health systems

Barriers to good pain rehabilitation


This is a long…… read

ooops, sorry, not.

Low back pain is, we know, the greatest contributor to days lived with disability (Rice, Smith & Blyth, 2016). And no-one anywhere in the world has found a good mix of services to reduce the number of days lived with disability as a result of this problem. And yet billions of dollars are used to fund research into the many contributors to a shift from acute low back pain to ongoing disability associated with low back pain.

At the same time, treatments that directly target disability, rather than pain (a target considered the most important outcome by Sullivan and Ballantyne, 2016) are difficult to access, by comparison with surgical solutions (or pharmaceutical or procedural). Overuse of unhelpful treatments is thought to occur when treatments are offered that are ineffective, pose high risks of harm, or where the balance between harms and benefits varies considerably (Brownlee, Chalkidou, Doust, Elshaug, Glasziou, Heath et al, 2017). Non-pharmcological, non-surgical, and non-procedural treatments fall into the large amorphous group of treatments often delivered by allied health clinicians.

When asked, clinicians (aka doctors, in this instance) were found in one study, to fall into three groups: Multimodel/Aggressive = 14%; Psychosocial/Nonopioid = 48%; and Low action = 38% (Phelan, van Ryn, Wall & Burgess, 2009). On the face of it, this looks reasonably good except when we have a look at what the numbers mean. The Low action group were more likely to move from prescribing one opioid to another opioid, suggest that patients take their opioids more regularly and at a higher dose, use a short-acting opioid, treat with a long-action opioid on a fixed basis, and order more diagnostic testing. The Psychosocial/Nonopioid group were more prepared to use psychosocial clinical approaches, refer elsewhere, not use opioids, and were happier to discuss emotional assessments, refer for physiotherapy, mental health evaluation, refer to pain specialists and so on. But their rationale was not to prescribe opioids because of their concern about tolerance, diversion and addiction. Finally, the Multimodal/Aggressive group combined the approaches of both the previous groups and the authors thought this most closely represented “the multimodal treatment strategy endorsed by proponents of the biopsychosocial model” (Phelan, van Ryn, Wall & Burgess, 2009, p. 1274).

I’m interested in why these clinicians chose the kinds of treatments they did – The authors analysed various aspects of each group and found the belief that pain has a physical cause (MA: 56%; PNO: 35%; LA: 31%; P chi-square < 0.01) and positive attitudes towards using opioids (MA: 3.5 [0.8]; PNO: 3.2 [0.8]; LA: 3.4 [0.8]; Pf-test < 0.01) differed by class. Added to this, the physician’s concern also different by class: High concern about drug use/abuse (MA: 39%; PNO: 47%; LA: 19%; Pchi-square < 0.001), and Concern about effectiveness of opioids (MA: 3.6 [1.0]; PNO: 3.4 [1.0]; LA: 3.2 [0.9]; Pf-test = 0.02). Adequate consultation/referral resources also differed by class (MA: 59%; PNO: 65%; LA: 42%; Pchi-square < 0.001).

Let’s look at this – inaccurate beliefs about pain and attitudes towards opioids featured in the choices made by doctors, while feeling there were inadequate consultation and referral resources were especially featured by the very group using these resources most readily!

Treatments that are available but are underused are as problematic for our health system (and the people seeking help) as over-using inappropriate treatments. Glasziou, Straus, Brownlee, Trevena, Dans, Guyatt, Elshaug & Janett et al (2017) describe four stages where people may not get or use helpful treatment:

  1. a total or partial lack of access to health care (because the system does not offer coverage or patients are unable to reach or pay for available care, or both);
  2. unavailability of effective services within the local health-care system;
  3. a failure of clinicians to deliver or prescribe effective, affordable interventions; and
  4. a failure of patients to commence or adhere to effective, affordable interventions. ” (p. 171).

As I look at these four stages, I can see that in New Zealand, 1, 2, 3 and or course 4 all apply. The ones that concern me most are the first three. Why are effective treatments not widely available? Think about the lack of pelvic pain services for men and women (yes, men have pelvises too) where, instead of evidence-based treatment, these people are offered repeated ineffective, invasive and non-evidence-based “exploratory” endoscopic surgeries. Who makes decisions not to fund so-called conservative interprofessional pain rehabilitation?

