clark, Margison, Barkham, McGrath, Connell & Audin, 2000), PHQ-9 (Kroenke, Spitzer & Williams, 2001) and GAD-7 (Spitzer, Kroenke, Williams & Löwe, 2006) questionnaires and the qualitative Change Interview (Elliott, Slatick, & Urman, 2001, as cited in Frommer & Rennie, 2001). Quantitative findings show that positive Reliable Change on global distress, depression and anxiety has taken place within both the short-term and long-term treatment groups with some clients achieving Clinically Significant Change on these measures. Qualitative findings arising from thematic analysis (Braun & Clarke, 2006) indicate that a broad spectrum of therapist factors and psychotherapy process factors within the TA therapy delivered were beneficial for this particular client group. The negative influence of a number of psychosocial factors on the veterans' well-being is also discussed based on numerical data and interview responses. Overall, these results suggest that TA psychotherapy can be effective in the treatment of PTSD among combat veterans.


Introduction
The aim of this research project was to investigate whether transactional analysis (TA) psychotherapy can be an effective treatment for post-traumatic stress disorder (PTSD) in ex-servicemen and women.Among the authors' motivations for pursuing this extensive study were the encouraging results generated by an earlier pilot study (Harford, 2013) in the same setting, together with their identification of a clear gap in the available TA research literature.There was also an awareness that it is highly unusual for a private practitioner to secure the opportunity of providing long-term TA psychotherapy on an indefinite basis with no limit on the number of sessions available for each client and all sessions funded by an independent charitable organisation.In addition, there has been intense media coverage of the problems faced by military personnel returning from Afghanistan and Iraq in recent years, the Broken by Battle edition of the BBC's Panorama programme (BBC 2013) being one memorable example, and this increased public awareness, combined with an attendant increase in the number of veterans willing to seek assistance, and limited NHS provision, has brought the need for a broader range of effective treatments for PTSD sharply into focus.

Definition of Post-Traumatic Stress Disorder
PTSD is a psychiatric disorder which presents after the individual has experienced a severe traumatic event, such as an assault, or accident which involves injury or threat of death, or witnessing severe injury, death, or threat of injury or death for others, where the individual experienced emotions such as powerlessness, horror and terror.Symptoms include recurrent intrusive thoughts and images related to the trauma (flashbacks), avoidance of stimuli which remind the individual of the trauma in some way, intense distress and physiological reactivity, nightmares, persistent anxiety and dissociative phenomena (American Psychiatric Association, 1994).

Literature Review
At the time of writing, the only published research investigating the effectiveness of TA psychotherapy in the treatment of PTSD is the pilot study conducted by Harford (2013).Based on a sample of six veterans and quantitative data gathered across 16 sessions of treatment, he concluded that " . . .anxiety and, to a lesser extent, depression appear to gradually reduce as, within the non-intrusive safety of an empathic therapeutic relationship, the veterans re-experience previously repressed affect, obtain the longed-for attuned response to their pain and then slowly build their Integrating Adult . . .capacity 'to reflect upon and integrate their own archaic states as well as past introjects, and . . .draw on them in the service of present-centred relating' (Tudor, 2003, p.202)".(Harford, 2013, p.28) Harford (2013) goes on to suggest that these initial results, "offer an encouraging level of support for the working hypothesis that 'TA psychotherapy is an effective treatment for PTSD' " (p.28), though with the qualification that "further research investigating the process and outcome of [TA] therapy for PTSD is warranted" (p.28).The present paper is based on a twoyear research project which immediately followed this pilot study and is the first full-scale analysis of the effectiveness of TA psychotherapy with this client group.
The available TA theoretical literature on PTSD is also relatively slim, though this has been addressed to some extent by the July and October 2012 editions of the Transactional Analysis Journal, which feature several articles exploring TA approaches to the treatment of trauma in adults; the former collection, in particular, examining survivors of torture, intentional violence and other conflict-related extremities of human experience.Clarkson (1987) provides one of the earliest TA theories relevant to this study in her concept of the Bystander, who "By not challenging or intervening . . .[gives] tacit permission to the abuse of power occurring in their environment" (p.82); a position all too familiar to some of the participants in Harford's (2013) pilot study, who reported being haunted by profound guilt over their involvement "passively in violent, or oppressive situations" (Clarkson, 1987, p.82).A useful entry point for any consideration of the self-protective function of dissociative defences in the face of traumatic experience is supplied by Erskine (1993), who suggests that dissociation "allows a person to remove [themselves] cognitively and emotionally from the experience and to physically adapt and behaviourally conform to external demands" (Erskine, 1993, p.184).Of particular significance with regard to PTSD, he adds that "Continuing the dissociation after a traumatic event enables a person to disengage from [relational] needs and emotions and to evade the memory and its devastating impact" (p.184).He goes on to offer an integrative approach to treatment that advocates empathic inquiry, attunement and involvement with the aim of "integration of affect-laden experiences…intrapsychic reorganization of the [veterans'] beliefs . . .and the integration of the dissociated parts of the personality " (p.190).Such integrative models of aetiology, symptomology and treatment (Erskine, 1993;Erskine & Trautmann, 1996) informed key aspects of the TA psychotherapy delivered to the veterans participating in this study.Pomeroy (1998) suggests that neuroscience may be helpful in conceptualising trauma by considering how the limbic system has the "ability to pre-emptively activate the survival mode system when it believes it detects a threat" (p.332).She proposes that this has relevance to the aetiology and symptomology of PTSD, which she believes occurs "when the limbic system is unable to return [executive] control to the neocortex" (p.335).She also provides a treatment plan for adults affected by trauma and PTSD, which corresponds closely to treatment plans offered within other psychotherapy modalities (see below) and adds insights from neuroscience that can be used to underpin the reparative models of the integrative school (Erskine, 1993;Erskine & Trautmann, 1996).Though focussing primarily on dissociative identity disorder (American Psychiatric Association, 1994), Korol (1998), meanwhile, offers a concise summary of the goal of treatment as "twofold: 1) to become aware of and accept disowned parts (i.e., ego states) and 2) to become able to contact other people while maintaining a sense of self" (p.115), a dual emphasis on the need for intrapsychic and interpersonal contact (Erskine, 1993) that, again, played a significant role in the therapy examined by this research.
Building on Pomeroy's (1998) depiction of PTSD as "a limbic system disorder" (p.335), Stuthridge (2006) foregrounds the implicit, explicit and autobiographical memory systems and, combining these neurological insights with Tudor's (2003) view of "neopsychic function as a process of integration, reflective function, and narrative" (p.272), envisages trauma as precipitating a breakdown in the Adult capacity for "integrating disparate Parent and Child ego states into a unified sense of self" (p.271), resulting in a fragmentation of the traumatic experience and the subsequent intrusion of "dissociated memory fragments, smells, images, and sensations that are not located in time and space" (p.274).She moves on to outline her relational TA approach, which employs attunement within the therapeutic relationship to resolve transferential enactments and facilitate neopsychic integration of dissociated Parent and Child fragments, and concludes by warning that "Trauma therapies that focus on outside, recall of events, and abreaction are insufficient to bring about lasting change" (p.282), which makes for an interesting comparison with current National Institute for Clinical Excellence Guidelines (2005) (see below).
More recently, Caizzi (2012) provides an unsettling insight into the effects of torture, where "the power of the [injunctions and counterinjunctions] a victim receives during torture is so deep that it impacts and changes his or her script protocol" (p.167), which later manifests in various types of somatisation and intense Persecutor/Victim (Karpman 1968) transferential enactments.As this extremity of trauma is "embodied in the subsymbolic mode, by which the patient communicates to the therapist what happened in the there and then through what is happening in the here and now" (Caizzi, 2012, p.169), like Stuthridge (2006), Caizzi's treatment considerations favour a relational TA approach, whereby client and therapist mutually cocreate (Summers & Tudor, 2000) and facilitate Adult integration of their shifting patterns of transference and countertransference, but with a body-orientated focus on supporting the client to identify, understand and, thereby, exert neocortical control over their "affective, somatic, sensory, and motor modes of mental processing" (Caizzi, 2012, p.168).Picking up on her earlier suggestion (Stuthridge, 2006) of a link between childhood trauma and adult-onset PTSD, Stuthridge (2012) comments on her work with earthquake survivors that "in every case in which posttraumatic stress symptoms escalated over time rather than subsiding, the client revealed a history of childhood trauma" (p.239).In the ensuing material, the innovative blend of neuroscience and relational TA first advanced in Stuthridge (2006) is reframed within the overarching metaphor of the Christchurch earthquake, such that "Bridging the [dissociative and cotransferential] fault lines in the interpersonal realm increases continuity within the self and expands [neopsychic] possibilities for affect regulation and intimacy" (p.239).
Non TA perspectives include the work of Davies and Frawley (1994), who offer a concise psychoanalytic treatment plan involving "1.containment 2. recovery and disclosure of traumatic memories . . .3. symbolization and encoding of memory and experience; 4. integration of disparate self and object systems . . .and 5. internalization of a new object relationship" (p.202), which, although intended for adult survivors of childhood sexual abuse, bears a marked resemblance to both Pomeroy (1998) and Stuthridge's (2006Stuthridge's ( , 2012) ) TA models.Perhaps the most familiar account of the treatment of PTSD and other trauma-focused disorders, however, is found in Rothschild (2000), who, citing Van Der Kolk (2004), advocates that "Therapy needs to consist of helping people to stay in their bodies and to understand these bodily sensations" (p.3); an attention to the somatic dimension of trauma that she saw as undervalued in contemporary psychotherapy.As described by Caizzi (2012), Rothschild sees the goal of treatment as supporting clients to "transform negative implicit memories of relationships by creating a newly encoded positive experience of attachment" (p.172), or, in object-relational terms, internalizing "a new representation of a caring relationship in both mind and body" (p.172).
Moving on to the research literature, Schnurr and Friedman (1997) furnish a brief overview of contemporary PTSD research; most notably, that PTSD is '"associated with . . .early conduct problems, childhood adversity . . .poor social support after a trauma" (p.13) and that researchers must be aware of ethnocultural limitations of their conclusions, as "the diagnostic criteria for PTSD and most of the current data .At present, no solid evidence exists for the efficacy of humanistic therapies with PTSD.In terms of UK health policy, National Institute for Clinical Excellence Guideline CG26 (2005) states that "there is as yet no convincing evidence for a clinically important effect of [other therapies] on PTSD" (p.19) and instead, recommends "a course of trauma-focused psychological treatment (trauma-focused cognitive behavioural therapy [CBT] or eye movement desensitisation and reprocessing [EMDR])" (, 2005, p.4).As Harford (2013) highlighted, there remains "a strong case for building the evidence base for TA psychotherapy in the treatment of PTSD and, thereby, influencing health policy and strategy within the NHS and at local and national government levels" (p.27).
A meta-analysis conducted by Benish, Imel and Wampold (2008) found no evidence of superiority of one therapy over any other and found no significant difference in efficacy between therapies included in their meta-analysis.They concluded that "bona fide psychotherapies produce equivalent benefits for patients with PTSD." (p.746).In light of this, it would seem reasonable to expect that TA therapy would also produce broadly comparable effectiveness in the treatment of PTSD.
Despite its chronicity, recovery rates amongst people receiving psychotherapy for PTSD are promising.A multidimensional meta-analysis of 26 studies on efficacy of psychotherapy (usually cognitive-behavioural therapy variants or EMDR, as discussed above) for PTSD conducted by Bradley, Greene, Russ, Dutra and Westen (2005) found that 67% of clients with PTSD who complete treatment will not meet diagnostic criteria at termination of therapy and 54% of clients would be considered 'improved' (indicating Reliable Change, but not Clinically Significant Change).These figures offer a useful benchmark for comparing the effectiveness of the TA delivered in this study with existing empirically supported therapies.Notwithstanding these encouraging measures, of the studies examined by Bradley et al. (2005), veterans with combat-related trauma tended not to fare so well from therapy and typically demonstrated a slightly lower recovery rate.This might be accounted for by the distinctive character and high prevalence of other harmful influences on veterans' mental health, not least of which are high rates of alcohol and substance addiction, perhaps exacerbated by a heavy drinking Cultural Parent (Drego, 1983) influence within the armed forces, personality and attachment problems arising from difficult childhoods prior to joining up, homelessness following discharge, relationship breakdowns whilst on duty and a lack of suitable treatment, recognition and reparation from the authorities in the immediate aftermath of their traumatic experiences.

