Changes

'''Systematic error''', or 'bias' is of particular importance in any epidemiological investigation, and should be avoided wherever possible. As with random error, biases will reduce the '''accuracy''' of any results obtained, but they also have the potential to reduce the '''validity''' of results. Although [[confounding]] is considered by many authors as a form of bias, it is covered separately.<br>

'''Systematic error''', or 'bias' is of particular importance in any epidemiological investigation, and should be avoided wherever possible. As with random error, biases will reduce the '''accuracy''' of any results obtained, but they also have the potential to reduce the '''validity''' of results. The forms of bias covered here can only be minimised through careful study design and execution - they cannot be accounted for in the analysis. Although [[confounding]] is considered by many authors as a form of bias, it can be accounted for in the analysis stage, and so is covered separately.<br>

There are a number of types of bias, which may be classified broadly as either selection bias or information bias, and which will differ in the case of different [[Study design|study designs]].

There are a number of types of bias, which may be classified broadly as either selection bias or information bias, and which will differ in the case of different [[Study design|study designs]].

===Descriptive or cross sectional studies===

===Descriptive or cross sectional studies===

In these studies, lack of compliance ('non-response') may result in selection bias, as participation amongst selected individuals is very rarely 100% in these cases. It is for this reason that compliance should be maximised as much as possible, through the use of reminders or incentives to participate.  

In these studies, lack of compliance ('non-response') may result in selection bias, as participation amongst selected individuals is very rarely 100%. It is for this reason that compliance should be maximised as much as possible, through the use of reminders or incentives to participate.  

===Case-control studies===

===Case-control studies===

==Information bias==

==Information bias==

Information bias results from errors in measurement or classification of exposures or outcomes of interest amongst the individuals included in the study. In the case of analytic studies, this may be classified as '''differential''' or '''non-differential'''. Non-differential bias will tend to reduce the strength of any association present, and will increase the probability of a [[Random variation#Hypothesis testing and study power|type II error]], whereas the effect of differential bias cannot be predicted.<br>

Information bias results from errors in measurement or classification of exposures or outcomes of interest amongst the individuals included in the study. In the case of analytic studies, this may be classified as '''differential''' or '''non-differential'''. whereas the effect of differential bias cannot be predicted.<br>

===Non-differental information bias/Measurement error===

===Non-differental information bias/Measurement error===

Non-differential bias occurs when the chance of bias is not affected by the group the individuals belong to. '''Measurement error''' is a common example of non-differential bias - for example, if scales are not correctly calibrated, they may incorrectly record the weight of all items weighed, regardless of their 'exposure' status.

Non-differential bias occurs when the chance of bias is not affected by the group the individuals belong to. '''Measurement error''' is a common example of non-differential bias - for example, if scales are not correctly calibrated, they will incorrectly record the weight of all animals weighed, regardless of their 'exposure' status. Non-differential bias will tend to reduce the strength of any association present, and will increase the probability of a [[Random variation#Hypothesis testing and study power|type II error]],

===Differential information bias===

===Differential information bias===