Paracetamol for low back pain

Review question

To see how well paracetamol works for non-specific low back pain (LBP). Non-specific LBP is back pain for which there is no identified disease or condition.


Paracetamol is one of the most commonly prescribed medicines for people with LBP, and it is recommended in the guidelines that are issued to help doctors manage different illnesses. However, recent evidence has called into question how effective it is.

Search date

The evidence is current to August 2015.

Study characteristics

We included three trials with a total of 1825 participants in this review, two trials with participants whose back pain occurred suddenly and recently (acute) and one trial with participants whose pain lasted for longer than six weeks (chronic). Most of the people in the study (90%) were middle-aged and came from a single trial that looked at acute back pain. All of the trials tested paracetamol against a placebo (which contains nothing that could act as a medicine). The treatments ranged from a single 1 g dose (given intravenously) up to 4g in a 24 hour period for up to four weeks (oral tablets). Participants were followed between one day and 12 weeks. The main outcomes we studied were pain and disability; we also looked at quality of life, how easily people could go about their daily lives, unpleasant or unwanted side effects, how well people felt they had recovered, sleep quality, whether participants had taken the medicine as prescribed, and if it had been necessary to take ‘rescue medication’ because the paracetamol had not worked. We combined the findings from two of the trials into a single analysis (meta-analysis) that compared paracetamol to a placebo; the third trial did not report the results for the placebo, and so it could not be included.

Key results and quality of evidence

We found high-quality evidence that paracetamol (4 g per day) is no better than placebo for relieving acute LBP in either the short or longer term. It also worked no better than placebo on the other aspects studied, such as quality of life and sleep quality. About one in five people reported side effects, though few were serious, and there was no difference between intervention and control groups. As most of the participants studied were middle-aged, we cannot be sure that the findings would be the same for other age groups.

There appears to be no difference between paracetamol and placebo in immediate reduction of chronic LBP, although the evidence is of very low quality, and the single study on which it is based has been withdrawn by the journal.

Authors' conclusions: 

We found that paracetamol does not produce better outcomes than placebo for people with acute LBP, and it is uncertain if it has any effect on chronic LBP.

Read the full abstract...

Analgesic medication is the most frequently prescribed treatment for low back pain (LBP), of which paracetamol (acetaminophen) is recommended as the first choice medication. However, there is uncertainty about the efficacy of paracetamol for LBP.


To investigate the efficacy and safety of paracetamol for non-specific LBP.

Search strategy: 

We conducted searches on the Cochrane Central Register of Controlled Trials (CENTRAL, which includes the Back and Neck Review Group trials register), MEDLINE, EMBASE, CINAHL, AMED, Web of Science, LILACS, and IPA from their inception to 7 August 2015. We also searched the reference lists of eligible papers and trial registry websites (WHO ICTRP and

Selection criteria: 

We only considered randomised trials comparing the efficacy of paracetamol with placebo for non-specific LBP. The primary outcomes were pain and disability. We also investigated quality of life, function, adverse effects, global impression of recovery, sleep quality, patient adherence, and use of rescue medication as secondary outcomes.

Data collection and analysis: 

Two review authors independently performed the data extraction and assessed risk of bias in the included studies. We also evaluated the quality of evidence using the GRADE approach. We converted scales for pain intensity to a common 0 to 100 scale. We quantified treatment effects using mean difference for continuous outcomes and risk ratios for dichotomous outcomes. We used effect sizes and 95% confidence intervals as a measure of treatment effect for the primary outcomes. When the treatment effects were smaller than 9 points on a 0 to 100 scale, we considered the effect as small and not clinically important.

Main results: 

Our searches retrieved 4449 records, of which three trials were included in the review (n = 1825 participants), and two trials were included in the meta-analysis. For acute LBP, there is high-quality evidence for no difference between paracetamol (4 g per day) and placebo at 1 week (immediate term), 2 weeks, 4 weeks, and 12 weeks (short term) for the primary outcomes. There is high-quality evidence that paracetamol has no effect on quality of life, function, global impression of recovery, and sleep quality for all included time periods. There were also no significant differences between paracetamol and placebo for adverse events, patient adherence, or use of rescue medication. For chronic LBP, there is very low-quality evidence (based on a trial that has been retracted) for no effect of paracetamol (1 g single intravenous dose) on immediate pain reduction. Finally, no trials were identified evaluating patients with subacute LBP.