Data collection in clinical toxicology: are there too many variables?

Abstract

The evidence base of clinical toxicology suffers in comparison to other clinical disciplines. There is an excess of case reports and case series with little in the way of case control or cohort studies, and very few randomized controlled trials. While randomized controlled trials are rightly regarded as the gold standard for interventional studies, they have limitations that are particularly evident in the practice of clinical toxicology. Properly conducted observational studies using quantitative, epidemiological methods [nonrandomized trials, cohort studies (prospective and retrospective), case control methods] can provide answers that may be impossible to obtain from randomized controlled trials. Development of a strong evidence base is essential for progress in clinical toxicology. Whether that evidence base is derived from randomized controlled trials or observational studies, it is essential to collect data. Important observations can be made from basic clinical data and systematic collection of those data into some form of electronic database has siginificant advantages. A clinical database provides accurate information in the areas of clinical practice, quality assurance (audit), and research. In the area of research, an appropriately designed database can be both a source of hypotheses as well as a vehicle to test them. It can also serve as a repository of research data in subsequent randomized controlled trials.