Ans. b. Berksonian biasRecruiting participantsTypeDefinitionExamplesStrategy to reduce biasSelection bias*. Nonrandom sampling or treatment allocation of subjects such that study population is not representative of target population (e.g., study participants included based on adherence or other criteria related to outcome).*. Most commonly a sampling bias.*. Berkson bias--study population selected from hospital is less healthy than general population.*. Healthy worker effect--study population is healthier than the general population*. Non-response bias-- participating subjects differ from nonrespondents in meaningful ways*. Randomization*. Ensure the choice of the right comparison/reference groupPerforming studyTypeDefinitionExamplesStrategy to reduce biasRecall biasAwareness of disorder alters recall by subjects; common in retrospective studiesPatients with disease recall exposure after learning of similar casesDecrease time from exposure to follow-upMeasurement biasInformation is gathered in a systemically distorted manner*. Association between HPV and cervical cancer not observed, when using non-standardized classifications*. Hawthorne effect--participants change their behavior in response to their awareness of being observed*. Use objective, standardized, and previously tested methods of data collection that are planned ahead of time*. Use placebo groupProcedure biasSubjects in different groups are not treated the samePatients in treatment group spend more time in highly specialized hospital unitsBlinding and use of placebo reduce influence of participants and researchers on proceduresObserver-expectancy biasResearcher's belief in the efficacy of a treatment changes the outcome of that treatment (aka Pygmalion effect; self-fulfilling prophecy)If observer expects treatment group to show signs of recovery, then he is more likely to document positive outcomesand interpretation of outcomes as neither are aware of group allocation