| Early treatment is associated with improved outcomes in patients with rheumatoid arthritis (RA), suggesting that a ‘window of opportunity’, in which the disease is most susceptible to disease-modifying treatment, exists. Autoantibodies and markers of systemic inflammation can be present long before clinical arthritis, and maturation of the immune response seems to coincide with the development of RA. The pre-arthritis phase associated with symptoms such as as joint pain without clinical arthritis (athralgia) is now hypothesized to fall within the aforementioned window of opportunity. Consequently, disease modulation in this phase might prevent the occurrence of clinically apparent arthritis, which would result in a persistent disease course if untreated. Several ongoing proof-of-concept trials are now testing this hypothesis. This Review highlights the importance of adequate risk prediction for the correct design, execution and interpretation of results of these prevention trials, as well as considerations when translating these findings into clinical practice. The patients' perspectives are discussed, and the accuracy with which RA development can be predicted in patients presenting with arthralgia is evaluated. Currently, the best starting position for preventive studies is proposed to be the inclusion of patients with an increased risk of RA, such as those identified as fulfilling the EULAR definition of ‘arthralgia suspicious for progression to RA’.