Country house poetry of the English Civil war: a metaphorical exploration of contemporary debates in UK child and adolescent psychoanalytic psychotherapy research and an update on TIGA-CUB

ABSTRACT

Using the country house poetry of the English Civil War as a metaphor, this article traces some of the key debates in child and adolescent psychoanalytic psychotherapy (CAPPT) research over the last 25 years. It identifies three main trends: initial confidence that child and adolescent psychoanalytic psychotherapists (CAPPTs) could apply their existing theoretical methods to empirical research; recognition of the need to use qualitative methodology and to engage with quantitative approaches; and a period of retrenchment, with particular questioning of the use of randomised controlled trial (RCT) methodology. The article also reports the results of a service evaluation undertaken subsequent to the feasibility Trial on Inter-Generational Attachment and Children Undergoing Behaviour problems (TIGA-CUB) to determine whether to proceed to a large-scale, definitive trial, and the inclusion of TIGA-CUB in an on-going, innovative research project on RCT methodology. The article concludes that, if CAPPTs want RCT methodology more appropriate to their ways of working, the profession and the Association of Child Psychotherapists (ACP) will need to develop a more pro-research environment, as well as career pathways which combine clinical work and publicly funded research.

KEYWORDS:

• Child and adolescent
• research
• psychoanalytic psychotherapy
• randomised controlled trial (RCT)
• TIGA-CUB

‘To know can only wonder breede and not to know is wonders seede’from the carol, Lord, when the wise men, Sidney Godolphin (1610–1643)

‘I beseech you, in the bowels of Christ, think it possible that you may be mistaken.’¹Oliver Cromwell, Captain General and Commander in Chief of the Army, addressing the General Assembly of the Kirk of Scotland, 3rd August, 1650, shortly before the Battle of Dunbar

Introduction

When I first started reading the research debates in child and adolescent psychoanalytic psychotherapy (CAPPT) over a decade ago, I found them evocative of 17^th century English country house poetry.² This – often little-known – genre includes a small number of eulogistic poems which echo across the years prior to, during, and following the English Civil War period (1642–1651), and reflects the cultural changes of that turbulent era. Although feminist scholarship is now unearthing earlier examples of the genre, such as Lanyer’s (1611) ‘The Description of Cook-hame’ (see, for example, Lewalski, 2009), the most typical poems usually include Jonson’s ‘To Penshurt’ (1616a) and ‘To Sir Robert Wroth’ (1616b), Carew’s ‘To My Friend G.N. from Wrest’ (1640a) and ‘To Saxham’ (1640b), Herrick’s ‘A Panegerick to Sir Lewis Pemberton’ (1648), and Marvell’s ‘Upon Appleton House’ (1681). The poems draw on classical poetical models and tropes, such as Horace’s Ode Book II, 18 (circa 725–730) (in which the speaker reflects on the modesty of his abode in contrast to grander dwellings, and the folly of worldly possessions in contrast to the wealth of the mind, in the face of mortality), or Virgil’s 4^th Eclogue sponte sua motif (Fowler, 1986; literally ‘of one’s own free will’, in relation to the idea of nature being spontaneously bountiful). They pay homage to the poets’ patrons or friends by praising their estates and thereby comment, often indirectly, on the tumultuous historical, political, and cultural changes of the period.

I was initially puzzled as to what the link between CAPPT research debates and these elegant panegryrics might be, but have gradually come to understand why these poems have come to constitute for me a means of exploring some of the key epistemological and ontological debates about CAPPT research in the UK, tracing as they do a route from confident, almost complacent authority, through more troubled, defensive retreat, to a quietly argued appeal for continued engagement.

The contemporary relationship between UK CAPPT and formal academic research has not always been an easy one. An initial self-assurance in the profession’s ability to make use in CAPPT research of existing approaches to the development of psychoanalytic theory from clinical material has perhaps now largely given way to a greater acceptance of the need to make use of methodology and methods usually employed in qualitative social sciences/humanities research. This gradual – if sometimes apparently grudging – acceptance nevertheless appears to have led recently to a more entrenched resistance to quantitative methodology, and to the randomised controlled trial (RCT), in particular (Schmit Neven, 2018; Waggett, 2018). Although Midgley and Kennedy’s (2010) systematic review and Midgley et al.’s (2021) updated systematic review show that numerous CAPPT RCTs have recently been undertaken, careful reading reveals that much of this progress results from the inclusion of research conducted on psychodynamic interventions, rather than specifically psychoanalytic treatment, and that most has been conducted outside the UK.

Whilst a brief overview of some key research findings and broad methodological approaches is now part of all UK CAPPT trainings, the picture, then, is not quite as rosy as might at first appear. Traditional attitudes still prevail and are, in some quarters, perhaps gaining ground. This matters because, whilst qualitative research can have an indirect influence at a policy level, the NICE (National Institute for Health and Care Excellence) guidelines – based predominantly on RCTs which must usually be conducted in the UK – are increasingly being used for NHS commissioning in Child and Adolescent Mental Health Services (CAMHS).

So why does this reluctance, and at times resistance, towards research – and to the RCT in particular – still exist? Does quantitative research represent an attack on CAPPT, constitute a threat to psychoanalytic clinical work, or preclude other forms of research, practice-based evidence, learning from experience, or the development of theory? Might resistance belie the traditional CAPPT claim to privileging ‘not knowing’? Or is it a healthy attempt to protect what is most distinctive in the profession’s approach, a means of retaining that which child and adolescent psychoanalytic psychotherapists (CAPPTs) most value?

This article aims to explore these questions by using English country house poetry as a metaphor characterising some of the shifts in key debates in UK CAPPT research over the past 25 years. Limitations of space preclude exploration of the fascinating psychoanalytic literature on metaphor (e.g., Aragno, 2009; Borbely, 2011; Britton, 2015; Holmes, 2004; Kirshner, 2015; Melnick, 2000; Ogden, 1997; Oppenheim, 2015; Wallerstein, 2011) and necessitate focussing on what are perhaps the three best known English country house poems – ‘To Penshurst’, ‘To Saxham’, and ‘Upon Appleton House’ – and what I consider to be the most influential debates. The first section draws parallels between ‘To Penshurst’ and the debates from 1996 to 2005; the second compares ‘To Saxham’ with those from 2006 to 2016; while the third links ‘Upon Appleton House’ with more recent discussions. The final section of this article relates to the TIGA-CUB feasibility RCT (Edginton et al., 2017, 2018), providing an update on the outcomes of a subsequent case series undertaken to inform the decision as to whether or not to proceed to a large-scale, definitive trial, and reports an exciting development on how CAPPT might begin to contribute to the development of a more nuanced and appropriate RCT methodology.