Why are ineffective services available despite evidence showing their lack of usefulness? Who makes decisions to continue funding ineffective treatments? What vested interests, what power relationships value the status quo even when it’s not helping people?

Why are clinicians failing to deliver effective, affordable treatments? Why is that GPs in New Zealand still don’t follow back pain guidelines, continue to offer imaging “for reassurance” (despite evidence that this kind of reassurance is not reassuring, Linton, McCracken & Vlaeyen, 2008), and fear talking about the psychosocial impact of persistent pain on people? (Darlow, Dowell, Baxter, Mathieson, Perry & Dean, 2013).

Before I have a bunch of aggrieved doctors jump on me for “blaming them”, let me assure you that I have incredible respect for general practitioners. They have an extraordinarily difficult job to do – and we don’t really understand why adopting guidelines is so difficult. Except… Webster, Courtney, Huang, Matz and Christiani (2005), Coudeyre, Rannou, Tubach, Baron, Coriat, Brin et al (2006); Crawford, Ryan & Shipton (2007), Hush (2008), Somerville, Hay, Lewis, Barber, van der Windt, Hill & Sowden (2008), Corbett, Foster, & Ong (2009), Finestone, Raveh, Mirovsky, Lahad & Milgrowm (2009), Fullen, Baxter, O’Donovan, Doody, Daly, & Hurley (2009), Phelan, van Ryn, Wall, M & Burgess (2009), MacNeela, Gibbons, McGuire & Murphy (2010), Williams, Maher, Hancock, McAuley, McLachlan, Britt, Fahridin, Harrison & Latimer (2010), Azevedo, Costa-Periera, Mendonca, Dias & Castra-Lopes (2013)…. OK I got tired of listing the many, many studies showing how deficient primary care of low back pain is. And that’s not even up to 2019!

Of course I could point the finger at many other health professionals as well – similar findings for physiotherapy, orthopaedic surgery, even pain services…

So WHY are unhelpful approaches so sticky? Why doesn’t practice change?

Me… wondering why

Here are my off the top of my head ideas (ideas based in reading a lot). Funding and policy for funding is, in New Zealand, governed by two main agencies: ACC (accidental injury) and Ministry of Health but devolved to local DHBs. Who sits on the decision-making committees? How many of these august people come from allied health backgrounds? What is the understanding of pain within these committees?

How do people get to be appointed to the committees and policy-making positions? Who makes those decisions? What are the hidden (and maybe not so hidden) vested interests within those groups? How many have holdings in medical companies? What exposure do these people have to evidence and to allied health?

Why do clinicians carry on with unhelpful approaches? I think there’s a lot to uncover: limited time, the push to replace face-to-face interaction with apps (when people really crave interaction!), lack of funding, fear of how to handle difficult conversations (possibly based on lack of training, but maybe because of a fear that the conversation will be misinterpreted – in turn, maybe that fear is because the clinician still holds a dualist model of pain), limited knowledge of what other clinicians offer, ties between companies stitching up who can be referred where, desire not to be different, and not to take more time….

Disheartened? Maybe it’s winter, but while there is a zeitgeist amongst physiotherapy believing that “we’ve reached a turning point” and aha! we will now move forward as enlightened pain rehabilitation beings! I’m inclined to look at the waves of change that have occurred over my clinical lifetime. The enthusiasm for multi and inter-professional treatments, the fervour for early intervention to prevent disability, the explain pain and it will lead to great outcomes… All good things, yet still the fundamental problem is, I believe, adherence to a mainly biomedical or biophysical model for what is a problem experienced by humans in all their myriad complexity. Until that reality is firmly fixed within the system (not just GPs!) I think change will be incrementally slow.

Either that, or I need to learn how to play the political game and get appointed to one or more of the decision-making committees in ACC, MoH and DHB-land.

