OBJECTIVES: To compare Rheumatoid Arthritis Quality of Life questionnaire (RAQoL) acceptability and psychometric properties following translation using forward-backward (FB) and dual-panel (DP) methodologies.  METHODS: The 30-item RAQoL has been adapted for use in Sweden independently by two groups. One used FB and the other DP (translation by bilingual Swedes followed by lay panel linguistic review) methods. The methods produced identical wording for 7 items. The two versions of the remaining items were evaluated by 23 bilingual Swedes who indicated their preferred version for each item. Fifty people with RA reported their preference for each item (presented in a paired blinded fashion) and 2 lay panels evaluated item pairs regarding appropriateness. A postal survey of 200 people with RA was then conducted to test the psychometric properties of the two versions. Participants were randomly assigned to complete either the FB or DP version first and the other 2 weeks later.  RESULTS: The bilinguals showed no preference (p=0.196), whereas patients and lay people preferred DP over FB item versions (p<0.0001). Mean (SD) RAQoL scores were 10.2 (7.4) for the FB, and 11.3 (7.5) for the DP version (p=0.151). Floor effects were small (FB, 6.1%; DP, 4.4%) and ceiling effects were negligible (FB, 0%; DP, 0.7%). Reliability was 0.92 for both versions, which were able to discriminate between respondents according to perceived general health, RA severity and whether they had a current symptom flare-up.  CONCLUSIONS: The DP approach showed advantages over FB translation in terms of preference by the target population and by lay people, while there were no obvious psychometric differences. The preference of the DP approach by patients contradicts the commonly held view that FB translation is the "gold standard". Additional head-to-head comparisons using other scales, languages and target groups are needed.