Section 1

Although feminist scholarship has now identified Lanyer’s ‘The Description of Cook-hame’ from ‘Salve Deus Rex Judaeorum’ (1611) as the foundational text, Jonson’s ‘To Penshurst’ (1616a) has, until recently, traditionally been considered the first major example of the genre to which subsequent poems refer. The ambience is one of stability, tranquillity, plenty and social harmony. The eponymous country house is praised for its modesty, in reference to Horace’s condemnation of luxury in his Odes (Werlin, 2019) – ‘Thou are not, Penshurst, built to envious show’ (l. 1) and, drawing on classical tropes, such as Virgil’s sponte sua motif (Fowler, 1986), the estate is seen as one where nature willingly offers up copious quantities of everything from stock to game – ‘sheep, … bullocks, kine, and calves’ (ll. 22–23), ‘seasoned deer … conies … purpled pheasant … painted partridge’ (ll. 23–28) – and from fish to fruit – ‘Fat aged carps … pikes … /Bright eels … ’ (ll. 33–37), ‘The early cherry, with the later plum/Fig, grape, and quince … /The blushing apricot and woolly peach’ (ll. 41–43). As Williams (1973) points out, however, this supposedly natural fecundity denies the existence of agricultural labour. Jonson does imply domestic work ‘ … some that think they make/The better cheeses bring them’ (ll. 52–53) – but only in the context of local tenants paying apparently willing tribute to the estate’s owners: ‘And though thy walls be of the country stone/They’re reared with no man’s ruin, no man’s groan’ (ll. 45–46). An impromptu royal visit finds the estate fully prepared (ll. 76–80) but praises only the lady of the house (ll. 83–85), with this harmonious plenitude seemingly enabling all – including the visiting poet – to partake equally: ‘Where comes no guest but is allowed to eat/Without his fear, and of thy lord’s own meat … ’ (ll. 61–64). Despite landlords being encouraged (and eventually compelled) to return to their country estates to enforce royal authority during the social unrest resultant of a shift from feudalism to capitalism in this period (Fowler, 1986; Werlin, 2019), what is portrayed here, then, is a sense of quiet, rural belonging and established, yet obscured, social hierarchy.

Drawing on Serres’ (1982) theory of the parasite, Remien (2014) argues that the poet’s participation in a system of aristocratic patronage nevertheless belies Jonson’s subtle disruption, through criticism of the advantages gained by his poetic contemporaries. Rather than subversion containing and reinforcing established social hierarchies, however (Greenblatt, 1990), Remien (2014) argues that ‘To Penshurst’ represents a radical critique of feudalism, which was to come to a head so dramatically with the English Civil War. How, then, does this first major poem of the genre link to key debates around CAPPT research in the late 1990s and early 2000s?

As good a place as any to begin are Fonagy’s (1996) paper at the British Journal of Psychotherapy 3^rd Joint Conference with the Freud Museum, and Rustin’s British Journal of Psychotherapy Annual Lecture, published a year later (1997). Contemporary debates about research in adult psychoanalysis had been ongoing for some time and, as Midgley (2004) points out, had started with Freud.

Despite some common ground, Fonagy and Rustin take very different positions regarding what constitutes research in psychoanalysis and how the profession might respond. Fonagy (1996) argues for a scientific approach, ‘an attitude of scepticism, the repeatability of observations’ and their accessibility to ‘mathematical and statistical tools’ (p. 106). Such an approach, he contends, would lead to helpful clarification and development in the field and to wider acceptance by other disciplines, because ‘science brings credibility’ (p. 106). As was appropriate to medical science in the 19^th century, Fonagy argues, Freud’s means of communicating developments in both theory and clinical technique was the clinical case study. Whilst appropriate for ‘illustrating a theoretical or technical point’, however, Fonagy contends, ‘for deriving the theory or technique, it is deeply flawed’ (p. 109), requiring only the iteration of supporting evidence, not the citation of that which is contradictory – a criticism of psychoanalysis, Fonagy reminds us, once infamously made by Grunbaum (1984). Whereas medical science largely rejected the clinical case report with the development of positivism, psychoanalysis ‘clung to the classical case-study method in the face of opposition, in part, to preserve unity and identity’ (p. 110). Clinical work, he argues, ‘involves following a thread through the minutiae of a highly charged relationship between two individuals’, while clinical research requires the ability ‘to stand back and discern patterns across large numbers of cases … [which] … must for this purpose be described and analysed quantitatively’ (p. 110). Fonagy illustrates the methods that might be used with a retrospective study of 763 cases from the Anna Freud Centre (Fonagy & Target, 1994; Target & Fonagy, 1994a, 1994b) and a study operationalising Fraiburg et al. (1975) ‘ghosts in the nursery’, also citing smaller studies. He concludes that ‘such systematic and rigorous methods’, when used in addition to clinical case reports, ‘might lead to coherence and progress in this field’ (1996, p. 116).

In contrast, Rustin (1997) draws on Latour’s work on the laboratory (1983, 1986, 1996) to argue that psychoanalysis has its own scientific method based on what occurs in the consulting room (p. 531), as it is ‘only here that relevant observations and experimental interventions in the form of interpretations can be made … and their effects studied’ (p. 531). He argues that, through the establishment and maintenance of rigorous conditions and ‘routinization of procedures’ (p. 532) for trainees, psychoanalysis has been able to use the clinical case study effectively to identify the points at which existing concepts fail to explain what is observed, thereby generating new theory. Rustin concedes that for Latour, the sciences achieve power through their ‘capacity to condense and accumulate evidence over time’, including by means of statistical summary of multiple empirical cases (p. 536), and that what is discovered in the laboratory becomes known beyond it largely by building both intellectual and political alliances outside that specific discipline (p. 531). In psychoanalysis, in contrast, he argues, clinical skills are essentially learned ‘by “craft apprenticeship” in personal analysis and supervision’, whereas the ‘“public” forms of communication involved in writing and publication are less influential … ’ (p. 537). Rustin acknowledges that, ‘as the community grows larger, so consensus will become harder to maintain’ (p. 537), but he does not – here, at least – offer a solution. For Rustin, then, it is not a question of needing to begin research in psychoanalysis, since the clinical case study has always been used in research. Rather, it is merely a question of ‘making … data more transparent’, and ‘making more visible its core practices for research purposes’ (p. 539).

Following these initial forays, both Fonagy and Rustin turn their attention specifically to CAPPT. In a direct appeal to the profession, Fonagy (2003b) calls for the generation of evidence about the ‘specific patient groups’ for whom CAPPT might be most effective and ‘assessment systems’ to identify them, for ‘sensitive measurement systems’ to identify changes that might ‘go beyond symptomatic improvement’ (p. 133), and the development of ‘new adaptations of psychoanalytic therapy’ to achieve ‘greater generalisability’ in terms of symptom relief or prevention (p. 133). He points out that a variety of research methodologies will be required to fulfil such an agenda (p. 133) but, critical of what he terms the ‘undeniable arrogance’ of psychoanalysis (p. 131), Fonagy concludes that, ‘[r]ather than fearing that fields adjacent to ours might destroy the unique insights offered by long term intensive individual therapy’, the profession might better consider ‘ending our splendid isolation and undertaking active scientific collaboration with other disciplines’ (p. 134).

