Want to stay up to date with the evidence?

Sign up here for short evidence-based summaries and other essential info

The Therapeutic Alliance and Low Back Pain

Written by Kieran Macphail on . Posted in Uncategorized

The therapeutic relationship has been defined as;

“A helping relationship that’s based on mutual trust and respect, the nurturing of faith and hope, being sensitive to self and others, and assisting with the gratification of your patient’s physical, emotional, and spiritual needs through your knowledge and skill” (Pullen and Mathias 2010).

All practitioners of the “healing arts” have experienced the importance of this relationship in helping to get the patient to comply with assessment, explain what they feel and want and helping them turn up for follow ups! In recent years more research has been conducted on the therapeutic alliance and just how powerful a positive one can be for improving care.

One systematic review has been conducted. However, Manzoni et al (2018) conducted their systematic review too soon, as by the author’s admission there is insufficient studies to date on the therapeutic alliance. All studies they reviewed examined low back pain patients. Two studies examined therapeutic alliance incentive measures and produced significant improvements in pain. The remaining four studies without alliance incentives showed no clear relationship.

Fuentes et al (2018) conducted an interesting four-arm trial on Interferential, sham interferential, with a good therapeutic alliance and poor therapeutic alliance with low back pain patients. Each session was thirty minutes. In the good therapeutic alliance during the first 10 minutes, each participant was questioned about his or her symptoms and lifestyle and about the cause of his or her condition. Therapists were encouraged to listen actively by repeating the patient’s words and asking for clarifications, tone of voice, nonverbal behaviors such as eye contact, physical touch, and empathy. For example saying, “I can understand how difficult LBP must be for you.” The therapist then stayed in the room during the entire treatment and during the measurement of outcomes. During this time, verbal interaction between the therapist and participant was encouraged. Finally, at the end of the session, a few words of encouragement were given. In the poor therapeutic alliance group, interaction was limited to about 5 minutes during which the therapist introduced herself and explained the purpose of the treatment. Participants were told that this was a “scientific study” in which the therapist had been instructed not to converse with participants. After setting up the interferential the therapist left the room and returned 15 and 30 minutes into the treatment to be present when the tester arrived to conduct outcome assessment. Mean differences on the post intervention NRS were 1.83 cm active interferential and poor therapeutic alliance, 1.03 cm for sham interferential and poor therapeutic alliance, 3.13 cm for the active interferential and good therapeutic alliance, and 2.22 cm for the sham interferential and good therapeutic alliance. Mean differences on PPTs were 1.2 kg, 0.3 kg, 2.0 kg, and 1.7 kg respectively. Thus therapeutic alliance was more effective at reducing NRS and pressure pain thresholds than the interferential. In looking at this study its remarkable how simple the instructions were to develop therapeutic alliance, and likely therapists went further naturally. Nonetheless the results are very impressive when you see sham interferential with good alliance outperform active interferential therapy and poor alliance.

The results of Fuentes et al (2018) suggest that by merely developing an effective therapeutic alliance, a sham treatment or very ineffective one may be beneficial. This may well be part of why many practitioners of methods established as ineffective can protest the effect of their method. They may be seeing the beneficial results of the therapeutic alliance they are developing and attributing their excellent results to their ineffective techniques. Although not discussed here it is likely a better therapeutic alliance will lead patients to over report on improvements in their symptoms as well.

Given that human interactions are difficult to quantify, there’s an argument qualitative research is especially useful in exploring the therapeutic alliance. Stenner et al (2018) conducted a qualitative investigation in to what matters most in physiotherapy consultations to patients. They interviewed 15 patients and physiotherapists. The key themes that emerged were that communication within physiotherapy is underexplored and frequently overlooked, understanding what is important to a person is vital to ensure a positive outcome, people’s issues are often simple but are sometimes voiced in an unstructured way, clinicians need to better support people to elicit ‘what matters to them’ and meaningful conversations may encourage an active role for people in their care.

Holopainen et al (2018) conducted a phenomenographic study to identify and describe the contextual nature of the conceptions of patients with low back pain of their encounters in the health care system. Seventeen patients with chronic or episodic low back pain classified as “high risk” were interviewed in open recall interviews, using videos of patients’ initial physiotherapy sessions that had been recorded previously. They classified themes in four levels: “non-encounters”, seeking support, empowering collaboration and autonomic agency. The key differences between the first and second categories were professionals “being present” and patients starting to understand their low back pain. Between the second and third category, the key aspects were strong therapeutic alliance and the active participation of the patient. Finally, the key differences between the third and fourth categories were the patient being in charge and taking responsibility while knowing that help was available if required. It’s possible to see this as a model for developing the therapeutic alliance as shown in figure 2. .