Azevedo, L. F., Costa-Pereira, A., Mendonça, L., Dias, C. C., & Castro-Lopes, J. M. (2013). Chronic pain and health services utilization: is there overuse of diagnostic tests and inequalities in nonpharmacologic treatment methods utilization? Medical Care, 51(10), 859-869. doi:10.1097/MLR.0b013e3182a53e4e

Brownlee, S., Chalkidou, K., Doust, J., Elshaug, A. G., Glasziou, P., Heath, I., . . . Korenstein, D. (2017). Evidence for overuse of medical services around the world. The Lancet, 390(10090), 156-168. doi:https://doi.org/10.1016/S0140-6736(16)32585-5

Corbett, M., Foster, N., & Ong, B. N. (2009). GP attitudes and self-reported behaviour in primary care consultations for low back pain. Family Practice, 26(5), 359-364.

Coudeyre, E., Rannou, F., Tubach, F., Baron, G., Coriat, F., Brin, S., . . . Poiraudeau, S. (2006). General practitioners’ fear-avoidance beliefs influence their management of patients with low back pain. Pain Vol 124(3) Oct 2006, 330-337.

Crawford, C., Ryan, K., & Shipton, E. (2007). Exploring general practitioner identification and management of psychosocial Yellow Flags in acute low back pain. New Zealand Medical Journal, 120(1254), U2536.

Darlow, B., Dowell, A., Baxter, G. D., Mathieson, F., Perry, M., & Dean, S. (2013). The Enduring Impact of What Clinicians Say to People With Low Back Pain. Annals of Family Medicine, 11(6), 527-534. doi:10.1370/afm.1518

Finestone, A. S., Raveh, A., Mirovsky, Y., Lahad, A., & Milgrom, C. (2009). Orthopaedists’ and family practitioners’ knowledge of simple low back pain management. Spine, 34(15), 1600-1603.

Fullen, B. M., Baxter, G. D., O’Donovan, B. G., Doody, C., Daly, L. E., & Hurley, D. A. (2009). Factors impacting on doctors’ management of acute low back pain: a systematic review. European Journal of Pain: Ejp, 13(9), 908-914.

Fullen, B. M., Baxter, G., Doody, C., Daly, L. E., & Hurley, D. A. (2011). General practitioners’ attitudes and beliefs regarding the management of chronic low back pain in Ireland: A cross-sectional national survey. The Clinical Journal of Pain, 27(6), 542-549.

Hush, J. M. (2008). Clinical management of occupational low back pain in Australia: what is the real picture? Journal of Occupational Rehabilitation, 18(4), 375-380.

Linton, S., McCracken, L., & Vlaeyen, J. (2008). Reassurance: Help or hinder in the treatment of pain. Pain, 134(1–2), 5-8. doi:http://dx.doi.org/10.1016/j.pain.2007.10.002

MacNeela, P., Gibbons, A., McGuire, B., & Murphy, A. (2010). “We need to get you focused”: General practitioners’ representations of chronic low back pain patients. Qualitative Health Research, 20(7), 977-986.

Phelan, S. M., van Ryn, M., Wall, M., & Burgess, D. (2009). Understanding primary care physicians’ treatment of chronic low back pain: The role of physician and practice factors. Pain Medicine, 10(7), 1270-1279. doi:http://dx.doi.org/10.1111/j.1526-4637.2009.00717.x

Rice, A. S., Smith, B. H., & Blyth, F. M. (2016). Pain and the global burden of disease. Pain, 157(4), 791-796.

Somerville, S., Hay, E., Lewis, M., Barber, J., van der Windt, D., Hill, J., & Sowden, G. (2008). Content and outcome of usual primary care for back pain: a systematic review. British Journal of General Practice, 58(556), 790-797.

Sullivan, M. D., & Ballantyne, J. C. (2016). Must we reduce pain intensity to treat chronic pain?. Pain, 157(1), 65-69.

Webster, B. S., Courtney, T. K., Huang, Y. H., Matz, S., & Christiani, D. C. (2005). Physicians’ initial management of acute low back pain versus evidence-based guidelines. Influence of sciatica. Journal of General Internal Medicine, 20(12), 1132-1135.

