Stats Flashcards
Define Incidence
Number of new cases of disease per year
Define prevalence
Overall proportion of population with disease
Define bias
Anything that statistically influences the conclusions about a group and distorts comparisons
What is Berkson bias
Arises when study includes hospitalised patients. These patients may have exposure to risk factors than are significantly greater than the population at large which can disrupt association between exposure and outcome
What is allocation bias
Systematic difference in how participants are assigned to groups
What is apprehension bias
When a study participant behaviours differently due to being observed
What is ascertainment bias
- When data for a study are collected such that some members are less likely to be in the final result than others
What is attrition bias
When individuals drop out of groups unequally
What is availability bias
Researchers use information that is readily available rather than collecting all
What is chronological bias
Participants recruited at different time points may vary in exposure
What is compliance bias
Participants that comply with an intervention differ in someway from those who don’t
What is data dredging
Occurs when multiple sub-group analyses are performed on the same data. Increases probability of type 1 error
What is the Hawthorne effect
Individuals modify an aspect of their behaviour in response to their awareness of being observed
Hot Stuff Bias
When a topic is fashionable investigators may be less critical in their approach and editors want to publish
Hypothetical bias
When individuals stated behaviour differs to their real behaviour
Industry sponsorship bias
When methods and results of a study support interests of funding organisation
Information bias
Bias that occurs due to systematic differences in collection, recording or handling of information used in a study
Observer bias
Systematic difference between value observed - due to observer variation, and the true value
Performance bias
Systematic differences in the care provided to members of different study groups
Popularity bias
Difference in uptake of healthcare as a result of public interest in a disease or condition - this leads to a selection bias
Publication Bias
Likelihood of study being published affect findings of the study
Neyman bias
Also known as prevalence-incidence bias.
Excluding patients with severe forms of the disease - will make disease appear less severe.
Excluding patients who have recovered - will make the disease seem more severe.
Greater time between exposure and investigation increases this risk. More common in long-acting conditions.
Case-control studies are most susceptible.
Recall bias
Systematic error due to differences in accuracy of recall of past events
Selection bias
Occurs when individuals in a group or study differ systematically from the population at interest
Unacceptability
Systematic difference in response rates or uptake of tests due to their unacceptability. (eg. measurements that hurt or embarass may be refused)
Volunteer bias
Participants volunteering to take part in the study have intrinsically different characteristics to population of interest