In reference to psychoanalysis more generally, Fonagy (2003a) goes on to point out that, although ‘[s]ymptom change as a sole indicator of therapeutic benefit’ is ‘crude in relation to the complex interpersonal processes which evolve’ in long-term work, the importance of ‘ambiguity and polymorphy’ is such that the profession might be ‘threatened by the systematization of clinical reasoning’ (p. 73). ‘Psychoanalytic theory’, Fonagy contends, ‘precludes the possibility that psychoanalysts can be adequate observers of their clinical work’ (p. 73) and he is critical of the narrative approach to clinical case studies of the traditional psychoanalytic method, asserting pithily that ‘data is not the plural of anecdote’ (p. 73). Fonagy outlines some of the early indications from neuroscience of the potential impact of psychoanalytic treatment but asserts that other disciplines are not likely to be interested unless psychoanalysis ‘can actually contribute to directing or to informing data collection in those disciplines’ (p. 75). He acknowledges the limitations for psychoanalysis of RCTs (p. 77) – short-term work [because of the nature of the funding], limited relevance to clinical practice, treatment of a single disorder, screening to maximise homogeneity, and a focus on ‘easily measurable’ outcomes – but nevertheless argues that embracing such research selectively only when the results are favourable will create an ‘illusion of participation’ in research, rather than genuine engagement (p. 78, my emphasis).

In Rustin’s seminal piece on CAPPT research (2003), he turns to a shorter editorial by Emde and Fonagy (1997) where they argue that, despite initially generating ‘an extraordinarily innovative body of knowledge’, psychoanalytic training culture ‘is one that values confirmation and closure rather than exploration and systematic investigation of what we do not know’ (pp. 643–651). Rustin (2003) encourages conversation between what he terms ‘those engaged in “traditional” psychoanalytic methods and those trying to give these a more verifiable, replicable and publicly accountable basis’ (p. 138). He argues that the interaction between patient and therapist being ‘unscripted and unpredictable’ (p. 141) makes structuring prior order into the process of data collection inappropriate. Because the unconscious is the object of study, it is ‘best … accessed where it manifests itself spontaneously – and most richly and informatively – within the transference relationship’ (p. 141). Like Fonagy, Rustin argues for interdisciplinarity, but suggests that other disciplines be used to corroborate existing methodology. He argues that ‘more routinised methods of data collection and analysis’ are unlikely to replace the consulting room as a means for generating new ideas in psychoanalysis (p. 143), and that the profession therefore needs ‘to hold on to two research methodologies, and not one’ (p. 143).

Midgley’s (2004) contribution helpfully expands the debate, drawing on McLeod (2001) to outline three main stages in the historical development of psychotherapy research: the formative period, where the main approach was the single case study; the period from the 1950s to the 1980s, the era of experimental designs; and from then onwards, a period of ‘uncertainty and opportunity’, which Midgley sees as sailing between ‘the Scylla of large-scale quantitative research’, ‘the dominant ideologies of evidence-based practice and the logic of the randomized controlled trial’ and ‘the Charybdis of the clinical case study’ and ‘an exclusive reliance on our traditional methods of “clinical wisdom”’ (pp. 90–92). He goes on to outline and illustrate different means of qualitative data collection and analysis, before summarising some of the key challenges and criteria for assessing the quality of such research. Midgley diplomatically bridges the gap between Fonagy’s rejection and Rustin’s defence of the traditional case study and provides a more detailed explanation of how rigorous use of qualitative research methods might offer a way forward. He nevertheless acknowledges that there are ‘real questions about how qualitative approaches can contribute to evidence-based practice, and whether they will have the political “clout” to be taken seriously by policy makers’, and recognises that qualitative approaches are ‘time-consuming, intensive and challenging’ (p. 106). Nevertheless, Midgley does not address how both qualitative and quantitative approaches to CAPPT research might be able to co-exist within the profession.

As with ‘To Penshurt’, then, during this initial stage in recent CAPPT research debates, there is, for the most part, a sense from the established order of a quiet confidence that things can remain much as they are and that demands for ‘research’ can be met without too much additional work or threat to existing practices. Authority is essentially maintained through inter-generational emulation and the exercising of hierarchy and patronage and there is, apparently, no real need to get to grips with qualitative methodology or methods, with who is to undertake the work of any adjunctive studies remaining happily unspecified. On the other hand, however, there is simultaneously an implied trope of et in arcadia ego – with Fonagy’s urgent, openly critical warning about the profession’s reliance on the clinical case study and lack of engagement with other forms of research or scientific disciplines. Midgley’s patient encouragement of qualitative methodology and methods, whilst not denying the likely labour involved, pushing for a full engagement with quantitative outcome studies, or disturbing the foundations of CAPPT, nevertheless introduces the subtlest of Jonsonian internal critiques.

Section 2

In contrast to the atmosphere in ‘To Penshurst’, that of ‘To Saxham’ (1640b) is somewhat less confident. Faith in the ability of the country house and all it represents to return to its former glory is maintained, but the scene is a wintry landscape, such that memory, rather than direct observation, evidences the estate’s beneficence. Written by the Royalist Carew, most probably shortly before Charles I came to the throne in 1625, the outdoor scene is viewed from within, and the estate’s abundance and plenitude concealed: ‘Though frost and snow lock’d from mine eyes/That beauty that without door lies’ (ll. 1–2). There is an air of insouciant self-sufficiency – ‘ … neither from, nor to thy store/Winter takes aught, or spring adds more’ (ll. 9–10); a sense that the estate is favoured – ‘The season hardly did afford/Coarse cates unto thy neighbour’s board/Yet thou hadst dainties, as the sky/Had only been thy volary’ (ll. 15–18); and a claim that the house has nothing to fear from outsiders – ‘Thou hast no porter at thy door/T’examine or keep back the poor’ (ll. 49–50). ‘To Saxam’ was published only after Carew’s death in 1640, at a time when Charles I had for some time been ruling by decree and was skirmishing with the recently reconvened Parliament, in the context of religious discontent and the threat of a war with Scotland. There is an increasingly oppressive mood of siege and, despite an optimistic metaphorical reference to the monarch, also a sense of foreboding: ‘ … they deride/The night and cold abroad; whilst they/Like suns within, keep endless day/Those cheerful beams send forth their light/To all that wander in the night … ’ (ll. 32–36). Again, however, the denial of threat attests to its very presence – here greater – with direct reference to ‘the poor’, ‘strangers’, ‘fear’ ‘foe’ and ‘thieves’ (ll. 50–57). There is, then, a feeling of holding together in the face of a more immediate menace, and an air of denial being used as a mechanism of defence.

