Intended for healthcare professionals

Practice Practice Pointer

Low back pain in people aged 60 years and over

BMJ 2022; 376 doi: (Published 22 March 2022) Cite this as: BMJ 2022;376:e066928
  1. Adrian C Traeger, research fellow1,
  2. Martin Underwood, professor of primary care research23,
  3. Rowena Ivers, general practitioner, associate professor4,
  4. Rachelle Buchbinder, rheumatologist, professor56
  1. 1School of Public Health, Faculty of Medicine and Health, University of Sydney, Sydney, Australia
  2. 2Warwick Clinical Trials Unit, University of Warwick, Coventry, UK
  3. 3University Hospitals Coventry and Warwickshire, Coventry, UK
  4. 4Graduate Medicine, University of Wollongong, Wollongong, Australia
  5. 5Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia
  6. 6Monash-Cabrini Department of Musculoskeletal Health and Clinical Epidemiology, Cabrini Health, Melbourne, Australia
  1. Correspondence to: A Traeger adrian.traeger{at}

What you need to know

  • Non-specific low back pain is common in people ≥60 years old; however, the index of suspicion for specific causes is higher than in younger people

  • Unfavourable benefit-harm profiles of pain medicines for non-specific low back pain is likely worse for people ≥60 years old than younger people

  • Consider individual patients’ previous treatment experiences, values, and preferences

A 63 year old woman presents with a three day history of severe low back pain. She has a 20-year history of intermittent back trouble, but this episode is more severe than previous episodes and came on without provocation. She is physically active and usually swims most days but is unable to do so with the current pain. Previously, over-the counter ibuprofen provided some relief, but this time it has had little effect. She asks about investigations, possible causes, and pain management options.

People aged 60 years and over are more likely to experience incapacitating and persistent episodes of low back pain than younger people.12 In most cases, no specific cause can be identified, hence the label “non-specific” low back pain, and the person is managed symptomatically.3 Specific causes are less common but still important to consider.

Contrary to current clinical guidelines45 and despite drug side effects and interactions being more common in older people,6 a population based cross sectional study in two UK regions found that a person in their 70s is less likely to be offered non-drug therapies such as exercise programmes and more likely to be offered pain medicines than someone in their 30s.7 Recent increased focus on the need for clinicians to better manage low back pain “without recourse to the prescription pad”8910 prompted us to describe how this might work in general practice for people ≥60 years.

Why does low back pain matter in people ≥60 years old?

Globally, low back pain is the third commonest cause of years lived with disability in people aged ≥70 years (behind diabetes and hearing loss), at 7.4% of the total.11 In higher income countries, low back pain is the leading cause of years lived with disability in ≥70 year olds, 8.8% of the total.11

In a systematic review of 34 cross-sectional studies of people ≥60 years old in community, hospital, or aged-care settings the one-year prevalence of low back pain ranged from 21% to 75%.12 Its prevalence increases with age up to 80 years and then slightly decreases; the prevalence of disabling low back pain also rises with increasing age.113 In a prospective cohort study of 969 people (mean age 43.3 years) who presented to primary care with acute low back pain, increasing age was independently associated with increased risk of persistent pain three months later.2

For older people, back pain can spark fears of physical deterioration, cause social isolation, and threaten independence.14 Activities of daily living and social interaction may become more difficult or impossible.15 The risk of falling almost doubles in people with low back pain (cross-sectional study of 522 community-dwelling adults aged ≥62 years).15 Immobility from back pain can also rapidly lead to physical deconditioning which may be more consequential in older people.16 Depressive symptoms and poor sleep are more commonly reported in older adults with low back pain compared with those without.1718

For a small proportion of people aged ≥60 years with low back pain, cognitive function assessment (including the ability to report pain, develop coping strategies, and adhere with treatment) and frailty are relevant; and pain management in the context of cognitive impairment and/or frailty remains a critical evidence gap.19 However, consideration of these aspects is beyond the scope of this paper.

How is low back pain assessed in people ≥60 years old?

Focused history and physical examination are largely the same as for younger adults, but the index for suspicion of some disorders will be higher (such as vertebral fracture, malignancy, and infection) and some will be lower (such as spondyloarthritis).

What is the difference between “non-specific” low back pain and pain from specific causes?

