What works (and what doesn’t) for lower back pain

Margaret, a 54-year-old woman, presents to your practice with lower back pain that began a few weeks ago. Once active and independent, she now avoids routine tasks like gardening and walking the dog, worried that she’ll worsen the pain.

Margaret’s presentation is common, with one in six Australians reporting back pain at any one time. Back pain is the leading cause of years lived with disability and the most common symptom-related cause for seeking care in Australian general practices. It is a complex condition associated with reduced physical and mental capacities and a higher risk of comorbid chronic disease, which can influence treatment decisions.

Margaret, like many individuals experiencing back pain, was concerned about the possibility of a serious underlying condition and explored treatment options on Google and social media platforms before consulting her GP. Patients often arrive at consultations with preconceptions, expectations and preferences regarding the treatment they believe they should receive.

Clinical practice guidelines for back pain management routinely recommend initial screening for any serious conditions. This includes identifying “red flags” such as recent trauma, unexplained weight loss, history of cancer, neurological deficits, or signs of infection. While serious specific pathology is rare — accounting for less than 5% of cases — it is far more common for GPs to encounter non-specific lower back pain where no specific cause can be reliably identified clinically. In these cases, treatment decisions shift towards reassuring the individual about the favourable prognosis of back pain and low risk of serious underlying disorder, providing personalised information and advice, and prescribing non-pharmacological (first-line) and pharmacological (second-line) interventions.

Which treatments work and which don’t 

In recent weeks, we published two comprehensive reviews summarising the evidence for back pain treatments in comparison to placebo or no treatment/usual care.

A recent overview published by the Cochrane Collaboration synthesised findings from 31 Cochrane systematic reviews involving nearly 97 000 individuals. It examined the effectiveness of non-drug, non-surgical treatments for lower back pain, providing a robust evidence base to support most first-line care decisions.

In another review published in BMJ Evidence-Based Medicine, we synthesised the evidence from 301 randomised controlled trials that compared non-surgical and non-interventional treatments with placebo.

We summarised the treatments supported by at least a moderate certainty of evidence — that is, findings likely to be close to the actual effect, based on assessments of risk of bias, consistency of findings, precision of estimates, and risk of publication bias.

The most certain treatments for (sub)acute back pain

Advise to stay active despite back pain

There is a natural response to encourage patients to rest when they experience recent onset (acute) back pain. However, we now know that excessive protective behaviours like prolonged rest can make it harder to return to meaningful activities later. Advising patients to stay as active as tolerable can reduce pain and improve function compared to rest in bed. However, prescribing a structured exercise program is not necessary.

This recommendation might be counterintuitive for patients, requiring GPs to provide personalised information and advice to reassure the patient that gradually increasing movement is safe and important for recovery.

Spinal manipulative therapy

Although effects may vary, spinal manipulative therapy is likely to provide small reductions in pain for some individuals compared with placebo, but it is probably less effective in improving function.

Multidisciplinary care

For pain lasting six to 12 weeks, multidisciplinary treatment is likely to reduce pain compared to standard care.

This approach involves coordinated care from GPs, physiotherapists, and psychologists to address movement patterns, thoughts, emotions, environmental factors, and sensitisation of the pain system to manage back pain and minimise the risk of it becoming persistent.

Multidisciplinary care might not be accessible for all patients. GPs may consider timely referrals to other health professionals and follow up with patients to monitor their recovery progress.

Medications

Non-steroidal anti-inflammatory drugs (NSAIDs) are likely to reduce pain compared to the placebo and may be considered for short term adjunctive treatments, considering their potential adverse effects and relevant contraindications.

The most certain treatments for persistent (chronic) back pain

Exercise therapy

Exercise — especially programs designed to match the patient’s needs and preferences — likely reduces pain and disability compared with placebo, no treatment, or standard care. This could encompass aerobic activity (eg, walking), strength training (eg, weights or resistance exercise), yoga, or Pilates-based movements.

Multidisciplinary care

As with subacute back pain, coordinated care addressing physical, psychological and environmental factors offered by two or more professionals likely works better than usual care alone.

Psychological therapies

The use of psychological therapies aimed at helping people change their thinking, feelings, behaviours, and reactions likely reduces pain; however, it may not be as effective in improving physical function.

Acupuncture

While some debate remains about how acupuncture works, the evidence suggests it probably reduces pain and improves your function compared to placebo or no treatment.

Medications

NSAIDs and transient receptor potential vanilloid 1 agonists (topical cayenne pepper) may provide small reductions in pain compared to placebo. Given their potential adverse effects and relevant contraindications, they may be considered for short term adjunctive treatments. There is mixed evidence for other medicines, including antidepressants, muscle relaxants and opioids, which may be associated with increased risk of harm.

Treatment decision making

Evidence-based management of back pain requires active patient involvement when discussing the many treatment options available. This includes consideration for the potential benefits and harms of treatments, patient preferences, relevant comorbidities and contraindications, available health resources, and relevant contextual factors. For Margaret, having received a comprehensive assessment, personalised information and advice about her back pain, and a short course of NSAIDs, she felt reassured about her recovery and empowered to gradually increase her activity and resume her meaningful activities as quickly as possible.

Dr Rodrigo Rizzo is a Postdoctoral Research Fellow at the Centre for Pain IMPACT, Neuroscience Research Australia and a Lecturer at the School of Health Sciences, University of New South Wales.

Dr Aidan Cashin is a Senior Research Fellow and Deputy Director at the Centre for Pain IMPACT, Neuroscience Research Australia and Conjoint Senior Lecturer at the School of Health Sciences, University of New South Wales.