Figure 1. Based Holopainen et al’s (2018) Findings

In using this framework to evaluate clinical relationships some interesting practical insights can be attained. For example in shadowing a private pain consultant I saw a patient attend who was desperate for the consultant to give her an injection or a different medication to relieve her chronic sub occipital pain. She had an MRI, CT spect, botox, steroid injections and had tried various medications. In putting her through my interpretation of Holopainen’s et al’s (2018) findings it was clear she did not understand her condition and in many ways she was not really present in that she was not ready to listen and engage but knew what she wanted and did not want to hear anything else. There seemed to be no real positive therapeutic alliance between her and any of the many healthcare providers she had seen. She was clearly an active participant in that she was actively seeking out new specialists to try and find a solution but she was unwilling to do any exercises for herself as she had tried that and it did not “work”. She was very much in charge and had seeded no control elsewhere. In terms of locus of control she had an internal locus of control over care seeking but not in treatment. Interestingly though with her care seeking behaviour, she was very autonomous but not in a positive way. Thus with this example it was clear how much education was needed to help her understand her condition, her treatment options and give her a positive therapeutic relationship to help her manage symptoms. Thus Holopainen et al’s (2018) findings could be better constructed as a pyramid where education on the patient’s condition, prognosis and options is a key foundation for a healthy therapeutic alliance, which will lead to a positive autonomy versus an unhealthy autonomy.

Figure 2. Layered Approach To Developing The Therapeutic Alliance Based On (Holopainen et al 2018)

Further understanding of the nature of this education can be gleaned when considering a hypothetical example the patient who has seen a well meaning therapist that educated them on the importance of bracing their core and avoiding rounding or over extending their back. Furthermore their pain consultant had told them they had the spine of an 80 year old. This patient has been educated, may develop a “positive” therapeutic alliance with their therapist and become autonomous, a model patient, in applying this “faulty” model which may in some cases do more “harm” than good. Thus therapeutic alliance must be placed within an, evolving, evidence-based medicine, bio-psycho-social approach.

This understanding helps to shed further light on the findings of Trager et al (2018) who found no benefit for two 30-minute pain education sessions over placebo in acute LBP patients when added to usual care in their RCT. However, placebo was active listening for 30 minutes. It is likely this active listening was developing a strong rapport whereas the education sessions are less likely to. Both of these interventions are part of the process of developing the therapeutic alliance. It is hard to see either as individual isolated interventions when viewed within this context.

In my MSc research (Macphail 2018) I drew the distinction between the clinical reasoning of what is best for the chronic low back pain patient and getting the patient to do what is best for them. Part of this is the therapeutic relationship. Figure 3 shows the iterative cycle that emerged from my interviews with a variety of experienced physiotherapists when describing how they would manage a typical chronic low back pain patient with yellow flags. Building rapport is essentially developing the therapeutic alliance, whilst the whole cycle is the workings of the therapeutic alliance.


Figure 3. Iterative cycle to help patients do what is best for them

All interviews placed a strong emphasis on understanding the patient. There were two elements to this one more focused on the technical aspect and another more focused on understanding the patient’s expectations, beliefs and past experience so as to develop rapport and adapt the assessment and treatment to better suit the patient. For example interview 3 outlined how she would vary her treatment approach based on the patient’s previous experiences.

Interviewee 3: He would say, “I have been to 2 different physio’s and they all
manipulated my back and each time it was terrible afterwards.” So of
course I would go over what I need to do, so I guess again it has a
lot to do with what the patient tells me.

There was a wide range of styles displayed for building rapport. For example interview 5 placed particular emphasis on rapport using humour.

Interviewee 5: I would probably say, “You just keep going until I get bored”, or something like that, to make a bit of a joke of it. Because if I can get that kind of jokey link with him I always feel that I can get better compliance with things.

All physiotherapists interviewed mentioned re-assurance and using the patient’s language. Broadly these were all strategies under rapport building which helped the physiotherapist to get the patient to do what the physiotherapist felt was technically best for them.

The simplest example of interviewees adapting is that they frequently stated they may only give a patient one or two exercises to gain compliance, despite that fact they felt the patient would benefit from more. This same theme shows up through the assessment and treatment process. For example interview 3 explained how she adapted her approach to delivering questionnaires depending on the patient.

Interviewee 3: I would not just want to give them a questionnaire, I
would want to explain to them that I am interested to understand their
problems from a lot of different aspects.