When a specific cause of low back pain cannot be identified at presentation and follow-up, the term “non-specific” low back pain is often used. Many physicians and patients dislike or distrust this label as a diagnosis, and may instead pursue a specific pathoanatomic cause for low back pain—such as “discogenic pain” or “facet joint osteoarthritis.”20 However, tests that identify these structure-based diagnoses as the source of low back pain—such as manual compression of the facet joints, or lumbar disc protrusion/degeneration on MRI—have poor validity.21 A diagnosis may also provide justification for the use of structure-based treatments that have been found to be ineffective—such as intra-articular facet joint injections,22 radiofrequency denervation procedures,23 and spinal fusion surgery.24

When is imaging considered?

We recommend that imaging is reserved for those with features suggestive of specific pathology (box 1, fig 1).33 Most people will not require further investigation.33 Any strategy for ordering routine imaging needs to be mindful of limits on available resource.

Box 1

When to consider diagnostic imaging (adapted from National Institute for Health and Care Excellence (NICE)2526 and American College of Physicians27)

Immediate imaging; refer to emergency care for imaging (MRI/CT/ultrasound as appropriate)

  • Suspected cauda equina syndrome

  • Suspected unstable vertebral fracture with neurological compromise

  • Suspected leaking or rupturing abdominal aortic aneurysm

Urgent imaging

  • Suspected vertebral infection—MRI if available or bone scan

  • Suspected metastatic disease in patient with high risk cancer—Whole spine MRI if available; otherwise blood test (ESR and full blood examination) will be a useful screen

  • Suspected acute vertebral fracture (high suspicion)—X ray; MRI if continued suspicion and X ray normal

  • Suspected pancreatic cancer—Abdominal CT

Delay imaging, trial of primary care management

  • Suspected malignancy (low suspicion)—X ray with blood test (ESR and full blood examination); MRI if continued suspicion

  • Suspected acute vertebral fracture (low suspicion)—X ray; MRI if continued suspicion

  • Suspected radiculopathy or lumbar canal stenosis in candidate for surgery—MRI

No imaging

  • Non-specific low back pain

  • Radiculopathy or lumbar canal stenosis (unless surgery is being considered)

  • MRI = magnetic resonance imaging; CT = computed tomography; ESR = erythrocyte sedimentation rate.

Fig 1
Fig 1

Alerting features for specific causes of low back pain in older adults presenting to primary care.3262829303132

Notable specific causes of low back pain in older people

Figure 1 shows the estimated prevalence and clinical picture of specific pathologies that are commoner in older adults presenting to primary care.

Cauda equina syndrome

Occurs due to compression of the lumbosacral nerve roots. It is rare, but the incidence in older adults is unknown. Aetiologies include severe central lumbar canal stenosis, a condition which is commoner in older people, and malignancy (see below).34

Key features—New impairments of bladder or bowel control, perineal or saddle numbness, and other neurological symptoms.

Management—Immediate referral to emergency care for imaging.

Vertebral fractures

These are more common in older populations. In the “Back Complaints in Elders” (BACE) study (n=669), vertebral fracture incidence in people over 55 years with back pain seen in primary care was 5%,3 whereas an inception cohort study showed a prevalence of less than 1% (8/1172 participants) in the general primary care population.28

Key features—Acute pain arising after minor trauma. In the BACE study, age >75 years, known osteoporosis, pain score ≥7, and thoracic back pain were associated with fracture, but trauma had the strongest association (positive likelihood ratio 6.2 (95% CI 2.8 to 13.5)).

Management—Most improve quickly over weeks irrespective of treatment and can be managed in primary care with simple analgesia, early mobilisation and physical therapy if needed.35 Immediate imaging is indicated only if there is suspicion of an unstable fracture. Unstable fractures, although rare in the absence of major trauma, may present with diffuse neurological symptoms such as worsening bilateral leg or “stocking” numbness.27 In those with a history of fractures or if suspicion of an unstable fracture is low, imaging can be deferred and performed if symptoms do not improve after several weeks.27

Central lumbar canal stenosis

A quarter of people attending primary care for low back pain have radiological evidence of lumbar canal stenosis; this proportion increases with age (5-35% in the 60-69 age group, 30-40% in the 80-89 age group).29 It is also a common imaging finding among people without symptoms. A Japanese study (n=983, mean age 66.3 years) found that only a sixth of those with “severe” radiological lumbar stenosis on MRI (50/285, 17.5%) had clinical symptoms.36

