A data set provides k-anonymity for the data subjects represented if the information for each person contained in the data set cannot be distinguished from at least k-1 individuals whose information also appears in the data set. For example, a data set has 5-anonymity if, for every record in the data set that describe characteristics of a data subject, there are at least four other individuals also represented by records in the data set who share the same characteristics described by the record.
Page 39 https://digital.nhs.uk/binaries/content/assets/legacy/pdf/b/o/1523202010spec.pdf
A data set provides k-anonymity for the data subjects represented if the information for each person contained in the data set cannot be distinguished from at least k-1 individuals whose information also appears in the data set. For example, a data set has 5-anonymity if, for every record in the data set that describe characteristics of a data subject, there are at least four other individuals also represented by records in the data set who share the same characteristics described by the record.
Page 39 https://digital.nhs.uk/binaries/content/assets/legacy/pdf/b/o/1523202010spec.pdf