Background: Indigenous young people are disproportionately exposed to risk factors for poor mental health. Methodologically rigorous research will be critical in the development and evaluation of prevention and treatment programs. Research examining the mental health of Indigenous young people may have been undermined by poor measurement. The extent to which research has used measures with adequate psychometrics is unknown. Methods: MEDLINE, PsychINFO and PUBMED databases, were systematically searched to identify papers published between 1998-2008 measuring the mental health of Indigenous young people from Australia, Canada, New Zealand or the US. Data extracted included type of mental health instrument, psychometric analyses reported and results. Results: Fifty-four relevant studies were identified. Seventy-nine mental health instruments were used, and 18% were bespoke. Only 14% of instruments had been validated for the relevant Indigenous population. Few studies reported assessment of the reliability or validity of instruments. Data about both the reliability and validity of 10 measures were reported. None of the measures met the standards set by the review. Evidence of at least one type of reliability and validity was demonstrated for six measures. Conclusions: From 1998-2008 few studies of mental health in Indigenous young people used measurement instruments with previously determined reliability and validity.