Method Participants
The participants were 15 white, British, male armed forces veterans between the ages of 32 and 64 years, all of whom had been medically diagnosed with PTSD by a physician or psychiatrist prior to the commencement of this study in accordance with DSM IV (American Psychiatric Association, 1994), or ICD 10 (World Health Organisation, 1992) criteria, and had endured a variety of traumatic experiences while serving in the British armed forces.
All were living within a dedicated residential care facility with 24-hour staffing provision operated by an independent charity responsible for their accommodation, social care and day-to-day needs.This intensive support was necessary due to the severity of the veterans' PTSD symptoms, which, on account of the destructive impact on their relationships, livelihoods and ability to function safely in society, had rendered them homeless.Consequently, all the therapy examined by this study was conducted in a dedicated room within this same residential facility.8 of the 15 veterans had recent histories of substance or alcohol addiction, which were managed by qualified professionals from other specialist agencies and monitored by the charity's key workers.Furthermore, all 15 men in the sample had additional medically diagnosed comorbid conditions (see Tables 2  & 3).This is perhaps unsurprising, given that they were all living in supported accommodation and exhibited severe functional impairment.
It is also worth noting that comorbidity, particularly with Axis One disorders (American Psychiatric Association, 1994) such as depression, anxiety and substance abuse, tends to be highly prevalent amongst individuals with PTSD, with up to 83% of individuals having at least one other comorbid condition (Breslau, Davis, Andreski & Peterson (1991).The presence of severe comorbidity needs to be taken into account when considering therapeutic outcomes with this sample.Individual veterans were referred for psychotherapy based on non-clinical assessments of need by the operations manager and three key workers employed by the charity.Following the successful pilot project with an initial cohort of 6 male veterans (Harford, 2013), the charity decided to extend the TA psychotherapy provision indefinitely and, in addition to the pilot cohort, a further 9 male veterans agreed to participate in this fullscale study.In accordance with current ethical guidelines, all 15 participants were aware that their questionnaire responses would be collected and used for both fundraising and research purposes and gave their full written consent for this process.Additional written consent was received from 8 of the 15 veterans to undergo the Change Interview (Elliott, Slatick, & Urman, 2001, as cited in Frommer & Rennie, 2001), having been given "clear and explicit" (Institute of Transactional Analysis, 2008, p.5) assurances that any information that might identify them would be omitted or anonymised in order to maintain complete confidentiality.All research was conducted using a naturalistic protocol within the same designated therapy room provided by the charity within their premises.
As can be seen in Table 2, 2 veterans terminated treatment involuntarily at the behest of the charity's management for reasons it is inappropriate to explain here.In addition, there were 8 veterans who received psychotherapy under the same arrangements and consented to provide quantitative data, but were excluded from this study: three did not meet criteria for a diagnosis of PTSD; one individual provided anomalous responses indicative of marked over-adaptation (Schiff & Schiff, 1971) to the therapist; four attended insufficient sessions for their responses to be deemed statistically valid.Another 5 veterans declined to participate in any aspect of this study from the outset of their treatment.

