Question

1. Each of the following statements is descriptive of situations which may be likely to present one of the major types of bias or confounding. If bias or confounding is probable, select the type which is most likely to occur by writing ‘confounding’, ‘information bias’, ‘selection bias’, or ‘no bias’, and a brief justification (in one sentence) to support your choice.

a). In a cohort study of the association of diet and gallbladder disease, the exposed subjects are evaluated twice yearly by physicians involved in the study by the use of ultrasound, and the non-exposed subjects are evaluated yearly by their physicians.

b). In a cohort study of breast implants and the risk of connective tissue disease, 40% of patients receiving implants were lost to follow-up.

c). In a comparison study of the association of socioeconomic status and infant mortality, low birth weight is associated with SES and infant mortality.

d). In a case control study of the association of Hansen's disease (leprosy) and the hunting and cleaning of armadillos, comparison subjects are obtained from close friends of the cases.

e). Medical records were used to validate the hospital discharge diagnoses of Crohn's disease and ulcerative colitis. By using this validation process instead of relying on hospital discharge coding alone, the authors were hoping to reduce the certain sources of error.

Just looking to double-check my work, and would like explanations so I can learn how to solve these on my own - thanks!

Answer 1

a). In a cohort study of the association of diet and gallbladder disease, the exposed subjects are evaluated twice yearly by physicians involved in the study by the use of ultrasound, and the non-exposed subjects are evaluated yearly by their physicians.

Answer- confounding

Improper diet may lead to gallbladder disease.

b)In a cohort study of breast implants and the risk of connective tissue disease, 40% of patients receiving implants were lost to follow-up.

Answer- selection bias

sample obtained is not representative of the population intended to be analyzed

c). In a comparison study of the association of socioeconomic status and infant mortality, low birth weight is associated with SES and infant mortality.

Answer-confounding

socioeconomic status can be a major reason for infant mortality.Here socioeconomic status is independant variable and infant mortality and low birth weight are dependent variable.

d). In a case control study of the association of Hansen's disease (leprosy) and the hunting and cleaning of armadillos, comparison subjects are obtained from close friends of the cases.

Answer-information bias’

e). Medical records were used to validate the hospital discharge diagnoses of Crohn's disease and ulcerative colitis. By using this validation process instead of relying on hospital discharge coding alone, the authors were hoping to reduce the certain sources of error.

Answer-selection bias

sample obtained is not representative of the population intended to be analyzed