In the research debates between 2006 and 2016, there is a growing recognition of the importance in CAPPT research of using established qualitative methodology and methods. Midgley (2006) takes a more robust approach, acknowledging the contribution of the case study, but outlining three key weaknesses: the ‘data problem’, the ‘data analysis problem’, and the ‘generalisability problem’ (pp. 125–126). In ‘the data problem’, the issue is the accuracy of the therapist’s notes, and whether ‘the clinical fact’ (O’Shaughnessy, 1994) is an independent entity or a construct of the therapist’s mind. Audio and video recording, Midgley argues, have traditionally been resisted as likely to impact on therapist and patient alike (p. 128), when some of the most important clinical data in the encounter – namely the transference and counter-transference – cannot be captured by this type of recording. The ‘data analysis problem’ refers to the construction of case studies to ‘persuade others of the validity of a particular point of view’ through use of aesthetic and narrative completeness, often with the omission of conflicting evidence (pp. 131–132). The ‘generalisability problem’ is the extent to which generalisations can be made from a single case study, given that statistical methods of ensuring generalisability in quantitative research tend to lack ‘external validity’ for clinicians working in CAMHS. What single case studies can be used for, Midgley argues, employing Popper’s (1963) famous term, is ‘falsification’ of a specific theoretical idea, or if carefully and systematically selected and aggregated, to create a series of comparable case studies from which theory can be deduced more robustly.

Overall, he concludes, the clinical case study has a potentially important role to play in CAPPT research, but care needs to be taken both to rediscover some of the rigour with which Freud used the method and to address the ‘limitations … identified by the wider research community’ (p. 139). An important milestone in this period is also reached in 2009 with the publication of Child Psychotherapy and Research: New Approaches, Emerging Findings (Midgley et al., 2009), where some key debates of the time are rehearsed and a variety of methods illustrated.

In his 2010 article, Rustin points out that outcome research and the development of psychoanalytic theory and technique have often been defined in opposition to each other, with the term ‘empirical research’ being abrogated by the former but being of ‘little intrinsic psychoanalytic interest’ (p. 382). He argues that this is a false dichotomy which forces psychoanalysts to choose ‘between a commitment to scientific research and a commitment to psychoanalysis’ (p. 382). Rustin acknowledges that outcome research can answer questions about treatment-related change and can also, where sessions are recorded, constitute a valuable source of data for determining types of change and reasons for them (p. 383). He goes on, however, to argue that psychoanalysis has been compelled to believe that the appropriate sciences to emulate are ‘the medical sciences which are in the main uninterested in mental life, and academic psychology, which in Britain has often made the disavowal of the unconscious a first principle’ (p. 386). Rustin rejects ‘radical proposals for the relocation of psychoanalysis within academic institutions’ (Kernberg, 1986), and argues for an incremental ‘“English” mode of advance’ in the development of psychoanalytic knowledge and academic accreditation (p. 389). Like Fonagy, Rustin outlines several areas of potential development for psychoanalysis, encompassing greater involvement in empirical research, more planned and purposeful scientific agenda-setting, and more interconnected and interdisciplinary dialogue between academics and psychoanalytic societies. He diverges, however, in his relative lack of emphasis on outcome research, which he attributes to ‘the unfamiliarity and indeed distastefulness of this instrumental way of thinking … and the difficulties and discomforts involved in framing the goals of psychoanalysis in utilitarian terms’ (p. 380). Hinshelwood’s (2013) important book, Research on the Couch: Single Case Studies, Subjectivity and Psychoanalytic Knowledge provides particular food for thought in this period with regards to how the single case study can be used in psychoanalytic research to provide robust (if limited) evidence, without violating psychoanalytic principles.

A further turning point occurs with Rustin’s (2016) article arguing for the use of grounded theory as a qualitative research method in CAPPT and Kegerreis’ (2016) response. Despite rejection of psychoanalysis by critics such as Popper (1963) and Grunbaum (1984) because of the impossibility of external ‘falsification’ and the difficulties of generalisation, Rustin argues (p. 182) that psychoanalysis is often seen as ‘a hybrid field of knowledge, part science, part humanity, because its concerns are not only with laws of cause and effect, but also with interpretation, with subjective meanings, and with individual differences’ (Ricoeur, 1970). The way in which psychoanalysts write about clinical cases is, Rustin suggests, ‘paradoxically quite close to the “falsificationist” model’, in that ‘hypotheses derived from existing theories were tested and abandoned specifically because they were disconfirmed by clinical evidence’ and ‘[a]lternative hypotheses were then adopted which better corresponded to the clinical facts’ (p. 186). A good deal of clinical writing, however, only describes or interprets existing theory, Rustin contends, and although this provides further ‘empirical evidence of validity’ in a field where ‘the sequential accumulation and comparison of such cases’ provides the means for testing theories, it does not always challenge existing theories or lead to their modification (p. 186). Citing Anderson (2006), Rustin argues that grounded theory is close to the ‘line by line’ analysis of psychoanalytic supervision, but that through its justification of the selection of clinical material and method of analysis, it is more transparent and ‘accountable’ to the reader (p. 188). He acknowledges that such qualitative work, at least in the short term, is unlikely to improve the scientific reputation of the field (p. 190). It has, nevertheless, he claims, enabled doctoral programmes to be established, and the demands that these impose – such as systematic reviews of the literature, explicit research questions, and use of widely recognised methods – have, for Rustin at least, made psychoanalytic research comparable with other theoretical paradigms. It is the central place of the unconscious in psychoanalysis that has made research more challenging (p. 190), he concludes, and this is where literature and the arts have a greater affinity with psychoanalysis than the sciences, because both psychoanalysis and the arts ‘recognise that the understanding of minds depends on the interpretation of meanings, and not merely on the correlation of observed facts’ (p. 191).

In her response to Rustin (2016), Kegerreis (2016) acknowledges Rustin’s point that ‘the practice of psychotherapy has been the “place” where the research is done’ (p. 199). She cites Perron’s (2006) impassioned response to Kernberg (2006) that psychoanalysis should be researched ‘using methods that do not kill their very object’ (Perron, 2006, p. 931) but also invokes Scott’s (2015) characterisation of some psychoanalysts’ attitudes to research, in reference to Steiner (1993), as essentially constituting ‘a psychic retreat’ (p. 199). Despite the development of well-designed outcome studies and of research methods which ‘can capture more effectively the kind of internal changes that are the central concern of psychodynamic practitioners’, the ‘concern that the essential qualities of the psychoanalytic encounter are at risk from rigorous research methodologies’, Kegerreis argues, ‘is still very much alive’ (p. 199). She points out that, whilst CAPPTs learn in training about how ‘theory informs practice’ and then work ‘in ways dictated and informed by these theoretical frameworks’, on qualification, the tendency is to ‘cease to question or interrogate our ways of working’ (p. 199). Kegerreis reports that, in teaching other CAPPTs how to undertake research, she has been shocked ‘to discover just how different the clinical and research mindsets tend to be’ (p. 200). She puts this down to the very challenging work that CAPPTs do, and the ‘need to have an extremely solid sense of our theoretical foundations to embark on the work at all’, and to guard against enactments or acting out (p. 201). CAPPTs do therefore, she contends, ‘hold on even more tightly to our setting and our rules in order to protect both our patients and ourselves’ (p. 201). Nevertheless, she concludes that CAPPTs need to become ‘much braver in questioning our underlying assumptions, to become less defensive in the face of scepticism and, to this end, to embrace the need to record our work for research purposes’ (p. 205) – something which, she argues, CAPPTs in the IMPACT study who recorded sessions found ‘much less intrusive than they expected and extremely helpful in examining their practice more truthfully’ (p. 204).