Therapy
As with the original pilot study (Harford 2013), the form of TA psychotherapy chosen was rooted primarily in the integrative TA model advanced by Erskine (1993) and Erskine and Trautmann (1996), with an emphasis on facilitating intrapsychic contact with dissociatively "encapsulated traumatic experiences, hidden needs" (Erskine, 1993, p.185) and repressed affect, providing an attuned response to that affect and strengthening veterans' Integrating Adult (Tudor, 2003) capacity "to reflect upon and integrate their own archaic states as well as past introjects, and . . .draw on them in the service of present-centred relating" (p.202).However, the authors were mindful of Cornell and Bonds-White's (2001) critique of the limitations of an over-reliance on the integrative model, with its attendant risk of promoting "a temporary, mutually gratifying narcissistic merger" (p.72) between therapist and client.Consequently, inspired by their support for Bollas' (1989) stance, which emphasises the need for "a balanced therapeutic process serving the dual functions of soothing and disturbing the client" (p.80), the integrative foundations of the treatment were augmented with a two-person (Stark, 2000) relational focus on the cotransferential (Summers and Tudor, 2000) domain, as pioneered by Hargaden and Sills (2002) and, later, applied to the treatment of trauma and PTSD by Stuthridge (2006Stuthridge ( , 2012) ) and Caizzi (2012).
Here, the therapist observes, participates in and makes available for mutual analysis the "transferential enactments [that] provide a crucial voice for implicit relational patterns and excluded [or dissociated] ego states" (Stuthridge, 2006, p.277) and, through an iterative "process of attunement, rupture, and repair" (Stuthridge, 2006, p.277), slowly challenges and rewrites the client's traumatic script (Stuthridge, 2006).In this way, aided by strict observance of ethical and professional boundaries, the therapist (Harford) in this study was able to provide "the 'safe container' in which the [veterans could] begin to integrate" (Hargaden & Sills, 2002, p.29) their fragmented selves, rediscover lost skills and resources and begin to build mutually beneficial interpersonal relationships and more rewarding and meaningful lives.
There was also a strong psychoeducational component to the chosen therapeutic approach; this, in part, reflecting Erskine (1993) and Erskine and Trautmann's (1996) attention to depathologising, or normalising clients' symptoms as a subset of therapeutic involvement, whereby Adult assurance is provided that "the client's experience [of PTSD] is a normal defensive reaction . . .that many people would have if they encountered similar" (Erskine & Trautmann, 1996, p.325) traumatic events.In addition, several veterans had commented during their initial assessments that, although prescribed breathing exercises in the past, they had been disinclined to practice them independently as the clinicians involved had not explained their underlying neurophysiological purpose.Responding to this overt request for Adult information, a combination of Pomeroy's (1998) summary of neocortical versus limbic system function and Stuthridge's (2006) account of implicit, explicit and autobiographical memory systems was used to underpin a range of simple breathing and grounding exercises with the intention of supporting veterans to cultivate a mindful capacity to "observe the contents of their awareness without judgement and without letting themselves get caught up in or identified with any particular content of awareness" (Safran & Muran, 2003, p.210) and gradually "returning executive control to the Adult ego state and the neocortex" (Pomeroy, 1998, p.338).

Data Collection
Quantitative A battery of measures was used to determine baseline severity of symptoms and their improvement over the course of therapy.These were the CORE-OM (Evans et al, 2000), PHQ-9 (Kroenke et al, 2001) and GAD-7 (Spitzer et al, 2006) questionnaires, which were administered every four sessions commencing with a pre-treatment ('Session 0') set of responses completed between the initial assessment and first full treatment session.The CORE-OM is a "34 item generic measure of psychological distress, which is pan-theoretical . . .pan-diagnostic" (Core ims, 2003) and, therefore, readily applied to any TA psychotherapy context as a means of assessing mental well-being before, during and after treatment.This tool is widely used across the NHS and, also, among independent counselling and psychotherapy providers to evaluate the impact and effectiveness of a variety of mental health interventions.Gathered concurrently were total scores from the nine point PHQ-9 Depression Severity and seven point GAD-7 Anxiety Severity indicators; again, both widely used in a wide range of NHS and private settings and the latter, in particular, having been trialled successfully for clients presenting with PTSD (Kroenke, Spitzer, Williams, Monahan & Löwe, 2007).As well as generating numerical indicators of change in global distress and functioning, depressive symptoms and levels of anxiety for individual veterans, a mean figure was calculated for each questionnaire type across all 15 veterans and, in addition, for the 7 veterans identified as short-term clients (those receiving up to 24 sessions of treatment) and the other 8 veterans composing the long-term cohort (those receiving up to 52 sessions of treatment).Due to the small sample size, it is not possible to conduct detailed statistical analysis, or present inferential statistics.Instead, simple descriptive statistics will be used; more specifically, thresholds for Reliable Change and Clinically Significant Change for CORE-OM, PHQ-9 and GAD-7 which correspond with the established precedents detailed in the Appendix.

Qualitative
Of the total sample of 15 veterans, 8 individuals agreed to undertake a recorded semi-structured interview, separate to their ongoing treatment and conducted by the therapist at the same location.This consisted of a simplified version of the Change Interview (Elliott et al, 2001, as cited in Frommer & Rennie, 2001) lasting no longer than one hour.Compared with purely quantitative measures, the Change Interview "can be more sensitive to negative or unexpected effects while also allowing [researchers] to understand . . .[clients'] experience of therapy [via] questions about changes over therapy, and also questions about the processes that may have brought about change" (Elliott, Watson, Goldman & Greenberg, 2003).Of the 8 veterans who agreed to be interviewed, only one individual originated from the shortterm (24 sessions) cohort, whereas 7 of the 8 veterans from the long-term (52 sessions) cohort gave their consent for the process.Once the therapist (Harford) had transcribed the 8 digital recordings and carefully anonymised, or omitted any identifying details that might compromise client confidentiality, the second author (Widdowson) examined the qualitative data using thematic analysis (Braun & Clarke, 2006).This is an inductive approach to data analysis whereby themes are extracted from the collated transcript material and is similar to grounded theory in that the data is not distorted to fit pre-existing hypotheses, or theories.Transcripts of the Change Interviews were read several times to immerse the researcher in the available data.The transcripts were then annotated with identifying codes which captured discrete concepts, or meanings evident in the participants' replies to the interviewer's questions.Once labelled, these codings were subjected to a process of constant comparison, whereby they were grouped repeatedly according to similarity so as to generate a number of salient themes.As this iterative process unfolded, progressively higher order themes were generated until all data had been accounted for, no further themes emerged and the data analysis was considered complete.

Data Auditing
In an effort to address the potential for bias and deficits in experimental validity highlighted by Wampold et al (2010), including those unintended distortions arising when "therapists have an allegiance to . . .treatments or are . . .supervised by researchers who have an allegiance" (p.930), and although the authors are both qualified practitioners in TA psychotherapy, only one (Harford) was involved in the clinical treatment, collection of questionnaires and interviewing of the veterans, while the other's (Widdowson) role was limited to carrying out the qualitative data analysis and co-authoring this paper.In the interests of enhancing the credibility and trustworthiness of the findings even further, the thematic analysis (Braun & Clarke, 2006) used to examine the qualitative data was checked and audited by Professor Caroline Hollins Martin from the University of Salford, who was selected on account of having neither personal, or professional allegiances to the topic or outcomes of this research.With respect to Wampold et al's (2010) latter concern, all clinical supervision for the therapist was obtained from an unconnected third practitioner.

