Making rapid decisions is a key component of everyday clinical practice. Informed decision-making such as this requires physicians to combine their own clinical expertise and training with high-quality scientific evidence (Guyatt et al., 2000). New scientific knowledge is emerging all the time and all health workers need to update their knowledge continuously. Most busy practising clinicians simply do not have sufficient time to read all the important primary research reports. Over two million articles are published every year in 20,000 biomedical journals (Mulrow, 1994) and even if a...

Making rapid decisions is a key component of everyday clinical practice. Informed decision-making such as this requires physicians to combine their own clinical expertise and training with high-quality scientific evidence (Guyatt et al., 2000). New scientific knowledge is emerging all the time and all health workers need to update their knowledge continuously. Most busy practising clinicians simply do not have sufficient time to read all the important primary research reports. Over two million articles are published every year in 20,000 biomedical journals (Mulrow, 1994) and even if a clinician restricts his/her reading to high-yield clinical psychiatry journals, he/she would need to read over 5000 articles a year (Geddes et al., 1999). The clinician therefore needs a reliable system of knowledge management, that is, systematic reviews and meta-analyses. In this chapter, we will outline the main features and limitations of systematic reviews and will suggest ways in which further syntheses of primary trial data may yield clinically useful information.