As in ‘To Saxham’, then, in this middle period, there is perhaps a sense of more traditional approaches being somewhat ‘under siege’, but a harking back, nevertheless, to periods where there was greater confidence and certainty of approach and methods. The ‘threat’ posed by external others – here, perhaps, quantitative approaches – is more discernible, but there nevertheless remains an independence, a self-sufficiency, and a sense that what cannot currently be observed is nevertheless present and can be held in mind. There remains a self-assurance – albeit a somewhat shaky one – that there is still no need for ‘defence’ against possible ‘threats’, but nevertheless, also a sense of foreboding and retreat. Meanwhile, there is an acknowledgement of the centrality of qualitative research and hope for the future, but nevertheless advocacy for change.

Section 3

In ‘Upon Appleton House’ (1681), thought to have been written in the early 1650s at the end of the English Civil War, Marvell urges Thomas Fairfax – Lord General of the New Model Army, of which Oliver Cromwell was Lieutenant-General – to return to engagement in public and political life. Fairfax had resigned his commission in 1650 at a critical point in the English Civil War and the start of the Anglo-Scottish Wars – a move which left Cromwell in control – and had retired to his Nun Appleton estate in Yorkshire. In this complex and subtle use of the English country house genre, Marvell explores the changing psycho-political geography of the English country estate and associated aristocracy, as well as those of the poet and of poetry itself, in this post-regicide era (Werlin, 2019).

The poem opens with the now familiar reference to rural modesty in comparison with more ostentatious, modern dwellings, ‘Why should of all things Man unrul’d/Such unproportioned dwellings build?’ (ll. 9–10), with Marvell emphasising Nun Appleton’s diminutive scale, ‘When larger sized men did stoop/To enter at a narrow loop’ (ll. 29–30). Whereas in ‘To Penshurt’, the rural estate is seen as the appropriate location from which the aristocracy is urged to govern on behalf of the monarchy, and then, in ‘To Saxham’, as a place of temporary refuge, it is now characterised as an almost womb-like space – ‘But when he comes the swelling Hall/Stirs, and the Square grows Spherical’ (ll. 51–52), and one, it is implied, in which Fairfax should himself only reside temporarily if he is not to lose his masculine qualities: ‘And for an Inn to entertain/Its Lord a while, but not remain’ (ll. 71–72). As Marvell interweaves both family history and legacy with various landscapes, ‘fragrant Gardens, shaddy Woods,/Deep Meadows, and transparent Floods’ (ll. 79–80), Werlin (2019) argues, he also insinuates links between the Catholic, the royalist, the domestic, and the feminine, on the one hand, the Protestant, the republican, the public, and the masculine, on the other. Masculine political power, Werlin argues, is not ‘at the centre of domestic life’, as in previous country house poems, ‘but rather in direct antagonism to it’ (Werlin, 2019, p. 9). In parallel with Fairfax’s dilemma, Marvell also considers his own position as a poet, and that of elite English poetry, until then largely associated with, and supported by, the English Court. Recognising that poetry ‘had no place on the battlefield’ but also ‘could not make its home in the domestic sphere’, Werlin (2019, p. 15) points out that Marvell concludes the poem with, ‘Let’s in’ (l. 775), as if about to return inside the house, and as if poetry itself were ‘falling silent on the threshold’ (2019, p. 15).

What, then, are the parallels here with the research debates from 2016 to date? First and foremost, this was the period in which the results of the IMPACT trial were published (Goodyer et al., 2017), followed by an ongoing succession of related articles on secondary and associated data analysis (see, for example, Calderon et al., 2019; Krause et al., 2021; Midgley et al., 2018; O’Keeffe et al., 2019, 2020). The results of the IMPACT trial were in fact positive for CAPPT, in as much as Short Term Psychoanalytic Psychotherapy (STPP − 28 sessions, with 7 sessions of parent work) was found to be no less clinically effective than Cognitive Behavioural Therapy (CBT − 20 sessions) in reducing depression and preventing relapse in young people aged 12–17 with moderate to severe depression, as well as being no less cost-effective (Goodyer et al., 2017). There nevertheless seemed to be a sense for some that IMPACT had been ‘unsuccessful’ because it had not shown STPP to have superior outcomes to CBT at one year follow-up, therefore not demonstrating the ‘sleeper effect’ (Midgley & Rustin, 2017, p. 7).

By 2018, there seemed to be a further shift in attitudes, with both Schmit Neven and Waggett publishing pieces in The Bulletin critical of RCT methodology. In her piece, Schmit Neven (2018, p. 12) describes RCT methodology as ‘reductionist’, arguing, in reference to IMPACT, that it is ‘this methodology which is likely to have been a more significant contributor to the disappointing research outcome’ (Schmit Neven, 2018, p. 12). Asserting that the RCT is ‘solely concerned with trying to eliminate problems within the individual child and adolescent and espouses a de-contextualised epistemology … which discounts the complexity of human experience and multiple variables that should legitimately have been part of the research inquiry’ (2018, p. 12), she argues that ‘the child/adolescent/parent relational interaction should be at the heart of all psychotherapeutic treatment’ (2018, p. 12). Schmit Neven contends that IMPACT is ‘another example in which clinical practice is compromised when in the service of adhering to inappropriate research methodology; in other words, trying to keep the therapeutic work with the young person separate and “uncontaminated” from the therapeutic work with parents’ (2018, p. 12). She goes on to argue that the young person and their difficulties need to be understood in the context of their family situation, and that the ‘most successful and enduring outcomes in therapeutic work with children and young people, involve not only a shift or change within the child, but also an essential shift and change in the parents’ perspectives of themselves’ (2018, p. 13). Schmit Neven hopes ‘in the longer term rather than being attached to the coat tails of a limited model of empirical science, that child psychotherapy positions itself differently’, and that qualitative research can potentially impact the policy making and political levels (2018, p. 13). She argues for ‘forming strategic collaborations and alliances with related disciplines and like-minded professional groups’ (2018, p. 13), concluding that the ‘need for a paradigm shift in how we conduct research is increasingly acknowledged by the larger professional group of mental health practitioners’ such as the Division of Clinical Psychology of the British Psychological Society (2018, p. 13).

