Ibuprofen was the most commonly used painkiller in the study
Long-term use of ibuprofen may reduce the risk of developing Alzheimer's disease, a large US study reports.
Data from almost 250,000 veterans showed those who used the painkiller for more than five years were more than 40% less likely to develop Alzheimer's.
The study in Neurology reported that some other similar painkillers may also have a protective effect.
Dementia experts said the results were interesting but warned against people taking ibuprofen to reduce their risk.
It is not the first time an association between non-steroidal anti-inflammatory drugs (NSAIDs), such as ibuprofen, and Alzheimer's disease has been reported but results have been conflicting.
The researchers from Boston University School of Medicine said one possible reason for inconsistent reports may be that different NSAIDs have different effects.
They looked at five years of data in 49,300 people over the age of 55 years who had developed Alzheimer's disease and almost 200,000 controls.
Overall, use of NSAIDs for five years was associated with a 24% reduced risk of developing Alzheimer's.
But the effects of ibuprofen were the most profound and some other NSAIDs, such as celecoxib, had no effect.
Study leader Dr Steven Vlad said: "There are theoretical reasons to think that not all NSAIDs would behave the same way in terms of Alzheimer's disease."
He said ibuprofen had been shown in animal models and the laboratory to reduce levels of protein deposits associated with Alzheimer's in the brain.
The results are probably partly due to direct effects of the drug and partly due to the fact ibuprofen is the most commonly used NSAID so the finding would be picked up more easily, he added.
But he added: "All NSAIDs have well known side-effects that can be very serious and we still need trials to make sure the risks and benefits are very clear."
Professor Clive Ballard, director of research at the Alzheimer's Society, said: "This interesting research builds on evidence of the protective effects of long-term use of NSAIDs against Alzheimer's disease.
"Whilst this is important research it does not mean that people should start taking ibuprofen to reduce their risk of developing dementia.
"Long-term use of NSAIDs is associated with a number of very significant side-effects."
He added it was now up to researchers to use the results in the development of future treatments for the condition.
Rebecca Wood, chief executive of the Alzheimer's Research Trust, said the results were promising.
"The apparent connection between ibuprofen and a reduced risk of dementia is no silver bullet, but indicates an exciting direction for future research," she said.
A separate study also published in Neurology showed people with shorter arms and legs may be at a higher risk for developing dementia later in life.
The US researchers said poor nutrition in early life may be the link between the two.