In a case control study, systematic sampling design for case (with disease) and purposive sampling design for control (without disease non -case healthy individual) can both design be used?
Hi
selection of controls is the most important point in case control studies.
•The way of control selection determines: What is estimated (risk ratio, rate ratio, odds ratio) and the validity of estimate.
Three basic tenets of comparability underlie attempts to minimize bias in control
selection. These are the principles of study base, deconfounding, and comparable accuracy.
Comparable accuracy principle. The degree of accuracy in measuring the exposure
of interest for the cases should be equivalent to the degree of accuracy for the controls, unless the effect of the inaccuracy can be controlled in the analysis.
Study base principle. Cases and controls should be "representative of the same base experience".
Deconfounding principle. Confounding should not be allowed to distort the estimation of effect.
in this case if you have many cases you can select them systematically but it is better you select the new cases. selecting of control is important. you must first identify the source population that cases arise from it. If your study is a population-based case-control study, that uses a primary base. when ascertainment of all cases in a primary base is difficult or impractical, it may be preferable to use a secondary base. The base is defined as the source of the cases, and controls are individuals who would have become study cases if they had developed disease
during the time of the investigation. For example, in a hospital-based study, the
cases might be all patients diagnosed with the study disease at one hospital; the individuals contributing to the (secondary) base would be all subjects who would be diagnosed at that hospital had they developed the study disease.
Finally i think purposive sampling without attention to source population of cases can disturb your results.
Purposive sampling design is not a probabilistic sampling design. From here, it is not possible to conclude statistical decisions based on statistical inference principles.
Some references in Spanish:
http://www.lulu.com/spotlight/Mariano_Ruiz_Espejo
I agree with Ahmed. The cases and controls must come from the same population. And the population in the first place is the one to which you'd like to apply your results.
The point is to randomlyselct controls in the same population as the cases. If they are chosen randomly from that population and are disease free then the study will not be biased by selection and will be random with respect to the hypothesis that is being tested.
How odd to be answering a reply that was offered 5 years after the original question was answer.
- Badges
- Science topic