Last update:
10-Sep-2012

Epidemiological studies include experimental and non experimental studies. Experimental studies include clinical trials, field trials and community intervention trials, while non experimental studies include quasi experimental, followup, "case-control", cross-sectional and ecological studies. Clinical trials provide the most widely accepted approach for comparing the effectiveness of different forms of treatment. A clinical trial is an experimental study with patients as subjects, all of whom have been diagnosed with a specific illness, but that illness is not the outcome under study. Instead, the outcome is some consequence of that illness, such as cure, death, etc. The aim of a clinical trial is to evaluate the incidence of the outcome under study in the groups assigned to the different forms of treatment. In most clinical trials, the forms of treatment are assigned by randomization, using random numbers tables. With randomization, the determination of treatment group assignment is based upon chance and is not influenced by physician or patient preference. In addition, randomization tends to produce comparability between the groups with respect to various characteristics (confounders) that might affect the incidence of the outcome under study. According to "intention to treat" analysis, all participants who are randomized should be included in the data analysis of a clinical trial and patients should be left in the treatment group originally assigned by the protocol of the study, even if they received one of the other forms of treatment after the original treatment regimen failed.

Key words: Clinical trials, Confounding by indication, Data analysis, Randomization, Survival curves.