Whilst interview 5 described using manual therapy if that was what the patient expected so that the patient would comply with his more active approach.

Interviewee 5: Just to see if I can get the jump on them in terms of if they expect a manual approach, and you don’t give it to them, then I feel like they don’t listen as well to my active approach.

There was a strong emphasis throughout the interviews on empowering patients to manage their own symptoms, for example.

Interviewee 1: The homework is the most important bit

In particular clinicians described using the clinical audit process to get buy in for a patient to complete their home exercises. Furthermore education was used to empower patients to understand their symptoms. For example interview 3 went as far as to state it was probably the biggest influence on the therapeutic outcome.

Interviewee 3: What I believe is that if I can manage this problem of pain
equals harm, if I can address this successfully then usually it can
be very successful in these patients. But it is probably mainly can I
get his trust and compliance. I guess this is the interaction between
myself and him. It is probably the main thing which I believe makes a
successful intervention or not.

Holopainen et al (2018) focused on understanding the patient’s experience of the relationship, whereas my research (Macphail 2018) focuses on the practitioner’s role. Between these two papers the issues raised by Stenner et al (2018) are addressed and a combined understanding of the two may be useful. Holopainen et al’s work provides a framework for assessing where the relationship with that patient is at, and where it should look to go next. My research provides aspects for the therapist to consider to develop this relationship and how to structure what to work on in a session. Furthermore this may stimulate therapists to consider other factors not covered that improve the therapeutic alliance. Practically this might mean that a session plan could involve, pain education, re-assurance, exercise and an aim for developing the therapeutic alliance. For example if the figures presented here are consulted before the session and you feel the therapeutic alliance is at the seeking support stage. You may then consult figure 3 and consider where you are lacking. For example, do you need to adapt your communication, assessment and treatment better to the patient? Were they expecting hands on treatment, and you “just sat and talked” about pain in the first session? In the second session you could adapt this to do an active manual therapy technique whilst discussing pain education. Alternatively you might feel you don’t fully understand the patient’s beliefs and expectations yet and look to understand this better. Tentatively I suggest a plan for developing the therapeutic alliance should be in the session plan for each patient.

There is a long way to go before we can really objectively assess the therapeutic alliance, understand it algorithmically and develop it optimally. It is clearly a powerful intervention in itself and may explain part of the benefit and success of some interventions based on clearly ineffective methods. Figure 2 based on Holopainen et al’s (2018) findings gives a method of evaluation of the therapeutic alliance. Furthermore this can be used as way of planning how to develop the therapeutic alliance in the clinical setting. Figure 3 outlining part of my MSc research (Macphail 2018) provides an overview of the different factors a therapist consider in improving their therapeutic alliances. It is tentatively suggested a combination of the two approaches is used to assess and plan the development of therapeutic alliance as part of session planning.



Fuentes, J., Armijo-Olivo, S., Funabashi, M., Miciak, M., Dick, B., Warren, S., Rashiq, S., Magee, D.J. and Gross, D.P., 2014. Enhanced therapeutic alliance modulates pain intensity and muscle pain sensitivity in patients with chronic low back pain: an experimental controlled study. Physical therapy, 94(4), pp.477-489.

Holopainen, R., Piirainen, A., Heinonen, A., Karppinen, J. and O’Sullivan, P., 2018. From “Non‐encounters” to autonomic agency. Conceptions of patients with low back pain about their encounters in the health care system. Musculoskeletal care.

Macphail, K. (2018) The Why, How and What, Physiotherapist Management of Chronic Low Back Pain Patients with Yellow Flags. In Press

Manzoni et al (2018) The role of the therapeutic alliance on pain relief in musculoskeletal rehabilitation: A systematic review. Physiother Theory Pract. 2018 Feb

Pullen Jr, R.L. and Mathias, T., 2010. Fostering therapeutic nurse-patient relationships. Nursing made incredibly easy, 8 (3), 4.

Stenner, R., Palmer, S. and Hammond, R., 2018. What matters most to people in musculoskeletal physiotherapy consultations? A qualitative study. Musculoskeletal Science and Practice, 35, pp.84-89.

Traeger, A.C., Lee, H., Hübscher, M., Skinner, I.W., Moseley, G.L., Nicholas, M.K., Henschke, N., Refshauge, K.M., Blyth, F.M., Main, C.J. and Hush, J.M., 2018. Effect of intensive patient education vs placebo patient education on outcomes in patients with acute low back pain: A randomized clinical trial. JAMA neurology.

Trackback from your site.

Follow us