In healthcare the term is used in two ways. To refer to a hospital patient accommodated in a different ward from that specialising in the relevant medical diagnosis. The term may be used in the examination of data where results fall well outside the pattern for the majority of NHS Trusts with similar units.
values in a set of data that are more than 1.5 times the interquartile range above the upper quartile or below the lower quartile
Are probabilistically remote events. Often viewed as statistically independent as well. Various techniques can test if actual data differ in a statistically significant manner from the benchmark or normal distribution. Also, independence assumptions can be tested to see if they are accurate representations of the underlying processes.
Data points which do not appear to follow the characteristic distribution of the rest of the data. These may reflect genuine properties of the underlying phenomenon (variable), or be due to measurement errors or other anomalies which should not be modeled.
Observations that do not appear to follow the characteristic distribution of the rest of the data. These may be genuine observations but could be due to measurement errors or other anomalies. All outliers should be carefully checked.
Services or costs that differ substantially from the standard established in a statistical profile of cost or usage.
Technically, outliers are data items that did not (or are thought not to have) come from the assumed population of data -- for example, a non-numeric when you are expecting only numeric values. A more casual usage refers to data items that fall outside the boundaries that enclose most other data items in the data set.