Williams, C. M., Maher, C. G., Hancock, M. J., McAuley, J. H., McLachlan, A. J., Britt, H., . . . Latimer, J. (2010). Low back pain and best practice care: A survey of general practice physicians. Archives of Internal Medicine, 170(3), 271-277.

Wait and see…when do we “escalate” care for low back pain?


Prompted by reading a paper by Linton, Nicholas and Shaw (in press), today’s post is about various service delivery models for low back pain and not the content of back pain treatment.

Service delivery in New Zealand is assumed to be based on getting most bang for the buck: we have a mainly socialised healthcare system, along with a unique “no fault, 24 hour” insurance model for accidents whether at work or elsewhere, which means market forces existing in other countries are less dominant. There are, however, many other influences on what gets delivered and to whom.

Back to most bang for buck. With a limited healthcare budget, and seriously when is there ever NOT a limited budget in health, it would make sense to a thinking woman for healthcare to focus on high value treatments. Treatments that have large impact and are low cost. In low back pain, the techno-fix has limited application. Things like costly surgical approaches (synthetic disc replacements, fusions to stop vertebral movement) should be reserved for only those with clear indications for the procedure, and given on the basis of clinical need rather than in response to a distressed person. The outcomes just are not all that great (see Maher, Underwood & Buchbinder (2017) for a good review of nonspecific low back pain).

High value and low cost treatments are typically delivered by low status clinicians. Those “nonmedical” people like occupational therapists, physiotherapists, osteopaths, chiropractors and the like. Maybe it’s for this reason that these treatments are relatively poorly funded. We lack lobby power.

Back to service delivery models. Currently in Christchurch, where I live, there is a health pathway (in other words, a service delivery model) developed in collaboration with GP’s, physiotherapists, osteopaths and secondary care. The model adopted applies to ALL episodes of low back pain, and uses the STarTBack tool to triage those who may need more intensive treatment under a biopsychosocial model (mainly because of the additional risk psychosocial factors pose for these people), and to continue with treatment as usual for those with lower risk as measured by this tool.

After about six weeks, if the person hasn’t responded to treatment, clinicians are meant to refer the person to a team for review and to see whether additional treatments or another pathway might be appropriate. Unfortunately, there is no indication of the makeup of that team, and no obligation for the clinician to send the person to it. I’m not sure about clinical audit of this pathway, and again this isn’t clear.

One of the problems (amongst many) with this approach is that six weeks without responding to treatment and the time needed after this to review the file, then be referred elsewhere is a very long time to someone experiencing back pain. A very long time. By six weeks it wouldn’t be surprising if the person’s sick leave is gone. If they’re receiving ACC the processes will have kicked in, but for the person who has typical grumpy back symptoms without an “accident” initiating it, there may be nothing.

Linton, Nicholas and Shaw point out that all of the triaging approaches for low back pain hold assumptions. The three are stepped care (begin with low intensity, once that hasn’t helped progress to more intensive and so on); stratified (triage those with high risk, and treat them accordingly, while low risk get lower intensity treatment); and matched care (treatments are administered according to an algorithm based on grouping people with similar characteristics).

Stepped care

The assumptions of stepped care include that people with basic acute low back pain will recover relatively easily, while those who need more help will be fine waiting for that additional level of care. There’s an assumption that factors leading to chronic disability occur in stages – the longer a person waits the more risk factors will appear – but this isn’t actually the case. Many people present with risk factors from the very beginning (and they can be identified), while waiting only allows those problems to be cemented in place. At the same time, we know acute low back pain is quite a rare thing: most people will have their first bout of back pain in adolescence, and will have learned good and not so good habits and attitudes from that experience. Another assumption is that duration of back pain doesn’t harm, but we know delayed attention to risk factors for chronicity is harmful. Stepped care can be useful because it’s efficient, easy to implement and overtreatment is less likely – but what about the person who appears with all the risk factors evident from the beginning? These people may not get adequate or appropriate treatment from the outset.

