Research into humanistic counselling has been given a major boost by the findings of two randomised controlled trials (RCTs) published earlier this year. The PRaCTICED trial,1 which was funded by BACP and conducted at the University of Sheffield, established that person-centred experiential therapy (PCET) does achieve comparable results with CBT when delivered in an IAPT setting – as analysis of practice-based data had already indicated. The ETHOS RCT,2 which was supported by funding from the Economic Social Research Council (ESRC) and led by a team at the University of Roehampton, found clear benefits for children receiving person-centred school-based counselling in terms of achieving their goals.

Yet Mick Cooper, Professor of Counselling Psychology at the University of Roehampton and lead researcher on the ETHOS trial, admits to some disappointment. The benefits of counselling did not reach the levels of effect that he had hoped for and expected, given the findings in a previous pilot. Moreover, the extra spend (some £400 per child) made school-based counselling more expensive than usual pastoral support, with no evidence of short-term compensatory savings elsewhere in the health system in terms of GP consultations and other medical costs.

Similarly, the ‘equivalence’ that the PRaCTICED trial established between CBT and PCET faded away after the initial six-month assessment. By one year, people who had received CBT were doing better than those who had PCET and more of those in the PCET group were looking to return to therapy, while those in the CBT group were more likely to be using the techniques they had learned in therapy to help them through recurring difficulties.

So, what do these results mean for the counselling profession? Professor Michael Barkham, who led the PRaCTICED trial team at the University of Sheffield, says the levelling off in benefit from the PCET interventions by one-year follow-up suggests that practitioners need to examine how they can adapt their interventions to better equip people to cope in the long term. For Cooper, these findings highlight a need for the person-centred counselling profession – and that means most counsellors raised on Carl Rogers’ theories of personality change and the six ‘necessary and sufficient conditions’ – to re-examine their model of practice. ‘I think there needs to be a rethink around what school counselling looks like. For some clients, it may be of greater value if it delivers more direction, more structure and more skills around developing their [young people’s] coping abilities.’

Drilling down into the data

There is no doubt that ETHOS showed that counselling was effective – after six months, the participating young people were significantly less distressed and a lot closer to their personal goals for what they wanted from counselling – to increase their self-esteem, reduce their anxiety and be better able to concentrate on their schoolwork. But there are also some very interesting and potentially challenging preliminary findings emerging in the qualitative interviews conducted with 50 of the participant young people, says Cooper. ‘Most of the young people were happy with the counselling – only a small minority didn’t like it, and there were a few who thought it was OK but didn’t find it that helpful. Clearly counselling isn’t going to be suitable for everyone. But there were two negatives in particular that stood out for me. One was a lack of input, structure and advice from the counsellor – a significant minority of the young people were saying they wished there had been more guidance, more techniques, more activity. And the other thing, equally common, was that they found the silences awkward.

‘If we really are listening to the voices of young people, a number of them are saying “I really like the counselling and that opportunity to talk about what’s going on for me, but I wish the counsellors would be a bit more active and a bit less silent”. They didn’t know how to use the space. They would have liked more direction.’

For Barkham, development of the experiential components – the more process-guiding, emotion-focused aspects of the therapy – also seems to hold most promise for enhancing the impact of PCET and making its benefits longer lasting. ‘Clients leave feeling better having told their story, but when the depression comes back, they don’t have the means to process and move forward again on their own – all they can think of is going back to therapy again. The findings are really demanding of practitioners to take that extra step in process-guiding, which is a step up and beyond the traditional person-centred model. The model has to compensate for the absence of some components, and I think the next step is for clinicians to come together around that in the various training programmes.’

