CASE CONTROL STUDY 4
Acase-control study is an observational study that helps in evaluatingbetween exposures and diseases. This study traces backwards from theoutcome to the exposure (Merrill, 2010). For instance, in the outcomeof a disease, a case-control study will look backward in time ofexposures that might have caused the disease. Observational studiesyield useful scientific results with little effort, time and money ascompared to other study designs. Food borne diseases outbreaks areexamined using case-control studies (Merrill, 2010).
Themain purpose of observational studies is to find out if there is anyrelationship between a particular disease and the risk factor. Thisis the first step in testing hypothesis. The second step in thedesign of a case study is defining a study population (Merrill,2010). This means that the study should be from the same population.Defining and selecting cases is an important step. This is made fromthe general population and data from a particular medical setting. Inaddition, one should select controls and measure the exposure. Thesemeasures help prevent researcher bias. Finally, there is theestimation of disease risk associated with the exposure (Merrill,2010).
Differentsteps have been adopted that help minimize different types of bias.To start with, selection bias occurs when a particular group differsin outcome as compared to other groups (Merrill, 2010). This bias canbe minimized through defined criteria of selection. For instance,define criteria of selection of both diseased and non-diseasedindependently in a case-control study and use random clinical trials
Secondly,information bias occurs when information is collected in differentgroups causing an error in the conclusion that lead tomisclassification. These errors are minimized through a standardizedprotocol of data collection by making sure that the methods andsources of data collection are common in all study groups (Merrill,2010). Further, one should adapt a strategy to assess a potentialinformation bias.
Confoundingoccur when the results between exposure and disease differ from thereal truth due to third variable influence (Merrill, 2010). Forinstance, the mortality rate of two cities can differ but with anadjustment for age, the rates do not differ. Confounding leads tosystematic errors that cannot be fixed. However, with known variablesthe effect is easily fixed.
Merrill,R. M. (2010). Introductionto epidemiology.Sudbury, Mass: Jones and Bartlett Publishers.