CORE-OM:
Pre-treatment levels of symptoms among all 8 veterans undergoing longer-term treatment were in the clinical range of severity, with two being classified as 'severe', three as 'moderate to severe', two as 'moderate' and one as 'mild' under the original scoring criteria (CORE ims, 2003).However, this rather mixed starting point is potentially misleading, as two of the three veterans beginning treatment with 'mild', or 'moderate' symptoms deteriorated to the 'severe' level within 8 sessions.As in the earlier pilot study (Harford, 2013), these results appear to accord with the researchers' prior expectations that some emotional "turbulence [would] occur as part of the change process" (Widdowson, 2010, p.203); more specifically, that veterans would feel somewhat worse before they began to improve as they made internal contact (Erskine, 1993) with previously repressed affect, leading to a temporary intensification of their distress.(Harford, 2013, p.28)      shows a dramatic improvement and attains Clinically Significant Change with a reduction of 18.9 to a level well below the clinical cut-off.Even those veterans whose pre-treatment and final scores are ostensibly less encouraging display marked improvements after reaching their peak scores, a pattern which might be accounted for by the gradual "dissolution of fixated contact-interrupting [dissociative] defenses" (Erskine, 1993, p.190) and reintegration of repressed affect alluded to in Harford (2013) above.From this latter perspective, the results for all but one veteran (H) could be interpreted as evidencing Reliable Change, with Client C remaining the sole example of Clinically Significant Change.Generally speaking, then, it appears that the veterans' overall mental well-being has improved significantly to a level beyond that which could be accounted for by nonclinical factors (e.g.measurement error), but only in one case to the extent of a nonclinical score comparable with the so-called "normal" population.
PHQ-9: Pre-treatment symptoms of depression for the long-term veteran cohort were in the clinical range of severity in all but one case, with three veterans meeting the scoring criteria for 'major depression (severe)' and another four sitting within the 'major depression Although these results are less convincing in their support for the authors' hypothesis that 'TA psychotherapy can be an effective treatment for PTSD in ex-servicemen and women', the shortfall is minimal and it is worth considering that consistent Reliable Change, or even Clinically Significant Change may yet be attained by further veterans, as their course of treatment in five cases has continued beyond 52 sessions.PHQ-9: Pre-treatment symptoms of depression among the 7 veterans in the short-term cohort were all in the clinical range of severity, with four veterans meriting a 'major depression (severe)' designation, one falling within the 'major depression (moderate)' category and the other three exhibiting 'dysthymia, minor depression, major depression (mild)', as defined by Kroenke et al (2001).However, as with the clinical CORE-OM results for the long-term cohort, the individual veterans' scores deserve closer inspection, as Client I commences treatment in the 'dysthymia, minor depression, major depression (mild)' band, but deteriorates to the 'major depression (severe)' level within 4 sessions.Though less prominent than that present in the long-term CORE-OM scores, this slight variance in the short-term PHQ-9 results would appear to offer a modicum of additional support for Harford's prediction " . . .that veterans would feel somewhat worse before they began to improve as they made internal contact (Erskine, 1993) with previously repressed affect, leading to a temporary intensification of their distress."(Harford, 2013, p.28)After 24 sessions, the mean PHQ-9 score falls just short of Reliable Change with a reduction of 5.0, rather than the necessary 6.0 required to surpass the reference threshold.This appears to be the consequence of an increase in depressive symptoms from Session 20 onwards, which coincides with a slowing of respective recovery rates for the short-term mean clinical CORE-OM and GAD-7 measurements.In parallel with the PHQ-9 figures for the long-term veteran cohort, these short-term results provide somewhat less support for the authors' research hypothesis, but, again, the deficit is relatively small and it is possible that Reliable Change, or even Clinically Significant Change may have been attained by more veterans had their course of treatment continued up to 24 sessions and beyond.

GAD-7:
Pre-treatment symptoms of anxiety for the shortterm cohort were all in the clinical range of severity, with five veterans occupying the 'severe' category, one at 'moderate' level and one evidencing 'mild' symptoms based on Spitzer et al's (2006)   CORE-OM: After 52 sessions, the mean clinical CORE-OM scores for all 15 veterans demonstrate Reliable Change with an improvement of 8.0 over the course of treatment, though with several setbacks between Sessions 20 and 36 -an intriguing phenomenon echoed in the mean clinical CORE-OM scores for the 8 veterans receiving long-term treatment.Overall, compared with the results for the long-term and short-term veteran cohorts, the combined figures for overall mental wellbeing also signify a marked improvement beyond that which could be accounted for by nonclinical factors, and one broadly consistent with both the rate and extent of positive change for the 24 session and 52 session groups.
PHQ-9: Unlike the results for the short-term and longterm cohorts, the mean PHQ-9 scores for all 15 veterans after 52 sessions meet the criteria for Reliable Change convincingly with a reduction of 7.0.Furthermore, the reduction in depressive symptoms is, again, reasonably consistent and features the same uneven phase between Sessions 20 and 36 found in both the mean PHQ-9 scores for the long-term cohort and the clinical CORE-OM scores mentioned above.These temporary anomalies may also help to explain the apparent deterioration of depressive symptoms from Session 20 onwards in the PHQ-9 results for the short-term veteran cohort and, consequently, the authors were inclined to speculate whether some of the negative extratherapeutic influences outlined in the various Change Interview (Elliott et al, 2001, as cited in Frommer & Rennie, 2001) responses might have contributed to this episodic variance.

GAD-7:
As with the encouraging amelioration of symptoms indicated by the clinical CORE-OM and PHQ-9 results above, the mean GAD-7 scores for all 15 veterans across 52 sessions reduced by 6.4, which more than satisfies the chosen conditions for Reliable Change.
Once again, there is an unsettled period of renewed anxiety between Sessions 20 and 36, which could be accounted for by extra-therapeutic factors, such as the destabilising impact of benefits assessments, conflict between staff and service users within the residential care facility and other detrimental influences alluded to in the following qualitative data.

Qualitative Results
Qualitative Results: Client Changes, Outcomes of TA Psychotherapy As an integral part of their Change Interviews (Elliott et al, 2001, as cited in Frommer & Rennie, 2001), the 8 participating veterans were asked to describe the predominant intrapsychic and interpersonal changes they had noticed since commencing TA psychotherapy.
The collated transcripts revealed considerable depth, range and specificity to the positive changes the veterans had experienced and, having been subjected to thematic analysis (Braun & Clarke, 2006), were grouped into conceptual categories by the second author (Widdowson).As depicted in Table 13, three of these conceptual categories referred to 'Interpersonal Changes', though, within this overarching classification, discrete clusters of subsidiary changes were also identified; all of which were sufficiently distinct to warrant separate sub-categorisation and consideration on their own merits.
Note: numbers after the items relate to the number of veterans who specified that particular change

Interpersonal Changes: Increased Assertiveness
Assertiveness and willingness to challenge others appropriately ( 4) Asking for what I want and asking for help   Braun & Clarke, 2006) .These findings in relation to the structural characteristics of therapy provision are likely to be particularly useful to practitioners who may be planning on developing such services themselves and might serve as a guide to the preferred format for delivering treatment.
1.1.3.Theme: Pacing and Intensity: All 8 veterans mentioned that the pacing and intensity of the therapy were crucial to their engagement, their sense of safety and, thereby, the likelihood of beneficial outcomes emerging from the process.This appears to underline Rothschild's (2000) cardinal advice that a competent and safe therapist should "know where the brakes are, and how to use them, before one applies the accelerator" (p.79).One key sub-theme was the therapist's ability to regulate the client's affect by, in turn, heightening and tempering the emotional intensity of the therapy, which, as referred to earlier, accords with Bollas' (1989) endorsement of "a balanced therapeutic process serving the dual functions of soothing and disturbing the client" (Cornell & Bonds-White, 2001, p.80).A further subtheme alluded to the therapist's attention to carefully pacing and regulating the veterans' disclosures.As evidenced below, this cautious approach was particularly welcome when applied to the disclosure of traumatic events witnessed, or experienced under combat conditions and the resultant obligation of the veterans to conceal official military secrets: This extract also highlights the need for appropriate pacing and intensity of therapeutic interventions so as to "minimize the potential for reinforcing [trauma] through iatrogenic shaming" (Widdowson, 2010, p.240) and, further, the need to create an individually tailored therapy for each client.

