Main themes
The three themes identified were:
1.Integrating the Keele STarT MSK tool and IT template within consultations
2.Acceptability and suitability of tool items
3.Use of matched treatment options in guiding decision-making
Integrating the Keele STarT MSK tool and IT template within consultations
All GPs described initial difficulties in integrating stratified care into their patient consultations. This was partly due to challenges in completing the tool and deciding the matched treatment options within the typical 10-min consultation timeframe alongside other tasks such as defining the patient’s problem(s) and examination:
It is a little bit clunky, there’s no two ways about it. I was interested to see how long the (video-recorded) consultations were. I noticed they were 13 min and that’s a long consultation for me; I’m usually 9–10 min and wrap up. (Female GP 1).
The additional length added to some consultations by the tool, in particular, was also acknowledged by patients; however, some patients felt that consultations longer than the standard 10-min timeframe were in keeping with the usual ‘thorough’ approach of certain GPs:
It went on a bit longer, probably because of that [i.e. the use of the tool]...but if you see him, he is always thorough, he doesn’t rush you. It’s quite normal for him.
(Male patient, aged 54).
Despite GPs and some patients highlighting the time the tool added to consultations, most of the GPs also reported that having become more familiar with the tool items and scoring within the IT template, the process became faster and more streamlined:
Now that I’m much more familiar with using it, I think I can get through the questions better. To begin with, sometimes I was hitting the wrong box or getting the ones in the wrong spaces, but now it’s fine. (Female GP 2).
There was variation observed amongst patients in that some in fact perceived the purpose of the tool as being to help GPs to move more quickly through consultations, which was seen as being reflective of broader pressures in the healthcare system:
I think generally doctors have got to do that now [i.e. use decision-aid tools], they’ve got to cut to the quick…that is just the state of the NHS [UK National Health Service] isn’t it…[the tool] has obviously been put together with a view to being helpful.
(Female patient, aged 71).
Some GPs felt that the tool distracted them from the identification of potentially serious and urgent ‘red flag’ symptoms, and also was of lesser value given its focus on prognostic indicators rather than diagnosis:
Trying to work out a diagnosis is fundamentally what you are doing. Whenever you get somebody who comes along with pain anywhere in the body, you are working out what’s underneath that pain…the vague sort of questions on the template don’t really help with formulating that diagnosis. (Male GP 1).
There was some variation in views, however, as other GPs reported that the use of stratified care had shifted their management of patients with MSK problems, from thinking diagnostically to taking a more functional approach, taking the onus away from the GP to find a ‘solution’ to the patient’s pain problem:
I feel very strongly that we are trained in a biomedical diagnostic model in general practice…whereas in musculoskeletal problems we are probably better off moving towards a functioning model, which [stratified care] is very much pushing us to do. It encourages people to not think there must be answer and therefore a solution, a one-off thing that a doctor can do. (Female GP 1).
GPs suggested that the tool could sometimes interrupt the flow of the consultation and they reported having to adapt their consultation style to fit around its use:
Because if I use the tool first and then go on the [patient] history, it doesn’t really help me…you are fitting the tool then around what’s happening really, and sadly, it’s just got in the way of the consultation. (Male GP 4).
Some patients felt that the closed question format of the tool’s items restricted them from being able to open up discussion with the GP:
I would have probably liked to have spoken back more but I didn’t, I just said yes or no. I would have liked to have taken part in a conversation...when I’m speaking to [the doctor] normally without this questionnaire, I have more of an input.
(Male patient, aged 61)
However, other patients perceived clinical tools like the Keele STarT MSK tool as being a routine part of consultations. Rather than seeing it as presenting a barrier to pursuing key consultation goals, they saw potential added value in its use:
It’s not surprising they use IT as a tool because you can set all the questions up, can’t you? To me it seemed quite normal, not something I wouldn’t expect or worry about… It’s probably better actually because it’s a structured approach.
(Male patient, aged 65).
Acceptability and suitability of tool items
Five of the nine items included in the development version of the Keele STarT MSK tool were generally seen as acceptable and useful by both GPs and patients. Patients highlighted, in particular, the importance of GPs asking about psychological concerns when patients report severe pain:
Any amount of pain that it’s so severe that you feel it’s 10/10, is going to affect you and is going to make you worry, so those are good questions...the doctor probably needs to know it’s causing anxiety. (Female patient, aged 38).
GPs felt the tool items facilitated the opportunity for patients to raise psychological issues that may be linked to their pain, but which they may otherwise have been reluctant to mention:
I think the tool naturally gives the patient more focus and almost permission to admit that there have been things they’ve been stewing over. (Female GP 5).
Whist GPs reported finding the tool useful in informing their understanding of the patient’s pain, all of the GPs highlighted certain items that they felt did not work as well in the consultation as others; for instance, cases where the wording appeared awkward and did not fit within the natural flow of the conversation. The first of these was: “Do you have any other important health problems?”. GPs reported that patients often replied as though they felt the GP should already know this, or responded by simply listing all of their health conditions without indicating which they considered important:
‘Have you got any other important health problems?’ That’s sometimes tricky because the patient has mentioned a list of things, but that qualitative measure of importance is quite tricky. The patient doesn’t necessarily say which are important, but they just produce a list. And then you’re left qualitatively deciding whether or not you think the patient thinks that’s important. (Male GP 3).
