Quality of research projects
For new studies to be useful to health risk assessments and for standard setting, the research must be of high scientific quality. This requires clearly-defined hypotheses, estimates of the ability of the study to detect small effects, and use of protocols that are consistent with good scientific practice. Quality assurance procedures should be included in the protocol and monitored during the study.
Details needed to ensure well conducted studies are given in NTP (1992), Valberg (1995), Repacholi (1998). Further discussion on this topic can be found on several web sites, including the following: