A study on more than 76,000 women found consuming tea raised the risk while drinking coffee had no impact.
Tea-lovers who enjoyed more than four cups a day had the highest risk – being 78 per cent more likely to develop rheumatoid arthritis than those who drank none. But drinking any amount of tea increased the chance by 40 per cent, compared with people who never drank tea.
The findings were presented at the Annual Congress of the European League Against Rheumatism in Rome.
Professor Christopher Collins, from Georgetown University Medical Centre in the US, said he was surprised by the differences between coffee and tea. He said: "We set out to determine whether tea or coffee consumption, or the method of preparation of the drinks was associated with an increased risk of (rheumatoid arthritis].
"It is surprising that we saw such differences in results between tea and coffee drinkers. This does make us wonder what it is in tea, or in the method of preparation of tea, that causes the significant increase in risk of developing rheumatoid arthritis."
The researchers also examined whether filtered coffee versus unfiltered coffee affected the results, and also looked at the impact of caffeinated and decaffeinated coffee. However, they found no significant associations with rheumatoid arthritis in these drinks. The women in the study were aged 50 to 79 and filled in questionnaires on their daily intake of coffee and tea.
Rheumatoid arthritis is a chronic, progressive and disabling auto-immune disease which affects around 350,000 people in the UK. Three times more women get it than men and it usually starts between the ages of 40 and 60.
The disease can cause swelling and damage to the cartilage and bone around the joints, most commonly the hands, feet and wrists. About 12,000 children under the age of 16 have the juvenile form of the disease.
Prof Collins added: "This is one more piece in the puzzle of why people develop rheumatoid arthritis. This was an unusual and complex finding, but, from the data we have, there is a relationship between tea and the disease.
"There have been other studies which have either stated that tea has no effect on risk of arthritis, and one study which found tea had a protective effect."
He said the team had looked for supporting information that could explain their own findings: "We found a study which said that an increased intake of flavonoids – which are in tea – from various sources resulted in an increased risk of rheumatoid arthritis."
The team had analysed data on caffeine separately and had found no link, suggesting it was something specific to tea, he added. "It's definitely an association, but the risk is very small. Nevertheless, when you look at enough people, a very small relationship can still be meaningful."
Prof Collins said women were only asked about their consumption of tea, and so no detail was available on whether they had drunk black tea, tea with milk or herbal teas.