Targeted therapy for acute whiplash gets it in the neck (again)

What to do about whiplash? Trials have historically produced disappointing results across the board for our management strategies. As is so often the case the interpretation of those results can be broadly divided into 2 camps. One camp (often rather small) who accept that current treatments are not really doing the job, the other who find fault with the trials and feel that they have not provided the fair test to which they claim.

The arguments are familiar, not wildly unreasonable, and by no means unique to whiplash. Trials did not find an effect because they applied “one size fits all” treatments to a wildly varied patient group, because they failed to consider subgroups in the population, because the treatment did not reflect best practice or “what I would do”. Luckily most of these are hypotheses that can be tested.

Prof Gwen Jull and colleagues have done just that. Hypothesizing that many of these arguments against existing trial results may be valid, they have just published an RCT of individualised targeted care versus usual care for acute whiplash. The targeted care group received a 10 week course of medical & pharmacological management, physiotherapy (including individualised exercise, manual therapy, and other modalities), and psychological care, with a specific emphasis on targeting post –traumatic stress reactions. All treatment was clinician-led and individually targeted,  no “one size fits all” here. The usual care group sought care as usual via their GPs, including visits to physiotherapy, chiropractic etc.

The findings are something of a bombshell for this model of “best practice”. There was simply no difference between the two groups at 6 month or 12 month follow-up. No difference in recovery rates, no difference in disability levels, no difference in pain levels.  This study quite reasonably hypothesized that multi-modal care, delivered in a way unique to each patients needs would result in better outcomes than the norm. The results do not support this.

Is there an escape route in the design or process of the trial itself?  Not really, since the trial was robustly designed, the treatment package was well reasoned and comprehensive and while the trial is a little small (n=101), there was no trend to be seen towards an improvement in the intervention group (in fact the trend appears to move in the opposite direction).  It is also worth remembering that these pragmatic trial designs tend to introduce an unavoidable bias which favours the intervention.

These results are not entirely unique. The recent MINT trial found that an early “active management” intervention delivered in accident and emergency had no beneficial impact and that a course of physiotherapy for those who remained symptomatic 3 weeks later had only a modest and short term benefit that disappeared quickly. These are two well designed and well conducted trials, with a familiar message. Then there is the context offered by some larger population based studies which have variously shown no improvement in recovery rates from multi-disciplinary interventions and that more care may lead to worse outcome (see here and here).

This result is important because it reveals, even more starkly, an uncomfortable truth. What we think may be “the right thing” for the management of whiplash is not resulting in the right outcomes.  While we have some predictors of poor outcome, our efforts to modify them are not particularly successful. Current therapy for whiplash does not appear to be impressively therapeutic. More does not always mean better.

About Neil O’Connell

Neil O'ConnellAs well as writing for Body in Mind, Dr Neil O’Connell, (PhD, not MD) is a researcher in the Centre for Research in Rehabilitation, Brunel University, West London, UK. He divides his time between research and training new physiotherapists and previously worked extensively as a musculoskeletal physiotherapist.
He also tweets! @NeilOConnell
Neil’s main research interests are chronic low back pain and chronic pain more broadly with a focus on evidence based practice. He has conducted numerous systematic reviews including some for the Cochrane Collaboration. He also makes a mean Yorkshire pudding despite being a child of Essex.
Link to Neil’s published research here. Downloadable PDFs here.


Jull G, Kenardy J, Hendrikz J, Cohen M, & Sterling M (2013). Management of acute whiplash: A randomized controlled trial of multidisciplinary stratified treatments. Pain PMID: 23726933

Côté P, Hogg-Johnson S, Cassidy JD, Carroll L, Frank JW, & Bombardier C (2007). Early aggressive care and delayed recovery from whiplash: isolated finding or reproducible result? Arthritis and rheumatism, 57 (5), 861-8 PMID: 17530688

Côté P, Hogg-Johnson S, Cassidy JD, Carroll L, Frank JW, & Bombardier C (2005). Initial patterns of clinical care and recovery from whiplash injuries: a population-based cohort study. Archives of internal medicine, 165 (19), 2257-63 PMID: 16246992