Is there a significant message here for the counselling profession? There is, says Cooper, an understandable wariness among person-centred counsellors about questioning Rogers’ six conditions and the fundamental client-led ethos of the person-centred approach. But, he argues, the research tells us this may not be what some clients want or find useful, and ethical practice demands that we listen to what clients say they want. ‘It is often assumed as a fixed position that it is the relationship that heals, but that’s not what the evidence always shows. It’s great if a therapist wants to be non-directive and work from a wholly relational stance, and there is evidence that it can really help some clients, but those practitioners need to be clear that it may not suit all clients, just as CBT doesn’t suit all clients.

‘It can be painful when you have trained in an approach and someone comes along and says “actually, the evidence challenges that”,’ Cooper agrees. ‘It may not feel great. But I don’t think the evidence says “abandon person-centred practices”. It is saying some clients maybe would benefit from more of something else. Of course, we can only offer what we are trained in, but having the skills and willingness to refer on may be an important element of a therapist’s work.’

So what works in counselling?

The evidence on ‘what works’ in counselling, derived from a vast amount of research over the past several decades, can be summarised very swiftly. First, no one model has been proven to be significantly more effective than any of the others. This then begs the question, what is it about all the talking therapies that makes them helpful to people in resolving their distress and regaining their resilience to life’s vicissitudes and an ability to cope? This is where the PRaCTICED and ETHOS trials have much to contribute.

The supreme compendium of data on the effectiveness of relational factors in the talking therapies is Psychotherapy Relationships That Work,3 which is based on the findings of the Third Interdivisional APA (American Psychological Association) Task Force on Evidence-based Relationships and Responsiveness. Now in its third edition, it is in two volumes: Evidence-based Therapist Contributions, edited by John Norcross and Michael Lambert, and Evidence-based Therapist Responsiveness, edited by John Norcross and Bruce Wampold. The book brings together findings from 18 vast meta-analyses of data on what makes talking therapies effective.

Norcross and Lambert offer several ‘take-home points’ from this massive body of research: ‘One: patients contribute the lion’s share of psychotherapy success (and failure). Two: the therapeutic relationship generally accounts for at least as much psychotherapy success as the treatment method. Three: particular treatment methods do matter in some cases, especially more complex or severe cases. Four: adapting or customising therapy to the patient enhances the effectiveness of psychotherapy probably by innervating multiple pathways – the patient, the relationship, the method, and expectancy. Five: psychotherapists need to consider multiple factors and their optimal combinations, not only one or two of their favourites.’ And, they add: ‘… the patient’s perspective of the relationship proves more important to their treatment outcome than the therapist’s. The patient’s experience of the alliance, cohesion, empathy, and support relate and contribute more to their success than the practitioner’s experience.’ In other words, the client knows best when it comes to how they feel about the therapy relationship.

But they also take pains to stress that treatment method is important: ‘It remains a matter of judgment and methodology on how much each contributes, but there is virtual unanimity that both the relationship and the method (in so far as we can separate them) “work”. Looking at either treatment interventions or therapy relationships alone is incomplete. We encourage practitioners and researchers to look at multiple determinants of outcome, [and] particularly client contributions.’

Norcross and Lambert have also helpfully summarised what it is about the therapy relationship that is ‘demonstrably effective’: the therapist-client alliance, collaboration, goal consensus, therapist empathy, positive regard and affirmation, and the recording of client progress data.4

BACP recently published a (much shorter) good practice guide on What works in counselling and psychotherapy relationships,5 compiled by Ani de la Prida. This concludes that ‘a clear understanding of what a client wants from therapy, and explicit agreement on how to work together is essential’. It goes on: ‘Regularly asking for feedback, paying attention to the development and maintenance of the therapeutic relationship and working non-defensively to repair ruptures promote improved outcomes. Self-awareness and cultivating an active curiosity and appreciation of difference are essential to ensure ethical, inclusive, non-discriminatory, culturally sensitive practice.’

To sum up: ‘... a pluralistic perspective of good practice that is inclusive of all modalities is essential. The therapeutic relationship is key to effective therapy, and a focus on ingredients such as collaboration, empathy, and responding to client preferences is vital to ensuring ethical and effective therapeutic practice. The recognition that different clients need different things promotes a more pluralistic provision of therapy services.’

