Match rules are set by a system administrator for each source subscription. These rules use varying levels of confidence, from fuzzy Matching that is not exact; a comparison of similar values (for example Smith/Smyth) depending on match rules set by the customer administrator. to strict, to determine which fields in a record are considered a possible match. Together they define the threshold at which a match is considered suspect.
Specifically, match rules might allow slight variations in names, nicknames, or specialties, while insisting on exact matches for key identifying information, such as a license number.
A number of scenarios apply when name matching takes place:
|Matches will consider potential misspellings as a result of additional or substituted letters.
|Smyth/Smythe, Phillip/Philip, Stephen/Steven
|Matches will consider typos as a result of transposed letters.
|Matches will allow initials used for the first name or middle name, as well as for nickname initials.
|James E Gardner/James Elliot Gardner/JE Gardner
|Titles and honorifics
|Matches will occur regardless of titles or honorifics.
|Dr James Gardner/James Gardner, Ms E Spenser/E Spenser
|Matches will occur regardless of middle name inclusion.
|Emily Anne Spenser/Emily Spenser
|Matches will consider segmentation, including names that may or may not include a space or hyphen between them.
|Matches are considered less likely if a generation indicator is not the same, or is omitted. Names with numeric indicators (for example, "the 2nd" or "II") are also considered.
|Fred Smith Jr/Fred, Smith/Fred Smith Jnr, King George Hospital/King George V Hospital
|Matches are considered for names that consist of two components but are not always shown.
|Jean-Claude Decloux/Jean Decloux/Claude Decloux
|Matches consider abbreviations in an organization's name.
|Matches consider names that may have been truncated.
|Baptist Medical Center/Bapt Med Ctr
|Matches consider place names that have been abbreviated.
|Pennsylvania Hospital/Penn Hospital/PA Hospital
|Matches distinguish between an organization's name and potentially extraneous information, such as a specialty, in the name.
|Sick Kids Toronto - Surgery/Sick Kids Toronto
The following scenarios are considered for address matching:
|Street address similarity
|Matches consider slight variations in street names, including spelling differences and directional information.
|Main Street/Main Street E/Main Street West
|Matches consider rules specific to the country to which the address belongs. Information such as postal code or house number would be considered particularly important for matching in certain locales.
The following scenarios are considered for identifier matching:
|Matches are required to have an exact match for certain identifiers, with potentially minor differences.
|Matches will consider identifiers with minor instances of transposition.
|Matches will consider only matches for identifiers against other identifiers of the same type.
|NPI vs NPI, AMA vs AMA