Bio Stats

Question 1

Lead-time bias

Answer 1

An artificial increase in survival time among tested patients who actually have an unchanged prognosis. Pts screened with more sensitive test only appear to live longer only bc the disease was detected earlier that it would have been clinically. Overall length of time from disease onset to death actually remains the same

***think of lead-time bias when you see “a new screening test” for poor prognosis disease

Question 2

Incidence vs prevalence

Answer 2

Incidence is the measure of the appearance of new cases.
Prevalence is the measure of those w the disease in the population at a particular point in time.
In a stable population:
Prevalence= incidence x time

***time can refer to factors that prolong the duration of the disease (eg improved quality of care)

Question 3

Observer bias

Answer 3

When the investigators decision is affected by prior knowledge of the exposure status

Question 4

Recall bias

Answer 4

Inaccurate patient recall of past exposure

Question 5

Selection bias

Answer 5

When subjects are selected for a study or from selective losses during follow up

Question 6

Matching

Answer 6

Method used in the design stage of case control studies to control confounding. Initial step involves selecting variables that could be con founders (i.e. Age, race). Cases and controls are then selected based on the matching variables so that both groups have similar distribution in accordance with the variables