Key features—Bilateral pain radiating from the back into the buttocks, thighs, and legs with “pseudoclaudication”; and symptoms that are provoked by standing or walking and relieved by sitting or bending forward.37

Management—Immediate referral is indicated if features of cauda equina are present. In some people, symptoms improve over time. Primary care management comprises advice and education, non-drug treatments and simple analgesia if needed (table 1). Occasionally, referral for surgical advice is indicated if unremitting pain and disability persist. The benefits surgical treatment of central lumbar stenosis are uncertain.55

Table 1

Evidence for benefits and harms of common low back pain treatments in the general population and considerations for older adults438

View this table:


Although malignancies are more common among older adults with low back pain, their prevalence as a cause of the pain remains low. In the BACE study the incidence of spinal malignancy as the cause of back pain was just 0.6% (n=4).3 However, 10% of all malignancies present with spinal pain as an initial manifestation.56

Key features—Usually presents in the context of previous diagnosis of malignancy as progressive or unremitting low back pain, local spinal tenderness, and spinal pain preventing sleep or aggravated by straining.

Management—The National Institute for Health and Care Excellence (NICE) recommends urgent investigation (preferably whole spine MRI within 24 hours if there are features of cord compression, and within one week for those without features of cord compression but where malignancy is suspected).25 Investigation urgency is informed by the propensity of a primary malignancy for bone metastases and the presence of neurological features such as motor weakness.57 For example, risk of bony metastatic disease is much greater in a woman with recent history of invasive HER2-positive breast cancer than a woman with HER2-negative breast cancer detected by screening and successfully treated many years earlier.58

For people aged over 60 years, NICE recommends an urgent CT scan to exclude pancreatic cancer in those with weight loss and low back pain,26 and blood tests (full blood count, blood tests for calcium and plasma viscosity or erythrocyte sedimentation rate) to exclude myeloma for those with persistent back pain or new vertebral fractures.26

Intra-abdominal pathology

Conditions such as abdominal aortic aneurysm (AAA) and cholecystitis can cause acute back pain.

Management—Immediate referral to emergency care is required for suspicion of a pending aortic aneurysm rupture (for example, acute onset back pain in person with known AAA). Abdominal ultrasound and blood tests including full blood count, liver function and C reactive protein tests are appropriate for suspected cholecystitis.

Infective causes

Examples include vertebral osteomyelitis, septic disciitis, facet joint septic arthritis, and epidural abscess

Key features—Immunosuppression, recent sepsis (such as urinary tract infection or endocarditis), or spinal procedures (such as epidurals, nerve root injections), intravenous drug use. People with an infective cause may be generally unwell, with or without a fever.

Management—Urgent referral to spinal surgeon; urgent MRI with blood test (C reactive protein and full blood examination) or bone scan if no access to MRI.

Inflammatory causes

Examples include axial spondyloarthritis, but new presentations are rare in older people.3

Radicular leg pain and radiculopathy

Lumbar nerve root compression can cause radicular pain. Radiculopathy refers to radicular pain with other neurological features associated with nerve root compression.

Key features—Unilateral leg pain, with or without low back pain, following a dermatomal pattern. Reflex loss, myotomal weakness, and dermatomal numbness are features of radiculopathy. Like non-specific low back pain, symptoms can be short lived, and most will improve within six weeks.

Management—Most people can be managed as per non-specific low back pain. People with progressive neurological signs require MRI and spinal surgeon review.

What are the management options for non-specific low back pain?

Primary care clinicians can generally provide the key aspects of non-specific low back pain care (self-management advice, reassurance, timely review). Most international guidelines recommend reassurance that there is no serious cause and that symptoms are likely to improve over time regardless of treatment.4 However, one prospective cohort study suggested that the risk of persistent non-specific low back pain at three months increases with age.7

NICE recommends using screening tools to determine risk of poor outcome.5 Alternatively, safety-netting and review over six weeks (by which time most recovery should occur) can determine progression and whether additional support is required. NICE also recommends providing information about the nature of low back pain to support self management, including advice to continue with normal activities.5 More detailed advice about how to cope with daily activities (such as lifting and carrying shopping, self care, sleep) and about the role of over-the-counter medicines (including the potential for drug-drug interactions) might also be required (box 2). Some older people with complex multimorbidity and/or polypharmacy may benefit from a review by a geriatician.

