The smallest worthwhile effect on pain intensity of nonsteroidal anti-inflammatory drugs

Key points

1. 01For acute low back pain, the smallest worthwhile reduction in pain from NSAIDs was 30% (IQR 10 to 40%).
2. 02For chronic low back pain, the smallest worthwhile reduction from NSAIDs was 27.5% (IQR 10 to 50%).
3. 03For chronic low back pain, the smallest worthwhile reduction from individualised exercise was 20% (IQR 10 to 40%).
4. 04A notable share of people would not consider the treatment at all: 18% declined NSAIDs (acute), 30% declined NSAIDs (chronic), and 9% declined exercise (chronic).
5. 05These patient-defined thresholds are often higher than the average benefit shown for NSAIDs in trials, where NSAIDs reduce pain by about 0.7 points on a 0-to-10 scale.

How it was conducted

Design

Benefit-harm trade-off study (online survey), protocol registered on Open Science Framework before data collection

Participants

English-speaking adults in Australia aged 18 or older with current non-specific low back pain of at least 1 day, recruited via social media (15 July 2021 to 30 August 2022)

Groups

People with acute LBP (under 3 months) and chronic LBP (over 3 months)

Outcome measure

Smallest worthwhile effect, the lowest percentage reduction in pain intensity participants considered worthwhile in addition to no intervention

Analysis

Median and IQR of the smallest worthwhile effect; regression for associations with baseline pain, pain duration and exercise level (sample size of 182 required, alpha 0.05, power 0.8)

What they found

• 116 people with acute LBP and 230 with chronic LBP provided data, with an 84% completion rate.
• For acute LBP, the smallest worthwhile effect of NSAIDs was a 30% reduction (IQR 10 to 40%); 20 people (18%) would not consider NSAIDs, leaving n = 96.
• For chronic LBP, the smallest worthwhile effect of NSAIDs was a 27.5% reduction (IQR 10 to 50%); 70 people (30%) would not consider NSAIDs, leaving n = 161.
• For chronic LBP, the smallest worthwhile effect of exercise was a 20% reduction (IQR 10 to 40%); 20 people (9%) would not consider exercise, leaving n = 200.
• For acute LBP NSAIDs, higher baseline pain (b = -2.6%, 95% CI -4.2 to -1.2), longer pain duration (b = -1.3%, 95% CI -2.4 to -0.2) and lower exercise level (b = 3.9%, 95% CI 0.6 to 7.2) were associated with a slightly lower smallest worthwhile effect.
• Mean baseline pain intensity was 4.9 (SD 2.2) for acute and 4.9 (SD 2.3) for chronic LBP.

Limitations

• Participants were self-selected English-speaking Australians recruited through social media, which may not represent all people with low back pain.
• The smallest worthwhile effect was based on hypothetical trade-off scenarios rather than actual treatment experience.
• Most participants had prior experience with NSAIDs (70 to 78%) and exercise (80 to 85%), which may have shaped their expectations.
• The wide interquartile ranges (for example 10 to 50%) show large variation between individuals in what they consider worthwhile.

Why it matters

For patients

It is reasonable to expect a meaningful pain reduction, around 20 to 30%, before judging anti-inflammatory drugs or exercise worth the effort and any side effects.

For clinicians

Patient-defined thresholds of about 30% for NSAIDs and 20% for exercise can frame realistic conversations about expected benefit, especially since the average NSAID benefit in trials is modest.

For readers

This study quantifies how much pain relief patients themselves consider worthwhile, a useful benchmark for interpreting whether trial results are clinically meaningful.