Theme: Environmental Factors:
Several veterans commented on extra-therapeutic factors deriving from their living environment which exerted a negative impact on their mental well-being and, consequently, inhibited their progress in treatment.In relation to their fellow veterans and, more specifically, the effects of institutionalisation, four of the men described how conflict between residents could have a destabilising effect on everyone living in the unit.In addition, despite generally positive comments about the residential care facility, several of the men expressed major concerns relating to the standard of care provided within the unit.A significant problem raised by three veterans was the regular occurrence of breaches in confidentiality, whilst the second most common issue (voiced by two veterans) was the perception that some of the charity's staff needed more training on mental health awareness and best practice.
Client The worrying picture emerging from the qualitative analysis corresponds markedly with fluctuations in the numerical data generated by the CORE-OM (Evans et al, 2000), PHQ-9 (Kroenke et al, 2001) and GAD-7 (Spitzer et al, 2006) questionnaires.Among those participants in this study who have encountered problems with their benefits claims, or have been required to be reassessed for proof of continued eligibility, all experienced a pronounced deterioration in their mental well-being during the process, together with an attendant increase in their PTSD symptoms.
A wider political and policy debate regarding the appropriateness of these benefits assessments and the manner in which they are conducted is beyond the scope of this paper.However, there is plentiful evidence in both the quantitative data (which shows 'spikes' of symptom deterioration that coincide with the timing of involvement with the benefits agencies) and the qualitative interviews conducted for this study, to lead the authors to believe that such assessments and their associated correspondence are significantly detrimental to the mental welfare of this client group.

Superordinate Theme 2: Therapist Factors:
Positive influences originating from the therapist were cited by all 8 veterans as a second superordinate factor contributing to their beneficial experience of TA psychotherapy.Following thematic analysis (Braun & Clarke, 2006), these therapist factors were categorised into the following themes:

Theme: Therapist Relational Qualities
Providing further reinforcement to established evidence (Luborsky, Singer & Luborsky, 1975;Gaston, 1990;Assay & Lambert, 1999;Paley & Lawton, 2001) that the quality of the therapeutic relationship is the paramount curative factor in all effective psychotherapy, the following excerpts were typical: Client G6: "You've also got that personality as well, where, there's sometimes, I don't want to go there, but I will, because I know there's that kind of two-way process.I think you would stop me if I went down some bad avenue . . .There's that kind of trust element".
Client B7: "But you're understanding, just even listening and then tweaking and helping me try and make sense of what I'm going through".
Client B10: "Comfortable . . .As soon as I walk in the room and sit down with you, I feel comfortable . . .I can sense when somebody's agitated.I can sense when somebody's peaceful", "I always feel comfortable with you and talking to you . . .that I've not experienced with anybody else in my whole life before".
In particular, reflecting guidance from the integrative school of TA that genuinely empathic "inquiry begins with the assumption that the therapist knows nothing about the client's experience and therefore must continually strive to understand the subjective meaning of the client's behavior and intrapsychic process" (Erskine & Trautmann, 1996, p.318), the following excerpt suggests that the therapist was deeply attentive to the client and sought to respect their own wisdom, whilst also maintaining a process-directive stance: Client A12: "You would actually sit there and listen", "I was the teacher and you were learning from me", "It was being understood and listened to . . .and acting on it and telling me what I could do to help that situation . . .and I acted on it.So, I was taking that on board".
The healing function of therapeutic dialogue in "helping the individual regain control over dissociative processes . . .accessing both the limbic system and the neocortex to repattern the traumatic experience" (Pomeroy, 1998, p.338), including the deconstruction and reconstruction of perspectives and stories, so as to "develop the Adult capacity for reflective function and self-narrative" (Stuthridge, 2006, p. 281), was highlighted by one of the veterans as follows: Client E1: "I've felt it's been extremely helpful in my present case . . .I could come and speak and we'd put different perspectives on things, different words, different ideas, because I would think something and just put my own thought to it and it would probably be a blinkered thought and that one thought only".

Theme: Individually Tailored Therapy
Another important therapist factor which emerged from the qualitative analysis was that the veterans recognised and valued that the therapy was individually tailored to their unique needs and preferences.
Client G5: "I had a bit of a knowledge from psychology and that in the background.So, I think it was . . .[my] interest in the transactional [analysis theory] as well, which made it a bit easier to sit and chat with you".
Client G8: "The . . .personalisation, that you're not a number, or anything like that".

Theme: Techniques
A number of therapeutic techniques were alluded to in the transcripts, although many such references consisted merely of vague descriptions, rather than precise attributions of efficacy.For instance, the following extract highlights the benefits of reflective journal writing: From a TA perspective, the reflective journal writing and mindfulness techniques were integrated into the therapy to promote both the recovery of Adult functioning and, consequently, the "capacity to reflect upon and integrate . . .archaic [or dissociated] states as well as past introjects" (Tudor, 2003, p.202) and, also, to assist veterans with self-regulation of their "affective, somatic, sensory, and motor modes of mental processing" (Caizzi, 2012, p.168).

Theme: Catharsis
All 8 veterans described their psychotherapy treatment as a cathartic experience, which they characterised as a painful, but ultimately necessary phase in their healing.
In addition, recalling Schnurr and Friedman's (1997) observation that PTSD is '"associated with . . .early conduct problems, childhood adversity . . .poor social support after a trauma" (p.13) and, similarly, Stuthridge's (2006) remark that "disorganized attachment in children predicts dissociative symptoms in adults" (p.274), veterans often mentioned the resolution of fixated trauma deriving from their early lives as essential to their recovery.Subsidiary themes of "letting go" and forgiveness also featured in the transcripts, as the veterans recalled a process of coming to terms with painful memories held in their earliest Child ego states, as well as subsequent military extremities of experience.1), the veterans' sense that they were directing the course of therapy was a notable theme and extended to the application of specific therapeutic techniques (e.g.cognitive behavioural problem-solving strategies).These strategies were offered to veterans in the form of Adultenhancing options and then, modelling "collaborative agreement about tasks and goals in therapy" (Wampold at al, 2010, p.931), the final choice over whether to pursue them was left entirely to each individual.The following excerpts describe just such a positive Nurturing Parent experience, where the client's autonomy (Berne, 1964) was respected: Client C7: "When I have got a problem, you've . . .reiterated the procedure that I could use to resolve the issue . . .but you've left the decision to me.You've not forced me to do it.You've left the decision for me to do, and the fact that you were willing . . .you did that meant that I can accept it more . . .I've learnt from that, and I use the same procedure with my kids . . .I will advise them and then I will be there for them when they make the wrong decision, or if they make a decision and it goes wrong, I'm still there . . .I've not forced them to act".Client G16: "I know I've got the kind of like safety word . . .I can hold my hand and say '[Therapist name], we're not going in this direction' . . .You will not push me in a direction I don't want to go, and that's the good thing".
In the following quotation, a veteran recounts another empowering experience of directing the agenda, but also acknowledges the need for any effective therapist to "challenge habitual assumptions and relationship patterns and create sufficient turbulence for new structures to emerge" (Holmes, 2001, p.17): Client E11: "I can bring anything to the forum, so to speak, to talk about, but you do question [challenge] me back, obviously, which is important . . . to help".

Theme: Sense of hope
For two veterans, there was a distinct sub-theme centred around developing and maintaining a positive vision for the future and, in particular, the hope of moving towards a meaningful and contactful (Erskine & Trautmann, 1996) return to civilian life.Client F18: "I'm extremely grateful and thankful for therapy . . .Hopefully, I can get back to a sense of normality within my life.Well, I feel like I'm getting there now, so I feel, if I can get this far in the short space of time since I've seen you, over the next while . . .I can . . .even get more better, so I've got good expectations".
Client A26: "[On returning to independent living in the community] It's no use going back if I'm not ready, I could . . .end up back here . . .And now I'm looking forward to it, which . . .I wouldn't even have done this, going back thing on its own without the therapy . . .I thought I was here for life.I thought, 'This is it' ".