Cassidy JD, Carroll LJ, Côté P, & Frank J (2007). Does multidisciplinary rehabilitation benefit whiplash recovery?: results of a population-based incidence cohort study. Spine, 32 (1), 126-31 PMID: 17202903

Lamb SE, Gates S, Williams MA, Williamson EM, Mt-Isa S, Withers EJ, Castelnuovo E, Smith J, Ashby D, Cooke MW, Petrou S, Underwood MR, & Managing Injuries of the Neck Trial (MINT) Study Team (2013). Emergency department treatments and physiotherapy for acute whiplash: a pragmatic, two-step, randomised controlled trial. Lancet, 381 (9866), 546-56 PMID: 23260167


  1. Just to thank all of the contributors to this discussion – it is, as Neil observes, a really difficult one – and we need intelligent reasonable people discussing it, which is exactly what we seem to have here. I am very grateful to all of you and, Neil – another cracking post.

  2. Neil O'Connell says

    Hi Jan,

    I guess what I meant was that after ruling out serious specific pathology in the acute stage we would give reassurance , advice to keep moving, advice re: analgesia. As I state this doesn’t seem to make much positive difference but it seems a bare minimum that we could offer.

  3. Jan Carstoniu says

    I should clarify. I did not mean to lump whiplash (or chronic pain) together with DM and RA as fitting into the same medical/biological model and thus should be treated with a similar medical /biological approach.

    I was referring to the idea that chronicity is their shared attribute and that this should inform whatever kind of intervention we apply, regardless of the model of care we choose. In fact some of the success in DM and RA can be attributed to non-medical/biological interventions.

    It is indeed ironic to go back to looking at treatment again since bad treatment led to a search for prevention but without more understanding of chronic WAD and in light of the recent findings I’m not sure what other options we have but more research into basic mechanisms and trying to determine what helps.

    I agree with Neil’s suggestion about acute care following whiplash but I’m interested in what constitutes” good information and the tools they need”. In my own practice I rarely see acute injuries but see many with chronic pain. I’d like to know your thoughts, Neil. Thanks.

  4. Neil O'Connell says

    Thanks all for some really thoughtful comments. Its refreshing to see some genuine reflection on this most difficult finding, where often we would see efforts to avoid the possible truth emerging from these data.

    John, if I read you right and interpret “incorporating neuroscience” as dropping the focus on tissue damage and structural causes, and instead delivering pain education, get moving etc then I take your point, to a point. The problem is that some of these studies just gave good advice about moving, reassurance etc: key aspects of such an approach, yet still the outcome was bad.

    I do agree though that other chronic illnesses such as DM and RA may not serve as great models (bringing in your point Jan). The difference is for those conditions we have a much stronger biological model and much more effective strategies for ongoing management that have really improved prognosis. We haven’t managed that for whiplash, or indeed most chronic pain yet. Ironically the call to try to prevent chronicity before it develops arose in part from our failure to improve those already chronic.

    Geoff, perhaps STaRTback does offer a model. I personally feel the modest effects do not suggest a similar approach would make a big impact though it would be great to be proven wrong. I totally agree that for now less is more.

    This really raises the issue of what to do when nothing seems to work. More so if the very act of offering even enlightened care, might plausibly negatively impact on outcome (I realise I am stretching the evidence a tad here).

    Perhaps more than any other chronic pain condition whiplash is a medical-cultural-social-economic mess, full of bad information, perverse incentives and a barrel load of industries enjoying a feeding frenzy. Stuck in the middle are patients, made worse by the industry itself. Maybe the real challenge for physio (and medicine) is to boldy make sure we are not part of the problem, encourage a minimum of contact, enough to give patients good information and the tools they need (even if that isnt more effective than usual care).

    The evidence suggests this would be no worse than what currently occurs, but it would also take the big leap of removing a substantial link from a complex and unhelpful chain. Maybe at this point thats the most positive thing we could do.