Mike Moss, an experienced person-centred counsellor and supervisor currently working in schools, sees nothing in all of this to challenge his approach to working with clients: ‘Whether they’re eight years old or 80, for me it’s the quality of the relationship that heals.’ Non-directivity doesn’t mean never actively intervening or taking a lead, he argues; when Carl Rogers developed his theories, it was in contrast to the highly directive approach common in counselling and psychological practice where the practitioner was the expert who knew best what the patient needed in order to feel better. ‘I applaud CBT if it can give people a set of tools that they can use for the rest of their life, but that’s not what I do. I am trying to create a place where their own, innate actualising tendency can flourish; to give the client a greater sense of their own potentiality. If I discover something that is really useful, I will offer it, especially to young people – online tools, for example, which a lot find really useful – but it’s not directing, it’s offering.’

And if there’s an uncomfortable silence? ‘I’ll explore that with them. I’m interested in why. But I carry some cards with me, and I’ll sometimes bring them out and we’ll just play a game if a young person is feeling particularly anxious or uncomfortable. It’s all part of building the relationship.’

Goals and outcomes

What many in the counselling profession are now saying is that Rogers’ six conditions certainly are necessary for most clients, but the six are not in and of themselves completely sufficient for everyone. More may be needed, primarily because clients are telling us so, often demonstrably with their feet. ‘What the evidence also shows is that Rogers’ core conditions are just some of many different relationship factors that are associated with positive outcomes,’ Cooper says. ‘For instance, alignment on the goals of therapy also seems to be important, and the therapist’s capacity to deal with ruptures in the alliance, and the use of systematic client feedback. Added to that, there’s some very good evidence, both quantitative and qualitative, that clients can really value and benefit from non-relational interventions, like normalisation through the therapist’s expert knowledge, or behavioural activation. Maybe the relational elements of these therapeutic encounters are sufficient to bring about some degree of change, but to just focus on them would be to ignore what some clients themselves are saying matters most.’ Broadly, research findings point to a need for greater attention to collaborative goal-setting, progress monitoring and feedback, use of more active and creative ways to engage clients, and more attention to client preferences – ways of working that the client themself finds helpful.

Listening to and learning from the client is a no-brainer for Barry McInnes, a counsellor in private practice and founder of the website Therapy Meets Numbers, which aims to bridge therapy research, evaluation and practice.6 ‘For me, the take-away from these two trials for person-centred therapists is to try to expand your repertoire and listen to what the client is asking for – do they have a preferred way of working and is what you are offering hitting the spot? – before they drop out because they aren’t getting what they want. For me, this is a wake-up call for the profession,’ he says.

He argues that collecting and feeding back outcome data to the client should be integral to the therapeutic process: ‘First, I think that working with the client to capture the essence of what they are seeking can, of itself, be a therapeutic process. It also provides a clear shared understanding of the aims of the work. Second, it helps both me and the client to hold those goals clearly in mind. Third, it helps us both to know how well we are progressing toward the goals. We know the contribution of the therapeutic alliance to therapy outcomes. We also know that one of the cornerstones of the alliance is agreement on therapeutic goals. Why then would we not seek to make their clarification a more systematic process?’

Taking outcome measures can help validate a client’s sense of distress – that they really do need help, which for some is important if they are to engage with therapy. Continuing to do so then provides a measure of progress, or an alert that progress has stopped. But McInnes’ own main reason for doing it originally was curiosity, he says. ‘I wanted to know how I compared with the overall effect rate. What’s my effect size? I’d be disappointed if therapists had no curiosity to know what difference they are making.’

