Background Complaints of arm, neck and/or shoulders (CANS) affects millions of computer office workers. However its prevalence and associated risk factors in developing countries are yet to be investigated, due to non availability of validated assessment tools for these countries. We evaluated the 1-year prevalence of CANS among computer office workers in Sri Lanka and tested the psychometric properties of a translated risk factor questionnaire. Methods Computer office workers at a telecommunication company in Sri Lankan received the Sinhalese version of the validated Maastricht Upper Extremity Questionnaire (MUEQ). The 94 items in the questionnaire covers demographic characteristics, CANS and evaluates potential risk factors for CANS in six domains. Forward and backward translation of the MUEQ was done by two independent bi-lingual translators. One-year prevalence of CANS and psychometric properties of the Sinhalese questionnaire were investigated. Results Response rate was 97.7% (n = 440). Males were 42.7%. Mean age was 38.2 [+ -] 9.5 years. One-year prevalence of CANS was 63.6% (mild-53.7% and severe-10%). The highest incidences were for neck (36.1%) and shoulder (34.3%) complaints. Two factors for each domain in the scale were identified by exploratory factor analysis (i.e. work-area, computer-position, incorrect body posture, bad-habits, skills and abilities, decision-making, time-management, work-overload, work-breaks, variation in work, work-environment and social-support). Calculation of internal consistency (Cronbach's alpha 0.43-0.82) and cross-validation provided evidence of reliability and lack of redundancy of items. Conclusion One year prevalence of CANS in the study population corresponds strongly with prevalence in developed countries. Translated version of the MUEQ has satisfactory psychometric properties for it to be used to assess work-related risk factors for development of CANS among Sri Lankan computer office workers.