Investigating which painkillers give the best pain relief
From the 39 Cochrane reviews of oral painkillers, which involved around 50,000 people in 450 studies, the team found reliable evidence on the effectiveness of 53 pairs of drug and dose in all types of surgery. They also found that for some medicines the evidence is unreliable or lacking, which is also important to know.
The effectiveness of the drugs was assessed by the number of people who had at least half the maximum possible pain relief, compared with placebo (dummy pill), expressed as the ‘number needed to treat’ (NNT). The NNT is an estimate of how many people need to have the treatment for one person to have a good outcome. The lower the NNT, the more effective the painkiller is considered to be. This can be helpful when considering treatment options. How long pain relief lasted was also reported.
For the detail of all the drugs investigated you’ll need to go to the full overview, but here are some key points:
- Fixed dose combination painkillers and fast acting formulations were among the more effective drugs e.g. ibuprofen 200 mg plus paracetamol 500 mg had an NNT of 1.5. Ibuprofen 200 mg plus caffeine 100 mg, ibuprofen 200 mg in a fast acting formulation and diclofenac potassium 50 mg (also fast acting) all had an NNT of 2.1
- Even the most effective drugs fail to give good pain relief to a proportion of people, more than half the people treated in many instances
- For 17 drugs, the evidence is absent or unreliable; none of these are in common use to treat acute pain
- Useful head-to-head comparisons between drugs were not available
Which are most and least likely to cause adverse events?
Evidence from around 35,000 people in 350 studies provides evidence on adverse events (unwanted effects). The PaPaS team notes that most people in the studies had wisdom teeth removed, were relatively young and fit, and likely to take painkillers only occasionally. Adverse events may be different in people who are less well, older, or who take painkillers for several days or longer. They also tell us that patient diaries recorded more adverse events than other methods of collecting this information, but that “meta-analysis should provide good information on relative increases or decreases compared with placebo”.
Here’s what they found:
- Similar rate of adverse events with most non-steroidal anti-inflammatory drugs (including ibuprofen, aspirin and disclofenac), paracetamol, and combinations of drugs that do not contain opioids (like codeine) as with placebo
- More adverse events with aspirin 1000 mg, opioids, or fixed dose combinations containing opioids, than with placebo
- Fewer adverse events with a combination of paracetamol and ibuprofen than with placebo
- Serious adverse events were rare (about 1 in 3200 people)
Where does that leave us?
There is reliable evidence about the effectiveness of a large number of drugs for acute pain. Simple drug combinations and fast acting formulations can give good pain relief at relatively low doses and serious adverse events are rare. It should be remembered that even the most effective painkiller won’t work for some people but that a different one might!