Box 2

Advice for older people with non-specific low back pain

  • Define non-specific low back pain as “pain that is probably caused by muscles, joints, and ligaments”

  • Reassure people with recurrent pain, no current malignancy, no features of specific causes, and no relevant findings on previous imaging that further investigation is unlikely to change management. Arrange a time to review if symptoms continue

  • Reassure people concerned about degenerative changes on previous imaging that these are common and almost as common in people without pain, making it difficult to know whether these are responsible for their pain. Open discussions about pain management options

  • Recommend heat packs, structured exercise programmes, and, depending on the patient’s safety profile, short term use of over-the-counter medicines such as non-steroidal anti-inflammatory drugs (if not contraindicated) for someone with mild to moderate subacute pain of <6 weeks’ duration. Advise on the gastrointestinal side effects of such drugs, taking them with food, and the possibility of a proton pump inhibitor prescription

  • Discuss pharmacological options (as well as movement and heat packs) with people with severe acute pain. Depending on patients’ safety profiles, an anti-inflammatory or opioid medicine might be offered—in both cases these drugs may have limited benefits and should be prescribed at lowest effective dose for the shortest period. Discuss the drugs’ side effects, including reliance.

  • Discuss multidisciplinary rehabilitation with someone with chronic pain who feels that they have tried everything


With limited evidence from randomised trials of management of non-specific low back pain in people aged ≥60, management is often subject to clinician bias, belief, and outcome perception.3859Table 1 summarises the benefits and harms of common therapies for low back pain, based on a 2020 systematic review (11 trials, 758 participants) of the effectiveness of non-specific low back pain interventions in older adults38and data from the general population.4

Non-drug treatments

Exercise and exercise programmes—Use patient preference to guide exercise type—such as yoga, tai chi, Pilates, general whole-body exercise (walking/running/swimming). In some cases, specific advice on safely performing or maintaining activities that the patient values could prevent deconditioning and be more beneficial than formal exercise programmes.38394041

Acupuncture—There is little reliable evidence suggesting sustained benefits,384243 but acupuncture could be appropriate for patients with previous positive experiences.

Spinal manipulation—A 2019 systematic review of the general population suggests that manual movement and pressure applied to the spine by a trained practitioner (such as physiotherapist or chiropractor) may have small benefits, although these were deemed likely to be clinically unimportant.44 Avoid high velocity manoeuvres in people with osteoporosis.

Multidisciplinary rehabilitation—Intensive programmes combining exercise, psychological, physical, and educational components are designed for patients with persistent pain or disability who do not respond to first-line options. In the general population and in those aged ≥65 years, long term pain outcomes seem better than with analgesia and physical therapy only.4760 Access in rural and remote areas may be limited. Pain services via videoconference may be possible but are untested.61

Referral for non-specific low back pain

  • Routine referral for non-pharmacological care does not improve outcomes for acute episodes of low back pain70

  • In healthcare settings without direct access, referrals for physical therapies could be reserved for patients with strong preferences, persistent symptoms, or those identified to be at high risk of poorer outcome, either through use of a screening tool (such as STarT Back, validated in the UK70; or PICKUP, validated in Australia71) or from lack of improvement at follow up

  • Decisions regarding when and where to refer for non-drug treatments may depend on pain severity, likely prognosis, patient preferences, local access options, and costs. Someone who is incapacitated with severe pain may be understandably reluctant to accept a referral for non-drug therapy; we provide advice on how to manage this situation in box 2

Drug treatments

Non-steroidal anti-inflammatory drugs (NSAIDs)—NICE and the American College of Physicians (ACP) recommend NSAIDs for the management of acute and chronic low back pain.45 However, estimates from systematic reviews in the general population suggest that the small benefits over placebo are unlikely to be clinically important.49 NSAIDs have well known risks and drug interactions even in well older adults without comorbidities who have previously tolerated oral NSAIDs (avoid in frail older adults19)—if found to be effective, they are prescribed at the lowest effective dose and used for the shortest possible time.1962

Paracetamol—High certainty evidence shows that paracetamol is unlikely to be effective for low back pain.50 However, some individuals report pain relief, and in the older population there may be no safer pharmacological alternatives. The ethics around recommending paracetamol may depend on whether a shared decision on its use includes a clear explanation of the likely benefits and harms of available treatment options.63

Opioids—Benefits are likely to be clinically unimportant and outweighed by potential serious harms (even from “weaker” opioids such as tramadol and codeine), especially in older adults and those with cognitive impairment19 who might be at higher risk of respiratory depression, falls, and fractures.1964 Prescribe as a last resort and at the lowest effective dose for the shortest period.

