Skip to main content

Table 4 Combinations of identifiers available on different record sources

From: An efficient record linkage scheme using graphical analysis for identifier error detection

Hospital Number NHS Number Name/date of birth jonah lims micro pas pashistory Total
+ + + 456553 (72.2%) 246326 (3.6%) 1494645 (28.5%) 1205042 (53.3%) 94935 (83.4%) 1494645 (16.2%)
+ + - 223 (0%) 1510 (0%) 5520 (0.1%) 19086 (0.8%) 161 (0.1%) 19086 (0.2%)
+ - + 174874 (27.7%) 2160475 (31.7%) 978448 (18.6%) 860916 (38.1%) 18372 (16.1%) 2160475 (23.4%)
+ - - 550 (0.1%) 30816 (0.5%) 36636 (0.7%) 177518 (7.8%) 366 (0.3%) 177518 (1.9%)
- + + 7 (0%) 103420 (1.5%) 813906 (15.5%) 0 (0%) 0 (0%) 813906 (8.8%)
- + - 0 (0%) 591 (0%) 2244 (0%) 0 (0%) 0 (0%) 2244 (0%)
- - + 95 (0%) 3883941 (57%) 1245979 (23.7%) 1 (0%) 1 (0%) 3883941 (42.1%)
- - - 3 (0%) 382490 (5.6%) 671076 (12.8%) 0 (0%) 0 (0%) 671076 (7.3%)
  1. Up to three identifiers, hospital number, NHS number and name & date of birth are available for each record, but they are not all present in each data set. Shown are the combinations of identifiers (- = absent,+ = present) for each dataset contributing to the database.