Attrition loss of participants during the course of a

Attrition: Loss of participants during the course of a study. Participants lost during the
course of a study are often called dropouts. Also called lost to follow up.
Attrition bias: Systematic differences between comparison groups in withdrawals or
exclusions of participants from the results of a study. For example, participants may
drop out of a trial because of side effects of the intervention. Excluding these
participants from the analysis could result in an overestimate of the effective- ness of
the intervention or an underestimate of side effect rates, especially when the proportion
dropping out varies by treatment group.

HIM 6844 Health Outcomes Research 18
Bayes’ theorem: A theorem used to update the probability of an event in the light of a
piece of new evidence.
Bayesian analysis: A statistical approach based on Bayes’ theorem that can be used in
single studies or meta-analyses. Bayesian analysis involves the use of existing and new
information to estimate the risk that a person will experience an event.
Beta error: See Type II error.
Bias: A systematic error in study design or conduct that results in a distorted
assessment of the intervention’s impact on the measured out- comes.1 In clinical trials,
the main types of bias arise from systematic differences in study groups that are
compared (selection bias), exposure to factors apart from the intervention of interest
(performance bias), participant withdrawal or exclusion (attrition bias), or assessment
of outcomes (detection bias). Reviews of studies may also be particularly affected by
reporting bias, where a biased subset of all the relevant data is available.
Bias prevention: Aspects of study design or conduct intended to prevent bias. In
clinical trials, such aspects include randomization, blinding, and concealment of
allocation.
Blinded study: An experimental study in which participants do not know the treatment
they are receiving; investigators may also be blind to the specific treatments. Double-
blind means that neither participants nor investigators know which treatment the
participants receive. However, the terms single-blind, double-blind and triple-blind are
not used consistently and are ambiguous unless those who are blinded are specified.
Blinding: See blinded study.
Case-control study: An observational study that compares individuals with a specific
disease or outcome of interest (cases) to individuals from the same population without
that disease or outcome (controls) and seeks to find associations between the outcome
and prior exposure to particular risk factors. This design is particularly useful where the
outcome is rare and past exposure can be reliably measured. Case-control studies are
usually retrospective, but not always.
Causality: An association between two characteristics that can be demonstrated to be
due to cause and effect (i.e., a change in one causes change in the other). Experimental
studies such as randomized controlled trials can be used to support causality. However,