Putting learning into practice

University practitioner Afra Turner has followed a path from psychodynamic to CBT in her professional development and orientation over the past 27 years. Now a senior therapist and supervisor in the counselling and mental health service at King’s College London, she has worked in seven different university counselling services. ‘The decision to do post-qualification training in CBT came about in response to the student voice and my own sense of what I’d been trained for. A lot of the literature was on long-term, in-depth work, but in the university counselling context, often the student only wanted or stayed for four to six sessions. Students wanted a place to be supported emotionally but they also wanted support to navigate the practical aspects of their academic work, and CBT really stood out for me as a possible way to combine the two. And that is exactly what I found. A personal crisis may bring the student to the service, but the reality is they do have these deadlines and commitments and will falter if they don’t have that support.’

King’s College London is a partner in a large-scale data collection initiative across the higher education sector called SCORE (Student Counselling Outcomes Research and Evaluation), which is supported by both BACP and UKCP and is amassing and publishing outcomes data from several student counselling services. Turner’s professional experience is that outcome measures are a helpful addition to her work and to the higher and further education sector. For her, collecting and pooling outcome data are essential for individual professional development and the development and status of the profession. ‘Large datasets ensure counselling services are fit for purpose. The data can help services strategically align themselves better to student need and practitioners shape their interventions to make them more useful and viable. I think studies like PRaCTICED and ETHOS are a huge opportunity for practitioners to look outside their own clinical practice and maintain our professional development in the way key stakeholders expect us to. These studies offer evidence of our value as a specialist service and that what we are doing is valuable from a more externalised measure.’

Carolyn Mumby, who currently chairs BACP Coaching, followed a similar professional trajectory, starting out as a person-centred counsellor working with young people and progressing to train in coaching and leadership. She relates strongly to the ETHOS finding that young people aren’t always comfortable with the person-centred approach. ‘What I found is that they didn’t respond to only being told “this is your space” – they were often coming with a particular problem that they wanted to solve. Some young people do need more of a holding space where they can have that sense of relationship, but what I found when I was running a service for young people was that practitioners were beginning to engage them in a more proactive way – working with them on how to make decisions and resolve problems, giving them information and techniques that they could use to help them move forward.

‘I think young people are often not well supported in terms of their autonomous thinking. School is often really prescriptive and that doesn’t help young people to think for themselves. I approach working with young people with the attitude that they are the expert on themselves and I have some potentially helpful ideas and information that I might share with them with their permission. It is a subtle difference; I am offering information, not giving advice.’

She says coaching is frequently misunderstood as highly directive and technique based. ‘Yes, coaches often have a toolbag of frameworks to use, but we also need to have the discipline to be focused on the person in the room and give them space to think – not reach for this or that tool and think this will fix them. That’s not our job. The tools are a way of helping them to explore things further for themselves. And you are always listening for their strengths and for the resources within them as well as the pain and suffering.’


Keeley Taverner, who runs a flourishing counselling service in west London called Key for Change, says people generally come wanting CBT because that is what they’ve heard about in the media and they want that quick fix. A lot of her clients are aspirant young black professionals who are very focused on resolving whatever is blocking them from achieving their goals in life. But she likes to keep the space open. ‘We are very goal-oriented socially and culturally these days. I am always mindful of how the person perceives therapy and to meet them at that level. But quite often I’ll find they don’t do their homework and then they start feeling uncomfortable about coming to therapy, so I say, “That’s fine, let’s just have a talk”.’

Taverner originally trained in person-centred therapy but had already worked for some time in prisons, running CBT-based programmes for prisoners. ‘That equipped me to see the limitations of that approach. CBT was fine as a starter, but then the prisoners would start opening up and wanting to talk about their lives, where they were coming from, and CBT didn’t recognise that bio-psychosocial context. My training enabled me to think much more broadly and I knew that often people want more.’