In his piece, and in response to Fonagy et al.’s (2005) statement that ‘the logic of RCTs is unassailable’, Waggett (2018, p. 14) argues that ‘the logic, and practice, of randomised control [sic] trials (RCTs), especially when applied to mental illness in children and young people, is disputable’. He asserts that the position of RCT methodology comes from ‘the dominance of the discourses and power relations surrounding them’, and that the ‘so-called hierarchy of evidence, with RCTs as the gold standard, is in fact a hierarchy of research paradigms which de-legitimises other approaches to knowledge generation’ (2018, p. 14). Waggett continues that ‘we need to have a fully informed but critical relationship with the evidence base, and to begin to develop a new language of effectiveness in collaboration with others who share our world view’ (2018, p. 14). He contends that, with RCTs being constituent of the paradigms of evidence-based medicine and the positivist/quantitative approach to medical research, disciplines which work from a practice-based evidence and more qualitative, non-medicalised approach are ‘excluded from the resources provided for research and become increasingly less credible’ (2018, p. 14). Waggett argues that researching brief psychoanalytic psychotherapy as a means of ‘saving the longer and more intensive forms of practice’ is ‘questionable’, and that it might at best lead to the provision of ‘only those treatments, such as STPP, that have some backing from RCTs’ (2018, p. 14). He argues that it is ‘unlikely that there will ever be funding for a large-scale RCT of long-term intensive psychotherapy so in EBM [evidence-based medicine] terms it is never going to be validated’. Waggett states that one of the claims made for RCTs is that they are ‘free of bias and ethically neutral’ (2018, p. 14) and concludes that ‘the logic of RCTs is flawed’ because it does not lead to the ‘best available treatments’ being provided for children and young people, ‘especially those with long-standing, complex, severe and co-morbid conditions that are significantly less amenable to this form of research’ (2018, p. 14). He argues that there are ‘strong parallels’ between ‘psychoanalytic child psychotherapy’s struggle to have its voice heard, and its work validated and that of qualitative researchers in general’ and concludes that the profession needs to ‘reclaim our own form of science and do our own form of research rather than adapting that of others as this entails accepting their framework and language, including inherent criticisms of the underlying principles and methods of psychoanalysis’ (2018, p. 14). There is, then, during this more recent period, a sense of those using RCT methodology to research CAPPT being under fire, if not quite yet in retreat. Perhaps such sallies are not surprising, coming as they do after the perceived ‘disappointment’ of the results of the IMPACT trial. The publication in 2019 of Rustin’s (2019) influential book, Researching the Unconscious: Principles of Psychoanalytic Method, in which he reiterates and further develops his highly nuanced argument of the previous two decades on the philosophical compatibility of qualitative and psychoanalytic epistemologies, nevertheless adds to this sense of critique of the RCT.

The arguments of Schmit Neven and Waggett with regards to RCT methodology merit careful consideration. With regards to Schmit Neven’s piece, it is unclear what the evidence is for her claim that it is RCT methodology per se ‘which is likely to have been a more significant contributor to the disappointing outcome’ (2018, p. 12), or why she feels that randomisation would not deal with ‘the complexity of human experience and multiple variables’ (2018, p. 12) that are indeed a feature of any research population in mental health. It is, after all, randomisation which, given an appropriately large sample size, enables most, if not all, variables to be distributed approximately equally across all treatment groups, which is precisely Fonagy’s argument as to why the logic of the RCT is, indeed, ‘unassailable’ (Fonagy et al., 2005, cited by Waggett, 2018, p. 12). Similarly, whilst Schmit Neven is certainly right that it is important to consider the wider family context of children and young people coming for treatment, as she herself points out, the misgivings raised about the potential for ‘contamination’ with the involvement of parents in a trial on adolescents came from the IMPACT CAPPTs themselves, rather than being to do with RCT design (Whitfield & Midgley, 2015, cited by Schmit Neven, 2018, p. 12).

Along similar lines, whilst Waggett is careful to make it clear that his piece is written in a personal capacity, it is unclear what the evidence is for his claim to be asserting the profession’s ‘world view’ or indeed for what constitutes the ‘best available treatment’ (2018, p. 14). Moreover, whilst it is certainly true that some RCTs are poorly designed and CAPPTs would do well to be able to analyse such weaknesses, it is far from the case that all RCTs are considered equally valid or free of bias: indeed, bodies such as NICE are very much concerned, when drawing up treatment guidelines, with the careful evaluation and ranking of RCTs in terms of the strength (or otherwise) of the research design and thus of the resulting evidence. There is not simply one RCT design, and, as anyone who has been closely involved in the design of an RCT will know, discussions as to how to try and eliminate as many potential biases as possible are lengthy and complex – indeed, that is why RCTs usually have to be designed from scratch, rather than employing a ‘one size fits all’ model. This does not mean, of course, that all RCTs get it right all the time – there is poor practice in the design and conduct of RCTs as there is in any endeavour – but now, quite rightly, each RCT’s methodology is rigorously scrutinised by the peer review process at both grant awarding and publication stages, a process which will hopefully, over time, lead to fewer poorly designed or executed trials. It is also perhaps worth pointing out that the RCT essentially came from agriculture, and that it is only in the last 70 years or so that RCT methodology has been used by the medical profession which, prior to that, relied predominantly (and some would argue perhaps, at times, it still does) on the kind of ‘folk knowledge’ and ‘craft’ skills to which Rustin refers (1997, p. 537).

Section 4 – TIGA-CUB

It was in this increasingly difficult CAPPT research climate that the TIGA-CUB feasibility RCT (Edginton et al., 2017, 2018) took place – a trial which Waggett himself requested in his then capacity as Operational Director at the Northern School of Child and Adolescent Psychotherapy (NSCAP). As a result of this request, with the help of colleagues at NSCAP and the University of Leeds, a service evaluation was undertaken across five CAMHS in the North of England to establish that there were sufficient cases of children aged 5–11 with treatment-resistant conduct disorders to merit further research, and this, in discussion with CAPPTs at NSCAP and colleagues at the then Clinical Trials Unit (now Institute) at the University of Leeds, eventually led to the development of the grant application and initial protocol of the TIGA-CUB feasibility RCT (see Edginton, 2013).

TIGA-CUB feasibility RCT results summary

The results of TIGA-CUB (Edginton et al., 2018) demonstrated that, with refinement to recruitment, teacher follow-up, and the treatment manual, conducting a large-scale, fully powered RCT to determine the clinical effectiveness of 12 weekly sessions of manualised Psychoanalytic Child Psychotherapy for children aged 5–11 with treatment-resistant conduct disorder and 12 concurrent weekly sessions for parent(s)/carer(s) would be feasible. TIGA-CUB used the Child Behavior Checklist (CBCL) and associated Teacher Report Form (TRF) (Achenbach & Rescorla, 2001; Thomas, 1991) as its primary outcome measure so as to make it comparable with other RCTs on this patient group, and indicated that four potential outcomes could be used to determine clinical effectiveness: the CBCL total score or externalising behaviour score as rated by the child’s primary carer, or the CBCL total score or externalising behaviour score as rated by the child’s teacher on the TRF.

Although this was a feasibility RCT and so not designed to detect a statistically significant change in the primary outcome if there was one, were an application for funding for a large-scale, definitive trial to be made, the feasibility RCT results would be considered broadly indicative of potential efficacy. Whilst the results demonstrated some positive changes, they were not as strong as they might have been. Although some families were followed up to eight months, it is possible that longer term follow-up or indeed longer-term treatment might have indicated more potential benefit, but limitations of funding precluded this. Only 50% of teachers returned the questionnaires that the Clinical Trials Unit issued at the end of treatment and, whilst those who did so indicated that the treatment had been effective, it was not clear if the other 50% had not returned their questionnaires because they had not observed much change in the child’s behaviour, or for some other reason. It was therefore advised that a further series of 5–10 cases be undertaken as a service evaluation at one of the feasibility study sites (comprising two CAMHS) to see if asking teachers to return their questionnaires to a named CAMHS would result in a higher teacher questionnaire return rate and if so, whether the data would indicate a positive behaviour change.

TIGA-CUB further case series

In a non-randomised service evaluation (categorisation being determined by using the Health Research Authority decision tool), eligible dyads were identified by the author screening initial referrals and re-referrals within CAMHS according to the inclusion and exclusion criteria from the feasibility RCT (see Edginton et al., 2017). Following participant identification and confirmation of CAPPT capacity, consent was gained for completion of the CBCL and TRF pre- and post-intervention. Both qualified CAPPTs and senior (fourth year) trainees were eligible to participate and trained in the TIGA-CUB manual using a standardised set of slides (with the manual having been revised in terms of structure and clarification of key points in the light of feedback from CAPPTs who had participated in the TIGA-CUB feasibility RCT). The intervention was otherwise as previously, except that the child and primary carer were seen by two CAPPTs in all instances, as was the practice in the participating CAMHS. Quantitative analyses were performed but, as this was a service evaluation, no hypothesis testing was conducted. Descriptive statistics are reported for baseline characteristics, retention in treatment, and baseline and four-month follow-up outcome data (see Table 1).

Table 1. Descriptive statistics pre and post Intervention.

Variable	Pre Mean (SD) N	Post Mean (SD) N	Difference* Mean (SD) N
CBCL Internal Score	14.86 (12.42) 7	16.86 (9.35) 7	2.00 (10.52) 7
CBCL External Score	31.57 (9.05) 7	29.29 (6.60) 7	−2.29 (10.86) 7
CBCL Total Score	88.86 (27.72) 7	86.71 (18.16) 7	−2.14 (34.76) 7
CBCL Internal T Score	63.29 (13.33) 7	66.14 (9.99) 7	2.86 (9.87) 7
CBCL External T Score	75.29 (5.47) 7	73.43 (4.28) 7	−1.86 (6.44) 7
CBCL Total T Score	74.29 (4.46) 7	74.00 (3.11) 7	−0.29 (6.02) 7
TRF Internal Score	6.71 (5.56) 7	7.43 (6.45) 7	0.71 (7.48) 7
TRF External Score	17.71 (11.63) 7	18.86 (11.22) 7	1.14 (5.55) 7
TRF Total Score	53.43 (32.85) 7	57.00 (32.06) 7	3.57 (24.89) 7
TRF Internal T Score	54.57 (9.95) 7	55.57 (1.50) 7	1.00 (9.59) 7
TRF External T Score	65.43 (7.39) 7	65.86 (7.43) 7	0.43 (3.78) 7
TRF Total T Score	63.00 (9.38) 7	63.86 (9.06) 7	0.86 (7.56) 7

*Post–Pre.

Between September 2019 and December 2020 a total of seven dyads was recruited, with treatment being aligned with school half terms. As previously, recruitment was limited by CAPPT availability. 57% (four) of the children recruited were male and 43% (three) were female. The mean child age at CBCL completion was 6.9 years (SD 2.12), median 7, range 5–10 years. Additional health conditions of the child and parent(s) were not recorded.

Four qualified CAPPTs and one senior fourth year trainee participated. CAPPTs were allocated to dyads in combinations that were determined by CAPPT availability in each of the two participating CAMHS. All CAPTs were offered monthly rather than fortnightly peer group supervision (in line with feedback from participating CAPPTs in the feasibility RCT), with additional service supervision if required, while the senior trainee was supervised by their usual service supervisor. Manual adherence was monitored via supervision.

Of the seven dyads, five (71%) attended all 12 child and parent sessions while one (14%) attended 11 child and parent sessions respectively and one (14%) attended 10. All primary carers and all teachers completed the CBCL and the TRF respectively at both baseline and on completion of the intervention. Loss to follow-up for both primary carers and teachers was therefore 0%. No deaths, related unexpected serious adverse events, or expected serious adverse events were reported.

The results from the additional case series demonstrate that it is possible to increase the return rate of the TRF by asking teachers to return the questionnaire to a named CAMHS rather than to the Clinical Trials Unit. This would, however, require substantial Clinical Studies Officer support in each CAMHS in any large-scale, definitive RCT. The male-to-female ratio was slightly below that found in the clinical population according to the statistics available at the time of the TIGA-CUB grant application, when approximately 70% of cases were male (Green et al., 2005). Current prevalence suggests that approximately 63% are male (Sadler et al., 2018), so the service evaluation was in line with this more recent estimate. With a return rate on the TRF of 100%, the mean differences (pre to post) and therefore effect size for the total score on the TRF were nevertheless very small (Cohen’s d = 0.14) (Cohen, 1988). This was a small-scale service evaluation and it is therefore possible that the results were due to chance, but they do not indicate a sufficiently effective intervention to warrant applying for funding for a full-scale, definitive trial.

Discussion

Given that the IMPACT RCT did not yield the results that some CAPPTs might have hoped for and that the TIGA-CUB feasibility RCT did not produce sufficiently promising results to warrant a large-scale, definitive trial, does this mean that CAPPTs interested in RCTs should, like Fairfax in Marvell’s ‘Upon Appleton House’, retreat from the fray and retire to the quiet solitude of the qualitative single case study or, as Waggett (2018) and Schmit Neven (2018) argue, seek alternative methodologies? I am certainly in favour of CAPPTs making contact with a wide variety of disciplines and exploring a range of methodologies and methods, but would nevertheless argue for continued engagement with the RCT. Although Waggett (2018) appears to take a pessimistic view, describing disempowered CAPPTs struggling to be heard as they adopt others’ quantitative methodologies and undermine their own clinical approach, might there be a way of engaging with quantitative methodology that, whilst avoiding naivety, is nevertheless more optimistic? If so, what might this alternative way forward encompass?

First, it is worth reiterating that the vast majority of RCTs in health are likely to have started from a clinician or scientist noticing something of interest going on in first one case, and then a small series of cases – making what might broadly be called ‘qualitative’ observations, prior to going on to explore these in a more quantitative manner. Secondly, as Rustin argues, RCTs have the capacity to generate large quantities of data which can, with the appropriate permissions, provide considerable data for qualitative analysis. The quantitative/qualitative dichotomy is, then, in this sense, a false one. Thirdly, it is important to acknowledge Waggett’s argument about funding because, with scarce resources, a powerful medical lobby now largely embracing RCT methodology, and policy makers wanting to know ‘what works’, the RCT does remain the gold standard and is likely to do so for the foreseeable future. There is nevertheless a small group of clinicians – in mental health but also in areas such as surgery – who engage in what are termed ‘complex interventions’ (Skivington et al., 2021), comprising either several components (e.g., parent work and child work, as in CAPPT) and/or involving the pre-assessment of cases (as often happens not only in CAPPT but also, e.g., in many types of surgery), and/or multiple ways of indicating a ‘successful’ outcome (RCTs tend to stand or fall on one, or occasionally, a small number of outcomes, designated in advance). For these groups, standard RCT methodology needs considerable refinement, if not downright re-design.