This was also reflected in patients’ views on the same tool item, however they also reported finding it acceptable as they trusted their GP to only ask them things that were relevant:
I did wonder why he was asking those again because he does know me very well. He knows all the problems I’ve got with my health…for many years. But I knew they would be relevant because I know he’s a good doctor. (Female patient, aged 58).
Another tool item highlighted by GPs as problematic was: “In the last 2 weeks, have you stopped enjoying all the things you usually enjoy?”. They reported that patients often responded to this as a functional question, i.e. whether they are physically capable of doing the things they enjoy, rather than one about their mood. This perception was reflected in how patients discussed their views about this item, as in interviews they oriented to this question in terms of function:
It’s bound to affect what you can do isn’t it…I’ve been fairly active but I’m 85 so you can’t expect a lot can you really…I find even going shopping you can’t go without a stick (Female patient, aged 85).
Yeah of course it [the pain] stops you doing things, of course. I particularly noticed it with the dogs because I do try to walk them but I can’t always.
(Female patient, aged 71).
The item asking patients: “Do you think your pain condition will last a long time?” was also highlighted as problematic because patients often responded with “I hope not”:
I would say 70% of people say ‘I hope not’ for that one. That leaves you wondering whether that’s a ‘yes’ or a ‘no’. I think that’s probably my most challenging question…because I think ‘I hope not’ is possibly a positive and so I ask whether that means ‘yes’? (Female GP 2).
Finally, the item: “Do you feel it is unsafe for a person with a condition like yours to be physically active?” was identified as not working well, because GPs were concerned that this puts the idea of safety in the patient’s mind, and could have a nocebo effect:
‘Do you think it’s unsafe?’ sometimes the patient will non-verbally jolt at that one. From our side, I guess we might worry that we’ve introduced the idea that it might be unsafe. That’s one that does occasionally feel clunky. (Male GP 5).
Use of matched treatment options in guiding decision-making
Many GPs reported that they felt the recommended matched treatment options were useful generally in informing clinical management; either as a check-list to confirm they had considered all suitable management options, or in some cases providing suggestions that they may have otherwise overlooked:
So a lot of the time it’s just confirming you’ve been through all the options that are available… it’s an aide memoir. Other times you’re kind of, ‘Actually, no, hadn’t thought of that treatment option’. (Male GP 2).
However, in most of the recorded consultations, the use of the matched treatments was not explained to patients, and this was reflected in SRIs with patients, who generally reported being unaware as to how their answers to the tool items specifically informed subsequent treatment decisions. However, most did feel that their responses to tool items could have value in aiding the GP’s decisions, despite not being explicitly aware that their responses were informing the GP about which treatments to recommend:
I definitely felt the approach was more thorough [when compared to previous consultations], I think it’s an additional extra…it’s good if it helps the doctor to make the correct decision. (Female patient, aged 42).
Several patients reported that they did not feel it necessary to know explicitly how the tool informed the GP’s decision-making, and were happy to rely on the GP’s judgement in recommending treatment options:
You trust your doctor and they’re using the questions to try and find out some degree of how bad your back is and then to decide what action we need to take.
(Male patient, aged 44).
In some SRIs, GPs reported that they had already made up their mind about treatment prior to completing the tool; they therefore completed the stratified care template mainly because they were participating in the pilot RCT, rather than using the tool to guide their decision-making about which matched treatment option(s) to recommend:
Would it change what I do? I don’t think it probably will because I’ve usually decided what I’m going to do with them. [In the recorded consultation] I knew the tool wasn’t probably going to change anything that I did, it was sort of, ‘oh now I’ve got to fill this in’ [the Keele STarT MSK template] (Female GP 1).
In a few of the consultations, GPs opted to refer the patient for an MRI scan (which, given this is an imaging request rather than a treatment option, was not included in the recommended treatment options) and not to select one of the recommended matched options. Referring patients for a scan did not necessarily indicate GPs acting against the recommendations of the tool – as the prognostic stratification was intended to supplement, not replace, a diagnostic approach; nonetheless, some GPs highlighted that in certain instances they saw a scan as being more appropriate than the recommended treatment options:
I guess it wouldn’t always feel terribly comfortable for this type of patient to opt down an immediate management pathway rather than a diagnostic pathway. Not least when she’s already said that she’s worried about something sinister going on. I felt duty bound to deal with that anxiety. She’s expressly asked about imaging, so I felt…it was probably better to image her. (Male GP 3).
In relation to the patient discussed in this extract, the matched SRI with the patient highlighted her sense of reassurance and satisfaction in being referred for an MRI scan following discussion with the GP:
This is a different kind of pain that I’ve got at the moment. It’s much more severe. I’ve [previously] had an x-ray on my hips and my knees and my spine. But he’s sending me for an MRI scan, he explained the difference to me. That will be even more helpful, it’s going to be better. Yes, I’m happy with that. I felt relieved and reassured. (Female patient, aged 61).
Some matched treatment options, e.g. referral to pain management clinics, were identified by both patients and GPs as being of lesser value given the difficulty of accessing such services either because they were unavailable in the local area or because of long NHS waiting lists:
If you’ve got a bad back it’s annoying and it drags you down, it’s painful and you can’t walk properly but you’ve got to wait months and months [for an appointment at the pain services]; it’s a long time to wait. (Male patient, aged 51).
There is the pain management clinic referral on there, isn’t there? I can’t remember really the last time I’ve used that simply because the waiting times are so horrifically long. (Female GP 4).