Linking primary and secondary healthcare databases in New Zealand

Abstract

Aim: To investigate the effectiveness of linking primary and secondary care data using an encrypted national health index (ENHI).

Methods: Primary care patient registers from 106 practices collected by the Dunedin Research Unit of the Royal New Zealand College of General Practitioners (RNZCGP) for the year 2001 were linked to the entire National Minimum Data Set (NMDS) for 2001 using an ENHI. The success of matching using the ENHI was measured, and primary and secondary care populations were compared in terms of hospital utilisation (number of discharges, length of stay, major diagnostic category [MDC]).

Results: 86,608 unique general practice ENHIs were successfully linked to the NMDS for the study period. Date of birth was matched on ENHIs (96.6%), sex (99.1%), and ethnicity (84.0%). Hospital morbidity and hospital utilisation (in terms of number of discharges and length of stay) were similar for general practice patient admissions and the entire NMDS admission data set.

Conclusions: Data collected in general practice linked well to secondary care data using the ENHI. Linked primary and secondary care data sets will provide a sound basis for research into publicly funded healthcare.