Targeted back care 'better and cheaper'

Targeted back pain care is “cheaper and better for patients”, BBC News has reported. The BBC says that a new model for assessing the severity of back pain has shown significant improvements compared with current methods, and could also save more than £30 per patient.

In cases of lower back pain not caused by a disease, GPs usually use an escalating approach, in which patients are given a succession of more intensive treatments when no improvement is seen. Patients are initially taught self-management techniques, and may be instructed to take a short course of painkillers. If there is no improvement, doctors will then usually discuss options with the patient and refer them for the physical therapy that they think is most appropriate for them (for example, physiotherapy). In this new trial doctors compared existing processes with a new model that used a screening tool to help decide whether or not patients should be referred for further therapy, and if so, which treatment this should be.

The trial demonstrated that the model was slightly more effective at improving patients’ symptoms, and also produced small cost savings compared with using standard practice. The doctors are quoted in the news as saying the research is “very promising”, particularly as the economic assessment demonstrates the approach to be cost effective. However, further testing of this screening tool in clinical practice will now be needed. Also, further follow-up is required to see whether use in wider numbers gives the expected longer-term benefits of reduced disability and improved quality of life for back pain sufferers.

Where did the story come from?

The study was carried out by researchers from the Arthritis Research UK Primary Care Centre at Keele University, the School of Population and Public Health at the University of British Columbia, and Vancouver Coastal Health Research Institute. Funding was provided by Arthritis Research UK. The study was published in the peer-reviewed medical journal The Lancet.

In general, BBC News has reflected the findings of this research paper well, although some of the terms used in its news report could be misinterpreted. For example, it is not very accurate to say that current general practice management of lower back pain is a “one size fits all” approach. The approaches used in this trial (for example physiotherapy with or without a psychological component) are currently included in the pathways of care recommended by the National Institute for Health and Clinical Excellence (NICE) and used in practice.

However, the approach tested in this trial was different in that it used a screening tool to identify which treatment was most appropriate, rather than the current practice where doctors use their clinical judgement when deciding which service they think is most appropriate to refer to. The screening tool used in this trial operates on the principle of stratifying patients into three risk groups and assigning those at greater risk of developing chronic problems to receive more intensive therapy.

What kind of research was this?

This was a randomised controlled trial (the STarT Back trial) designed to compare current general practice management of lower back pain with an intervention of “stratified primary care”. In this stratified primary care, people would receive one of three levels of care depending on their perceived prognosis – low-, medium- or high-risk.

Lower back pain is a chronic health problem that not only places a considerable burden on the healthcare system, but also causes a high level of persistent disability among those affected, reduces capacity to work and significantly affects quality of life. The research paper reports that 6-9% of UK adults consult their GP about lower back pain each year, and that 60-80% of them will still be suffering from pain one year later.

The study was concerned with lower back pain that would sometimes be medically termed “non-specific” lower back pain. This means that the cause of pain, tension or stiffness of the lower back is not clear. It is a diagnosis based on the exclusion of specific disease causes of pain, such as cancer, fractures, inflammatory conditions, infections or spinal cord compression. These are all serious causes of lower back pain that must be excluded by a doctor during initial assessment.

Current medical practice follows a step-by-step approach to non-specific lower back pain, initially focusing on self-management and then considering referral for further therapy if back pain persists. The first step tends to be encouraging the person to remain as active as possible, with the use of short-term painkillers (paracetamol or an anti-inflammatory drug) to control pain if required. If the person does not improve, the GP may then refer them for physical therapy, such as physiotherapy or an exercise programme. In some cases referral may be made for combined physical and psychological treatment. Referral to an orthopaedic consultant for consideration of surgery would be a last resort.

Under current guidance, people who have lower back pain that is associated with nerve root compression or entrapment (for example from a herniated, or ‘slipped’, disc) may sometimes be given an earlier referral for orthopaedic assessment depending on their clinical features. Nerve root compression causes pain going into the legs along the course of the nerve. This is termed radiculopathy (sciatica is the term commonly used when there is compression of the sciatic nerve).

In this trial, the main hypothesis was that using a stratified approach to decide on the most appropriate management option for lower back pain (with or without radiculopathy) would result in better clinical and economic benefit compared with current best practice.

What did the research involve?

In 10 GP surgeries near to Stoke-on-Trent in the UK, medical records were searched to identify patients who had consulted their doctor about back pain between June 2007 and November 2008. The researchers excluded patients with any pain caused by serious disease (including those mentioned above), those with serious medical illnesses or mental health conditions, pregnant women and people currently receiving non-GP management of their back pain.