    But I am speculating massively, and also dreaming. Money, after all, talks.

    ian stevens Reply:

    Perhaps more than any other chronic pain condition whiplash is a medical-cultural-social-economic mess, full of bad information, perverse incentives and a barrel load of industries enjoying a feeding frenzy. Stuck in the middle are patients, made worse by the industry itself. Maybe the real challenge for physio (and medicine) is to boldy make sure we are not part of the problem, encourage a minimum of contact, enough to give patients good information and the tools they need (even if that isnt more effective than usual care).

    Neil , you have nailed it ……agree with you totally and MIllenson (whiplash and other useful disorders) . Much of the ‘spinal industries’ are as Gordon Wadell suggested years ago an iatrogenic disaster . The stock car driver research and lack of disability shows what a highly sociological disorder WAD is. I am not suggesting for a minute t hat people are not suffering but until you attempt to disentangle the mess ,nothing much is likely to change …Highly skilled therapists will see and ‘treat’ dysfunction to the neck stabilisers and trigger points, others may see psychological adjustment disorders , others centralised pain syndromes and like the blind men and the elephant all may be partially correct but fail to see the elephant ….
    I was told litigation makes no difference and working in the east end of glasgow people used to drive into each other (years ago for claims) , others due to over investigation probably devleoped iatrogenic disability and many did and probably still bounce between providers ….
    Many people are making a good living out of the disorder and I agree patients are often stuck in the middle …

  5. John Ware, PT says

    I think we should be careful about lumping those with non-pathological pain in with those who have documented medical conditions, like RA and DM. It seems counter-intuitive to me to make a charge of medicalization and then turn around and suggest to approach chronic pain from a traditional medical vantage point as we have other conditions associated with chronic pain.

    I think we need to start by educating the professionals who encounter patients with these painful conditions. That’s when a lot of the misconceptions and “thought viruses” are planted and re-inforced. There have been some good studies looking at the woeful neuroscience education with respect to pain in both medical and physical therapy programs. Here’s one on the pain curriculum in North American medical schools that was published just a couple of years ago: The Journal of Pain, Vol 12, No 12 (December), 2011: pp 1199-1208.

  6. Jan Carstoniu MD FRCPC says

    As a clinician, I have no arguments with the conclusion that current medical practice for whiplash and chronic pain in general is terrible for the most part. This study is really discouraging however because it indicates that nothing seems to help, at least over the short term of 10 weeks.

    What seems rational to me at this point is that following whiplash we should be guided by the principle of first doing no harm and then doing our best to relieve pain. We now have a lot of evidence suggesting that other treatment does not prevent the development of chronic WAD.

    Perhaps we need to look closer at better treatment for those who go on to develop chronic pain since we are doing such at bad job at preventing it. I know this seems nihilistic but given current evidence I’m not sure what else we can do.

    Finally, given that chronic pain is, well, chronic, perhaps we need to shift our mindset away from curing someone’s pain with our various treatments to determining what helps affected individuals achieve optimal relief and functional capacity. That’s how we deal with diabetes and rheumatoid arthritis. We’re doing a better job with those incurable conditions than we are with whiplash.

  7. Another great post Neil. The sub-grouping argument makes some sense, but when I think of targeted treatment, I think more about sub-grouping to eliminate the noise in a sample. Individualized multi-modal treatment does not necessarily eliminate the noise from patients who will not respond to treatment due to good prognosis (i.e. they will respond to anything, even nothing). The Hill et al study in Lancet got to this, albeit with only modest results, but I appreciate the strong (though not new) message that a substantial proportion of patients require minimal intervention. The body of literature suggests exquisite methodologies is unlikely the answer. It would be nice to see more critical perspectives such as yours where instead of assuming every patient requires individualized treatment, I wonder if it better to assume a patient needs minimal intervention until proven otherwise.
    I really appreciate your critical lens!

  8. Niamh Moloney says

    Neil, thanks for highlighting this issue once again. It seems to me we really need to shift away from our traditional practice models in many cases or we’re going to continue going around in circles. In my mind, one clear need is to identify those who actually dis-improves with our current intervention strategies and perhaps by striving to improve our understanding of these patients we can develop new models of care. But I wholeheartedly agree, more is not always better and we certainly need to be cognizant of the negative impacts of ‘medicalization’ and ‘over-servicing’ as previously outlined.