Skeletal muscle relaxants (non-benzodiazepine antispasmodics)—ACP recommends these as second-line therapy for acute pain in the general population.4 However, a 2021 systematic review suggests that the benefits are likely to be clinically unimportant,52 and the 2019 American Geriatrics Society (AGS) guideline strongly recommends avoiding them in adults ≥65 years old because of anticholinergic adverse effects, sedation, and increased risk of fractures.48

Antidepressants—Trials in the general population suggests there are no clinically important benefits,53 although ACP recommends duloxetine as second-line option. These medications can cause dizziness and sedation that may increase risk of falling.4 AGS makes a weak recommendation to avoid duloxetine in older people with a creatinine clearance <30 mL/min.

Anticonvulsants—Anticonvulsants are ineffective for back pain with or without sciatica in the general population.54 Gabapentinoids in particular can be dangerous if combined with opioids and can increase risk of death.48

How are shared decisions made?

Begin with discussions about the patient’s previous experiences with care and their preferences regarding treatment options (table 1). Despite two systematic reviews of shared decision making for musculoskeletal disorders finding no evidence of improvement in patient outcomes,6566 respect, understanding, adequate time to discuss options, and timely follow-up all increase engagement.676869

How this article was created

We used recommendations from the National Institute for Health and Care Excellence (2016),5 American College of Physicians (2017),4 and the American Geriatrics Society (2019).48 We supplemented these with searches of the Cochrane Database of Systematic Reviews from inception to 1 March 2022. We searched for age-specific evidence using PubMed (NLM database) to located systematic reviews in older adults published from August 2011 to March 2022. Search terms were “low back pain” AND “older” OR “elder” OR “senior” AND “systematic review.”

How patients were involved in the creation of this article

Our patient advisor, Jan Donovan, and two of the authors (RB and MU) are older adults with recent experience of low back pain. We used this combination of lived experience to inform the content of this article. Jan asked us to increase emphasis on the importance of GP advice regarding the role of exercise, everyday activities, and rest, and use of over-the-counter medicines. RB and MU highlighted the impracticality of some non-drug treatments for severe pain and the judicious use of tests to avoid overdiagnosis and overtreatment.

Education into practice

  • To what extent do you use watchful waiting to confirm a diagnosis of non-specific low back pain?

  • How do you discuss the benefits and harms of treatment options for low back pain?


We thank our patient partner, Jan Donovan, for sharing her story and for her helpful advice on the content and presentation of this article.


  • Contributors: AT conceived the article and is guarantor for it. All authors wrote and reviewed the article. RB was the contact for patient involvement. J Donovan contributed her personal story, provided feedback on the article plan, and reviewed and revised box 2.

  • Competing interests: AT is funded by an Australian National Health and Medical Research Council (NHMRC) Early Career Fellowship (ID1144026). MU is chief investigator or co-investigator on multiple previous and current research grants from the UK National Institute for Health Research, Arthritis Research UK and is a co-investigator on grants funded by the Australian NHMRC and Norwegian MRC. He was an NIHR Senior Investigator until March 2021. He has received travel expenses for speaking at conferences from the professional organisations hosting the conferences. He is a director and shareholder of Clinvivo, which provides electronic data collection for health services research. He is part of an academic partnership with Serco, funded by the European Social Fund, related to return to work initiatives. He receives some salary support from University Hospitals Coventry and Warwickshire. He is a co-investigator on three NIHR funded studies receiving additional support from Stryker Ltd Until March 2020 he was an editor of the NIHR journal series, and a member of the NIHR Journal Editors Group, for which he received a fee. He has published multiple papers on low back pain some of which are cited in this paper. RI has no conflicts to declare. RB is funded by an Australian NHMRC Investigator Grant (APP1194483) and is a chief investigator or co-investigator on multiple previous and current research grants funded by the Australian NHMRC. She has published multiple papers on low back pain some of which are cited in this paper.