Objectives were to evaluate the psychometric properties and to determine optimal scoring of the retinopathy treatment satisfaction questionnaire (RetTSQ) in a cross-sectional study of 207 German patients with diabetic retinopathy and a wide range of treatment experience. Forty patients (19%) also had clinically significant macular oedema. Principal components analysis was used to identify factor structures and Cronbach's alpha to assess internal consistency reliabilities. Two highly reliable subscales represented negative versus positive aspects of treatment (both alpha = 0.85). A highly reliable total score can be calculated (alpha = 0.90). Construct validity was examined by testing expected relationships of RetTSQ scores with visual impairment, stage of diabetic retinopathy, additional impact of macular oedema, SF-12 scores and scores of the RetDQoL measure of quality of life in diabetic retinopathy. Worse impairment, worse diabetic retinopathy and macular oedema were associated with less treatment satisfaction. RetTSQ scores correlated moderately with SF-12 scores (r: 0.33-0.53, p < 0.001) and RetDQoL scores (r: 0.43-0.51, p < 0.001). Answers to an open-ended question indicated no need for additional items. Repeating the analyses in a subsample with experience of more intense treatment showed very similar results. It can be concluded that the RetTSQ is valid and reliable for people with diabetic retinopathy with or without macular oedema who have experienced different treatments.