Expert practical advice on using the NSAID painkillers

Question: My GP says that the NSAID drugs including ibuprofen are unsafe. What pain killers are actually safe? Does this mean that GPs will now be recommending alternatives instead? What should patients know?


There is a very long answer to this question, for which one might need more time than for the purposes of a quick reply.

In the absence of any context, then, what I will try to do is to set out a series of bullet points that address the main issues, and explain, I hope, why the question is a daft one. The simple answer is the old Paracelsus one, that all drugs are poison, it just depends on the dose.

  1. Pain is very common – with 1 in 3 GP visits having pain as an issue, and as a consequence pain medicines are very common.
  2. Pain is associated with immense reduction in quality of life, more than any other chronic disease.
  3. Pain is associated with quantity of life – people with chronic pain lose a lot of life. This is usually because they tend to be inactive, and the consequence is a doubling of the risk of FATAL CV events if your arthritis disables you.
  4. All analgesics, of whatever class, have adverse events associated with them. They are no different than other drugs.
  5. Safe is a short word that takes an age to understand, so let’s make it simple by splitting it down into two categories of harm:
    • Harm that is common, predictable, reversible on discontinuation, and therefore temporary. It might lead to patients stopping taking the drug.
    • Harm that is uncommon, unpredictable, irreversible, and permanent – such as heart attack, stroke, death etc.
  6. These two get mixed up, but shouldn’t be. If the tablets make you sick, you’ll stop taking them, and will have no risk of the second type of harm.
  7. Analgesics produce massive reductions in pain in some patients who take them. Those who don’t get good pain relief usually get none – so it tends to be an all or nothing effect.
  8. The proportion who benefit is small – perhaps 20-25% more than placebo for osteoarthritis, 10-15% in diabetic neuropathy, but only 5-10% in fibromyalgia and chronic low back pain (CLBP). No one drug is sufficient, and patients need to try several to find one that works for them.
  9. With good pain relief comes a host of other benefits – sleep is better, fatigue and depression go away, quality of life improves to the norm, and work ability comes back to almost normal. So good pain relief produces huge QALY (quality-adjusted life year) gains, somewhere around 0.1 to 0.2 of a QALY – a prodigious benefit.

Right then, lets suppose we have osteoarthritis and have to decide whether to take an NSAID like naproxen or a coxib like etoricoxib.

Here are the arguments:

  1. NSAIDs increase GI (gastro-intestinal) bleeding, which comes with 20% mortality i.e. it can kill. So we should also take a PPI inhibitor, but most docs won’t prescribe it and most patients won’t take it if prescribed. Oh! and PPIs taken for a long time at high dose may increase the risk of hip fracture, which if you are old is 30% fatal, and has a 30% probability of not going back to independent living.
  2. Coxibs don’t cause GI bleeding (or at least so much). So, unless you had very high risk factors, you wouldn't need a PPI.
  3. NSAIDs and Coxibs may (may not) increase cardiovascular (CV) risk by 30% - not mortality, not stroke, but non-fatal heart attacks only.

Two scenarios:

Scenario A: The patient gets good pain relief with NSAID or coxib, with massive QoL (quality of life) benefits and is now active again. No common side effects, or at least if present they are tolerable. Some possible risk increase of 30% of non-fatal heart attack, but the associated increase in activity prevents a 100% increase in a fatal CV event. There is a tangible immediate gain, with some potential risk of harm in the future from the tablets, which is balanced by a larger and more important reduced risk of harm in the future by virtue of increased activity being protective for the heart.

Scenario B: The patient doesn’t get good pain relief and doesn’t take the tablets, or can’t take the tablets because of intolerable side effects. So no longer term risk of bad things happening.

And that, in short, is the argument. So trying to ask which are safe and which are not is daft because it depends on whether the tablets work for you or not. THAT is the issue. Without pain relief, you shouldn’t be taking the tablets. You can’t manage risk until you manage the pain.

Andrew Moore
Pain Research
Nuffield Division of Anaesthetics
Nuffield Department of Clinical Neurosciences 
University of Oxford  
Churchill Hospital, Oxford
Another term for painkillers. Full medical glossary
Inflammation of one or more joints of the body. Full medical glossary
blood pressure Full medical glossary
A disease of long duration generally involving slow changes. Full medical glossary
Feelings of sadness, hopelessness and a loss of interest in life, combined with a sense of reduced emotional well-being Full medical glossary
One of the three main food constituents (with carbohydrate and protein), and the main form in which energy is stored in the body. Full medical glossary
A chronic, disabling condition characterised by widespread muscle pain and fatigue. Full medical glossary
gastrointestinal Full medical glossary
The death of a section of heart muscle caused by an interruption in its blood supply. Also called a myocardial infarction. Full medical glossary
An animal or plant that supports a parasite. Full medical glossary
relating to the intestines, the digestive tract between the stomach and the anus Full medical glossary
Pain in the muscles. Full medical glossary
Damage to or disease of nerves. Full medical glossary
non-steroidal anti-inflammatory drug Full medical glossary

Non-steroidal anti-inflammatory drugs. A group of drugs that provide pain relief and reduce inflammation.

Full medical glossary
A disease mainly of the large joints of the body, as a result of wear and tear of the surface cartilage. Full medical glossary
septic arthritis Full medical glossary
Any sudden neurological problem caused by a bleed or a clot in a blood vessel. Full medical glossary
Relating to blood vessels. Full medical glossary