Introduction: Those conducting systematic reviews have to decide what type of trials to include. The absence of a sufficient number of 'high quality' RCTs evaluating a given intervention is widely regarded as indicating that there is no point in doing a systematic review.

Methods: Over the last three years, we have undertaken systematic reviews of 114 interventions for tardive dyskinesia.

Discussion: Because of the questionable quality of many of these studies, meta-analysis could only be performed in 10 of these treatments and the results of these meta-analyses should be interpreted with great caution. However, while performing these reviews we identified several "gaps" that warrant future research. Even when a systematic review does not reveal a large amount of quality data ideally suitable for synthesis, a rigorous review can reveal the strengths and weaknesses of the data supporting the use of a given treatment, and pointing the way for future research.