Theme: Psychoeducation
Various didactic aspects of the therapy were judged beneficial by the veterans interviewed in this sample.Of particular value were those psychoeducational interventions which assisted veterans to understand the intrapsychic process and psychobiological symptoms of PTSD, which justifies Pomeroy's (1998) emphasis on "education about the common symptoms of traumatic stress reactions so the person is assured, with regard to his or her safety, that he or she is not going crazy" (p.338).As demonstrated below, mention was also made of the CORE-OM (Evans et al, 2000), PHQ-9 (Kroenke et al, 2001) and GAD-7 (Spitzer et al, 2006) questionnaires as being intrinsically helpful in strengthening the veterans' capacity to conceptualise their fluid and, often, intense experiences: Client E3: "You've explained to me the processes that are taking place in the body, without me knowing whether . . .what 's wrong with me . . . and it's [his brain, his thinking] all been, like, defragmented like a computer, if you know what I mean, and things slowly, but surely are starting to probably find their place again.That is a big factor, yeah, making sense of it [PTSD symptoms]".Client E12: "I think my [questionnaire] scores have shown me I've . . .improved dramatically, then there's been a spike, and . . .I bet you they probably tie in to the dates when I've received really grotty [official] letters . . .They reflect directly, I think, the dates of . . .letters".

Theme: Value of TA theory
Five veterans made explicit reference to the fact that learning a number of TA models had played an important role in helping them to make sense of their phenomenological experience.This positive reaction to the teaching of relevant theory would appear to support established psychoanalytic and TA treatment plans which emphasise "symbolization and encoding of memory and experience" (Davies & Frawley, 1994, p.202) as an essential phase of "returning executive control to the Adult ego state and the neocortex" (Pomeroy, 1998, p.338)

Theme: Normalisation
The positive impact of normalisation was mentioned by all 8 veterans as an essential component of their treatment by way of facilitating a "change [in] the way [they] . . .categorize or define their internal experience or their behavioural attempts at coping from a pathological . . .perspective to one that respects archaic attempts at resolution of conflicts" (Erskine & Trautmann, 1996, p.325).Sub-themes were also evident in relation to helping the veterans make sense of their subjective experiences and "create a coherent self-narrative" (Stuthridge, 2006. p.282), as can be seen in the following excerpts: Client F4: "Your reassurance . . . the way you analyse things and feed it back to me, you know.The way I perceive it [his anxiety and PTSD symptoms] then is, you know, sounds normal . . .which I never got before on a professional level".
Client B7: "But you're understanding, just even listening and then tweaking and helping me try and make sense of what I'm going through".

Other Positive Outcomes
Changes in psychiatric medication: Almost all of the veterans undergoing the Change Interview (Elliott et al, 2001, as cited in Frommer & Rennie, 2001) reported significant alterations to their prescriptions for psychiatric medication during their psychotherapy treatment; five veterans had their dosages of various antidepressants, anxiolytics and sedatives reduced, or terminated altogether.
Table 14 provides a convenient summary of these changes, many of which might be attributed, at least in part, to the positive influence of TA psychotherapy.Transition to independent living: Out of the total sample of 15 veterans who participated in this research, 5 individuals have recovered sufficiently from their PTSD, other comorbid disorders and associated lifestyle problems to move out of 24-hour residential care and resume living independently in the local community.Veterans undergoing both long-term (52 session) and short-term (24 session) psychotherapy treatment achieved this notable milestone, with 2 individuals from the long-term cohort (Clients A and B) and 3 from the short-term cohort (Clients L, M and O) securing their own tenancy by the time the final quantitative scores were collected and all interviews completed.

Data Limitations
General: Notwithstanding the precautions taken by the authors to allay concerns raised by Wampold et al (2010) around the experimental validity of clinical research in the field, there remain a number of potential sources of bias and distortion that must be taken into account in any interpretation of the results of this study.Significant among these is that both authors "clearly have an allegiance to a particular treatment" (Wampold et al, 2010, p.930) and might be considered to have a vested interest in finding supporting evidence for their working hypothesis.There was also a notable "Duality of Roles" (Institute of Transactional Analysis, 2008, pp.10, 12) for one of the authors (Harford), as the practitioner delivering TA psychotherapy to the chosen sample of veterans and the researcher collecting their quantitative and qualitative responses.The potential for bias inherent in practitioner research was to some extent mitigated by the second author, who audited the quantitative results and conducted the qualitative data analysis.
The selection criteria for the veterans involved in the study are also open to question in that, although all participants had received prior formal diagnoses of PTSD from qualified medical professionals with reference to the DSM IV (American Psychiatric Association, 1994), the choice of which veterans were referred for psychotherapy was left to the management of the independent charity so there may have been other, nonclinical motives for the choices made, such as the organisation's own policies, priorities and concerns.
Returning to relevant TA theory, of particular note in the context of this research is Stuthridge's (2006) observation that "disorganized attachment in children predicts dissociative symptoms in adults" (Stuthridge, 2006, p.274), which led the authors to question to what extent the combat-related PTSD exhibited by veterans participating in this study, several of whom originated in highly dysfunctional families, had been exacerbated by insecure and, sometimes, abusive attachments during childhood.Perhaps, therefore, these individuals were exhibiting what Schnurr and Friedman (1997)   much extant research into the effectiveness of psychotherapy in treating PTSD, there is also the fact that all 15 veterans in this study were male and, mindful of Shadbolt's (2004) comments on the curative necessity of the twinship transference (Kohut, 1984) and sameness experiences in the therapeutic relationship, it is interesting to speculate how the results might have differed had there been female veterans available for study.
Quantitative Data: Moving on to possible weaknesses within the quantitative data, it must be noted that not all the veterans in the short-term and long-term cohorts completed their respective 24 and 52 sessions and, therefore, the individual CORE-OM (Evans et al, 2000), PHQ-9 (Kroenke et al, 2001) and GAD-7 (Spitzer et al, 2006) data sets are necessarily incomplete and the mean figures derived from them subject to proportionate statistical error.This issue is less marked for the longterm cohort, where 6 of the 8 veterans completed 52 sessions with only Client F terminating after 34 sessions and Client H declining to complete further questionnaires after 28 sessions and terminating altogether after 50 sessions.More problematic in this regard are the shortterm group in which Client O received only 11 sessions, Clients K and L received just 12 sessions, while Client I underwent 22 sessions.The explanations for these early terminations also contain some variation, with both Client F and H's endings being involuntary and imposed by the charity's management, Clients J and O leaving voluntarily, but without fulfilling their contractual agreement to undertake one final session following the decision to terminate, and Clients I, K and L finishing voluntarily and in accordance with their agreed contracts.
In an effort to mitigate the impact of these inconsistencies on experimental validity, mean CORE-OM (Evans et al, 2000), PHQ-9 (Kroenke et al, 2001) and GAD-7 (Spitzer et al, 2006) figures for all 15 veterans have been provided by the authors for analysis alongside those originating from the distinct long-term and short-term cohorts.A comparison of all three data sets shows them to be relatively consistent and provides a degree of reassurance that these variations in the numbers of sessions received have not unduly distorted the encouraging patterns and trends evident within the quantitative results.
Arguably more serious in terms of potentially compromising the experimental validity of this data is the influence of external psychosocial factors on the mental well-being of the participants at various stages of their treatment.Foremost among these are the impact of eligibility assessments for state benefits and the effects of interpersonal conflict within the resident veteran population and between participating veterans and the charity staff, both of which figure prominently in the collated Change Interview responses.The causal role and degree of influence of these psychosocial factors in exacerbating the veterans' symptoms and hampering their treatment were not measured in this study and, as discussed further below, could potentially be a source of anomalies in the quantitative data obtained.
Qualitative Data: Despite addressing the same themes and being governed by the same protocol, the Change Interview (Elliott et al, 2001, as cited in Frommer & Rennie, 2001) employed in this study is, by its very nature as a semi-structured interview, subject to some variance from one respondent to the next.Although the interview was limited to one hour in duration and conducted in accordance with the numbered order of topics with all 8 veterans in the sample, the exact wording, tone and emphasis of the researcher's questions, together with any intermittent comments and requests for clarification, introduce a degree of singularity to the process.The qualitative data obtained, therefore, may be subject to bias and distortion by the practitioner researcher, who, consciously, or otherwise, may have asked questions, or commented in a leading way, perhaps discounting (Mellor & Schiff, 1975), or, in turn, over emphasizing certain aspects of the clients' answers, so as to build a strong supporting case for the working hypothesis.That said, such biases are common to most forms of qualitative research and the thematic analysis (Braun & Clarke, 2006) utilised in this study is a wellestablished technique with a proven track record of support and critique.Furthermore, as the veterans sampled were by definition traumatised, extremely vulnerable and struggled with issues of trust in relationships, it was considered inappropriate for an outsider to be brought in for the purposes of interviewing, as this would likely cause undue distress and, perhaps, countertherapeutic iatrogenic shaming (Widdowson, 2010).Taken as a whole, then, these risks to the integrity of the results were partially mitigated by the presence of a second author conducting the qualitative data analysis and the additional safeguard of an unconnected and suitably qualified third party in Professor Caroline Hollins Martin (University of Salford) to audit the process of qualitative analysis.

