Virtual Diabetes Register and web tool

Each year Health New Zealand releases national estimates of the prevalence of diabetes based on the Virtual Diabetes Register (VDR). The VDR is a tool to monitor prevalence of diabetes and support national and local clinical quality improvements.

The VDR contains data about people suspected as having diabetes, identified through their use of diabetes related health services.

The VDR uses an algorithm to identify these people in data extracted from hospital inpatient and outpatient, laboratory test type, and pharmaceutical dispensing data collections.

The VDR is collated annually. National and regional diabetes prevalence estimates are calculated based on the number of people on the VDR during the previous calendar year – in this case, during 2023.

People suspected to have diabetes, who were alive and enrolled in a PHO at the 31 December of the VDR year, are included in the totals.

For further technical information about the VDR, see the Virtual Diabetes Register: Technical Guide.

The algorithm used to create the 2023 VDR has two output methods.

• Output 1 (traditional method): the diabetes prevalence estimates are based solely on the number of people alive and enrolled in a PHO, at 31 December of the VDR year. This output method is useful for the purposes of health service planning, for example. This information is available via the Virtual Diabetes Register web tool (see below).
• Output 2 (more inclusive version): the diabetes prevalence estimates are based on people who were alive and enrolled in a PHO at any point during the calendar year. This version can be used to:
  • better capture the population of people living with diabetes over the year of interest. This would be a useful output method when considering the cost and/or burden of disease over a year, for example.
  • include people that have died during the year. This allows more representative reporting of some outcomes, such as amputation rates for people with diabetes, and potentially other diabetes-related complications. 

The latest version of the VDR algorithm is v689. This VDR algorithm allows for either method of diabetes prevalence calculation to be used.  

The Virtual Diabetes Register web tool contains data using output 1 of the VDR.

If the user prefers to include people who were alive, those that have died, and PHO enrolled at any point during the year (output 2), this data is available to download below.

This web tool presents both estimated numbers of people registered as having diabetes, as well as the estimated age-standardised prevalence of diabetes per 1000 people, across different demographic groups in the New Zealand population.

The data presented in this tool can be explored by year, ethnicity, sex, district of residence, deprivation quintile and age group.

• In 2023, about 323,700 people in Aotearoa New Zealand were estimated as having diabetes. The estimated age-standardised prevalence of diabetes was 44.3 per 1000 population (95% CI: 44.2, 44.5).
• Over the last ten years, there was an increase in the estimated age-standardised diabetes prevalence, from 37.3 per 1000 population in 2014 (95% CI: 37.2, 37.5), to 44.3 per 1000 population in 2023 (95% CI: 44.2, 44.5).
• In 2023:
  • The Pacific population had the highest estimated age-standardised diabetes prevalence at 125.5 per 1000 Pacific population (95% CI: 124.3, 126.6), followed by the Indian population at 103.4 per 1000 Indian population (95% CI: 102.1, 104.6) and the Māori population at 71.6 per 1000 Māori population (95% CI: 71.0, 72.2). The European or other population had the lowest estimated age-standardised diabetes prevalence at 32.0 per 1000 European or other population (95% CI: 31.9, 32.2).
  • Counties Manukau district had the highest estimated age-standardised diabetes prevalence at 76.3 per 1000 population (95% CI: 75.7, 76.9), while Nelson Marlborough district had the lowest age-standardised diabetes prevalence at 27.4 per 1000 population (95% CI: 26.8, 28.1).
  • The estimated age-standardised diabetes prevalence increased with increasing socio-economic deprivation quintile. Those living in the most deprived socio-economic areas (quintile 5) had age-standardised prevalence of diabetes 2.7 times higher than those living in the least deprived socio-economic areas (quintile 1):
    - NZDep quintile 5: 74.3 per 1000 population (95% CI: 73.8, 74.8)
    - NZDep quintile 1: 28.0 per 1000 population (95% CI: 27.7, 28.2)
  • The estimated age-standardised diabetes prevalence of males was 48.0 per 1000 population (95% CI: 47.7, 48.2). This was higher than that of females, with 41.1 per 1000 population (95% CI: 40.9, 41.3).

Output 2 data is available to download below.