    Neil O'Connell Reply:

    Thanks Niamh,

    So you would have a priority trying to identify folk who our interventions actively make worse, and see if they can be identified? That would be truly revolutionary in the therapies I think.

    Niamh Moloney Reply:

    Hi Neil,
    Yes, I do think we need to acknowledge that we may be making some people worse and it seems negligent to not try to identify them. First, do no harm, right? But identifying what the characteristics of these people are, might allow us to develop new treatment models. I’m sure this line of thinking isn’t new as many researchers have been focusing on sub-grouping and understanding mechanisms of pain so we can target treatment more effectively.
    But, are we comfortable with accepting that some people not only don’t do well with physiotherapy but might actually get worse with it?

    There seems to be a lot of evidence that sensory hypersensitivity is an issue for at least some patients with whiplash (and other musculoskeletal conditions). One mechanism of central sensitisation is disrupted pain modulation e.g. to exercise; so sensitised patients can potentially become more sensitised with exercise (see review by Nijs et al. 2012). We routinely prescribe exercise but are we taking the presence of sensory hypersensitivity adequately into account?

    The same could be said for manual therapy; we have some evidence that manual therapy can stimulate the endogenous opioid system and produce an analgesic effect but what if that manual therapy is ‘noxious’? Are we then further sensitising an already sensitised system?

    Pivotal to this, I believe we should re-think the idea of ‘post-treatment soreness’. My understanding is that it is largely accepted as a consequence of treatment in physiotherapy; this might be acceptable if you want to mobilise a hypomobile ankle that is not particularly painful or sensitised but I believe it is unacceptable as a side effect in many patients, particularly those who are sensitised.

    Neil O'Connell Reply:

    Thanks Niamh,

    Considering subgroups you hit upon an oft ignored truth in the subgrouping argument. If no effect is observed and one group are improving then an equal size group must be being actively worsened by therapy. The alternative is that the therapy is essentially inert. Most subgroupers so the human thing of focusing on the positive (responders) but I agree a focus on worseners is interesting.

    With regards sensitization as a mechanism for worsening I also agree. However in this trial sensitivity was factored in to the intervention and where identified it was stipulated that all rehab methods must not be pain provocative. So perhaps not such a big player here..

    John Ware, PT Reply:

    You may want to take a look at the pilot RCT by Ryan et al ( where they compared pain neuroscience education to pain neuroscience education plus exercise for patients with chronic LBP. Interestingly, the patients who received the exercise did *worse* than those who received the education alone. In the discussion, they suggested that the attendance at the clinic may have provided a “medicalization” effect that nullified whatever improvements that had resulted from the education component. Makes sense to me.

    The exercises, as I recall, were based on recommendations compiled by the British NHS. So, when we talk about less being more, does that include government intervention in health care?

    Niamh Moloney Reply:

    Thanks John, will have a look.

  9. Neil O'Connell says

    There was no difference between the groups in the proportion of folk with compensation clains lodged. So it wouldn’t explain the lack of effect between the groups.

    It is however a clear real world issue that might affect prognosis in WAD, but even accepting that, our interventions would need to be effective despite that, in order to be considered genuinely effective.

    John Ware, PT Reply:

    I’ll see your “conservative therapies are not helping WAD” and raise it with “we’re professionally derelict in the treatment of persistent pain problems”. I think the useless and often expensive and protracted therapies foisted on patients with WAD is emblematic of how poorly modern medicine in general and the Physical Therapy profession in particular have incorporated current neuroscience into the clinical reasoning and treatment planning for patients with persistent pain problems.

    As Esther Williamson found in her thorough and incisive analysis within the MINT trial, it was the lack of confidence that a patient with WAD would improve that best predicted poor outcome at 6 months. This tells me that the impact of “medicalization” upon the WAD sufferer is profound, and it’s a wonder that anyone can make his way to recovery from whiplash with all this ignorance, good intentions, self-interest and outright greed providing obstacles every step of the way.

    We must do better.

  10. Did this study include payor source and whether or not there was third party litigation? I realize two therapy approaches were compared but perhaps there is another issue influencing whatever approach is utilized.