Objective: Back translation is considered a vital step in the generally accepted methodology for the translation of PRO measures (Wild et al. 2005). However, critics question its use (e.g. McKenna et al. 2005). There is agreement that more research is needed. Before embarking on more detailed research, it would be helpful to consider how back translation is currently being used as a step in the methodology for the translation of PRO measures. Methods: A review was undertaken of 50 back translation reports from 4 past projects at Oxford Outcomes. Examples of how back translation contributed to the translation process were gathered. Statistics were also gathered on how many items were questioned by the reviewer, and on how many of those items questioned were changed as a result. Results: Reviewers use back translation review as an opportunity to question what is in the translation, but do not force changes on the investigator. Review of the back translations can catch important misunderstandings/errors in the translation. The back translations can show investigators that their translation may be open to misinterpretation. Fourteen percent of items within the sample were changed as a result of back translation review. Although there are differences of style with different reviewers (e.g. questioning more items), the rate of changed items is around 15%, suggesting that the differing styles do not unduly influence the outcome of the reviews. The structure of the translated language can cause reviewers to question items that are not incorrect. Reviewers question more items in languages that are from cultures that differ widely from their own (e.g. Indic languages). Conclusion: Whilst back translation undoubtedly has its weaknesses and strengths, its users are aware of them and can consequently use it as tool to improve the quality of the translations of PRO instruments.