Case-controlstudy is an observational study design, which is mostly applicable inthe field of epidemiology. In designing a case-control study, thefollowing steps are applicable. The first step is to indicate a studypopulation, and this selection must be free of exposures ofinterests. The second step is to define as well as select cases. Inthis step, identification of cases can be from the health data andregisters. The third step is to define as well as select controls,which represents the population that cases originate from. The fourthstep is to measure exposure and then approximate disease risk linkedwith this exposure (Schultz & Grimes, 2002).
Biascan be regarded as a systematic error which occurs when estimatingcause and their effect. On the other hand, confounding are thosevariables apart from risk factors, where cases as well as controlgroups differ. To reduce confounding, matching cases, such as ages oruse of restriction, can be applied. Bias can be prevented by ensuringquality case choices and diagnostic criteria among other methods(Schultz & Grimes, 2002).
Ina research study, controls is a group which is separated from theexperimental group so that, they are not influenced by theindependent variables. The main importance of controls is that, itenables the researcher to easily ignore alternative explanations,especially when experimental conditions are difficult and complex toisolate (Schultz & Grimes, 2002).
Confoundingvariables are extraneous variables, which correlates either inverselyor directly to the independent and dependent variables. Confoundingvariable affects a research study as they lower internal as well asthe external validity of a research, thus making it hard for aresearcher to attain the set goals (Schultz & Grimes, 2002). Forthis study, some of problems and limitations arising from confoundingvariables include the inability to increase the number and type ofcomparisons carried out in analysis, the limitations of matchingcases such as years among other notable factors.
Schultz,K.F. & Grimes, D.A. (2002) "Case-control studies: researchin reverse". TheLancet,359, 431-34.