Stratified care

In stratified care, treatment is provided according to the category of risk the person presents with, maybe circumventing some of the problems from stepped care. Stratified care assumes we’re able to identify risk factors, and that they are stable from the outset rather than changing over time. It also assumes that risk factors exert a cumulative effect with more risk factors meaning greater risk. BUT while screening can identify some risks, and those at low risk get more adequate treatment while higher intensity treatment is given to those with more risk, this approach doesn’t identify underlying mechanisms, and more comprehensive treatments addressing specific issues may not be provided. This approach may not even consider the impact of workplace factors, family dynamics, social and recreational issues. It’s also pretty challenging to implement as I think the Christchurch example demonstrates.

Matched care

In matched care risk factors are identified and treatments are matched to the person’s needs, and like stratified care it assumes that risk factors can be identified, are stable, and that they can form a “profile” or subtype. This approach also assumes that tailoring interventions to individual risk will be more efficient than alternatives. There’s some support that screening can identify some risks, and that profiles can be constructed – but this continues to be a work under progress. Some of the limitations are the emerging nature of research into grouping people according to multiple indicators is complex, particularly at the beginning of treatment, and treatments matching profiles are therefore also under development. It’s a very complex approach to implement so I can understand why local health authorities may be reluctant to embark on this strategy. It’s also back to the problem of assuming that people’s risk profiles are stable over time.

What do we do?

One part of me thinks, well it doesn’t matter really because as a lowly nonmedical person I have very little influence over health systems, the perverse incentives that drive them, and absolutely no political clout whatsoever. BUT I know that the “wait and see” six weeks before reviewing progress is not helping. And the current considerations fail to integrate those important workplace, family, socio-economic and contextual factors that are hard to quantify.

We already know that low back pain guidelines are routinely ignored by most clinicians in favour of “what I do” and “it’s worked before” and “the guidelines are biased so I won’t follow them”. There’s also the fear that by identifying psychosocial risk factors we’re condemning people to the “back pain is really in your head” meme (it’s even something I’ve been accused of. FWIW I think low back pain is far more complex and is multifactorial. Psychosocial factors are certainly more useful at predicting disability than biomechanical or diagnostic ones, but this doesn’t mean the problem is purely psychological. <steps off soapbox>). Furthermore, it’s clear that not only do physiotherapists feel poorly-prepared to identify and work with psychosocial factors (Singla, Jones, Edwards & Kumar, 2015; Zangoni & Thomson, 2017), so also do medical practitioners although for different reasons (Coudeyre, Rannou, Tubach, Baron, Coriat, Brin et al, 2006). It’s difficult to open Pandora’s box when you only have 10 minutes with a patient.

As Linton, Nicholas and Shaw (in press) point out, training is needed before clinicians can feel both confident and efficient at screening and then managing low back pain via an integrated multidimensional model. “Role” delineation (who can contribute to the various aspects of treatment?) and the paucity of funding for allied health within primary care, especially in New Zealand makes this approach an aspiration. 

Naturally I’d like to see a range of different health professionals involved in developing health pathways. Not just professionals, but people well-versed in understanding the research literature and those with effective knowledge translation skills. I’d love to see high value and low cost treatments provided rather than techno-fix approaches, especially when the high value treatments are significantly safer and develop personal self efficacy and locus of control. Wouldn’t that be a thing to see?

  • Coudeyre, E., Rannou, F., Tubach, F., Baron, G., Coriat, F., Brin, S., … & Poiraudeau, S. (2006). General practitioners’ fear-avoidance beliefs influence their management of patients with low back pain. Pain, 124(3), 330-337.
  • Linton, S. J., Nicholas, M., & Shaw, W. Why wait to address high-risk cases of acute low back pain? A comparison of stepped, stratified, and matched care. Pain. in press
  • Maher, C., Underwood, M., & Buchbinder, R. (2017). Non-specific low back pain. The Lancet, 389(10070), 736-747.
  • Singla, M., Jones, M., Edwards, I., & Kumar, S. (2015). Physiotherapists’ assessment of patients’ psychosocial status: are we standing on thin ice? A qualitative descriptive study. Manual Therapy, 20(2), 328-334.
  • Zangoni, G., & Thomson, O. P. (2017). ‘I need to do another course’-Italian physiotherapists’ knowledge and beliefs when assessing psychosocial factors in patients presenting with chronic low back pain. Musculoskeletal Science and Practice, 27, 71-77.