Recall bias most commonly associated with which study design:-
Case control study:
- An epidemiological study where a group of individuals with disease (CASE) are compared with a group of individuals who are not suffering from disease (CONTROLS) in terms of specific disease causing exposures.
- Since the starting point is a group of people who already have suffered from the disease, this is labeled as a retrospective study. (BACKWARD LOOKING STUDY)
- Drawbacks of case control studies
- Prone to selection and recall bias.
BIAS IN ANALYTICAL STUDIES:
BIAS is any systematic error that occurs during any stage of the study, thereby resulting in mistaken estimate of exposure and/ or outcome. A case control study is more prone to bias as compared to a cohort study. Although included as a bias confounding is not a systematic error in measurement, it is a true phenomenon existing in nature.
Several types of bias can occur.
1. Selection bias – occurs when the case and controls selected for the study are not representative of the general population, hence the results can’t be generalized (loss of external validity).
2. Measurement bias – Occurs while data collection and occurs when the subjects can’t recall the events properly or the investigator is biased in collecting data because of prior knowledge of the association between the risk factor and the disease, e.g. if the investigator is studying the relationship between smoking and lung cancer he might take a more detailed questioning of the cases and not the controls and thus exaggerate the relationship.
3. Confounding bias – Occurs when the disease under study is associated with another factor apart from the risk factor under study. In such a scenario it might be difficult to interpret the results, e.g. if we are studying the association between smoking and CHD; it is possible that the cases also take alcohol, thus confounding the results. This bias can be minimized by matchingStatistical procedures like regression analysis also help in detecting potential confounders.
4. Berkesonian Bias – Differing rates of admissions to hospitals cause bias
5. Surveillance bias'- If a population is monitored-over a period of time, disease ascertainment may be better in the monitored population, than in the general population and may introduce a surveillance bias which leads to an erroneous estimate of RR or OR. E.g.-physicians monitored patients who had been prescribed OCPs much more closely than they monitored their other patients. As a result, They were more apt to identify cases of thrombophlebitis that developed in those patients who were taking OCPs (and who were therefore being more closely monitored) than among other patients who were not well monitored. As a result, just through better ascertainment of thrombophlebitis in women receiving OCPs, an apparent association of thrombophlebitis with OCPs may be observed, even if no true association occurs.