Discussion
It is heartening to observe that Wampold et al's (2010) checklist of effective treatments for PTSD matches the psychotherapeutic methods adopted with the participants in this research extremely closely, with only "Exposure . . . in vivo outside of [the] session" (p.931) absent from the chosen TA approach.Even so, many of the veterans agreed inter-session contracts to conduct controlled exposure experiments in their own time, which involved practicing breathing exercises and mindfulness techniques while in the presence of triggering stimuli, such as crowd situations, travelling on busy public transport, or walking along narrow alleyways overlooked by high windows.It is perhaps unsurprising, then, that, viewed as a whole, the data gathered in this extensive study points decisively to considerable improvements in the psychological well-being of the veterans receiving TA psychotherapy in both the short-term (up to 24 sessions) and longer-term (up to 52 sessions) formats and, there-fore, offers solid confirmation of Harford's (2013) pilot findings that both anxiety and, to a lesser extent, depression appear to gradually reduce as, within the nonintrusive safety of an empathic therapeutic relationship, the veterans re-experience previously repressed affect, obtain the longed, for attuned response to their pain and then slowly build their Integrating Adult to "reflect upon and integrate archaic states and introjects, and draw on them for present, centred relating" (Tudor, 2003, p.202).
Concentrating initially on the quantitative outcomes and measuring from peak scores rather than from pretreatment levels, 2 of the 15 veterans achieved Clinically Significant Change on their mean clinical CORE-OM (Evans et al, 2000) scores, 3 of the 15 attained Clinically Significant Change on their mean PHQ-9 (Kroenke et al, 2001) scores and 2 of the 15 finished with Clinically Significant Change on their GAD-7 (Spitzer et al, 2006) scores.By comparison with Bradley et al's (2005) benchmark of 67% from a meta-analysis on the efficacy of psychotherapy as a treatment for PTSD, this equates to an overall mean of 16% of veterans in this study no longer meeting the diagnostic criteria for PTSD by the end of their respective 24, or 52 sessions.These results appear disappointing, but can still be considered positive given the severity and chronicity of comorbidity, addiction and other lifestyle problems exhibited by the sample population, combined with the harmful effects of the various negative extra-therapeutic factors pinpointed by the qualitative data.As noted earlier, there is also the fact that 5 veterans in the long-term cohort went on to receive further treatment beyond 52 sessions and may well have attained Clinically Significant Change had it been possible to include this additional data in the current analysis.Furthermore, it is worth noting that the PHQ-9 (Kroenke et al, 2001) and GAD-7 (Spitzer et al, 2006) scales are not exclusive diagnostic indicators for PTSD, but, rather, a broad spectrum of Axis One disorders (American Psychiatric Association, 1994) and, consequently, the therapist's (Harford) judgement was that, based on the DSM IV (American Psychiatric Association, 1994) criteria alone, 8 of the 15 veterans sampled no longer met sufficient criteria for a diagnosis of PTSD at termination.This can be construed as a phenomenological form of 'clinically significant change' and presents a much better percentage score of 53% for consideration alongside Bradley et al's (2005) results.Less contentious is the fact that 14 of the 15 veterans achieved Reliable Change on their mean clinical CORE-OM (Evans et al, 2000) scores, 9 of the 15 attain Reliable Change on their mean PHQ-9 (Kroenke et al, 2001) scores and 10 of the 15 completed treatment with Reliable Change on their GAD-7 (Spitzer et al, 2006) scores.Employing the same comparison, this yields an overall mean of 73% of veterans in this study meriting a description of 'improved' (indicating Reliable Change), which compares very favourably with Bradley at al's (2005) benchmark of 54%.
The qualitative outcomes in Table 13 illustrate an impressive range, depth and specificity of beneficial outcomes reported by the veterans undergoing interviews as a direct consequence of their TA psychotherapy treatment.Echoing Korol's (1998) comments on the dual goals of treatment, these positive changes are evident in both the intrapsychic and interpersonal domains, with veterans enjoying new friendships and sexual relationships, improved interactions with authority figures, greater relational assertiveness and trust and an increased capacity for empathic contact and mentalisation (Bateman & Fonagy, 2006), as well as the internal advantages of reduced symptoms, a broader repertoire and deepened capacity for cognitive and affective self-regulation and a much greater understanding, acceptance and sense of normalisation (Erskine 1993;Erskine & Trautmann, 1996) in relation to the psychobiological dimensions of their phenomenological experience.Additionally, the authors have been very encouraged by a number of potentially far-reaching changes in the veterans' lifestyles, with 5 individuals achieving the transition to independent living in the community, 5 evidencing a reduction of dosage or a complete cessation of their prescriptions for psychoactive medication, several reducing their dependence on alcohol and recreational drugs and others commencing academic courses of learning, vocational training, or participating in voluntary work with a view to eventually securing employment.
Within this same qualitative data, a number of recurring themes emerged from the collated transcripts which, in the authors' opinion, deserve extensive research analysis of their own.The most prominent is the psychological impact on vulnerable adults with mental health issues of benefits assessments carried out by the UK welfare agencies, or various commercial concerns under the auspices of national government.Based on comparative analysis of veterans' qualitative responses with their quantitative data and the timing of several pronounced phases of deterioration in their symptoms, there appears to be a direct causal relationship between enforced attendance at benefits eligibility assessments and related health assessments, and between the arrival of related written correspondence and increased symptoms of anxiety and depression.Both forms of contact with officialdom appeared to be linked to an adverse influence on progress in therapy.
It is not possible to attribute these fluctuations directly to the impact of benefits and capacity work assessments based on the data presented in this study.The authors acknowledge that the results observed might be due to factors such as the veterans' perceptions of the assessment process, rather than the practical and relational reality of how they are conducted, and that the sample size of 15 represents a tiny fraction of the total number of individuals drawn from the client group undergoing such procedures across the UK.However, in recent years there has been extensive media criticism of the accuracy, care standards and levels of sensitivity employed during these assessments, along with the propriety of involving commercial interests in such delicate procedures with vulnerable people (Wintour, 2013;Fagg, 2012;Toynbee, 2014).Such adverse coverage has been mirrored by the UK Department for Work and Pensions' own Public Accounts Committee, which noted that "Poor decision-making causes claimants considerable distress . . .The Work Capability Assessment process has a disproportionate impact on the most vulnerable claimants.The standardised "tickbox" approach fails to adequately account for rare, variable or mental health conditions and this can lead to greater inaccuracies in decision-making for these particular claimant groups" (Department of Work and Pensions, 2012).This issue clearly warrants closer inspection; including analysis of how the present UK welfare system may not be in line with the guiding principles and codes of best practice underpinning both the Mental Health Act 2007 and Mental Health (Care and Treatment) (Scotland) Act 2003.
As noted in Harford's (2013) earlier pilot study, contrary to the practitioner-researcher's prior misgivings that the use of questionnaires with this extremely vulnerable client group "might present a countertherapeutic intrusion of bureaucracy into fragile therapeutic alliances" (p.28), their introduction has, in fact, proved an invaluable aspect of the veterans' treatment, as referred to explicitly in a number of interview responses.As well as providing a convenient rough measure of progress for the therapist and generating precious evidence of success for the charity's ongoing efforts to secure external funding for the psychotherapy programme, the sensitively "contracted use of [CORE-OM, PHQ-9 and GAD-7] questionnaires appears . . . to have gone some way to satisfying veterans' structure hunger (Berne, 1961) in the face of self-fragmentation, provided "a degree of emotional containment" (Widdowson, 2010, p.203) and afforded a way for these vulnerable clients to measure and conceptualise their labile phenomenological experience."(Harford, 2013, p.28).
In the light of these findings, the authors would invite all TA practitioners to consider incorporating these relatively straightforward statistical measures into their practice wherever possible, as there are clear benefits to be gained for all concerned in the process.

