Question

identifying the type of bias being described in each scenario.

Suppose that a new screening test has recently been developed to aid in the early identification of individuals with rheumatoid arthritis. However, the value of the test has not yet been established. In an effort to test the effectiveness of the new test in predicting a future diagnosis of rheumatoid arthritis, investigators enroll a group of volunteer participants, record their test result, and follow them forward for the development of the condition. It is subsequently determined that the new screening test is highly effective in predicting a definitive diagnosis of rheumatoid arthritis because the incidence of the condition among those who screened positive was much higher than that of the general population.

SHOW ALL WORK

Answer 1

Bias in research is an error in design or execution of a study that produces results that are distorted either in positive way or negative way. For example if i have a new bed test to detect diabetes mellitus and if i consider only diabetic patients in my hospital as sample, sensitivity and specificity that we obtain may not be accurate as we are not testing people who actually do not have the disease.

In the study that is mentioned in the question a screening test is used. An ideal screening test must be cheap, easy to use, cause less discomfort, valid (distinguish between those who have the disease and not have the disease), reliable.

Some of the bias seen are:

- Here the cut off value of the test is not yet established hence there is a Test interpretation bias: the value established by the test in study population is not reliable.

- Selection of Population for study: Since the population of study are from those who are at higher risk of develpong Rheumatoid Arthritis there is Spectrum Bias -Cases and controls are not representative of the population.

- There is also Length Bias Slowly progressive disease (RA) is over represented in the population than those which arise in normal population