All remaining eligible participants were then assessed using the STarT Back Screening Tool. This was a validated, simple, prognostic screening tool designed for this study which allocated patients into three defined risk groups of low-, medium- or high-risk. The assessment tool took into account factors such as the level of distress, anxiety, fear or depression that their back pain was causing them. A higher score indicated they were at higher risk of having chronic and persisting back problems.

Participants were randomised into either a control group receiving standard care (283 people) or an intervention group receiving care directed by the screening tool’s results (568 people). The control group received a 30-minute assessment and treatment session from a physiotherapist who gave them exercises and advice (for example about remaining active or about returning to work), with the option of onward referral for further physiotherapy (the decision made at the therapist’s discretion).

Those randomised to the intervention group (568) received the same initial physiotherapy assessment and treatment session, but decisions on further referral were made using the person’s risk classification on the STarT Back Screening Tool. Those patients identified as low-risk only received the initial physiotherapy session, but those in the medium- and high-risk groups were automatically referred for further therapy.

Further therapy was provided by therapists as follows:

In the control group it was given by physiotherapists who had received general training in physical therapies and some training in more complex psychology-based treatments, but who had received no additional training specific to this trial.
In the medium-risk intervention group, patients received treatment from physiotherapists who had been given three days of specific training on providing standard physiotherapy to address symptoms and function.
In the high-risk intervention group, patients received psychologically informed physiotherapy from therapists who were given nine days of specific training on providing therapy that addressed physical symptoms and function, in addition to tackling the psychological consequences of their back pain and psychological barriers that may hinder recovery.

The main clinical outcome was improvement of scores in the Roland and Morris Disability Questionnaire (RMDQ) score at 12 months. Scores range from 0 to 24, with higher scores indicating more severe disability.

To perform an economic evaluation, the researchers estimated the incremental quality-adjusted life years (QALYs) gained with the intervention. QALYs are used to measure the health benefits that an intervention gives over standard treatment. They take into account the quality of the person’s life rather than just how much the treatment may extend life. The researchers then looked at the cost of any QALY gained by the intervention.

What were the basic results?

The average age of participants in this trial was 50 years, and 59% were women. In the intervention group, 26% of patients were stratified as low-risk, 46% as medium- and 28% as high-risk. Across all people in the trial the average number of treatment sessions received was comparable: 3.8 in the control group and 3.9 in the intervention group. The basic results were as follows:

Overall, people in the intervention groups experienced an average (mean) improvement of 4.3 RMDQ points by 12 months, while those in the control group experienced an average improvement of 3.3 points. This small difference between the groups, equating to 1.06 points, was statistically significant (95% CI 0.25 to 1.86).
The researchers then calculated a factor called the “effect size”, which indicates the size of the difference between two treatment groups, with a larger effect size indicating a more effective treatment. The “effect size” for the difference in RMDQ score at 12 months was 0.19, which is quite small.
At 12 months, the intervention strategy of stratified care was associated with a mean increase of 0.039 additional QALYs compared with standard care and a saving of £34.39 (with the cost in the intervention group being £240.01 versus £274.40 in the control group).

How did the researchers interpret the results?

The researchers conclude that a stratified approach to care for lower back pain, which uses a prognostic screening tool to decide whether to refer someone for further physiotherapy (with or without a psychological element) “will have important implications for the future management of back pain in primary care”.

Conclusion

This was a large and well-conducted trial that has demonstrated a small effectiveness benefit and a small cost saving when people with lower back pain were stratified using a screening tool. Under the workings of this tool those with the highest levels of distress and problems associated with their pain would be placed in the high-risk group and so referred for physiotherapy with a psychological component, those with medium risk features would have a greater number of physiotherapy sessions and those with the lowest risk would have initial physiotherapy with advice on self-management.

It is important to note that this practice does not differ dramatically from standard general practice care of lower back pain, rather that it applies the use of a simple tool (rather than clinical judgement) to help decide which treatment option would be most suitable. It is incorrect to suggest that the current system is a “one-size-fits-all” approach as patients will already be offered different treatments based on their clinical features (taking into account other medical or mental health problems they may have), and their response to previous treatment. Instead it is probably more accurate to think of the proposed method as a tool that would guide the clinician to which treatment should be given, giving a more standardised approach to care than the current system.

This trial has demonstrated small benefits with this different system. Further testing and validation of this screening tool is now needed in clinical practice, along with further follow-up to see whether use in wider numbers gives the expected longer-term benefits of reduced disability and improved quality of life for back pain sufferers.