Conclusion
Over the two-year duration of this study, a compelling body of evidence has been gathered in support of the authors' initial hypothesis that 'TA psychotherapy can be an effective treatment for PTSD in ex-servicemen and women'.Indeed, yet more corroboration of the enduring efficacy of TA with this client group could still emerge, as, on account of the positive outcomes detailed in this paper, the charity concerned has committed to funding the psychotherapy programme for the veterans under their care and support indefinitely so further opportunities for research on related topics will be readily available.In particular, opportunities to explore the effectiveness of TA psychotherapy as a treatment for PTSD among female veterans, those from different social and cultural backgrounds and, also, further investigation of the predisposing influence of childhood trauma would be most welcome.

Figure 3 :
Figure 3: Table 6: Mean GAD-7 score: 8 long-term clients (moderate)' category, as defined byKroenke et al (2001).After 52 sessions, the mean PHQ-9 score falls short by a tiny margin of the chosen Reliable Change threshold with an improvement of 5.958333, rather than the required 6.0, assuming neither Session 0 nor Session 52 figures are adjusted to one decimal place.Again, Client C alone evidences Clinically Significant Change with a sharp reduction of depressive symptoms to a point well within a nonclinical classification, although not without pronounced spikes at Sessions 28 and 48, which, with reference to the Change Interview data summarised later, may reflect the negative impact of extra-therapeutic factors during these phases of treatment.Two veterans (E and F) show Reliable Change based onKroenke et al's (2001) criteria, while, if the previous point regarding measurement from peak scores onwards is taken into account, the results from a further two clients (A and B) could also be argued to demonstrate Reliable Change.

Figure 7 :
Figure 7: Mean Clinical CORE OM Score: all clients

Figure 10 :
Figure 10: Themes related to Safety . .are based primarily on research and clinical experience with North American or European individuals" (p.16).Meanwhile, after debating the relative bias and distortions within the available studies, Wampold at al (2010), supply a convenient summary table of "specific ingredients and . . .common factors" (p.931) found in all effective treatments for PTSD, as shown in Table 1.
Opportunity to talk about trauma (i.e., tell stories)Ensuring the patient's safety, especially if the patient has been victimized as in the case of domestic violence, neighbourhood violence, or abuse Helping patients learn how to avoid revictimization Identifying patient resources, strengths, survival skills and intra and interpersonal resources and building resilience With respect to clinical practice in the United States, four main approaches are indicated for the treatment of PTSD, of which three can be considered variants of Cognitive-Behavioural Therapy (CBT).At the 'classic'

Table 2 :
Characteristics of sample undertaking long term therapy

Table 3
Characteristics of sample undertaking short term therapy

Table 4 :
Mean Clinical Core-OM score: 8 long-term clients

Table 6 :
Mean GAD-7 score: 8 long-term clientsFigure 1: Table 4: Mean Clinical Core-OM score: 8 long-term clients Figure 2: Table 5: Mean PHQ-9 score: 8 long-term clients Spitzer et al's (2006) of anxiety for the longterm cohort were all in the clinical range of severity, with five veterans meriting a 'severe' classification, another two presenting at the 'moderate' level and one displaying 'mild' anxiety in accordance withSpitzer et al's (2006)scale.After 52 sessions, the mean GAD-7 score reduced by 5.8, which exceeds the chosen threshold for Reliable Change.Once again, Client C is the sole veteran GAD-7:CORE-OM: Pre-treatment symptom levels among all 7 veterans undergoing short-term treatment were in the clinical range of severity, with five being ranked as 'severe', one as 'moderate to severe' and one as 'moderate' under the established scoring criteria(CORE  ims, 2003).After 24 sessions, the mean clinical CORE-OM score demonstrates Reliable Change with an improvement of 7.8 over the course of treatment, though with a noticeable reduction in the rate of positive change from Session 12 onwards.Client L, in particular, exhibits a dramatic improvement and attains Clinically Significant Change with a reduction of 14.4 to a level far below the clinical cut-off.In summary, it appears that the veterans' overall mental well-being has been enhanced significantly over 24 sessions to levels beyond that which could be explained by nonclinical factors, but, again, only in one case to a degree comparable with the 'normal' population.

Table 7 :
Mean Clinical Core-OM score: 7 short-term clients

Table 10 :
Mean Clinical Core-OM score: all clients

Table 13 :
Conceptual Categories of Change for 8 Veterans' Change Interview Responses (after L14: "One [fellow veteran] in this whole building can create a hell for people . . .and stir things up and create bad, bad energies throughout the whole place . . . . . .I used coping strategies quite a lot during that period . . .It was never the money, it was just the unfairness of ATOS that was getting to me . . .It was bad . . .It was upsetting me . . .but with the coping strategies, again, I got through it".
A lot of [the veterans] are institutionalised ...So, any switch in their normal pattern is massive for them, you know?And then, of course, it's fear, anger, and it shows  itself ...[in]the amount of fights we've had".Client A16: "It's the confidentiality thing everybody's scared of here, because some of the[charity's]staff . . .You're scared in case they're talking about you, which we know they do".Client L11: "Certain staff . . .They're not trained well enough to deal with . . . the mental problems that the guys in here have . . .[Staff] can trigger off people [veterans' symptoms], and I've seen it on a regular basis".Client A18: "[Regarding benefit assessments] I couldn't believe the questions they [ATOS] were asking . . .You weren't getting a fair play . . .a fair shot at it, you knew that with ATOS Client B18: "It's been great . . .for me to write down every day what I do [in a diary brought to sessions], to tell you how . . .what my life is like and how I'm going through my life right now".
Client C14: "I'm feeling a lot of pain from [examining his childhood experiences].The more I open up to it, the more pain I'm feeling, but, I suppose, pain has got to be felt for understanding and all that to come . . .That's the only bad side to the therapy . . .For me to understand it, I've got to feel the . . . the pain again, and I would rather not . . .The need to . . .open up far outweighs the pain I'm getting".
don't know what it is, but, just, for me to talk to you to tell you what had actually happened, about what my Dad was feeling, and how it happened".3.1 Theme: Client in Charge of the Therapy Client E8: "What I like as well I'm not forced into saying anything, which is important . . .It seems as though I can pick the topic; what I want to talk about, what's important".Consistent with Wampold et al's (2010) table of "specific ingredients and . . .common factors" (p.931) found in all effective PTSD treatments (see Table

.
Client G14: "I think what I like about the transactional[analysis]is the way that you do look at it from a lot of aspects; from emotion, from the past, from family ...So, things that I thought in the past, sometimes, were irrelevant ...I can now put more into context".Client G17: "The fact about the transactional [analysis diagrams] as well . ..I'm a visualization kind of person .. .There's your Parental [ego state], there's this and . . .you can see how it overlaps and how it all fits together".

Table 14 :
Changes in Psychiatric Medication during TA Psychotherapy Treatment Certified Transactional Analyst (Psychotherapy) can be contacted on harfordtherapygmail.com