Although TIGA-CUB is not proceeding to a large-scale, definitive trial, it has recently been selected as one of eight RCTs on ‘complex interventions’ in the UK for a five year National Institute of Health Research funded study (NIHR301709) on developing statistical methods for empirically optimising complex interventions in health and social care, in order to create a step-change in the way in which RCTs for complex interventions are designed and statistically analysed. This means that the TIGA-CUB feasibility RCT will be re-designed using RCT methodology inspired by methods from the Design of Experiments field to reflect more closely the ways in which CAPPTs and similar groups work clinically, arguably leading to the possibility of undertaking trials which will provide more informative results. A simulation study will then be run to evaluate the properties of the new designs under a range of scenarios, including the potential need for fewer participants across a series of complex intervention trials – a development which, if it proves possible, might make it easier to demonstrate clinical effectiveness not only more quickly, but also more cheaply.

So, what would a refined RCT methodology suitable for CAPPT look like and what would it be able to do? It is important here to distinguish between the urgent need for development of appropriate measures for capturing quantifiable changes in factors that CAPPTs might consider important – e.g., a child’s psychic integration, the child–parent relationship, the transference relationship, overall progress as evaluated by the CAPPT and supervisor – versus the focus of the above mentioned five-year study, where the emphasis is on the re-design of RCT methodology to deal with the different components of complex interventions and/or the interaction between them – e.g., taking account of the fact that CAPPTs usually assess cases prior to treatment, or the kind of interaction between child and parent work to which Schmit Neven (2018) refers. Ideally, it might even be possible to consider including not just ‘crude’ behaviour change (Fonagy, 2003a) on one primary outcome, but rather, the synthesis of a range of different factors, such as those outlined above. Whilst the concerns raised by Waggett (2018) and Schmit Neven (2018) about the difficulties of evaluating CAPPT via RCT methodology are understandable, especially given the high co-morbidity and complexity of CAPPT cases, if revised RCT methodology were to facilitate the more nuanced analysis of complex interventions, it might eventually prove possible to design RCTs more suited to the complexities of CAPPT cases and ways of working. Revised RCT methodology might also ideally address the issue that Fonagy (2003a, p. 78) raises with regards to the mismatch between the homogeneity in cases required by RCTs and the heterogeneity of the usual CAPPT case load.

Perhaps the reason the country house poems of the English Civil War period came to mind when I first started reading the research debates – and that I have used martial metaphors here – is because methodology is so often treated as a battle. What might be required, then, in the non-prejudicial sense of the term that Rustin (2010, p. 389) intends, is a more ‘English’ approach, i.e., one of evolution rather than revolution. RCTs certainly have their fair share of the abuse of power, privilege, patronage, and politics (and yes, gender politics comes into it, too, and not just at the level of language with e.g., metaphors about ‘hard’ and ‘soft’ data). At their best, however, they create a transparency in data analysis which is unparalleled in the research world. Of course, the possibility should not be discounted that resistance to RCTs might in part stem from this very transparency, which might threaten some of the politics, power, privilege, and patronage that arguably exist in our own profession. But just as CAPPTs once argued that it was not possible to manualise treatment or to audio record sessions – both of which have recently proved to be both possible and broadly acceptable in large trials such as IMPACT (Henton & Midgley, 2012), so currently, some might feel that finding ways of articulating what goes on in the more ineffable areas of our practice in order to make them quantifiable might be equally unacceptable. It seems to me, however, that with developments in technology which make transcription and linguistic coding in, for example, conversational and narrative analysis much quicker and easier, and with exciting contemporary developments in neuroscientific measurement and understanding, such areas might also eventually become analysable in quantitative research terms in CAPPT, too. After all, in our clinical practice and when we write up, we are constantly noticing a myriad of subtle changes, however complex: the challenge is to articulate to ourselves and others exactly what these are and to find ways of analysing them not only qualitatively, but also quantitatively.

Conclusion

With a refined approach to RCTs taking into account the kinds of factors that complex interventions – including CAPPT interventions – involve, the plan will be to make use of this adapted RCT methodology in future clinical trials. In the longer term, if this approach is to be used for the evaluation of CAPPT, serious consideration will need to be given by the profession and by the Association of Child Psychotherapists, in particular, as to how to build on the establishment of doctoral programmes to create not only a climate of encouragement and support and the development of much-needed expertise, but also sustainable career paths for those wishing to combine clinical work with publicly funded research. Although we do not yet know what this new style RCT will look like, one of the key outcomes of the TIGA-CUB feasibility RCT will, then, hopefully be that, by engaging with RCT methodology, CAPPT will be able to influence RCT methodology’s future design and development. I, for one, cannot think of a better long-term outcome of TIGA-CUB than that.

Confidentiality and permissions statement

The additional cases in the above mentioned further case series were undertaken as a service evaluation in two CAMHS which comprised one of the original TIGA-CUB study sites, with the permission of the relevant NHS Trust, and with the kind permission of Professor Achenbach of the Research Center for Children, Youth & Families, Inc, for use of additional CBCL and TRF questionnaires under the original licence from the TIGA-CUB feasibility RCT.

Acknowledgments

This paper is dedicated to my parents, Ann and Jim Edginton, in love and gratitude, for teaching me to think as well as to feel. I would like to thank Dr. Rebecca Walwyn, lead methodologist and statistician on the TIGA-CUB feasibility RCT, for her statistical analysis of the results of the service evaluation cited in this article, as well as for her past and ongoing support and friendship. I would also like to thank those originally involved in the TIGA-CUB trial, whether as researchers, clinicians, patients, primary carers, and/or questionnaire respondents, and the CAPPTs, fourth year trainee, supervisors, primary carers, patients, and their teachers, who participated in the service evaluation referred to above, as well as Professor David Cotterell, mentor on the original TIGA-CUB project. Finally, I would like to thank Dr. Stephen Minta for his assistance in tracing the original publication dates of the English country house poems.

Disclosure statement

Dr. Elizabeth Edginton was chief investigator on TIGA-CUB, funded by the National Institute of Health Research (NIHR) under its Research for Patient Benefit (RfPB) Programme (Grant Reference Number PB-PG-0614–34002). ID: ISRCTN86725795/Registered 31 May 2016. The views expressed are those of the author and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care. The article was written by the author in her own time in a purely personal capacity and the research was not carried out by, or on behalf of, the Counselling Service of the University of Oxford.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Elizabeth Edginton

Elizabeth Edginton qualified as a child and adolescent psychotherapist at the Birmingham Trust for Psychoanalytic Psychotherapy (BTPP) in 2010. She was research lead at the Northern School of Child and Adolescent Psychotherapy from 2011 to 2016 where she was the chief investigator leading the TIGA-CUB feasibility RCT, funded by the National Institute of Health Research (NIHR Research for Patient Benefit Programme PB-PG-0614-34002). Since 2010, she has also worked in Sure Start, CAMHS, and at the Anna Freud Centre and the University of Warwick, and she currently works at the Counselling Service of the University of Oxford and in private practice.

Notes

1. A reference to mercy or tenderness, with these emotions previously thought to have been located in this part of the body.

2. For the full text of the poems mentioned, please see the websites given in the references, under the names of the respective poets.