However, she is not an advocate of passively sitting and waiting for the client to take the lead. ‘For the lay person, therapy can be an uncomfortable experience, particularly for people who are working class, because of the mystique around it. So it’s good to establish with them what would be a good outcome from therapy for them. But what we end up working on very often isn’t the issue that brings them through the door. Often it’s very much more flavourful – attachment issues, how they were loved or unloved, self-fulfilling prophecies, humiliation – how all of these feed into their personal lives. I articulate to clients that in unknowingness I have seen magnificence unfold that neither I nor the client could have predicted. I see what naturally emerges, which takes me into a space of spontaneity, listening as their lives are unfolded to me and reacting to what they bring.’


Although in some ways the results of both ETHOS and PRaCTICED could be seen as disappointing for the person-centred community, both are rigorous and robust randomised controlled trials, and they provide much-needed evidence of the effectiveness of counselling. PRaCTICED may provide the evidence needed to encourage NICE to give PCET its backing and reinforce its provision in IAPT and other NHS-funded settings. ETHOS shows school-based counselling is a valid and effective intervention with children and young people that helps them achieve their personal and academic goals.

Says Dr Clare Symons, BACP Head of Research: ‘Broadly for both trials, the findings remain very helpful and supportive because both underline the effectiveness of counselling and equip us with evidence that we are criticised for not having much of, by comparison with CBT. While we argue that RCTs are not the single most effective way of evidencing effectiveness, we are hampered if we can’t say we have RCTs that show this too. Our argument is that people should have a choice of effective, evidence-based therapies and are not confined to the one therapy for which there is a greater amount of RCT evidence, and we can do that now.’

However, she points out, ‘Ideally we would have a whole tranche of such studies to counterbalance the evidence amassed for CBT.’ And putting so much focus on comparative effectiveness means we tend to lose sight of the other learning to be gained from the trials, much of it hidden in the qualitative findings, which are still being analysed. ‘Like all good research, these trials throw up lots of different questions.

‘What NICE has created is a culture of dogmatism – “my approach is better than yours” – when the reality is that any theoretical approach to talking therapies involves numerous factors – client, practitioner, model and extra-therapeutic influences. We overlook something very important if we don’t use these findings to question how we work and what we do that is beneficial and where we might be letting our clients down. The PRaCTICED trial invites us to drill down to explore what are the active change agents, which aspects of person-centred counselling promote initial change and why this progress isn’t sustained, and in the ETHOS trial, what is it about the non-directive approach that young people don’t always find helpful,’ she says.

So, what can we conclude? Whether practitioners come from coaching, CBT, person-centred or psychodynamic foundations, there is a clear steer from the research towards a pluralistic approach. The research increasingly tells us that listening to the client and adapting how we work to the preferences of the client is what helps keep them engaged and the work relevant to their needs. Often clients prefer a more directional approach and like having a measure of their wellness or distress. But we need to stay open to the broader psychosocial, family and early environment influences and experiences and leave some space for the unknown and unpredicted. If not, we risk closing down potentially fruitful avenues for exploration and growth.

Next in this issue


1. Barkham M, Saxon D, Hardy GE et al. Person-centred experiential therapy versus cognitive behavioural therapy delivered in the English Improving Access to Psychological Therapies service for the treatment of moderate or severe depression (PRaCTICED): a pragmatic, randomised, non-inferiority trial. The Lancet Psychiatry 2021; 8: 487–499. S2215-0366(21)00083-3
2. Cooper M, Stafford MR, Saxon D et al. Humanistic counselling plus pastoral care as usual versus pastoral care as usual for the treatment of psychological distress in adolescents in UK state schools (ETHOS): a randomised controlled trial. The Lancet Child & Adolescent Health 2021; 5: 178–189.
3. Norcross JC, Lambert MJ (eds). Psychotherapy relationships that work: vols 1 & 2. Oxford University Press; 2019.
4. Norcross JC, Lambert MJ. Psychotherapy relationships that work III. Psychotherapy: 2018; 55(4): 303–315.
5. BACP. What works in counselling and psychotherapy relationships. GPaCP 004. Lutterworth: BACP, 2020.